transforming asian governance: rethinking assumptions, challenging practices (routledge research on...

cover next page >

Cover

title: Transforming Asian Governance : RethinkingAssumptions, Challenging Practices RoutledgeResearch on Public and Social Policy in Asia ;2

author: Ramesh, M.; Fritzen, Scottpublisher: Taylor & Francis Routledge

isbn10 | asin: 0415474787print isbn13: 9780415474788

ebook isbn13: 9780203884843language: English

subject Social change--Asia, Political culture--Asia,Asia--Case studies.--Social policy , Asia--Casestudies.--Economic policy , Asia--Case studies.--Politics and government

publication date: 2009lcc: HN652.5.T73 2009eb

ddc: 306.2095subject: Social change--Asia, Political culture--Asia,

Asia--Case studies.--Social policy , Asia--Casestudies.--Economic policy , Asia--Case studies.--Politics and government

cover next page >

< previous page page_i next page >

Page iTransforming Asian GovernanceThere are a multitude of hazards that confront attempts to change institutional orpolitical orders in pursuit of good governance. Even seemingly technical prescriptionsrun up against local political and social realities which make their adoption difficultand, if adopted, require significant modification of the original prescriptions. Moreover,the technical, rationalist and/or normative language employed in the good governancediscourse masks contests over power, rights, resources, and actors’ conflictinginterests. There is a definite need to situate the good governance debate in the localcontext rather than reflexively adopting a universalistic positing of the fact ordesirability of governance convergence across countries and sectors, because thereality is that the worldwide deployment of good governance rhetoric is notaccompanied by convergence in thinking or practices across nations. TransformingAsian Governance asks:• How do good governance principles translate into local settings?• How do local settings influence the conception of what is good governance and howis the debate over good governance deployed as a political or administrative strategy?Using case studies in governance from Thailand, the Philippines, Pakistan, Malaysia,India, Indonesia, Korea and Japan, this book will be essential reading for anyoneinterested in the public and social policy of Asia, and international and comparativegovernance more generally.Scott Fritzen is Associate Professor in the Lee Kuan Yew School of Public Policy atthe National University of Singapore.M. Ramesh is a Professor in the Department of Social Work and SocialAdministration, The University of Hong Kong.

< previous page page_i next page >

cover next page >

Cover

title: Transforming Asian Governance : RethinkingAssumptions, Challenging Practices RoutledgeResearch on Public and Social Policy in Asia ;2

author: Ramesh, M.; Fritzen, Scottpublisher: Taylor & Francis Routledge

isbn10 | asin: 0415474787print isbn13: 9780415474788

ebook isbn13: 9780203884843language: English

subject Social change--Asia, Political culture--Asia,Asia--Case studies.--Social policy , Asia--Casestudies.--Economic policy , Asia--Case studies.--Politics and government

publication date: 2009lcc: HN652.5.T73 2009eb

ddc: 306.2095subject: Social change--Asia, Political culture--Asia,

Asia--Case studies.--Social policy , Asia--Casestudies.--Economic policy , Asia--Case studies.--Politics and government

cover next page >

< previous page page_ii next page >

Page iiRoutledge research on public and social policy in Asia1 Ultra-Low Fertility in Pacific AsiaTrends, causes and policy dilemmasEdited by Gavin Jones, Paulin Tay-Straughan and Angelique Chan2 Transforming Asian GovernanceRethinking assumptions, challenging practicesEdited by M. Ramesh and Scott Fritzen

< previous page page_ii next page >

< previous page page_iii next page >

Page iiiTransforming Asian GovernanceRethinking assumptions, challenging practicesEdited by M. Ramesh and Scott Fritzen

New York London

< previous page page_iii next page >

< previous page page_iv next page >

Page ivFirst published 2009 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RNSimultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016Routledge is an imprint of the Taylor & Francis Group, an informa businessThis edition published in the Taylor & Francis e-Library, 2008.

To purchase your own copy of this or any of Taylor & Francis or Routledge’s collectionof thousands of eBooks please go to www.eBookstore.tandf.co.uk.© 2009 Editorial Selection and matter, M. Ramesh and Scott Fritzen. Individualchapters, the contributors.All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers.British Library Cataloguing in Publication Data A catalogue record for this book is available from the British LibraryLibrary of Congress Cataloging-in-Publication Data Transforming Asian governance : rethinking assumptions, challenging practices / edited by M. Ramesh and Scott Fritzen. p. cm.—(Routledge research on public and social policy in Asia ; 2) 1. Social change—Asia. 2. Political culture—Asia. 3. Asia—Social policy—Case studies. 4. Asia—Economic policy—Case studies. 5. Asia—Politics and government—Case studies. I. Ramesh, M., 1960- II. Fritzen, Scott, 1969-HN652.5.T73 2009 306.2095—dc22 2008026900ISBN 0-203-88484-1 Master e-book ISBNISBN10: 0-415-47478-7 (hbk)ISBN10: 0-203-88484-1 (Print Edition) (ebk)ISBN13: 978-0-415-47478-8 (hbk)ISBN13: 978-0-203-88484-3 (Print Edition) (ebk)

< previous page page_iv next page >

< previous page page_v next page >

Page vContents

List of illustrations vii

List of contributors ix

Acknowledgements x

1 Introduction SCOTT A. FRITZEN

1

2 Improving administrative performance in Malaysia: the more difficult next stepsin reform HONG-HAI LIM

19

3 India: getting governance right for inclusive growth SANTOSH MEHROTRA

42

4 Is foreign aid compatible with good governance? Theory and evidence from thePhilippines EDUARDO ARARAL, JR.

58

5 Public sector transparency and corporate accounting practices in Asia XUN WU

74

6 The Hara factor: some lessons from innovations in governance in Japan JORRIT DE JONG AND ARRE ZUURMOND

97

7 “Good governance” and the constraints of local conditions: Thaksin and thebreakdown of Thailand’s public integrity system ALEX M. MUTEBI

116

8 The judiciary’s role in good governance in Korea JOONGI KIM

135

< previous page page_v next page >

< previous page page_vi next page >

Page vi

9 Healthcare reforms in Thailand: rethinking conventional wisdom M. RAMESH

154

10 Restructuring government: an empirical analysis of agencification in the publichealth system of Rajasthan AVANTIKA SINGH

168

11 The structural transformation of public bureaucracy towards good governance:the case of team-based management in the Korean government CHANG KIL LEE

189

12 Technocratic solutions versus political realities: implementing governancereforms in the Balochistan province of Pakistan RAZA AHMAD AND SYED A.A. AKIF

206

Index 229

< previous page page_vi next page >

< previous page page_vii next page >

Page viiIllustrationsFigures 1.1 Nodes in governance analysis 14.1 The vicious cycle problem in irrigation and moral hazard problem in aid 614.2 The effect of aid fungibility on bureaucratic incentives 624.3 Growth of irrigation aid in the Philippines, million dollars (2002 prices) 644.4 Solution mechanism to the problems of moral hazard and aid fungibility 71

10.1 Linkages of agencies with the public health system and their service deliverynetwork

178

10.2 Service delivery networks in the public health system of Rajasthan withagencies as the focal organisation

180

10.3 Revenue generation of RMRS over the period 1997–98 to 2002–03 18112.1 Balochistan: global imperatives and national and local systems of governance 21412.2 PRMP policy goals 21512.3 Conceptual framework of DSSP policy reforms 217Tables 2.1 Representation by race in the Malaysian bureaucracy 2005 262.2 Index of representation by race in the Malaysian bureaucracy 2005 275.1 World Business Environment Survey (2000) coverage in Asia 785.2 Public sector transparency as perceived by firms across Asian countries 805.3 Asian firms adopting International Accounting Standards (IAS) and external

auditing of annual financial reporting (AUDIT) 83

5.4 Accounting practices in Asian firms: sales reported for tax purposes 84

< previous page page_vii next page >

< previous page page_viii next page >

Page viii5.5 Dependent and independent variables: description and descriptive statistics 865.6 Probit models: adoption of accounting standards 905.7 Ordered probit models and interval regression models: accounting practices 937.1 A simple typology of corruption 1197.2 Key “good governance” provisions in Thailand’s 1997 constitution 1238.1 Election-law-related prosecutions 1428.2 Election-law-related court judgments 1438.3 Sectors affected by corruption (2005) 1448.4 Corruption-related court judgments 1478.5 Corruption-related pardons and re-instatements 1509.1 Hospital admissions rate, percentage of total population 1599.2 Summary of the development of major health policy reform measures 165

10.1 Agencies selected for the study on agencification of the public health systemin Rajasthan

174

11.1 Change of team size and team number before and after the team system 19411.2 Performance indicators by balanced scorecard in MOGAHA 19511.3 Classification of three periods by adoption time 19911.4 Three influential drivers, by adoption periods 20212.1 Net fiscal transfers to local governments 218Graphs 8.1 Rule of law (World, 2005) 1368.2 Control of corruption (World, 2005) 1378.3 Voice and accountability (World, 2005) 1388.4 Korea, South (2005) 1388.5 Rule of law (World, 2005) 1468.6 Original prison sentence and actual sentences served for senior public officials 149

< previous page page_viii next page >

< previous page page_ix next page >

Page ixContributorsAlex M. Mutebi, Singapore Management UniversityArre Zuurmond, The Universiteit LeidenAvantika Singh, TERI UniversityChang Kil Lee, Sejong UniversityEduardo K. Araral, Jr., Lee Kuan Yew School of Public Policy, National University ofSingaporeHong-Hai Lim, Universiti Sains MalaysiaJoongi Kim, Yonsei UniversityJorrit de Jong, Ash Institute for Democratic Governance and Innovation, HarvardUniversityM. Ramesh, University of Hong KongRaza Ahmad, Asian Development BankSantosh Mehrotra, Planning Commission, Government of IndiaScott A. Fritzen, Lee Kuan Yew School of Public Policy, National University ofSingaporeSyed A.A. Akif, Director General Secretariat of the Sindh Provincial Ombudsman inPakistanXun Wu, Lee Kuan Yew School of Public Policy, National University of Singapore

< previous page page_ix next page >

< previous page page_x next page >

Page xAcknowledgementsPapers in this volume were first presented at the International Workshop onGovernance in Asia – Case Studies, held in Penang on 13–14 December 2006.The workshop was jointly organized by the Asian Political and International StudiesAssociation (APISA) and the Centre for International Studies (CIS), Universiti SainsMalaysia. Funding for the workshop was generously provided by the SwedishInternational Development Agency (Sida) through APSIA.We would like to express our gratitude to APISA, CIS, and Sida for their support.Special thanks is due to the APISA Executive Secretary Dr Hari Singh and the CISDirector and Professor Dato’ Syed Ahmad Hussein for their contribution to the successof the workshop. The CIS staff cheerfully provided excellent logistical support for theworkshop – thank you.Earlier versions of the papers by Eduardo Araral, Joongi Kim, Hong-Hai Lim, AvantikaSingh, and Raza Ahmad and Syed A.A. Akif were previously published in the journalPolicy and Society (vol 26, no 2). The article by Alex Mutebi has been previouslypublished in the journal. We are grateful to the journals for the permission to re-publish these papers.M. Ramesh Scott Fritzen

< previous page page_x next page >

< previous page page_1 next page >

Page 11 IntroductionScott A. FritzenThe term “governance” has migrated to the center of an overlapping set of conceptsdeployed across a broad range of settings, from academic theorists, to advocates forchange, to practitioners. In the process, phrases such as “good governance” havebecome commonplace: paradoxically, both overused and underanalyzed. The conceptof governance is so multifaceted for three reasons: it rose to prominence against abackdrop of myriad contemporary political and ideological changes and pressures; ithas been substantively deployed and defined in a number of contradictory ways, andin support of varied agendas; and the study of its “transformation” is associated witha broad array of methods that do not necessarily generate coherent findings. Viewinggovernance in these three cross-sectional views allows us to express some of thediversity of the contributions to this volume.A changed contextMost discussions of governance can be conceptualized in terms of how authority,resources and power are distributed among the public, private and “people” sectors,and across various levels of government (from a supranational to community level).Figure 1.1 provides a schematic representation of resulting nodes, where the scope ofthe concept of governance can be described, or mapped out, in terms of which of thenodes are included in the discussion, and how they interact. A common presumptionin the literature on public sector governance is that the last two to three decadeshave witnessed significant changes in the way public authority and resources arebeing deployed across the matrix in Figure 1.1, and that these changes are significantenough to warrant a new conceptual vocabulary (Peters 1996; Asian DevelopmentBank 1999; Kettle 2006).Figure 1.1 Nodes in governance analysis.Sector: Level of governance system Government Private ‘People’International National Intermediate Local/community



Page 2In assessing the contextual factors that have affected the governance debate, we maybegin with impacts on the public sector itself, the prominence and prestige of whichhave come under constant scrutiny over this period, as can be seen in severaldisparately related trends.One is a sense of “permanent fiscal crisis” (Osborne and Hutchinson 2004) that hasbuffeted governments around the world, and which has strained governments tomake difficult choices. The problems became apparent during the 1970s’ stagflation,characterized by high unemployment as well as inflation coupled with large publicsector deficits. A prominent body of thinking plausibly pointed the finger atoverextended government as the source of the problem. The “golden” 1950s and1960s, when social programs proliferated without an accompanying increase inproductivity, had laid the foundations for the fiscal crisis. Since it was hard to rollback the transfer programs, governments turned to extracting greater efficiency fromexisting resources (Brittan 2004). The fact that the “crisis” may be more related tothe astonishing range of roles that contemporary government continues to take ondoes not diminish the pressures associated with making the most of limited resources(Peters 2001).A second is growing citizen demand for higher-quality, and more responsive,government services, driven by a combination of increases in educational attainment,living standards and the power of the media (Frederickson 1996). Citizens throughouthistory have had low expectations from governments: they were content merely not tohave a rapacious and brutal ruler; but this changed with democratization as citizenvoters expected governments to serve them well and be thrown out of office if theydid not. Rising education and income buttressed these trends (Diamond 1994).A third trend involves changes in the level of general public trust in the competenciesof government; by one measure, for instance, in secular decline since at least theearly 1970s in the US (Pew Research Center 2008). Attempting to capitalize on and topropagate this widespread sentiment, the U.S. President Ronald Reagan memorablystated in his inaugural address that “government isn’t the solution to our problems;government is the problem.”For developing countries, a fourth trend concerns the pervasive sense ofdisappointment in the viability of state-directed economic and social transformation.This left many countries vulnerable to attempts to significantly pare back the reach ofthe state under the influence of the so-called “Washington consensus” beginning inthe 1980s. There have also, of course, been countervailing arguments and pressures.For instance, since at least the mid-1990s, the World Bank has advanced the idea thatdevelopmental effectiveness comes from strong state capabilities, properly directedand focused. In so doing, it has probably done more than any other singleorganization to promote the language of governance, seeing in it the conceptualunderpinnings of its expanded focus on both economic development and broader stateeffectiveness in promoting poverty reduction and corruption control (World Bank 1997,2001, 2004).Another set of trends is related to the changing contours of the so-called people andprivate sectors. The proliferation of Non-Governmental Organizations (NGOs),



Page 3both international and national in scope and in a bewildering diversity of sizes andcapacities, drew increasingly prominent attention and was in part fueled by thedemocratization trends noted above. Many claimed that this “third” or “people” sectoroffered unique advantages in delivering services in comparison to the state (Brownand Ashman 1996; Silliman and Noble 1998; Shigetomi 2002), even as others(Narayan 1999) feared that the combination of weak core government capacity andstrong civil society organizations would undermine state capacities and sector-widegovernance. Meanwhile, the size and prominence of the private sector has beendynamically growing as well. The last two decades in particular have seen anacceleration of international economic integration; trade and capital flows betweencountries have reached unprecedented levels, while all but a handful of countries inthe world have come to espouse the principles (if certainly not always the unfetteredpractices) of the free market as the driver of economic advancement (Wolf 2005).Substantive agendas and meaningsThe trends described above have set the stage for contestations over the question ofwhat makes for good governance. To a perhaps unusual degree, given the sometimestechnical and neutral language it deploys, the governance agenda has grown uparound a range of agendas, which could be projected onto the matrix presentedearlier. These agendas can be broken down into those that challenge the size of thestate, the locus of authority relationships, or the way government authority isdeployed. We briefly lay them out below.Contestations over the desirable size of the public sector have been provoked by theprivatization agenda. If “government is the problem” – due to its suffocating effect onother sectors, and its essentially self-regarding and empire-building mentality – then itis one that conceivably might be lessened by downsizing it. The privatization debatehas been summarized at length in a number of works (Netter and Megginson 2001;Shirley and Walsh 2001), and is far from over, with some analysts seeing areassertion of the importance of the state (Ramesh and Howlett 2006).A different logic seeks to change or supplement the national-level/public sectorquadrant of the matrix – the traditional locus of public authority and decision making.The decentralisation agenda represents, in one view, an attempt to “roll back thefrontiers of the state” rather than downsizing it literally (Devas 1997). Advocates ofdecentralisation (which itself takes administrative, fiscal, market-based and politicalforms) would see it serve a diverse range of objectives, from administrative efficiencyto democratic deepening (Andrew and Goldsmith 1998; Manor 1999; Tendler 1997;World Bank 2005). Along the same lines, but in a different direction of the verticalaxis of Figure 1.1, is the global (or supranational) governance agenda, which positsthe need to develop supranational institutions, both to provide a range of global publicgoods and to rationalize international decision-making and dispute resolution. Thevast expansion of the European Union’s size and powers over this period is but themost radical example of an ongoing attempt to craft institutions supplementing, and insome cases supplanting, authorities traditionally held by the nation state.



Page 4Perhaps the most characteristic movements in Figure 1.1 concern attempts to changethe way the public sector interacts with the people and private sectors. A participatoryagenda has sought to dovetail with both the democratization trends and theincreasing expectations and sophistication of citizens, noted above, in order to makethe state more responsive and transparent to, and more empowering of, its citizens(Peters 1996). Such participation and transparency may be initiated either because ofdemocratization (as a response to already articulated demands), or in some cases inmore authoritarian political settings as a substitute or shield against more far-reachingdemands of the self-same democratization (Grindle 2000; Fritzen 2005). A range ofreforms have sought variously to liberate business from the clutches of anoverweening bureaucracy, to tap into its capacities and dynamism throughcontracting, or to engage it as a partner in jointly financing and implementingcomplex schemes in collaboration with the public (and sometimes also people)sectors; these represent the deregulation, contracting and co-production agendas,respectively (Goldsmith and Eggers 2004; Kaul 1997; McCourt and Minogue 2001;Schick 1998; Camm 2005).In addition to the above, much discussion has focused on how to improvegovernment efficiency through the redesign of its own internal processes and incentivestructures. The reinvention agenda (Chisholm 1989; Rondinelli 1993; Osborne andPlastrik 1997; Rainey 2003) could at times espouse any of the strategies just noted,but at its heart was the attempt to make government more efficient and effective inits core processes, and thus presumably more like the private sector.Against this backdrop, the meanings taken on by the concept “governance” can beinferred from these agendas, or at least from the analytical efforts to capture thedynamics underlying the interactions implied by Figure 1.1. While specific definitionsvary, there are perhaps two broad ways of talking about governance.One category is that of “good governance principles.” A very broad range of suchpresumed principles (e.g. transparency, meritocracy, participation, and economicfreedom) are sketched out (often without much explanation or debate), and theanalyst then proceeds to examine the extent to which a particular country or casematches or deviates from the standard. A variation on the same theme involvesranking countries or other units of analysis along a continuum from good to poorgovernance quality according to some fixed criteria (Transparency International 2007;Freedom House 2007; Kaufmann et al. 2001, 2006). Groups adopting the principlesapproach may then develop specific indicators of the manifestation (or lack ofmanifestation) of the principle in a given environment; the percentage of budgetallocations going to relatively poor provinces, for instance, may be taken as a proxyfor the principle of “pro-poor governance.”The “good governance principles” approach is very popular, and is the basicconceptual apparatus underlying huge databases generating many circulated papersattempting to operationalize good governance (Kaufmann et al. 2006; Rotberg andGisselquist 2007). Of course, it has its critics. Its methodological flexibility allowscomplex concepts to become measurable and therefore gain the appearance ofobjectivity and overly precise measurement. The “principles” approach typically



Page 5combines different elements from the agendas introduced above, without having toconfront the difficult question of how the principles are interrelated and whether thereare any trade-offs among them that must be weighed. If the existing literature ongood governance often suffers from being overly deductive, this tendency is greatlystrengthened through the use of the principles vocabulary; such principles may turnout to be less universal than suggested, as in practice many have been found to beneither necessary nor sufficient for realization of good governance.A second broad manner of talking about governance makes it largely synonymouswith public management and administration. Specific techniques or approaches forlinking organizations, sectors, and levels of government are described and givenvarious interpretations as to their effectiveness. Even where analysts are predicting oradvocating fundamental changes in the performance of the public sector through theapplication of a new approach – as in some of the more breathless works ofReinventing Government literature (e.g. Osborne and Plastrik 1997) – the fundamentalanalysis concerns the drivers of organizational performance in a way that would notbe totally out of place in a much earlier literature on the principles of scientificmanagement (Rainey 2003). In this category, then, governance becomes defined aswhatever aspect of the administrative and managerial system and environment theanalyst is most interested in (Lynn 2006), and often takes on the emergent meaningof a “network” linking different nodes, as in the Figure 1.1 matrix.The contributors to this volume take a variety of perspectives, falling within both ofthe above categories. The next section explores their diverse understandings ofgovernance by examining the question of how they approach the question ofgovernance transformations.A great transformation?To say that the concept of governance is often defined by the agendas that analystsand advocates bring to the discussion is another way of focusing on another elementin the book’s title: transformation. One online dictionary (www.dictionary.com) definestransformation in a theatrical context as “a seemingly miraculous change in theappearance of scenery or actors in view of the audience.” Although many advocateswould wish this to be taking place in respect to governance, and would claim it isoccurring in some country contexts, here we take a much looser and less grandmeaning of the term as essentially synonymous with sustained change (or the attemptto introduce it). In every country examined in this volume, we find basic governanceparameters being questioned, or at the very least the deployment of supposed “goodgovernance principles” to question the gap between norms espoused by politiciansand societal leaders, and the realities experienced on the ground. Such pressures andchanges may in many cases not add up to a “great transformation” of the orderPolyani famously noted for England in the late 18th and early 19th centuries (Polyani2001), but they shed light on real and ongoing contestations for the future ofgovernance ideals and practices in each region.One of the abiding characteristics of the chapters in this collection is that they



Page 6do not take one perspective on the definition of governance or the way governanceparameters are shifting (or are failing to shift despite growing pressures to do so).Amid a complex set of changes and continuities observed from up close in these casestudies, contributors take on two questions related to “transformation” and thesequestions broadly separate the first and second sections of the book. The first sectionconcerns where the pressures for change are coming from. The second concerns whatthe trajectory of change in specific sectors and arenas has looked like. In whatfollows, we look at these in turn, in the process introducing each chapter of thisvolume.Part one: context, constraints and drivers of changeChapters in Part One of the book highlight instances in which local conditions eithercontradict supposedly universal good governance principles or severely constrain theirrealization. Context here includes a broad range of factors: who is at the table, whatideas are in the mix, and what other sources of change are operative. These fivechapters thus do not talk primarily about specific reforms, but about this brew fromwhich varied reforms might emerge. In the process, they highlight broader factorsthat may be presumed to be operative across a wide range of settings. Three broadfactors or drivers of change are highlighted in the five papers of this section, cross-cutting to some degree: macro- and micro-political configurations, institutionalincentives or games, and well-positioned individuals.One of the most obvious places to look for accumulating pressures for governancereforms is a country’s macro-political system and its impacts. Hong-Hai Lim’s piece onthe Malaysian administrative system underscores the doubleedged sword of thebargains struck – and contested – at this level. The Malaysian bureaucracy, Lim notes,has played an important role in driving that country’s impressive development since itsindependence in 1957. Yet two bargains struck on a macro-political scale in the earlyyears of the country’s political development have turned into serious, and to dateimpassable, structural obstacles for bureaucratic performance improvements (whetherconceived in terms of efficiency, equity or responsiveness).The first is the disproportionate share of civil service positions deliberately reservedfor (economically disadvantaged but politically ascendant) ethnic Malays. This hascreated a situation in which the bureaucracy is neither representative of the broaderpopulation (a condition Lim posits as essential in a multiethnic society), norincentivized to perform to a professional standard, “significantly sacrificingeffectiveness and efficiency.” Lim claims there are direct and indirect impacts of thisimbalance: “[a]n unrepresentative bureaucracy not only symbolizes unequal treatmentto under-represented races … it is also likely to lead to unequal substantive treatmentof under-represented races.”The second abiding feature of the early political order is disproportionate executivedominance and control. This imbalance was reflected in, and reinforced by, the so-called unbalanced strategy of bureaucratic development and capacity building (onestrongly advocated by influential American advisors at the time), where the



Page 7term “unbalanced” refers to its failure to strengthen extra-bureaucratic mechanismsfor exercising oversight of the bureaucracy. Lim highlights the path dependenceembedded in such a choice: “the role and capacity of most existing mechanisms ofpublic control have also been progressively diminished” by the government that hasdominated electoral politics in Malaysia since its founding. Weak public control, Limclaims, interacts with the Malay-dominant bureaucracy to stem meaningfulperformance improvements within the bureaucracy as a whole.Macro-level political configurations, which took root some half a century ago, thuseffectively explain the current impasse that Lim sees in the Malaysian bureaucracy.The fact that this story is far from static, however – that pressures for changecontinue to build under, and sometimes breathe through to, the surface – is evidentfrom the 2008 Malaysian election outcomes, which saw opposition parties gaining anunprecedented share of the national vote. Politics emerges repeatedly in this volumeas both a constraint and a resource for governance innovations and transformations.Santosh Mehrotra also takes up political institutions as a potential source ofgovernance reform. He focuses on India’s constitutional framework for localgovernment, and the effective political incentives structured into a supremelyheterogeneous local environment in federal India. While India has been feted as arising economic superpower and an example of “Asia rising,” Mehrotra notes that fully106 out of 607 districts in the country are facing a Maoist (or Naxalite) insurgency,“the result of the effective collapse of the state in these districts.” The criticalgovernance challenge facing India – a challenge that India has only begun to address– is to achieve more inclusive growth through the institutionalization of “much greateraccountability … to ensure which functionaries should be made accountable to locallyelected bodies.”To accomplish this, he argues, the key requirement is “deep democraticdecentralisation.” Decentralisation as Mehrotra conceives it is not any individualreform. It is instead the transferring of greater rights and resources to local levels ofgovernment, and notably to elected community-level bodies (the panchayats), coupledwith the deepening of the effectiveness and responsiveness of these bodies. Changesin accountability relationships at the local level are essential to ensure thatintermediaries – local bureaucrats and community elites – do not capture forthemselves the lion’s share of state resources designed for the poor, and that indeedthey are sufficiently incentivized to promote service delivery, on which theadvancement of the poor ultimately rests.The set of conditions for the “deep democratic decentralisation” Mehrotra describes isderived from a host of experiences around the world, summarized notably by theWorld Bank’s World Development Report of 2004, in which changes in a country’smacro-governance framework reinforce the incentives of central and local governmentrepresentatives, service delivery staff and citizens to deliver services to end-users inmore efficient and responsive ways. Case studies of education in two of India’spoorest states – Rajasthan and Madhya Pradesh – suggest that significant progress atthe sectoral level is possible. At the same time, the macro-political reform that couldfacilitate a broader replication of these positive



Page 8experiences across sectors and other local governments – the 1993 constitutionalresurrection of local (sub-state) governance – has been generally resisted by states,most of which “have not been willing to transfer funds and functionaries to the locallevels of government – undermining the whole purpose of the reform.” Thus, despitepromising signs in individual sectors and localities, Mehrotra’s analysis, much likeLim’s, leads to a pessimistic conclusion about the impact of macro-political institutionson prospects for needed systemic reform: he bewails the “tragedy” that “the federalstructure of the governance system in the country prevents the central governmentfrom exercising any sanctions against state governments when they fail to fulfil theirobligations under the constitutional amendment.” The best that can be hoped for, it isimplied, is a patchwork of diverse local reforms that may emerge unevenly in thespaces created by India’s federal system.If Lim and Mehrotra explore the way political dynamics influence developmentdynamics, the next two contributors focus on a partially overlapping but distinctmechanism – the configuration of incentives embedded in the institutionalenvironments that often have little to do with the macro-political rules of the game.Eduardo Araral answers the question, “Does foreign aid undermine good governance?”in the affirmative; his examination of the impact of foreign aid to the irrigation sectorin the Philippines finds that “aid is embedded in a perverse set of incentives thatundermine the application of good governance principles espoused by donors.”Elaborating approaches pioneered by Ostrom (1990) and North (1990), he finds thattwo problems in particular are endemic to foreign aid, and clearly evident in the casestudy he reports.The first is moral hazard. Aid recipients face little incentive to implement promises ofreform: he notes that “the injection of aid alleviates the immediate fiscal crisis of therecipient government and hence the urgency for change,” especially since theincentives faced by aid officials themselves mitigate against the strict enforcement ofconditionality. The second is the fungibility of aid, which allows expenditure on non-related items (Araral could, but does not, mention corruption as one possibility) to risewhile minimum standards in the intended, ostensibly “high-priority” public gooddeteriorate. Araral places these particular problems of foreign aid in the broadercategory of two “incentive problems of governance” – motivation and information –which arise primarily not due to macro-political characteristics, but rather due tospecific characteristics of the “game” being played: the institutional characteristics ofthe good in question (such as its excludability); the attributes of the relevant players(such as their history of cooperation); and the broader rules of the game (such aspatterns of enforcement and rule-making). Analyze these characteristics (which willvary from sector to sector, from game to game), Araral suggests, and one will have amuch better understanding of governance performance, pressures for change, andprospects for success. With its focus on micro-institutional configurations, Araral’sanalysis lends itself to a potentially more optimistic conclusion for reformers than thatoffered by the previous two authors; he shows how a specific, “game-changing”intervention can enhance system performance in the irrigation sector he explores.Xun Wu’s empirical paper provides another example of institutional constraints



Page 9on governance reforms, using corporate accounting practices as a case study. Thequality of governance in the private sector is clearly one of the key determinants ofgovernance in the public sector. At the same time, improvements in corporategovernance are constrained by public sector governance. Wu’s analysis suggests thatthe quality of corporate accounting practices is positively related to the quality ofpublic sector governance as measured by predictability of rules, laws, and regulations.There is little chance that adoption of International Accounting Standards will lead tohigh-quality accounting practices as long as firms and their political patrons havevested interests in keeping the public sector opaque to protect existing rent-seekingschemes. In an opaque operating environment, firms may have no choice but to adoptshady accounting practices to cope with various risks arising from informationasymmetries between government and business.Wu’s analysis based on the linkage between public and private sector governanceoffers two important policy lessons. First, policy makers in the region should takepublic sector transparency into consideration when designing and implementingreforms aimed at the private sector. Ignoring such interrelationships between publicand private sector governance may reduce the relevancy of reform policies. Second,his findings broaden the scope for effective measures for corporate governancereform by placing a new set of instruments, focused on corporate operatingenvironments, at the disposal of those charged with projecting and implementingreforms. For example, pointed efforts can now be made at improving both theeffectiveness of corporate governance reforms and the success of anticorruptioncampaigns by enhancing public sector transparency.Jorrit de Jong and Arre Zuurmond provide an assessment of innovations ingovernance that is both stylistically and substantively very different from the chaptersabove. They examine an extended case of successful innovation in the developmentand implementation of electronic medical record systems in health care networks inJapan’s Shikoku province. The authors put forward several competing clusters ofhypotheses purporting to explain the conditions under which management innovationis likely to take place. While they disavow any “modernist-positivist ambition” toconstruct a unified, comprehensive framework explaining innovation, they use thecase to poke holes in some of the more institutionalist explanations for innovationprevalent in the literature.Do institutional patterns and structures, such as the degree of centralization of thestate structure, or the funding regime in the sector in which innovation is to occur,matter to the success of the Japanese case? Did the pattern of managerial incentivesand human resource management offered by the superiors of those directly involvedin the case of innovation matter, as management theorists would contend? In bothcases, the answer is clearly no, according to the authors. Instead, de Jong andZuurmond find that innovators’ own social networks and skills in change management,coupled with both “serendipity” and the changing set of options posed bytechnological advancement, formed the set of necessary and sufficient conditions inthe case examined.De Jong and Zuurmond’s case straddles the two parts of this collection. Like others inpart two, it examines a specific (and in this case successful) reform



Page 10trajectory. But its main contribution within the volume is to highlight idiosyncraticfactors underlying governance transformations, at least at the level of specific sectoralpractices and management tools.Part two: Reform trajectories and transformationsThe six chapters in part two tell the story – the “trajectories” – of specific reformattempts in four different countries: Thailand, South Korea, India and Pakistan. Thegeographical and developmental diversity of the countries speak for themselves. Thereforms undertaken in them also include attempts to change governance parametersat the framework level (Thailand’s constitution and Korea’s judiciary); meso-levelreforms to specific sectors (Thailand and India’s health sectors); and micro-levelmanagerial reforms implemented across a range of government bureaus (Korea’sministerial reforms); and one case in which the scope and level of proposed reformsas applied to a sub-national government are both deliberately varied and vague(Pakistan). The chapters also vary in terms of the success these reforms havegarnered, from showing strong evidence of a positive outcome, to mixed success, toessentially complete failure (at least in the short term). The premise of part two isthat while it will never be possible in a single volume to capture or to theorize the fullrange of attempts to “transform governance” in Asia, the diversity of the piecesshould shed light on both the substance of a variety of reforms and the methods thatmight be employed to study them.Alex Mutebi’s chapter kicks off part two at the same level at which part one began:with macro-politics. Mutebi addresses what appears to be a puzzling and disappointingfailure. When promulgated in 1997, Thailand’s latest constitution was touted worldwide as a “good governance” constitution, one which expressed some of the bestpractices in crafting governance arrangements to include numerous checks andbalances and avenues for people’s direct participation, all in service of being bothaccountable and responsive to citizens. A battery of new institutional innovations wasestablished. These included, among others, an Election Commission (to investigateelectoral fraud and enforce strict rules, with broad powers banning political partiesfailing systematically to follow regulations); a National Counter Corruption Commission(to mandate asset disclosure of senior politicians and to bar from office politiciansdetermined by the commission itself to be corrupt); and an Office of the Ombudsman.All of these (among other institutions) “proved woefully inadequate, primarily becausethe type and levels of corruption that predominated during [Prime Minister ThaksinShinawatra’s] leadership simply overwhelmed the existing anticorruption measures.” Arange of additional measures – such as the stipulation that the Senate would bydecree be non-partisan, and the adoption of various anti-corruption regulations out ofthe Prime Minister’s office – complemented these constitutional changes. Advocates of“good governance principles” espoused by multilateral development agencies assessedthe emerging institutional configurations to be on the cutting edge of best practice incombating corruption.Mutebi chronicles the myriad ways in which an electorally successful and strong



Page 11prime minister bent on the accumulation of executive power was able to subvertvirtually all of these measures and institutions. Basically Mutebi’s explanation is thatThai reforms in the mid to late 1990s primarily employed “technocratic approaches”that proved incapable of preventing “state and regulatory capture” forms ofcorruption, and the undermining of the effective implementation of administrativecorruption control measures. Mutebi’s broader point is that a wide gulf can existbetween the formal rules of the game and the capacities and configurations ofincentives necessary to prevent these formal rules from being completely underminedin practice. His conclusion, that “we can at the very least appreciate the challenges ofaddressing both administrative corruption and the various forms of ‘capture’ in anyintegrity, anticorruption and ‘good governance’ reforms,” could stand for many casesin which broad political parameters undermine attempts to elevate governancepractices to the level of congruence with some imagined principles of goodgovernance, even where the rhetoric – or even the official institutions themselves –are adopted enthusiastically.Joongi Kim’s examination of the judiciary in South Korea similarly focuses on broadsystem-level contests over “good governance” and the degree of their effectiveimplementation. His case is in some ways the mirror opposite of Mutebi’s: publicpressure and institutional innovations to enhance the enforcement capacities of thejudiciary had exactly the intended effect over roughly a ten-year period, leading tostricter application of campaign financing, election laws and anti-corruption laws. Kimexplains the gains in terms of “a combination of legislative changes, strictersentencing, faster review, greater incentives for whistleblowers and, ultimately, judicialrecognition of the importance of establishing enforcement discipline,” and claims that“[t]he result has been historic progress” and a discernible movement in the directionof good governance ideals.The reform progress Kim documents is made all the more remarkable because of thedisadvantageous starting point in the recent past. Judicial enforcement against seniorpoliticians and civil servants who were found to be corrupt was noticeably lax up toabout the mid-1990s, and the country has been ranked generally more poorly against“rule of law” criteria than its income level would predict. Kim cites a range of factorsunderlying this poor performance. Somewhat in contrast to institutional explanationsof the sort offered by Araral and Mutebi, some of the factors cited by Kim werecultural factors, such as a bias against the strict punishment of white-collar crime, andfears on the part of judges; others, like the ability of well-connected and resourceddefendants to exploit legal loopholes and influence sitting judges (often formercolleagues and friends), are more direct and tactical in nature.Against this background, Kim’s explanation of emerging patterns of enhanced judicialenforcement centers on a virtuous circle of democratic maturation, legal andadministrative mandates, and the dilution of the power of an “old boys” network ofjudges. The first factor is perhaps the most decisive: the consolidation of democraticprocesses in South Korea from the mid-1990s onwards has created the preconditionsfor more effective checks and balances on the executive and more scrutiny over thejudiciary (including appointments). With several successful



Page 12transfers of power under electoral democracy, all parties stood to gain from theestablishment of clearer standards, effective mutual scrutiny of processes, and theperception that they are on the “public’s” side of the rising groundswell of anger overofficial corruption: hence the broad support for tough legal changes that have provensurprisingly effective, such as the establishment of sizeable monetary rewards (notjust protection from punishment) for whistleblowers, and prohibitions against retiredofficials working in areas related to their former responsibilities. Finally, as thejudiciary has grown in size, judges have become more diverse in background, helpingto “decentralize the concentration of power that revolved around a limited number ofjudicial elites.” It is remarkable to note that Kim is describing precisely the publiccontrol strategy and the democratization of the civil service which Hong-Hai Limclaims as essential and lacking from the start in the Malaysian civil service case. Thechapter’s findings also imply that Thailand’s case, as documented by Mutebi, mayhave turned out quite differently had the institutions had a longer time span tomature (in the absence of the Thaksin factor).Yet as if to remind us of Mehrotra’s point on the need to disaggregate governanceperformance in a variety of ways – to use both a wide-angle lens and a zoom whenassessing country experiences – M. Ramesh in Chapter 8 uses the case of Thailand’ssuccessful health sector reforms under the very same Prime Minister to “rethinkconventional wisdom” regarding policy design in the sector. Ramesh notes thatcurrent orthodoxy regarding policy design in the health sector would predict Thailand’shealth sector – characterized by government dominance of both provision andfinancing of services – to be “a policy disaster.” A tax-funded Universal Coverage (UC)healthcare program launched in 2001 only strengthened this dominance.Yet far from a disaster, early evidence regarding the effects of the UC reform“suggests a massive improvement: overall healthcare spending has gone down whileaccess has improved.” Ramesh’s explanation for UC’s success is both straightforwardand jarring to those caught up in the current orthodoxy’s enthusiasm for competitionand marketization of health services. A careful analysis suggests that it is the absenceof competition among providers of health services that has served Thailand’sdeveloping country setting well, since it has reduced wasteful capital investment andduplication of services. The pooling of financing has meant the ability to keep averageoutlays per patient low and appropriate to what the country’s level of economicdevelopment can bear. And the program’s very existence is due to the fact that itleveraged popular anxiety over an inability to cover basic or catastrophic health carecosts (particularly among the almost one-third of the population shut out of any socialinsurance mechanism) into a politically attractive and operationally and financiallysustainable package. While imperfect, “[t]he Thai healthcare reforms … defy theaphorism that good policies make bad politics, because UC is attractive in botheconomic and political terms,” he notes.In other words, judge policy design by the outcomes it is able to achieve rather thanby reigning orthodoxies, Ramesh’s analysis implies. A contingency approach is impliedhere: dominant policy prescriptions in health may be appropriate for increasing thequality of health care in developed country settings, but are likely to



Page 13be incomplete or entirely inappropriate where the dominant challenge is to increasethe quantity of basic health provision. Ramesh’s chapter suggests that policy design isa critical arena in which proposals for challenging governance parameters play out.Battles over policy design can reflect the interaction of ideologies of public governanceand management, macro-governance structures and constraints, the diagnosis ofspecific reform problems and identification of possibilities (often by individual reformchampions, as de Jong and Zuurmond noted), and the politics of managing policychange. Examining specific policy sectors is also promising because it challengesanalysts to look beyond processes presumed to reflect good governance towards theassessment of actual outcomes “on the ground” as the litmus test of goodgovernance.Avantika Singh and Chang Kil Lee (Chapters 9 and 10) move the discussion towardsthe public management end of the governance debate. Both develop in-depth, up-close analyses of specific New Public Management reforms as they have been appliedin settings far different in their institutional configurations and organizational culturesfrom those of the countries in which the reforms were initially theorized. Lee notesthat the New Public Management focuses attention onto “two distinctive methods” forenhancing organizational performance – agencification and empowerment – and theseare the subjects of Singh’s and Lee’s chapters, respectively.Singh continues Ramesh’s sector-wide scope of analysis in her chapter on the publichealth in India, but drills down on the observed impact of agencification in the healthsector. “Agencification” is defined as the carving out of separate, autonomousagencies to carry out specific activities within policy sectors that had previously beenconducted by formal government departments or ministries. It is a staple ofrecommendations for good governance coming out of the New Public Managementliterature; the assumption is that agencies freed from a ministerial hierarchy will beempowered to behave more innovatively, responsibly and efficiently – in short, asSingh notes in language reflective of the “reinvention” and “deregulation” agendascovered above, “agencies are expected to function in a more business-like manner.”Yet agencification as a specific reform proposal has its critics as well. These arguevariously that it has not worked, even in those developed countries (such as NewZealand and the UK) where it has been adopted, prevents the horizontal coordinationof policy implementation, obscures the evaluative and effective policy-making activitiesof agencies, and may represent, as in Christensen and Lægreid’s (2004) phrase, partof a set of “ideological imports rather than real innovations.”Singh’s detailed analysis of health reforms in the state of Rajasthan methodicallytackles several issues. The first is to show how agencification was introduced toRajasthan’s health sector, in different forms at both state and grassroots levels, as aresult of several different forces: mimetic factors (agencification was regarded andpromoted as a best practice in the literature), coercive factors (aid donorsnecessitating agencification for program implementation), and normative factors(forming agencies to promote the perceived competence and legitimacy of the healthauthorities). Another is to show the way in which the specific form of



Page 14agencification adopted in Rajasthan differs substantially from the “Westernconceptualization” of the reform, in ways that are critical for understanding theimpacts of the reform. Those impacts include the creation of space for both privatesector participation and creativity in the contracting and delivery of hospital services –completely consistent with the predictions of reform advocates – but also the failureto strengthen vertical “steering” and accountability relationships through performancemeasurement.Chang Kil Lee’s case of “team-based management” reforms across South Koreanministries is similar to Singh’s in its focus on the dissemination of an externallyinitiated New Public Management reform to a non-Western context. Team-basedmanagement is posited by Lee to be one of the foremost examples of the“empowerment” principle within the NPM, which is intended to delegate “bothauthority and responsibility to lower levels of the hierarchy,” resulting in improvedemployee motivation and organizational performance. The team approach in theKorean case reflects structural changes to delayer levels of organizational hierarchy;the appraisal of team performance to supplement individual performancemanagement; enhanced human resource flexibility in the recruitment of team leadersand the composition of members; and greater deployment of horizontalcommunication rather than vertical authority to manage activities and resolvedisputes. Lee shows how all of these principles, actively advocated by the Ministry ofGovernment Administration and Home Affairs for adoption by ministries, met withresistance in an administrative culture which had long emphasized authorityrelationships, centralization of decision-making and the elite status of seniorleadership. But the reform was not uniformly resisted: echoing Xun Wu’s method ofanalyzing the characteristics of differentially positioned units in a system, Lee’sanalysis is a careful modeling of the reasons underlying the differential speed ofadoption of team-based management approaches across ministries. He concludes thatthe reform has been generally positive in its impacts, but will be faced with importantchallenges for its sustainability, most notably the sustained leadership driving thechange.Singh and Lee’s studies thus represent approaches for studying the adoption ofreforms initiated or inspired by forces external to the implementing field, but givenlocal content through the trajectory of their implementation. We can understand themthrough the institutional literature. In this way, they serve as templates for how toanalyze the adoption of international management reforms across country contexts.Raza Ahmad and Syed Akif’s exploration of a large donor project to promote goodgovernance in a profoundly challenging environment – Pakistan’s Balochistan province– rounds off the collection. The Asian Development Bank’s Balochistan PublicResource Management Project, coupled with complementary social service deliveryprograms, has sought to strengthen the technical efficiency of the provincialgovernment apparatus and, more ambitiously, to enhance the responsiveness andeffectiveness of local service delivery institutions. Its scope emerges as extremelyambitious, combining technical, managerial reforms such as those described by ChangKil Lee and Avantika Singh with support for “deep



Page 15democratic decentralisation” advocated by Mehrotra for India. All of this takes place ina political and social environment characterized by high levels of rent seeking,corruption, ethnic polarization, elite capture of development resources and a longhistory of failed public sector reforms. The gap between formal and informalinstitutions, such as those implied by Araral and Wu, become here potentially fataltraps for those who would promote even modest governance reforms.Yet Ahmad and Akif’s narrative is one of heroic efforts coupled with “partial reformsyndrome.” While limited progress may be made in some dimensions along someindicators, technical dimensions of the reform project are typically overwhelmed bythe constraints and imperatives of political economy, which change only slowly andhave little to do with the broader good governance principles espoused by even thebest-intentioned foreign aid donors. They ultimately note a lack of currently availabledevelopment tools and instruments “that facilitate ‘transformational change’ throughinstitutional reform, and that recognize that development is a political process.” Theirrecommendation is not to cease aid efforts entirely in such settings, but to adopt asystemic approach in which donors deploy their resources in ways that potentiallywiden “spaces for political action in the medium term,” spaces characterized bygrowing linkages between local legislatures, voters and policy imperatives “beyond thetechnocratic frontiers of executive agents in national governments and aid agencies.”Ahmad and Akif’s chapter is a fitting end to the collection because it reinforces severalthemes of the collection: tensions that almost always exist between technical andideological conceptions of good governance; the ultimately political, highly contestedand reversible nature of governance reforms; and the ultimately vague andindeterminate nature of conceptualizations of good governance when these crossnational, cultural and organizational boundaries.Where, or what, is Asia?What, then, should one make of the “Asian” element of the collection? Is it anormative concept, denoting some model of governance that could and should standin contrast to “Western” assertions that transfer (as we have seen) uneasily acrosscultural boundaries? Probably not: in examining some nine countries across three sub-regions of Asia, we find no compelling, unified construct that would qualify as an“Asian governance” model, although claims for the same are from time to time putforward by regional thinkers and leaders.Nor do we see a particularly unified set of challenges and pressures facing Asiansettings with respect to governance reforms. While commonalities exist and set thestage for some of the comparisons across country contexts made by the presentvolume, the picture that emerges is that of a diverse mosaic, in which even themeaning of “deep democratic decentralisation” appears to differ markedly betweenneighbors India and Pakistan. The “Asia” in the title does not denote in a stringentway a descriptive set of pressures or trajectories underlying governance reforms, atleast not in a way that would facilitate sweeping generalizations across sub-regions ofAsia, let alone the continent as a whole.



Page 16What is left is Asia as a geographical marker only. To say this is not to shut downmeaningful comparisons, but to narrow their scope and to make them morecontingent. The same, ultimately, is true of the concept of governance itself. Fewusers of the term “governance” would meet the test specified by Frederickson(2005:293; cited in Lynn 2006:12): “[G]overnance theorists must be ready to explainnot only what governance is, but also what it is not [and] be up-front about thebiases in the concept and the implications of those biases.” That is clearly thechallenge for both proponents of good governance and analysts of regionaldevelopments with respect to the same.ReferencesAndrew, C. and M. Goldsmith 1998, ‘From Local Government to Local Governance –and Beyond?’ International Political Science Review, 19(2): 101–17.Asian Development Bank (ADB) 1999, Governance: Sound Development Management.Manila: ADB.Brittan, S. 2004, Against the Flow. London: Atlantic Books.Brown, L. D. and D. Ashman 1996, ‘Participation, Social Capital, and IntersectoralProblem Solving: African and Asian Cases.’ World Development, 24(9): 1467–79.Camm, F. 2005, ‘Using Public-Private Partnerships Successfully in the Federal Setting.’in Klitgaard, R. and P. Light (eds), High-Performance Government: Structure,Leadership, Incentive, Santa Monica, CA: Rand Corporation, 179–214.Chisholm, D. 1989, Coordination Without Hierarchy: Informal Structures inMultiorganizational Systems. Berkeley and Los Angeles: University of California Press.Christensen, T and P. Lægreid 2004, ‘Governmental Autonomisation and Control: theNorwegian Way.’ Public Administration and Development, 24(2): 129–35.Devas, N. 1997, ‘Indonesia: What Do We Mean by Decentralization?’ PublicAdministration and Development, 17: 351–67.Diamond, L. 1994, ‘Rethinking Civil Society: Toward Democratic Consolidation.’ Journalof Democracy, 5(3): 4–17.——2008, ‘The Democratic Rollback: The Resurgence of the Predatory State.’ ForeignAffairs, March/April 2008.Frederickson, H. G. 1996, ‘Comparing the Reinventing Government Movement with theNew Public Administration.’ Public Administration Review, 56(3): 263–70.Freedom House 2007, Freedom in the World 2007. Washington, DC. Available at:http:// www.freedomhouse.org/template.cfm?page=15 [June 4 2008]Fritzen, S. 2005, ‘The “Misery” of Implementation: Governance, Institutions, and Anti-corruption in Vietnam’ in N. Tarling, ed. Corruption and Good Governance in Asia.New York, NY: Routledge.Goldsmith, S. and W. Eggers 2004, Governing by Network: The New Shape of thePublic Sector. Cambridge, MA: Brookings.Grindle, M. 2000, Audacious Reforms: Institutional Invention and Democracy in LatinAmerica. Baltimore: The Johns Hopkins University Press.Kaufman, D, A. Kraay and P. Zoido-Lobaton 2001, ‘Governance Matters: FromMeasurement to Action.’ Finance and Development, June 2001.Kaufmann, D, A. Kraay, and M. Mastruzzi 2006, Governance Matters VI: Aggregateand Individual Governance Indicators 1996–2006. Washington, DC: World BankInstitute.Kaul, M. 1997, ‘The New Public Administration: Management Innovations inGovernment.’ Public Administration and Development, 17: 13–26.



Page 17Kettle, D. F. 2006, The Global Public Management Revolution: A Report on theTransference of Governance. 2nd ed. New York: Brookings Institution Press.Lynn, L. E. 2006, Public Management: Old and New. New York: Routledge.Manor, J. 1999, The Political Economy of Democratic Decentralization. Washington,DC: The World Bank.Narayan, D. 1999, ‘Bonds and Bridges: Social Capital and Poverty.’ World Bank PolicyResearch Working Paper 2167. Washington, DC: World Bank.Netter, J. M. and W.L. Megginson 2001, ‘From State to Market: A Survey of EmpiricalStudies on Privatization.’ Journal of Economic Literature, 39(2).North, Douglas C. 1990, Institutions, Institutional Change, and Economic Performance.Cambridge: Cambridge University Press.Osborne, D. and P. Plastrik 1997, Banishing Bureaucracy: The Five Strategies forReinventing Government. Reading, MA: Addison-Wesley.Osborne, D and P. Hutchinson 2004, The Price of Government: Getting the ResultsWe Need in an Age of Permanent Fiscal Crisis. New York: Basic Books.Ostrom, E. 1990, Governing the Commons: The Evolution of Institutions for CollectiveAction. Cambridge: Cambridge University Press.Peters, B.G. 1996, The Future of Governing: Four Emerging Models. Lawrence:University Press of Kansas.——2001, The Politics of Bureaucracy. New York: Routledge. Pew Research Center forthe People & the Press 2008, ‘The Federal Government’s Favorables Fall EvenFurther.’ Available at: http://pewresearch.org/pubs/836/opinion-federal-government-institutions [May 29, 2008]Polyani, K. 2001, The Great Transformation: The Political and Economic Origins of OurTime, Reprint (1944). Boston: Beacon Press.Rainey, H. 2003, Understanding and Managing Public Organizations. San Francisco,CA: Jossey-Bass.Ramesh, M. and M. Howlett, eds. 2006, Deregulation and Its Discontents: Rewritingthe Rules in Asia, Cheltenham, UK: Edward Elgar.Rondinelli, D. 1993, Development Projects as Policy Experiments: An AdaptiveApproach to Development Administration. New York: Routledge.Rotberg, R. I. and R. M. Gisselquist 2007, ‘Ibrahim Index of African Governance.’Belfer Center for Science and International Affairs, Program on Intrastate Conflict andConflict Resolution. Available at: http://www.moibrahimfoundation.org [25 September2007]Schick, A. 1998, ‘Why Most Developing Countries Should Not Try New ZealandReforms.’ The World Bank Research Observer, 13: 1123–31.Shigetomi, S. 2002, The State and NGOs – Perspectives from Asia. Singapore:Institute of Southeast Asian Studies.Shirley, M. M. and P. M. Walsh 2001, ‘Public vs. Private Ownership: The Current Stateof the Debate.’ World Bank Policy Research Working Paper No. 2420. Available at:http:// ssrn.com/abstract=261854 [4 June 2008]Silliman, G. S. and L. G. Noble, eds. 1998, Organizing for Democracy: NGOs, CivilSociety, and the Philippine State. Honolulu: UH Press.Tendler, J, 1997, Good Government in the Tropics. Baltimore: Johns Hopkins.Transparency International 2007, Corruption Perceptions Index. Berlin. Available at:http:// www.transparency.org/policy_research/surveys_indices/cpi/2007 [June 4 2008]McCourt, W. and M. Minogue, eds. The Internationalization of Public Management:Reinventing the Third World State, Cheltenham, UK: Edward Elgar.Wolf, M. 2005, Why Globalization Works. New Haven: Yale University Press.World Bank 1997, World Development Report 1997: The State in a Changing World.Washington, DC: World Bank.——2001, World Development Report – Attacking Poverty. Washington, DC: WorldBank.



Page 18——2004, World Development Report – Making Services Work for Poor People.Washington, DC: World Bank.——2005, East Asia Decentralizes: Making Local Government Work. Washington, DC:World Bank.



Page 192 Improving administrative performance in MalaysiaThe more difficult next steps in reformHong-Hai LimMalaysia has devoted almost continuous attention to improving administrativeperformance since its independence in 1957. In terms of performance, the Malaysianbureaucracy is generally rated high, even as a star, among developing countries(although this is not the most demanding comparison group). However, of late,Malaysia’s deficiencies have attracted increased concern. Besides the palpably risingcitizen expectations of administrative performance (and hence intolerance ofdeficiencies), the growing perception that the country is losing its edge in the morecompetitive global economy has significantly added fuel to this concern – notsurprising in a country where continued economic growth is seen as key to bothregime security and sociopolitical order. Sensing the need, Prime Minister AbdullahBadawi (who took over from Mahathir Mohamad in October 2002) made improvingadministrative performance and fighting corruption key pillars of his highly successful2004 election campaign. But there have followed widespread perceptions of tardinessin delivering on these promises.This paper is an attempt to look back in order to find ways to move forward. Despitethe risk of spurious learning, such an attempt at empirically based prescription iscrucial to a field that seeks not only to understand but also to improve practice.Where did past reforms fall short? What further reforms seem necessary for pushingadministrative performance to higher levels? These questions, and the issue of reformstrategy that they imply, are addressed in this paper.Reform as improving capacity and willMy search for answers is structured by some basic concepts and ideas. Very briefly,these follow.How do we judge the performance of public administration? What is good publicadministration? Good public administration is responsive, i.e. it endeavors to producethe services that we want and to do the things that we (or political leaders on ourbehalf) have decided they should do for us. It is effective, i.e. it succeeds rather thanfails in producing what we want. It is efficient, i.e. it produces what we want at lowerrather than higher costs (to the government or to its citizens). Responsiveness,effectiveness, and efficiency are widely used as basic criteria for judging publicadministration or its performance. Responsiveness is arguably the



Page 20most basic – effectiveness and efficiency in doing something that is not responsive toour wants is worthless at best. Responsiveness is also more complex thaneffectiveness and efficiency. This is because the “we” or “public” that public officialsare supposed to be responsive to often consists not of a single collective “we” but ofmany groups that want different and often incompatible things. The questiontherefore is which group and which of its wants public officials should be or areresponsive to and in what degree. This situation suggests the necessity andimportance of equitable responsiveness to all groups as a criterion or measure ofperformance. Responsiveness, effectiveness and efficiency are all reduced orperverted, and in a particularly objectionable way, by corruption on the part of publicofficials. Although logically implied by the three basic criteria, the level of corruption iscommonly and understandably used as a separate performance criterion.Good public administration would make our lives nicer in myriad ways. We are alsoincreasingly told, almost certainly correctly, that good public administration wouldmake our country more attractive to foreign investors and our businesses morecompetitive in the global marketplace, and this would create more high-salary jobsand make us richer. Conversely, bad public administration would cost us dearly; itwould burden us and injure our prosperity, as a Chinese phrase puts it. Muchtherefore depends on administrative performance. So what factors determine it?Whatever the tasks, administrative performance (to the extent that it is controllable)depends on the two broad categories of capacity and will – that is to say, thecapacity and will of those carrying out public administration; namely, bureaucrats andpoliticians in office. This formulation simplifies the matter, of course, but not by much;it is therefore one that I find useful. Improving administrative performance thereforerequires attention to the capacity and will of public officials to be responsive(including equitably responsive), effective, efficient, and non-corrupt.Capacity depends very largely on human resources (knowledge, skills, energy, andcreativity) and technology (methods, techniques, processes, and tools such ascomputers and ICT). These resources are scarce or limited and thus limit capacity.The will to perform, i.e. to perform assigned tasks and to perform them well or aswell as capacity permits, is also scarce. To put it more pointedly, it is precarious. Thisis because of what Selznick (1966:253) calls “the recalcitrance of the [human] tools ofaction.” The humans in public administration have values and beliefs that make themwant, or will, to do other things besides their job, even though these things conflictwith job requirements or take time and energy away from performing their job. And itis very difficult or costly to completely prevent them from doing what they want,whether this is their job or otherwise.All organizations must possess or develop sufficient capacity and achieve sufficientinfluence over the will of their members in order to perform or achieve goals. How toensure that they do so? The best one-word answer is pressure, not management.Managers of organizations must, of course, do their job to ensure that otherorganizational members have sufficient capacity and will to perform,



Page 21but pressure is first needed to ensure that managers do their job. This pressure onorganizations to perform comes, primarily and indispensably, from the environment,which differs very significantly between private and public organizations. For privateorganizations, the environment is the market, supplemented by governmentregulation. For public organizations, the environment is politics – possibly, but notalways, supplemented by the market – and primarily consists of various mechanismsof public control or accountability that mobilize public pressure on them to perform.Reform consists of measures to raise capacity and will, and hence administrativeperformance, above existing levels. These possible measures cover a wide range, i.e.all measures that directly or indirectly affect capacity and will, and include changeswithin the organization as well as its environment. For public organizations, orcollectively the bureaucracy, they include, notably, changes for strengthening themechanisms of public control. It should come as no surprise that reforms forimproving administrative performance often redistribute power within the politicalsystem, as well as the benefits and costs among members of society.Besides drawing upon established ideas in the study of public administration andorganizations, the above applies to the bureaucracy the ideas and principle thatJames Madison enunciated in Federalist 51 for the larger purpose of constitutionaldesign. The basic principle is to match capacity and will with function – in Madison’sterms, to match “constitutional means and personal motives” with “the constitutionalrights of the place” (Madison 1788; see Rossiter 1961:321–2). It may be noted thatcontemporary (and heavily private-sector-oriented) management theory continues toemphasize capacity and will in its basic prescription that managers secure the righthuman resources or employees, provide them with the right means, and create theright environment (for individual employees) to induce them to perform.Past reforms: achievements and shortfallsSince 1957, when Peninsular Malaysia became independent, the Malaysianbureaucracy has more than doubled its strength to about a million at present (for alllevels of government but excluding the police and armed forces). This is in line withthe vastly expanded role of government, especially after 1970 and with the launch ofthe NEP (New Economic Policy) to eradicate poverty and increase Malay participationin the urban-centered private sector. Beginning in the mid-1960s, the government hasalso carried out a program of reform to improve bureaucratic performance (for detailssee Lim 2006a). Based on a strategy of reform crafted by two American professors(Montgomery and Esman 1966), the reforms consist of changes within thebureaucracy and focus on personnel training, the budgetary system and otheradministrative technologies for strengthening bureaucratic capacity.Such a strategy has been characterized as unbalanced by scholars of administrativereform because it neglects the strengthening of extra-bureaucratic institutions forcontrolling the bureaucracy. A balanced strategy is one that



Page 22gives equal importance to developing control institutions. Strengthening controlinstitutions (which would also control politicians in power) is clearly much lesspolitically feasible, but proponents of balance argue that strengthening bureaucraticcapacity by itself would not increase performance by much without strong controlinstitutions capable of “extracting” performance from the bureaucracy. The debate hasrapidly led to a convergence of scholarly opinion, with most agreeing that while it isusually worthwhile and more politically feasible to begin with reforms for enhancingcapacity under the unbalanced strategy, movement towards a balanced strategy isneeded for substantially realizing the benefits of capacity-enhancing reforms (Riggs1963, 1971; Esman 1967; Braibanti 1971; LaPalombara 1963, 1971). This is clearlyreflected in the justification provided by Esman (a leading defender of the unbalancedstrategy) for the strategy he designed for Malaysia (Esman 1972:8, 286–7):Yet when societies are conflict-prone, especially along communal lines, thedevelopment of these [i.e. participative-control] capabilities should be secondary intime and emphasis to the capacity of authoritative institutions to govern and controlthem. [The strategy] emphasizes the priority of strengthening the operationalcapabilities of the Malaysian administration, recognizing that more attention must alsobe devoted to building participative capabilities and processes if administration itself isto become more effective. … In Malaysia externally induced cultural transformation orpolitical change were out of the question; strengthening the administrative apparatusappeared to be feasible and to offer early and assured returns in governmentaleffectiveness; attention to participative institutions, while complementary, wassecondary in time and importance.The reforms so far undertaken under the unbalanced strategy consist largely ofpersonnel training, the improvement of administrative processes and procedures, andthe utilization of new technologies. During the 1990s, some reforms then popularizedby the New Public Management have been added. These latter reforms give moreattention to the will or motivation of bureaucrats and include merit pay (or pay-for-performance), client charters, and privatization. The system of merit pay (called NewRemuneration System) introduced in 1992 was plagued by problems of measuringperformance – in line with experience elsewhere – and was replaced in 2002 by thepresent system (Malaysian Remuneration System) based on competency as indicatedby courses or examinations. Citizen charters setting out standards of service deliverywere also introduced and have remained, along with inadequate efforts to monitorand correct failure in meeting service standards. The privatization program begun inthe early 1990s (but recently slowed) has transferred a large portion of public utilitiesand infrastructure to the private sector. However, instead of fully bringing marketcompetition to bear, it has often created private monopolies and oligopolies for whicheffective regulation remains both an acknowledged need and a continuing search.Although attribution is hazardous and not all reforms have been adequately



Page 23or properly implemented, or even clearly useful, there can be little doubt that thereform program as a whole has helped to shore up performance in the rapidlygrowing Malaysian bureaucracy. However, dissatisfaction, including among thecountry’s leaders, continues unabated and is even palpably mounting. This reflects thepressure of higher domestic and foreign expectations, as noted earlier, as well asglaring and persistent performance deficits. These deficits include the following: delaysand shortfalls in implementing development plans and projects, unsatisfactory servicedelivery to citizens and other clients, weaknesses in financial administration, unequalresponsiveness and bias along ethnic lines, and corruption.Inheriting these problems, the present prime minister, Abdullah, has made improvingthe performance of the bureaucracy and fighting corruption high priorities of hisadministration and taken some steps (to be noted later) toward strengthening publiccontrol. However, performance has not noticeably improved. Recent ratings ofMalaysia by international bodies – now the object of public attention in the country –have rapidly followed one another in highlighting administrative shortcomings andcorruption as serious dampeners of good governance and national competitiveness.The Transparency International Corruption Perceptions Index for the countrycontinues to hover close to the average score of 5.0, with the ranking slippingdownwards. And seemingly confirming widespread feelings, including among membersof his own party, that his reform efforts have lost or simply lacked steam, Abdullahhimself has admitted that results have not met expectations (New Straits Times [NST]24 November 2006).The present Chief Secretary to the Government, the country’s top civil servant, hasacknowledged that “there is a lot to do” to check corruption and to get bureaucrats todo “what they are supposed to do” (NST 5 November 2006). Few would quarrel withhis assessment – which only reiterates what has been repeatedly said by the country’stop political and bureaucratic leaders in recent years. Yes, there is a lot more to do toimprove administrative performance – but what? My answer is presented in the nextthree sections.Strengthening capacity: a continuing and broader challengeStrengthening capacity is a continuing need not only because capacity has to bereplenished in order to be maintained, but also because new environmental demandsand technologies require new or increased knowledge and skills. Meeting thechallenges of globalization requires capacity to understand and track changes in theglobal environment and to craft policy responses to overcome obstacles and exploitopportunities. Exploiting new technologies for improving performance also demandsnew skills. To cite but two examples in Malaysia, the use of computers and ICT havethe obvious requirements of new skills, while privatization requires new capacities inregulation and contract management. The government has also acknowledged theproblem of poor management in the bureaucracy and recently hired consultants (threeHarvard professors) to recommend ways to address it. The report has not been madepublic, but discussion with the consultants indicates that



Page 24the main recommendation is likely to be (à la Esman) more training, i.e. trainingtailored for those in, or heading for, managerial positions.While the government continues to invest in ICT and personnel training, there areincreasing doubts about the ability of the bureaucracy to attract and retain qualitymanpower, which is needed to realize the full benefits of training, technology andother resources. To quote Abdullah, the present prime minister: “without the bestminds conceptualizing, adapting and driving public policies, good governance wouldremain an elusive ideal” (NST 23 June 2000). Many believe or fear that thebureaucracy is filled with “second-raters” or worse. Examination of the bureaucracy’sability to attract quality personnel must consider two general factors: theattractiveness of public vis-à-vis private employment and the pool of job-seekers fromwhich the workforce is drawn.The bureaucracy has to compete for talent with the private sector. In this regard, payis widely seen as an important (although not the only) factor – and a problem for thebureaucracy. Pay in the public sector has long lagged behind the private sector formost kinds of graduate personnel (Lucas and Verry 1999:234–7). In addition, thepractice of tying pay to “status” in the bureaucracy, long defended by elite generalistadministrators benefiting from it, has produced persistent shortages in qualityprofessional and specialist skills that are highly remunerated in the private sector. Thegap between public and private sector pay for various categories of worker has to bereduced in order for the bureaucracy to compete more effectively for qualitypersonnel.The public as well as other sectors depend on the educational system, or schools anduniversities, to meet their staffing needs. Their products supply the pool of availablestaff. The palpable decline in the quality of graduates, including in English languageproficiency, has caused concern, including among foreign investors. Improving theeducational system is now widely acknowledged as another pressing need. Increasingthe number of universities and their intake is relatively easy and continues apace, butincreasing the number of graduates at the expense of quality has dangers of its own.There are already tens of thousands – present estimates range from 40,000 to 80,000– of mainly Malay unemployed (and many believe low-quality) graduates, puttingtremendous pressure on the government to absorb them into the bureaucracy. Thegovernment has recognized the need and begun efforts to improve the educationalsystem. (The educational system is largely public and is hereafter included in thebureaucracy.)Besides the two general factors just noted, public-sector personnel policies andpractices also affect capacity, and need reconsideration. Reforms so far have left themain personnel policies unchanged on grounds of political feasibility. As Esman(1972:145), the main architect of Malaysia’s reform strategy, explains (italics inoriginal):The first principle was to work within the existing structure. This had two majorconsequences: to accept the legitimacy of an elite corps of … policy officers andgeneralist administrators, superior in power and status to other groups of officials; andto acquiesce in Malay control of this elite corps through



Page 25the existing quota system. As a practical matter, the report [i.e. the proposedreforms] would not have been accepted or even seriously considered if it hadchallenged either of these arrangements.The elite corps of generalist administrators is now called the PTD (PerkhidmatanTadbir dan Diplomatik; in English, the Malaysian Administrative and DiplomaticService). While accepting the superior status and policy-advising role of the PTD,Esman clearly saw the need to strengthen its policy capacity. Towards this end, heproposed that PTD officers specialize in broad functional or policy areas. However, theproposal has not been seriously implemented by the generalist-controlled centralpersonnel agency (Esman 1972:185–95). This oversight is increasingly insupportablein today’s world. The integrative role of generalist administrators remains valuable,but some functional-area specialization within the PTD is urgently needed forstrengthening the policy capacity of the Malaysian bureaucracy.The problem of improving bureaucratic capacity and performance generally inmultiracial Malaysia has importantly to do with race, or the need to balance theinterests of various races. As Malays lag behind Chinese and Indians in the privatesector, they (i.e. Malays) are given preferential access to public sector jobs. Malaycontrol of the PTD is an important part of this policy of preferring Malays in staffingthe bureaucracy. As Malays also lag behind Chinese and Indians in educationalachievement, this policy entails the dilution of the merit system in the bureaucracy, adilution that inevitably compromises bureaucratic performance in terms ofeffectiveness and efficiency.The policy of Malay preference has not only diluted merit but also resulted in apredominantly Malay bureaucracy. This has produced effects that are increasinglyworrisome, not only (certainly not least) to the racial groups, i.e. non-Malays, notsufficiently represented in it. To reduce these negative effects, the bureaucracy has tobe made more representative of the country’s multiracial society. This can be done –and can best be done – by strengthening merit, which would also improvebureaucratic effectiveness and efficiency. The need for, and the complexities of,strengthening merit in the country’s multiracial context is examined in more detail inthe next section.Strengthening merit and representativenessAccording to the latest census in 2000, Malaysia’s population of nearly 22 million ismade up of 53.4 percent Malays, 26.0 percent Chinese, 7.7 percent Indians, 11.7percent other bumiputra, and 1.2 percent others. Malays and other native groups inthe states of Sabah and Sarawak are classified as bumiputra (literally, “princes of thesoil”) and enjoy “special rights” over the other, non-bumiputra races. Theseconstitutionally provided special rights (see Article 153 of the Federal Constitution) aremeant to help the educationally and economically disadvantaged bumiputra to catchup with the other races. They allow quotas for reserving a “reasonable” proportion ofvarious kinds of opportunities, including positions within the bureaucracy, forbumiputra.



Page 26As Means (1986:105) points out, what has been crucial are “hiring practices, notformal quotas” (these exist for some important “services” or categories of personnelbut have long been exceeded): “As more Malays have acquired education andqualifications … the natural proclivity of the [multiracial coalitional but Malay-dominated] government, particularly after the NEP, has been to fill the positions withMalays if at all possible.” The result is a rapid decline in non-bumiputra representationas shown by the following figures from the Public Service Commission: “While Chineseaccounted for 29.7 percent and Indians 9.8 percent of civil servants in the 1980s,their numbers fell to 8.2 percent and 5.2 percent respectively in 2003.” CorroboratingMeans’ observation, the Commission’s secretary explains that this “drastic drop … wasdue to the retirement of the post-Merdeka [i.e. post-independence] batch of[non-bumiputra] civil servants, who were not being replaced” (NST 26 July 2005).There can be little doubt that the preference for bumiputra, particularly Malays, is themost important factor that accounts for the presently unrepresentative Malaysianbureaucracy. It is widely believed that, in practice, preference is carried beyondrecruitment into career development and promotion – even though this is clearlyconstitutionally prohibited (see Article 136 of the Federal Constitution). The policy andpractice of preference thus greatly limits the intake of non-bumiputra applicants aswell as the promotion of non-bumiputra bureaucrats to senior positions. These effects(together most likely with lower pay in the public as compared to the private sector)have in turn dampened the number of non-bumiputra applicants. Figures released bythe Public Service Commission show that non-bumiputra applications for public servicejobs have dropped to very low levels in recent years (for 2001–4 see NST 26 July2005).Table 2.1 presents data on the racial composition of the Malaysian bureaucracy forJune 2005, the latest available figures. The simple index of representation inTable 2.1 Representation by race in the Malaysian bureaucracy 2005

Ethnicgroup

Top managementgroup

Management andprofessional group

Support group All groups

No. Percent No. Percent No. Percent No. PercentMalay 1,370 83.95 155,871 81.65535,495 75.77692,736 77.04Chinese 151 9.25 17,896 9.37 66,248 9.37 84,295 9.37Indian 83 5.08 9,777 5.12 36,194 5.12 46,054 5.12Other 23 1.41 6,156 3.23 63,649 9.01 69,828 7.77bumiputra Others 5 0.31 1,203 0.63 5,129 0.73 6,337 0.70Total 1,632 100 190,903 100706,715 100899,250 100Source: Reply by Mohd. Johari Baharum, then Parliamentary Secretary in the PrimeMinister’s Department, to a question in the Senate (Dewan Negara) on 7 December2005. It was reported in the Chinese-language daily Oriental Daily News, 8 December2005. The fi gures do not include the police and armed forces.



Page 27Table 2.2 Index of representation by race in the Malaysian bureaucracy 2005Ethnicgroup

Top managementgroup

Management and professionalgroup

Supportgroup

Allgroups

Malay 1.57 1.53 1.42 1.44Chinese 0.35 0.36 0.36 0.36Indian 0.66 0.66 0.66 0.66Other 0.12 0.27 0.77 0.66bumiputra Others 0.25 0.51 0.59 0.57Table 2.2 is a racial group’s percentage in the bureaucracy (as in Table 2.1) dividedby its percentage in the population (given at the beginning of this section). Malays aresignificantly over-represented not only in overall terms. Their representation alsoincreases with level, from the support group to the top management group. (Indeed,Malays virtually monopolize all senior administrative positions – for example, inuniversities, from deans upwards.) All other races are significantly under-represented,including other bumiputra. The tables exclude the armed forces and the police. Theseservices are even more Malay-dominated, but under-represent only thenon-bumiputra races.Malay domination of the bureaucracy reflects the goals of providing jobs to Malaysand retaining administrative power in Malay hands. However, the pursuit of thesegoals has compromised merit. Comprising merit in recruitment results in recruits oflower capacity while compromising merit in promotion lowers will and effort toperform. “There can be little doubt,” observes Esman (1972:75), “that the countrypaid a price in reduced administrative effectiveness.” The price due to lower capacityis likely to be especially significant when merit is relaxed for a majority (as opposedto a minority) group, and for pushing its over-representation to high levels.In his “Malaysia: The Way Forward” (or Vision 2020) speech, the then prime ministerMahathir (1991: paragraph 68) stresses the importance of human resourcedevelopment for Malays so that Malay personnel would not be “a millstone around thenational neck.” “What may be considered a burden now” – i.e. vastly expandededucational opportunities for Malays – “can … be the force that lightens our burdenand hastens our progress.” In other words, it is hoped that the loss of capacity wouldbe largely ameliorated as Malays make good use of the vastly expanded educationalopportunities afforded them to improve their capacity or merit for employment.However, Malay and other leaders and educational staff continue to lament that Malaystudents in schools and universities have generally not strived hard enough to matchtheir non-Malay counterparts in educational performance. Preferential entry into thebureaucracy may even be a factor that dampens motivation among Malay students.Besides significantly sacrificing effectiveness and efficiency, the highlyunrepresentative Malay-dominated bureaucracy also has other effects that are



Page 28especially serious in a multiracial society. These effects are both symbolic andsubstantive (Mosher 1968:10–14). An unrepresentative bureaucracy not onlysymbolizes unequal treatment to under-represented races; in a racially dividedsociety, it is also likely to lead to unequal substantive treatment of under-representedraces. The reasons for this are various (for a systematic formulation see Lim 2006b)and can be recast in terms of capacity and will, i.e. the capacity and will of theunrepresentative bureaucracy to be equitably responsive to all races.In terms of capacity, bureaucrats have more knowledge or understanding of theneeds and views of their own race than of other races; they are also more able toattract clients, and induce desired behavioural changes in clients from their own racethan from other races. In terms of will, bureaucrats have greater will to serve theirown race than other races. This can obviously arise from their partiality or biastoward their own race. Even if they try to be impartial, bureaucrats share more valuesand beliefs with, and have greater empathy for, members of their own race. Thiswould shape their will and lead them to behave in ways that serve their own racebetter than other races. Bureaucrats from different races also influence and moderateone another, but the moderating influence of minority-race bureaucrats on majority-race bureaucrats is likely to be weak in a bureaucracy dominated by the majorityrace.For all these reasons, the substantive benefits that a racial group receives from thebureaucracy are likely to be significantly related to its representation in thebureaucracy. This proposition has been supported by empirical research in the UnitedStates (most of it reviewed in Lim 2006b) showing that public organizations with morerepresentation of racial minorities also produce more substantive benefits for thoseminorities. It follows that an unrepresentative bureaucracy would result in unfairtreatment of under-represented races and that all races have to be fairly representedin the bureaucracy if they are to be fairly served by it. This is, of course, the centralargument for a representative bureaucracy.Similar research on the relationship between representation and substantive benefitsfor a racial group has not been carried out in Malaysia. However, there are frequentcomplaints that the Malay-dominated bureaucracy has not treated non-Malays fairly.The police, in particular, have been a major target of such complaints. More recently,public concern has also focused on national schools, where the mainly Malay headsand senior staff are widely seen to be biased against, or at least insensitive toward,non-Malay students. The declining enrolment of non-Malay students in nationalschools has serious implications for national unity. The government’s policy oferadicating poverty regardless of race has also been given a strong racial slant by thebureaucracy. A recent study pointedly notes that national development plans since themid-seventies have contained provisions for helping the increasing number of poorIndians but “there has been a noticeable absence of programmes and budgetaryresources” for this purpose (Centre for Public Policy Studies, 2006). In other words,the provisions in these development plans, which depend on programme design andfunding requests by bureaucrats for their implementation, have been largely ignoredby the Malay-dominated bureaucracyWhat makes the above especially serious is the widespread belief, not just among



Page 29non-Malays, that the main reason for these problems is that Malay bureaucrats arepartial towards their own race and biased against other races – a belief that issupported by the observations of scholars. According to Means (1972:46–8), “thebuilt-in pro-Malay bias” of the bureaucracy has led to “the erosion of publicconfidence in the public service, especially among non-Malays.” A prominent Malayacademic (Shamsul 1996; 25) notes that the Malay-dominated Malaysian bureaucracyhas “inevitably become highly ethnicised if not deeply pro-Malay.” More generally,Esman (1999:355) observes that bureaucrats in racially divided societies “aresocialized into a sense of professional obligation to members of their own communityand fear the sanctions of their community if they fail to comply.” (An irrepressiblequery: Is obligation to one’s own race “professional” for a bureaucrat?)As explained earlier, unfair treatment of under-represented races is almost inevitableand likely to be significant when the bureaucracy is dominated by one race in aracially divided society, and this is not just because of bureaucratic partiality. Theproblem becomes more serious when public control is inadequate, as is the case inMalaysia. (Inadequate public control is the subject of the next section.) Occasionalreminders by government leaders – such as that by a Malay deputy minister to themainly Malay staff in public schools and universities “to be fair” to non-Malay studentsso as not to “alienate” the younger generation of non-Malays (NST 21 October 2003)– are at best inadequate for countering built-in values and inclinations, especially inthe absence of specific machinery for investigating and correcting racial bias.A highly unrepresentative bureaucracy, therefore, has serious implications for equity inadministration and consequently for trust in the bureaucracy, government legitimacy,and national unity (i.e. race relations) in the country. After reports of blatant bias byMalay police in handling an outbreak of interracial violence in 2001 (in KampungMedan), then prime minister Mahathir has reportedly acknowledged the need for a“sufficiently representative” bureaucracy for promoting national unity (NST 20 May2001):The Government wants an increase in non-Bumiputra representation in the civilservice to promote national unity. Prime Minister Datuk Seri Dr. Mahathir Mohamadsaid today there was a need to ensure all races were sufficiently represented in alllevels in the service. “We are a bit worried that there are not too many non-Malays –that is Chinese and Indians – in the government service.” … Fair representation of theraces should not only be restricted to government service. It should be reflected inthe private sector as well.Mahathir’s call for a sufficiently (rather than a fully) representative bureaucracy in theMalaysian context makes an important point that merits brief explication. Whileequitable treatment of all races by the bureaucracy is important, it should be seen notin isolation but in the larger perspective of overall equity in society. This requirestaking into account the representation of various races in all sectors and all the waysin which representation affects interracial equity. The reality in Malaysia is that,notwithstanding considerable progress under the NEP and successive



Page 30national policies, Malays still trail behind Chinese and Indians in wealth ownership andemployment in the private sector. This is an important reason why Malays aregenerally opposed to a representative bureaucracy. The present high level of Malaypreference and over-representation in the bureaucracy needs to be reduced, but somedegree of Malay preference and over-representation is still needed in the interest ofoverall interracial equity. Malay satisfaction, including with their economic position andemployment opportunities, is crucial for national unity. A fully representativebureaucracy would reduce jobs for Malays and increase their economic disgruntlementto a level that could pose an even greater threat to na tional unity than non-Malaygrievances with an unrepre sentative bureaucracy. Thus, a sufficiently representativebureaucracy (say, with Malays comprising no more than two-thirds in each of thethree groups in the civil service), is more appropriate than a fully representative onein the Malaysian context. In the longer term and as Malay participation in the privatesector increases, non-Malay representation in the bureaucracy should also beincreased so that eventually both sectors will become closely representative of thecountry’s plural society.To sum up, the case for strengthening merit does not rest solely on its contribution toeffectiveness and efficiency. No less important in Malaysia’s multiracial society is itscontribution to making the bureaucracy more representative and hence moreequitably responsive to all races. In the present Malaysian context, however, somelimits to merit and representativeness in the bureaucracy are still needed to promoteoverall interracial equity. All this supports the goal of a sufficiently representativebureaucracy.Strengthening public control: the will to performThe country’s past reforms and persisting performance shortfalls also strongly point tothe need to strengthen public control. Public control is used here generally to refer toall forms and mechanisms of public pressure on the government, including thebureaucracy, to improve performance. To repeat, performance requires not justcapacity but also the will to perform – the will to make full or adequate use ofavailable capacity and indeed to acquire or develop more capacity if it is needed (and,as already argued, it is almost certainly needed). Public control is as important forensuring the will to perform in the public sector as the market is in the private sector,or as regulation is in the highly imperfect markets faced by privatized entities.Madison (1788; see Rossiter 1961:322) famously stated both the need and the meansof public control in Federalist 51 as follows:In framing a government that is to be administered by men over men, the greatdifficulty lies in this: you must first enable the government to control the governed;and in the next place oblige it to control itself. A dependence on the people is, nodoubt, the primary control on the government; but experience has taught mankindthe necessity of auxiliary precautions.A major problem in Malaysian public administration is insufficient will to perform



Page 31among many public officials. This has been allowed to persist through – if it is notexactly caused by – weak public control or deficiencies in the mechanisms for obligingpublic officials to control themselves. Below I provide a brief explanation of why publiccontrol is weak, and the resulting negative effects (for more discussion see Lim2002a) before stating the case and noting key means for strengthening public control.In the fifty years of independence, the bureaucracy has more than doubled its staffand greatly increased its role and powers. However, the mechanisms of public controlhave failed to keep pace. Their development was not only neglected or deferred in theunbalanced strategy of administrative reform, but the role and capacity of mostexisting mechanisms of public control have also been progressively diminished by thegovernment. The weakening of public control reflects the design of central rulingpoliticians to create a strong executive-dominated government; that is to say, one inwhich the executive is dominant in relation to both other branches of government andthe governed.Regular elections make for government dependence on the people, Madison’s primarycontrol. However, this electoral dependence has been significantly lowered by twofactors. The first is the unbalanced party system that exists in Malaysia. The rulingcoalition – now called the National Front or Barisan Nasional in Malay and led byUMNO (United Malays National Organization), the main Malay party in the country – ishighly dominant vis-à-vis other political parties and has held power at the federal level(and in most states) since independence. In the absence of any viable alternative –such as another durable coalition and this has failed to materialize despite a fewattempts – the ruling coalition faces virtually no risk of electoral defeat.The second factor is the electoral system and process. The government has used itspower to shape the electoral system and process and make its grip on power evenmore secure and immune to challenge at the polls (for details see Lim 2002b; Limand Ong 2006). This “managed” electoral system and process significantlydisadvantages the opposition in both the campaign for votes and the translation ofvotes obtained into seats won. The opposition is disadvantaged in the campaign forvotes by legal restrictions, unequal media access, the ruling coalition’s use ofgovernment resources, and inadequate and poorly enforced laws on electoralspending. The translation of votes into seats is inherently non-proportional andfavours the largest party under the first-past-the-post electoral system used in thecountry. However, in Malaysia this feature of the electoral system is furtheraccentuated by the partisan delineation of electoral constituencies. This is done inthree notable ways. First, mal-apportionment in constituency delineation greatlydisadvantages the mainly non-bumiputra urban areas and advantages the mainlybumiputra rural areas. Second is the supplemental use of gerrymandering in theincreasingly racially mixed urban centres to carve out bumiputra-majorityconstituencies that favour the ruling coalition vis-à-vis the non-bumiputra opposition.The ruling coalition is thus sheltered not only from urban voters who probably caremore about good governance than their patronage-oriented rural counterparts, butalso specifically from the opposition-inclined non-bumiputra voters concentrated inurban areas. Third, the much-improved performance of the



Page 32Malay opposition party at the expense of UMNO in the 1999 election has led to theuse of gerrymandering against the Malay opposition party in the subsequent 2003delineation exercise (Ong and Welsh 2006).The fear of being ousted from power and being at the receiving end, therefore, hasnot been an effective brake to the government’s drive to increase executive powers(for a useful summary see Crouch 1996:77–95) and also to weaken and exert itscontrol over what Madison called the auxiliary mechanisms of public control –parliament, judiciary, NGOs and the press – that also control the executive and notjust the bureaucracy. To control public officials (and others), the executive has chosento rely more on agencies that are formally and wholly under its own direction orcontrol. This strategy may simply be called executive control in contrast to publiccontrol. Thus the government has set up the Anti-Corruption Agency and PublicComplaints Bureau and placed them directly under the Prime Minister. It rejected akey proposal made in 1990 to set up an independent, non-partisan commission formonitoring policy implementation, even though the proposal came from a widelyrepresentative body (National Economic Consultative Council) set up by thegovernment to deliberate on post-NEP national policy. To assist his anti-corruptioncampaign, the present Prime Minister has increased the staff of the Anti-CorruptionAgency and set up the Integrity Institute of Malaysia, which, like the others, areunder his office and staffed by bureaucrats. In line with executive control, hecontinues to resist calls to make the Anti-Corruption Agency independent orresponsible to parliament and to set up an ombudsman in place of or in addition tothe Public Complaints Bureau. The most important recommendation of the royalcommission on the police (discussed in detail later) set up by the Prime Minister is anindependent commission to deal with public complaints against the police. At the timeof writing, or more than a year after the proposal was made, the government is stillconsidering what to do with it.Not surprisingly, Malaysia has long been regarded by scholars as a semi- or quasi-democracy (Zakaria Ahmad 1989; Case 1993) – and now also as a “flaweddemocracy” by The Economist in its 2006 index of democracy (The Economist 2007).This index gives Malaysia a rank of 81 among 167 countries, the lowest placing in thecategory of “flawed democracies”: one slip and the country would lose its democracylabel altogether. And of the six indicators in the World Bank’s Worldwide GovernanceIndicators: 1996–2005, the indicator that is similar to public control, i.e. “voice andaccountability,” is also the one on which Malaysia performs the poorest in 2005, witha percentile score of only 34.3 compared to over 60 on all the other indicators (TheWorld Bank 2007).Government leaders have always contended – in fact, even before independence (Lim1989) – that a strong government, i.e. one endowed with wide powers and subject tominimum checks in its exercise of these powers, is needed for ensuring order in theracially divided and therefore volatile plural society. One does not need to doubt thisargument or leaders’ belief in it to note that weak public control also makes the livesof ruling politicians more comfortable. More probably than not, this is also part of thereason why ruling politicians in Malaysia have weakened, instead of strengthened,public control.



Page 33However, weak public control has predictable effects. It reduces pressure onbureaucrats to perform. It also reduces pressure on political leaders to perform,including in their important task of pushing bureaucrats to perform. The will of publicofficials (politicians and bureaucrats) to perform their duties and not to performprohibited acts may remain inadequate or be lowered, as they can neglect their dutiesand commit wrongdoing with a high degree of impunity or little fear of punitiveconsequences. Weak public control therefore allows and thus encourages publicofficials to slack, i.e. to not perform their duties or perform them poorly and slowly,and to indulge, i.e. commit wrongful actions such as racial bias and corruption.There are already ample signs of slack and indulgence among public officials inMalaysia. These manifestations of inadequate will – as opposed to capacity – accountfor many of the five main performance weaknesses identified earlier (and repeatedhere): delays and shortfalls in implementing development plans and projects,unsatisfactory service delivery to citizens and other clients, weaknesses in financialadministration, unequal responsiveness and bias along ethnic lines, and corruption.Any doubts that will rather than capacity is mainly at issue should be cleared by thefollowing words of the former Chief Secretary before his retirement (NST 5 November2003):I am still receiving complaints of officers who go missing from their seats, misuseoffice vehicles, chat incessantly, don’t answer telephones, deal roughly with clientsand ask for bribes. There are also many other negative activities going on ingovernment departments that the public are not aware of, such as cheating onpunch-in cards, truancy, avoiding work, doing personal work during office hours andincessantly being on the phone.The same Chief Secretary has also questioned, but more politely, the performance ofpublic managers by noting that “what happens behind the counter determines thequality of service at the counter” (NST 3 March 2004). Evidently, many publicmanagers are not managing (perhaps practicing “management-by-default”?) and areneglecting to ensure that their subordinates perform their assigned duties. Anotherformer Chief Secretary has lamented the reluctance of department heads to takedisciplinary action against errant staff (Ahmad Sarji 1996:253–8). Figures ondisciplinary cases from the Public Service Commission for the period 1989–98 bearthis out: the cases in each of these years amounted to less than a quarter of onepercent of total bureaucrats; they also decline in absolute numbers, even though thenumber of bureaucrats and of public complaints has increased over the period(Abdullah Sanusi Ahmad et al. 2003:137). As pointed out by a senior officer of theAnti-Corruption Agency (NST 3 February 2007), departmental heads even ignorerecommendations by the Anti-Corruption Agency to take action against staff believedto be corrupt (but not brought to court because of insufficient evidence). Failure toenforce basic discipline among subordinates largely reflects inadequate will ratherthan inadequate capacity on the part of public managers. Increasing managementtraining, it may be noted, addresses the latter cause more than the former.



Page 34One should also ask why public managers are slacking, and follow the chain ofcausation right up to their political superiors or ministers. Ministers have often publiclycomplained of poor management by public managers, and of bureaucrats under-performing relative to their capacity. In response to the latest call by the umbrellapublic sector union for more staff and pay, the Human Resources Minister haspointedly replied that, notwithstanding staff shortage in certain categories, the publicsector is generally overstaffed and that “the important thing is to increaseproductivity and not increase the number of workers” (NST 18 December 2006).Ministerial heads, past or present, cannot be unaware of corruption in the customsdepartment, the road and transport department or the police, to cite some prominentand persisting examples. Corruption is not confined to bureaucrats and may even bemore serious among ruling-party politicians. Money politics is common, especially inUMNO, along with self-serving political influence on most important distributive andregulatory decisions in the government. However, ministers have not made enougheffort to overcome these problems. Most Malaysians believe that some ministers maybe guilty of more than just omission in making public managers manage, andchecking corruption within their departments, although few have been as bold as thereporter who responded to a minister’s complaint about bureaucratic corruption byasking (and dumbfounding) her “how [the government] was supposed to eradicatecorruption when leaders themselves were allegedly corrupt” (NST 27 June 2003).The Report of the Royal Commission to Enhance the Operation and Management ofthe Royal Malaysia Police (2005; 119–20) attributes police corruption and abuse ofpower to the “culture of indifference, permissiveness and tolerance for corruption” andthe “culture of impunity” in the Malaysian police. There is little reason to believe thatthe level of permissiveness and impunity differs very much for other bureaucrats orpoliticians in government office – or even for nongovernmental actors engaged ingovernment projects, such as Malay businessmen who are favoured in governmentcontracting and whose poor performance is a major source of implementationweakness. All too commonly, administrative weaknesses and scandals, including thoseexposed by the Auditor-General and the press, are either ignored – by “closing oneeye,” to use the phrase made infamous by one such recent case – or “managed” inways that leave those responsible insufficiently punished and insufficiently deterred,and the problem unsolved. While higher-level officials often remind other officials toperform duties and not to abuse power, and even warn of punitive action, these arenot enough. As a newspaper editorially reacted to the most recent instance, suchwarnings only serve as admissions that “internal control mechanisms have been lessthan adequate” if there is no follow-up action (NST 25 January 2007) – and it isaction that has been lacking. The problem is systemic: lower-level public officialsslack and indulge with impunity because higher-level officials responsible for takingcorrective action fail to do so, as they too slack, if not also indulge, with impunity.And this situation is allowed to persist because of weak public control.The clear need is to replace impunity with consequentiality, i.e. negative sanctionsagainst those who slack or indulge, in order to deter such misbehaviour



Page 35in future. This in turn needs effective public control. Effective public control is neededto reliably visit adequate negative sanctions on those who slack and indulge. This canbe done in large part by pushing various responsible others in the political-administrative system – the Prime Minister, other ministers, executive controlagencies, bureaucratic managers – to act against those who slack and indulge. Thesepoints merit amplification with an important example: the campaign againstcorruption.The campaign was launched by the present Prime Minister prior to the 2004 electionwith the arrest of two prominent persons and indications of more to come. Since theimpressive electoral victory, the Anti-Corruption Agency has been given more staffand has increased its annual number of arrests from 300-plus to 400-plus – there istalk that this is its new quota. Glaringly, however, those arrested are almost all smallrather than big fish. The government has pleaded the difficulty of obtaining evidence.It has also stressed the primary importance of values, pleaded for more time tochange values through courses and seminars, and set up a new executive agency (theIntegrity Institute of Malaysia) primarily for this purpose. However, these measuresare widely seen as inadequate. The country’s rating by Transparency International hascontinued to slide and public disappointment has set in. There is even reason todoubt whether the campaign is on the right track. Posing “the question of whetherthe motives and ethics of public servants can be intentionally changed – re-designedby politicians and/or by management,” Pollitt (2003:146) provides the followingsobering summary of available evidence: “The short answer seems to be: ‘only by aconsiderable effort, on several fronts and using different tactics, over a sustainedperiod of time.’” (Even shorter answer: it’s tough and demands too much time andeffort. Note that the difficulty of changing values also underlies the need for makingthe bureaucracy more representative.) Pollitt (2003:147) also observes: “Rascals willbe rascals, and the existence and vigorous use of disciplinary procedures are likely tobe a much more important curb on their activities than codes of ethics.” In otherwords, only punitive action, not talk or preaching of values, can replace impunity withconsequentiality and stem corruption.Why is the campaign long on preaching values and short on punitive action? Onereason is belief that efforts to change values will work. A more probable reason is thePrime Minister’s fear of losing political support, i.e. the support of other leaders in hisparty and the support that most Malay bureaucrats have long shown for UMNO. AnUMNO leader, and also “close family member,” of the Prime Minister has explained asfollows: “We can’t take on the whole system – that’s too hard” (Vatikiotis andJayasankaran 2004:22). Even more pointed is the reply of the Prime Minister himselfwhen asked about “the need to reform the political culture in UMNO to combatcorruption”: “That’s easier said than done, and it’s just not practical. The personleading the government must also be leader of the party” (Vatikiotis andJayasankaran 2004:22). Whether the Prime Minister is in a position to act against thecorrupt within his party is a matter on which judgment can differ, but surely the 2004runaway electoral victory must have bolstered his position. Has the victory alsoweakened his will? In any case, the challenge is clear:



Page 36public control must be strong enough to overcome the Prime Minister’s reluctance orto make good any deficit in his political will, including overcoming the political forcesthat account for his reluctance.There is no adequate substitute for effective public control to overcome existingdeficiencies in will and hence overcome slack and indulgence among public officials.Strengthening public control may not produce results immediately, but it would do soin much less time than it takes to change values. For top leaders, strengthening publiccontrol is also more practical or feasible than taking punitive action against politicallyimportant offenders. More effective public control would then help top leaders tocheck these offenders and even strengthen the hand of top leaders in taking actionagainst them. However, top leaders must also be willing to pay a price. More effectivepublic control would also subject top leaders to more scrutiny and pressure toperform. Their lives too would become more difficult. This is a potentially powerfuldisincentive for all power-holders to strengthen public control – even if they believe init in principle and desire its benefits. However, not to strengthen public control entailsa big price for the country, a price that threatens to get bigger with globalization andhence to rebound upon the reputation and position of political leaders.In welcome contrast to his predecessors, Prime Minister Abdullah has taken somesteps towards strengthening public control. Since assuming the post in 2002, he hasacknowledged the role of NGOs as society’s watchdogs and sentinels, to inform thegovernment of weaknesses in public administration. Under him, the press has becomepalpably freer and even more muckraking. Some changes have also been made toreinvigorate parliament. Select committees have been set up to examine pressingnational problems – these are not to be confused with the select departmentalcommittees found in some parliamentary countries for shadowing and scrutinizingparticular government departments or ministries. For the first time the deputychairman of the Public Accounts Committee has been appointed from the opposition –although, contrary to widely established parliamentary convention, the committeechairman continues to be a government member.These changes are a useful start and much more needs to be done to make publiccontrol effective. Existing mechanisms of public control need to be strengthened orallowed to play their role by removing or relaxing the executive controls on them thathave accreted over the years and that inhibit their will and capacity. The present tightlegal controls on NGOs and the press should be relaxed. Parliamentary scrutiny ofadministration should be strengthened by various reforms within parliament, includingrevising rules and practices that restrict scrutiny by opposition members (Ong 1987),requiring ministers to provide more adequate answers, introducing bipartisandepartmental committees, and appointing an opposition member to chair the PublicAccounts Committee. Secrecy in government has to be reduced, as it seriouslyhampers public control, whether inside or outside parliament. On many matters, it isdoubtful whether secrecy does anything more than protect their handling from publicscrutiny. For example, in a recent controversy, the agreements between thegovernment and companies operating toll highways and public utilities under theprivatization programme are considered secret, and access



Page 37to them is denied to NGOs questioning tariff increases (The Sun 5 February 2007).The importance of reducing secrecy is also well explained by Madison (1822; seePadover 1953:337): “… a people who mean to be their own governors must armthemselves with the power knowledge gives. A popular government without popularinformation or the means of acquiring it is but a prologue to a farce or a tragedy orperhaps both.”Scholars of public administration typically confine their attention to auxiliary controls,but Madison reminds us that public control also includes the government’s primarydependence on the people through elections. As Scott (1996:425) has noted in hisexamination of the British parliament, auxiliary controls may have to rely for theireffectiveness primarily on informing voters when they exercise their primary controlover the government. In Malaysia, public control through elections is significantlyreduced by the lopsided party system or the lack of a viable alternative to the rulingcoalition. In addition, the coalition has “managed” elections so as to further enhanceits strength or majority in parliament. Since independence, the government has heldover two-thirds of parliamentary seats with between half and two-thirds of total votes.The last election in 2004 has increased its parliamentary majority to a record 90.9percent (199 of 219 seats) on the basis of 64.3 percent of total votes. With such adominant position, is it any wonder that the government has not felt compelled to domore to curb corruption and improve administrative performance, or to strengthenauxiliary controls for these purposes?Scholars generally agree that elections in Malaysia are free but not fair, and thatreform is needed to enhance fairness (see Lim 2005; Lim and Ong 2006). Makingelections fairer would not lead to a change in government because of the lopsidedparty system. However, it would reduce the ruling coalition’s majority in parliament.Two main approaches to making elections fairer in Malaysia have been identified. Thefirst, replacing the first-past-the-post electoral system with some form of proportionalrepresentation, would lead to a much bigger reduction in the government’s majority,i.e. from 90 to about 65 percent of seats in the last election in 2004. The secondapproach, “cleaning up” or removing unfair practices in the existing first-past-the-postelectoral system, would effect a much less drastic but still non-trivial reduction in thegovernment’s majority.Making elections fairer would improve government performance, besides providingother important benefits for the political system. The government’s reduced majorityunder fairer elections would mean more opposition members and therefore moreeffective scrutiny of government in parliament. It would also make the government’stwo-thirds majority – this is required for amending the constitution and thereforeconsidered important by the government – more vulnerable to the swings in votersupport that have occurred in the past. The government would thus be under morepressure to perform in order to safeguard its two-thirds majority. It would have toaddress, and address more effectively, any problem that concerns significant numbersof voters – including urban voters who care more about corruption, poor servicedelivery, inadequate accountability and weak auxiliary control mechanisms, but whoseelectoral importance is presently significantly reduced by mal-apportionment.



Page 38Conclusion: extending reform strategy and rebalancing competing values“The new reality of globalization is that a nation’s standard of living as well as itsindependence depends upon its capacity to compete successfully in the worldeconomy. Globalization is forcing convergence around the most competitivepractices.” “Convergence does not come easily; it is fraught with conflict.” A majorreason for this conflict is because it requires the “destruction of old and oftencherished ways.” These quotations from Lodge (1995:10–11) now enjoy widespreadacceptance, including by Malaysians and their leaders. Despite past progress, mostMalaysians realize that their country’s public administrative performance is still notgood enough and has to be brought up to par to help the country compete – even assome still cherish the old ways that benefit them. Malaysian leaders have to grapplewith the conflicts involved and decide on reforms for improving administrativeperformance.From a survey of past reforms and results, this paper argues the need to re-examinethe strategy of administrative reform in order to produce the needed improvement inMalaysia’s administrative performance. The reform strategy that has been followed iscalled unbalanced as it focuses on capacity-enhancing reforms within the bureaucracyand neglects public control. Continued support for this strategy has been expressed bya former head of the government’s reform agency (Malaysian AdministrativeModernization and Management Planning Unit): “Esman’s view that the Malaysian civilservice be strengthened in all dimensions … rings as true today as when it was firstasserted” (Muhammad Rais Abdul Karim 1999:32). Strengthening capacity requiresmore reforms to be undertaken under the unbalanced strategy. The government hasto strengthen the ability of the bureaucracy to compete for desired manpower andalso the ability of educational institutions to supply desired manpower. Theunbalanced strategy also has to be broadened by relaxing the constraints imposed inthe past, especially the preference for staffing the bureaucracy with Malays and thecompromise of merit that it entails. Greater emphasis on merit would significantlyboost capacity as well as the will to perform. It would also make the bureaucracymore racially representative and thus enhance the capacity and will of bureaucrats inways that would ensure more equitable treatment of the country’s various races.The unbalanced strategy not only has to be broadened; it also has to be transcendedor balanced by strengthening the institutions of public control. As Esman, the mainarchitect of Malaysia’s unbalanced strategy of administrative reform, acknowledges inthe passage quoted earlier, strengthening public control is eventually needed “ifadministration itself is to become more effective” in actual performance. Moreeffective public control is needed for improving public officials’ will to perform. Whilethe capacity of bureaucrats needs continuous improvement and renewal, existingcapacity is significantly underused and misused because of deficiencies in will. Even ifthe capacity of public officials is not increased, effective public control can provide theneeded boost to their will to perform and thus lift the system to a significantly higherlevel of performance. Improving public control is now arguably even more importantthan improving capacity.



Page 39This paper also identifies needed reforms for improving capacity and will. However,most of these reforms are harder than those already attempted, as they require therevision of old ways or existing political-bureaucratic arrangements that are stillcherished by those benefiting from them. While all Malaysians would benefit to someextent from the improved administrative performance produced by reform, someMalaysians, i.e. those who receive specific benefits from the old ways, would have topay the price. This arises from the fact that existing political-bureaucraticarrangements reflect the pursuit of other interests and therefore goals besidesadministrative performance as earlier and conventionally defined. The weakmechanisms of public control reduce pressure not just on ruling politicians but also onbureaucrats to perform. The preference for Malays and the compromise of merit instaffing the bureaucracy – and in contracting and privatization – also dampensadministrative performance, including in terms of equitable responsiveness to theother races. Both weak public control and Malay preference importantly loweradministrative performance. However, the performance costs of Malay preferencehave to be weighed against the contribution preference makes to the important goalof narrowing interracial economic disparities in the Malaysian context. Thebureaucracy therefore should be made sufficiently, rather than fully, representativethrough greater emphasis on merit.Formulating and reformulating political-bureaucratic arrangements for conductingpublic administration in Malaysia are thus good examples of what Vickers calls “multi-valued choices” that involve “optimizing-balancing” competing values or interests. AsVickers also notes, changing conditions and requirements mean that a satisfactorysolution or balance among competing values cannot be achieved “once and for all but,like the mariner’s course, must constantly be sought anew” (Vickers 1968:115–16).The challenge facing Malaysia and her leaders is that changing conditions, bothdomestic and global, require the existing solution – and the political-bureaucraticarrangements that embody it – to be revised to give greater weight to administrativeperformance in terms of all the identified dimensions of effectiveness, efficiency,responsiveness (including equitable responsiveness) and freedom from corruption. Thebalanced reform strategy and the specific reforms suggested for improving both thecapacity and will of public officials are important for this purpose.ReferencesAbdullah Sanusi Ahmad, Norma Mansor and Abdul Kuddus Ahmad 2003, TheMalaysian Bureaucracy: Four Decades of Development. Kuala Lumpur: PearsonPrentice Hall.Ahmad Sarji 1996, The Chief Secretary to the Government, Malaysia. Petaling Jaya:Pelanduk Publications.Braibanti, R. 1971, ‘Administrative Reform in the Context of Political Growth’ in F. W.Riggs, ed. Frontiers of Development Administration: 227–46. Durham: Duke UniversityPress.Case, W. 1993, ‘Semi-Democracy in Malaysia: Withstanding the Pressures for RegimeChange.’ Pacific Affairs, 66(2): 183–205.Centre for Public Policy Studies 2006, ‘Ensuring Effective Targeting of EthnicMinorities: The Case of Low Income Malaysian Indians’ in Proposals for the NinthMalaysia Plan. Kuala Lumpur: Asian Strategy and Leadership Institute.



Page 40Crouch, H. 1996, Government and Society in Malaysia. St. Leonards: Allen and UnwinAustralia.The Economist 2007, The Economist Intelligence Unit’s Index of Democracy. Availableonline at: http://www.economist.com/media/pdf/DEMOCRACY_INDEX_2007_v3.pdf [30August 2006]Esman, M. J. 1967, ‘The Ecological Style in Comparative Administration.’ PublicAdministration Review, 27(3): 271–8.——1972, Administration and Development in Malaysia. Ithaca: Cornell UniversityPress.——1999, ‘Public Administration and Conflict Management in Plural Societies: TheCase for Representative Bureaucracy.’ Public Administration and Development, 19(4):353–66.Federal Consitution, Malaysia. Available online at: http://cofinder.richmond.edu/admin/docs/malaysia.pdf [30 August 2006].LaPalombara, J. 1963, ‘An Overview of Bureaucracy and Political Development.’ in J.LaPalombara, ed. Bureaucracy and Political Development: 3–33. Princeton: PrincetonUniversity Press.——1971, ‘Alternative Strategies for Developing Administrative Capabilities in EmergingNations’, in F.W. Riggs, ed Frontiers of Development Administration: 171–226.Durham: Duke University Press.Lim, H.H. 1989, ‘The Eve-of-Independence Constitutional Debate on FundamentalLiberties and Judicial Review: A Window on Elite Views and Constitutional Governmentin Malaysia?’ Kajian Malaysia (Journal of Malaysian Studies), 7(1 & 2): 1–37.——2002a, ‘Public Administration: The Effects of Executive Dominance’ in Francis K.W. Loh and B. T. Khoo, eds. Democracy in Malaysia: Discourses and Practices: 165–97, Richmond: Curzon Press.——2002b, ‘Electoral Politics in Malaysia: “Managing” Elections in a Plural Society’ in A.Croissant, G. Bruns and M. John, eds. Electoral Politics in Southeast and East Asia:101–48, Singapore: Friedrich Ebert Stiftung.——2005, ‘Making the System Work: The Election Commission’ in M. Puthucheary andNorani Othman, eds. Elections and Democracy in Malaysia: 249–91, Bangi: UniversitiKebangsaan Malaysia Press.——2006a, ‘Malaysia: Distilling the Lessons of Forty Years’ in K.L. Ho, ed. Re-thinkingAdministrative Reforms in Southeast Asia: 1–53, Singapore: Marshall Cavendish.——2006b, ‘Representative Bureaucracy: Rethinking Substantive Effects and ActiveRepresentation.’ Public Administration Review, 66(2): 193–204.Lim, H. H. and K. M. Ong 2006, ‘The 2004 General Election and the Electoral Processin Malaysia’ in A. Croissant and B. Martin, eds. Between Consolidation and Crisis:Elections and Democracy in Five Nations in Southeast Asia: 147–214, Berlin, LitVerlag.Lodge, G. C. 1995, Managing Globalization in an Age of Interdependence. KualaLumpur: Golden Books Centre.Lucas, R. E. B., and D. Verry 1999, Restructuring the Malaysian Economy:Development and Human Resources. New York: St. Martin’s Press.Mahathir Mohamad 1991, ‘Malaysia: The Way Forward’, speech at the inauguralmeeting of the Malaysian Business Council, 28 February, 1991, Kuala Lumpur.Reproduced as an appendix in A.B. Shamsul 1992, Malaysia’s Vision 2020: Old Ideasin a New Package, Working Paper 92–4, Monash-Melbourne Joint Project onComparative Australian-Asian Development.Means, G. P. 1972, ‘Special Rights as a Strategy for Development.’ ComparativePolitics 5(1): 29–61.——1986, ‘Ethnic Preference Policies in Malaysia.’ in N. Nevitte and C. Kennedy, eds.Ethnic Preference and Public Policy: 95–118. Boulder, Lynne Reinner.Montgomery, J., and M. J. Esman 1966, Development Administration in Malaysia:Report to the Government of Malaysia. Kuala Lumpur: Government Printer.



Page 41Mosher, F. C. 1968, Democracy and the Public Service. New York: Oxford UniversityPress.Muhammad Rais Abdul Karim 1999, ‘Introduction: Two Decades of Managing Change’in Muhammad Rais Abdul Karim, ed. Reengineering the Public Service: Leadership andChange in an Electronic Age: 15–32. Subang Jaya: Pelanduk Publications. NST (NewStraits Times), an English-language newspaper in Malaysia.Ong, K. M., and B. Welsh 2006, ‘Electoral Delimitation: A Case Study of Kedah’ in M.Puthucheary and Norani Othman, eds. Elections and Democracy in Malaysia: 316–45.Bangi: Universiti Kebangsaan Malaysia Press.Ong, M. 1987, ‘Government and Opposition in Parliament: The Rules of the Game’ inZakaria Haji Ahmad, ed. Government and Politics of Malaysia. Singapore: OxfordUniversity Press.Padover, S. K., ed. 1953, The Complete Madison. New York: Harper.Pollitt, C. 2003, The Essential Public Manager. Maidenhead: Open University Press.Report of the Royal Commission to Enhance the Operation and Management of theRoyal Malaysia Police 2005. Kuala Lumpur: Government Printer.Riggs, F. W. 1963, ‘Bureaucrats and Political Development: A Paradoxical View’ in J.LaPalombara, ed. Bureaucracy and Political Development: 120–67. Princeton:Princeton University Press.——1971, ‘Introduction’, in F.W. Riggs, ed. Frontiers of Development Administration:3–37. Durham: Duke University Press.Rossiter, C., ed. 1961, The Federalist Papers. New York: Mentor.Scott, R. 1996, ‘Ministerial Accountability.’ Public Law, Autumn: 410–26.Selznick, P. 1966, TVA and the Grass Roots: A Study in the Sociology of FormalOrganization. New York: Harper and Row.Shamsul, A. B. 1996, ‘The Construction and Transformation of a Social Identity:Malayness and Bumiputraness Re-examined.’ Journal of Asian and African Studies, no.52: 15–33. The Sun, an English-language newspaper in Malaysia. TransparencyInternational, Corruption Perception Index 2007, Berlin. Available online athttp://www.transparency.org/policy_research/surveys_indices/cpi [30 August 2006].Vatikiotis, M. and Jayasankaran, S. 2004 ‘Softly, Softly Go Reforms’, Far EasternEconomic Review, June 3: 20–23Vickers, G. 1968, ‘The Multi-Valued Choice’, Value Systems and Social Process: 112–32. New York: Basic Books.The World Bank 2007, World Governance Indicators: 1996–2005. Available online at:http://info.worldbank.org/governance/wgi/index.asp [30 August 2006].Zakaria Ahmad 1989, ‘Malaysia: Quasi-Democracy in a Divided Society’ in L. Diamond,J. J. Linz and S. M. Lipset, eds. Democracy in Developing Countries, Vol. 3: Asia: 347–81. Boulder: Lynne Reinner.



Page 423 IndiaGetting governance right for inclusive growthSantosh Mehrotra1Introduction: The current conjuncture in India’s governance systemFor a nation that has suddenly emerged on the world stage as a major regional playerand attracted much global attention as one of the fastest growing emerging marketeconomies of the world – along with Brazil, Russia, and China – one might beforgiven for believing that India has also experienced some tremendous change (oreven revolution) in its governance system for this to have happened. Nothing of thekind has in fact happened. The World Bank’s Governance Indicators tables for 1996 to2004 do not show much change in the world ranking of India in any of the sixindicators: political stability/no violence; government effectiveness; voice andaccountability; regulatory quality; rule of law; and control of corruption (World Bank2006).The one big change in governance that did happen was the end of licence-raj in theearly 1990s, with the political-bureaucratic machinery’s role in economic decision-making being reduced in key areas, especially in relation to the state’s directinvolvement in productive sectors of the economy. This change, however, is notstrictly a change in the domain of governance but of macro-economic policymanagement, which is much easier to accomplish in a federal country with strongunitary features; the more difficult change to ensure is change at state level.India, in fact, faces a grave crisis of governance – which the ordinary citizen,especially if poor, must contend with each day. The most egregious manifestation ofthe looming predicament is that some 160 of the 607 districts of the country, or wellover a quarter, spread over 12 states of central and eastern India, are affected by aMaoist (or Naxalite) insurgency. This is the result of the effective collapse of the statein these districts. These areas are particularly those where indigenous tribes, or inconstitutional terms the Scheduled Tribes, reside. The Scheduled Tribes, whoconstitute 8 percent of India’s population of 1.1 billion, have been systematicallyoppressed, and their rights over their forest land (their source of livelihood) and theminerals located on their lands exploited by outsiders without regard to their humanrights. Such is the collapse of the state in these districts that the only recourse thatpoor people have is to rely upon the Maoists to run a parallel government.Differences have also increased between states in terms of per capita incomes. TheGovernment of India is attempting to address the rural/urban divide through centralgovernment-run programmes like Bharat Nirman (or building urban



Page 43infrastructure in rural areas); their effectiveness will depend upon whether stategovernments and the system of local governance (i.e. the panchayati raj institutions,or PRIs) can make the delivery of these programmes accountable to the localcommunity. The inter-regional differences have become the most obviousmanifestation of governance failure in India, with a swathe of territories in the northand east demonstrating the weakest social indicators (and continuing Maoist violence).On the other hand, the southern and western states usually fare much better, interms of social outcomes, per capita incomes and economic growth rates as well asgovernance. But the northern and eastern states have also tended to ignoredecentralisation of functions to the PRI system, while southern states like Kerala andKarnataka have been leaders in decentralisation.At the same time, the central government’s goal, as specified in its 11th Five YearPlan document, is to ensure “faster and more inclusive growth”. The inclusive ness ingrowth is likely to come from four sources, all of which have been rightly identified inthe Approach Paper to the 11th Five Year Plan (Planning Commission 2006). The 11thFive Year Plan (2007–12) went into effect in April 2007, and its Approach Paperappropriately identifies the following priorities for the Plan invest ment: health,education, agriculture and infrastructure. The Planning Commission not only preparesthe policies and programmes for the next five years in these sectors, but alsoallocates resources for capital investment in pursuance of those programmes, and cancontribute to the design of these programmes.However, the effectiveness of programmes in three of the four core areas identifiedwill be crucial for inclusiveness in growth – education, health and agriculture. Successis critically dependent upon much greater accountability in these areas to the peopleto ensure which functionaries should be made accountable to locally elected bodies.Perhaps the most important issue around decentralisation arises around distributiveconflicts, particularly at the local level.2 Some argue that the problem of capturemight be more serious at the local level (Bardhan and Mookerjee 1999, 2000;Bardhan 2002). Most developing countries are characterised by considerable socialstratification, income inequality, and disparity in ownership of assets. Under thecircumstances, local elites often “capture” the local governing agencies, leading tosevere problems and leaving the poor very vulnerable. Also, collusions among the elitegroups at the local level may be easier than at the national level.3 For instance, thereare fixed costs for organising resistance groups or lobbies, and the poor may be moreunorganised at the local level than at the national level, where they can pool theirorganising capacities. For instance, in socially homogeneous Japan or Korea,community-level organisations may work better in enforcing cooperative norms.Similarly, Putnam (1993) suggested, on the basis of regional differences in Italy, that“horizontal” social networks (i.e. among people of similar status and power) are moreeffective in generating trust and norms of reciprocity than “vertical” ones.Nevertheless, if there is extensive evidence of state capture at national or provinciallevels, as in India, there is also no evidence that the extent of local level elite captureis greater than at the provincial or state level. Hence, the argument that there will belocal elite capture in the case of decentralized



Page 44governance is a non-starter. The real clincher in favour of decentralisation underappropriate safeguards is that at the local level it is easier to enforce accountability ofpetty officials and elected functionaries than at the provincial or national level. It canbe argued that only when decentralisation takes the form of deep democraticdecentralisation, accountability of the bureaucrats responsible for delivering socialservices will increase.Section one briefly presents some of the fiscal and political imperatives of governancereform that state governments face in India, and argues that inclusiveness in thegrowth process in the 11th Plan period (2007–12) will crucially depend upon anemphasis on health and education. However, to realize the laudable goals in healthand education, areas in which India lags seriously behind China as well as other fast-growing East and Southeast Asian economies, requires serious governance reform.Section two discusses the organizational context of state provision of these socialservices, and suggests that the nature of the post-colonial state, including itscentralized character, is a factor underlying state failure in delivery of social services,including in India. Section three argues that while decentralisation has been globallypromoted over the 1990s, a certain kind of decentralisation alone will enable the stateto deliver basic services successfully: deep democratic decentralisation. Section fourpresents a brief case study of how decentralisation in two Indian provinces enhancesthe effectiveness of service delivery in the education sector and could successfully beapplied to the rest of the state. Section five suggests that this model is only partiallyin place in India. Section six concludes this paper.Section one: The imperative to reform governance at central, state and locallevelsThe inclusiveness imperativeAlthough a 10th Plan agenda for governance reform remained unfulfilled, theimperative to carry out reforms has not abated. If anything, it has become moreintense. We would argue that the goal of inclusiveness in the 11th Plan – “Towardsfaster and more inclusive growth” is the title of the 11th Plan Approach Paper and the11th Plan document – may be seriously compromised in the absence of governancereform. Health, education, and agriculture – which are meant to be the crucial areasof the 11th Plan in its pursuit of inclusiveness – require by their very nature a regularinterface between state functionaries and the citizenry. Even though agriculture islargely in private hands, the revival of agricultural extension will have to be based onthe state expanding its role. Public health (though not necessarily tertiary level,curative care) is dependent upon the rejuvenation of the public health system.Similarly, the vast majority of school-age children are still in government schools,especially in rural areas. Although the private system is expanding because of thepublic school system’s poor performance, the quality of the government school systemwill have to improve over the course of the 11th Plan if inclusiveness stands a chance.



Page 45The fiscal imperativeAn important role of the central government derives from the fact that stategovernments taken together have a fiscal deficit amounting to three to four percent ofthe country’s GDP. The central government, too, has a similar fiscal deficit, but itgenerates two-thirds of the country’s tax revenues.The combined fiscal deficit of the centre and states is such that Parliament legislateda fiscal constraint act in 2004, which commits the central government (and jointly thestates) to reduce the fiscal deficit. The Fiscal Responsibility and Budget ManagementAct requires the central and state governments to bring their fiscal deficit down to sixpercent of GDP by 2008–9 and the revenue deficit (i.e. the recurrent budget deficit)down to zero. At the same time, the 11th Plan Approach Paper makes a commitmentto the nation that the resources for health, education, agriculture and infrastructure –which are in dire need of additional public investment – must increase. Both healthand education are recurrent cost-intensive sectors, which will mean that the revenuedeficit reduction goal has to be achieved under severe constraints. Additional teachershave to be hired, as do additional para-medical staff, in order to meet the programmegoals in education (Sarva Shiksha Abhiyan) and public health (National Health RuralMission), both of which are flagship programmes of the central government, andcentral to the 11th Plan strategy.Therefore, the Finance Ministry at the centre and the states will have to control non-Plan expenditure (most of which is of a recurrent nature) and revenue expenditurewhich is not related to these 11th Plan priority sectors. Besides, better administrationand improved deployment of existing resources is a sine qua non of the simultaneousachievement of the dual imperatives of fiscal discipline and increased publicinvestment of the 11th Plan equitable growth strategy. All state governments –regardless of political affiliation – will need to commit to the dual objectives of fiscaldiscipline and increased public investment for health, education and infrastructure.Without this commitment, the objective of getting governance right for humandevelopment will be meaningless.The political imperative to reform governance at the state levelThe political process and major political actors in the states stand to gain hugely fromgood performance on human development outcomes by avoiding anti-incumbency atelection time. The political history of elections in India in the last 15 yearsdemonstrates unmistakably that non-performing governments face defeat at the nextelection – whether it is at the state level or the centre. Improving governance withthe objective of enhancing human development, therefore, will be a win-win situationfor all – the poor and the political leaders.In addition, the stability of governments over a five-year period, and even over a 10-year period, will enhance the international credibility of the respective stategovernment, therefore attracting greater Foreign Direct Investment relative tocompetitor states, and possibly also Externally Assisted Projects and thereby setting inmotion a virtuous cycle of economic growth and human development.



Page 46Section two: The most important governance reform to underpininclusiveness in development – deep democratic decentralisationIf India is to achieve the UN Millennium Development Goals (a list of which can befound in Mehrotra and Delamonica, 2007, chapter 1) by the 2015 target date, the pastfailures of state provision of social services will have to be addressed. In education,public health and water services there have been systematic problems of stateprovision – poor quality, low performance, limited responsiveness and weakaccountability. As a result, and in response to these problems of state failure, therehas been a recent move towards promoting the private sector and privatisation ofsocial services. However, this paper argues that there are recent developments indeep democratic decentralisation – all occurring in the 1990s – that provide evidencethat it is possible to devise effective policy responses to these problems of statefailure. The problems of state provision of basic services do not result from financialresource constraints and allocative inefficiencies alone, but also from the bureaucraticand organisational context of state provision (Mehrotra and Delamonica 2007). Thereis a considerable literature related to state failure, under the rubric of “new politicaleconomy” also referred to as the literature on rational choice and rent-seeking.4Regardless of the weaknesses of the state provision of social services and theconsequent growth of private sector financing and provision, the state will remain themost important provider of basic social services. Here we concentrate on the state asfinancier and provider of social services, and what kind of governance reforms arenecessary to reduce state failure in delivery of basic social services.In India, as in most developing countries, the state delivers development services in atop-down, bureaucratic manner through sectoral line ministries, down to the locallevel. But this manner of service delivery defeats one of the greatest sources oftechnical efficiency in the utilisation of resources – the synergy of interventions in thevarious social sectors. Without the state making a conscious effort to ensure synergybetween interventions in the spheres of health, education, water and sanitation,reproductive health, and nutrition within a geographic location, these latent synergiesmay not be realised. In turn, the failure of the state in this regard may leave thedelivery of such services to the vagaries of the market, and to the uncertainties ofdelivery by the private sector, which is likely to result in coordination failure – andhence the non-realization of these latent synergies. But the state is incapable ofdelivering these services effectively as long as it operates vertically. Inter-sectoralaction is best triggered through “voice” at the local level, with village-level planning. Ifa collective voice at the local level puts pressure on local-level functionaries torespond to local needs and demands, instead of delivering services merely based onresource allocation determined at a higher, bureaucratic and more disconnected levelof decision-making, two benefits can result: synergy between interventions acrosssectors and the effective delivery of individual public services.Elsewhere we have presented evidence from countries where this is happening(Mehrotra 2006; Tendler 1998); these are not “projects,” but whole territories,provinces or cities affecting the lives of millions of people. In other words, there



Page 47is growing evidence that these reforms involving deep democratic decentralisationhave been shown to “go to scale” and improve social service delivery, with tangibleresults in terms of improved school enrolments, better access to water and sanitation,and reduced infant/child mortality at the local level.Decentralisation and social service deliveryReforms within bureaucracy over the years have faced numerous design andimplementation problems. Staff at health centres do not turn up regularly, andteacher absenteeism is rampant in India, especially in the northern and easternstates. Incentive reforms run up against the civil service norms of guaranteedemployment, and salaries which are unrelated to merit or performance. Enhancedmonitoring has been difficult to implement, partly on account of political problems andthe tendency of hierarchical superiors to prevent punitive action. Centralizedgovernment, in which accountability is largely vertical and not horizontal (radiating outto the community), provides the environment in which such practices can flourish.The current post-colonial state structure in most low-income countries is highlycentralised, since it was inherited from colonial administrations. A centralised structureof administration served the purposes of distinctness for its personnel, and theobjective of surplus extraction (Dutt 1903). The post-colonial state inherited thefunctions of the colonial administration in Asia and Africa, but added developmentalfunctions to the hitherto minimal ones associated with surplus extraction. However, inthe best-case scenario, the accountability of the developmental state has been givenmainly to parliament.It is not as though the developmental state is not currently “accountable.” To a wholeseries of networks (the capitalist, the landed gentry, the labour aristocracy) wellrepresented in parliaments, the two key sets of personnel – bureaucrats and ministers– are to some extent already accountable. So the state structure is indeed embeddedin networks within society. This story largely applies to much of South Asia as well asto sub-Saharan Africa, naturally with certain variations on this broad theme (on SouthAsia, see Bardhan 1984; Wade 1985, 1989; on Africa, see Sender and Smith 1986;Bates 1988).5 But the kind of accountability that would interest poor citizens is that atthe community level of the lowest-level functionaries of the state – both electedfunctionaries (i.e. politicians), assuming local officials are elected (as in India since1993) not appointed from above, as well as civil servants (e.g. nurse/midwife, schoolteacher, water engineer). Without that accountability, services cannot be deliveredeffectively. Nor can the synergy of interventions in health, education, nutrition, andwater and sanitation be realised without that accountability.Meanwhile, intermediaries exploit the distance between the centre and the villagewhere the school (or health centre) is located, to foil the well-intentionedgovernmental objectives of the centre. Hierarchical control of functionaries is rarelyeffective under such circumstances (Paul 1992). In fact, without the state “enabling”collective voice and action, which emerge as a counterweight to the intermediaries,the delivery of services, and hence the functionings (in the sense



Page 48that Sen [2000] discusses them), cannot be realised. Since the poor have limitedchoice of “exit” (Hirschman 1970), “voice” alone works.Development is generally accompanied by decentralisation. It is well known that percapita income of a country is highly correlated with quality of government, measuredin whatever way, and is also strongly correlated with decentralisation. The abovediscussion suggests that decentralisation of social services is an appropriate responseto the problems of state failure.By the beginning of the 1990s there was increasing recognition globally that, sincehierarchical control was not successful in ensuring delivery of services, at the veryleast administrative decentralisation was needed. In addition, the 1990s had been thedecade of the spread of democracy in the developing world. It was also the decade ofthe spread of decentralisation. For instance, in Latin America, with the exception of afew small countries, virtually all legislative and executive authorities are now elected in13,000 units of local government. Similarly, India, Indonesia, and the Philippinesintroduced large-scale decentralisation efforts in the 1990s.Unfortunately, this effort at decentralisation in many countries often coincided with aninternational effort by powerful donors to promote decentralisation. The internationalfinancial institutions launched their “roll back the state” campaign after the fiscaldeficits of overly stretched developmental states grew to unmanageable levels. Thisnotion of the state is keenly informed by the literature on government failure (e.g. theWorld Development Report, see World Bank, 1997). Government failure, in this view,had characterized the pre-1980 state in most developing societies. However, statestructures which were inherited from the colonial state – created for entirely differentobjectives – were bound to suffer from “government failure.” The mere imposition ofWestminster-style parliaments in new states was not going to transform structuresmeant for surplus extraction and law-and-order maintenance into democratic forms offunctioning – least of all in largely illiterate societies with few institutions meant toenable the articulation of “voice” on a regular basis.Not surprisingly, decentralisation introduced in this strategy, and the institutionalenvironment recommended by international financial institutions in a context ofunresponsive, centralized states inherited from colonial times, continues to face manyproblems. While decentralisation spread rapidly during the 1990s, the experience withit has not been as notable for successes as one might have expected. Tanzi (1995)argues that local bureaucrats may be poorly trained and thus inefficient in deliveringpublic goods and services. This criticism recognizes a real issue, the lack of training,which hinders effective decentralisation. However, training can be imparted easily ifthere is a policy mechanism incorporated at a higher level of government.We have provided evidence from Latin America, India and sub-Saharan Africa (asnoted above in Mehrotra 2006) that certain types of decentralisation are indeed moreeffective than those practiced widely in the 1990s. Here we merely suggest, on thebasis of the emerging evidence on decentralisation, what such an effective modelmight be from the perspective of effective social service delivery.



Page 49Section three: A model of decentralisation that worksThere is increasing evidence now that where the centre acts to enable the articulationof voice by the local community, the functionaries of the state tend to respondpositively to such local-level pressure. There is a three-way dynamic between thestate government (central, or provincial government in a federal country), the localauthority, and civil society that ensures effective service delivery. We call this deepdemocratic decentralisation.Basic social services closest to the needs of the poorest people – primary healthcentres, schools, water services – are usually managed by bureaucrats andgovernment employees who report to their superiors within the vertical line ministries.Such government employees rarely feel a strong sense of accountability to theneighbourhoods they administer.Making states accountable with respect to delivery of basic services, we would argue,requires:1 a functioning state (not a weak, and certainly not a “failed,” one), and effectivestate capacity (both at central and local levels);2 empowered local authority to which functions, functionaries and finance have beendevolved by the central authorities; and3 “voice” articulated on a collective basis by civil society, through institutions enabledby the state.However, the mere existence of the three does not amount to an effective instrumentfor service delivery. The relationships between these three levels are crucial: localauthorities must experience pressure both from above (for accountability to nationalgovernments) and from below (for service delivery to local citizens). Hence successfuldecentralisation requires establishing a three-way dynamic among local governments,civil society and an effective central government (Tendler 1997; Mehrotra 2006) inorder to ensure effective delivery of services.Without effective state authority, the central government cannot devolve power tolocal authority. Decentralisation assumes coordination between levels of governmentand requires more regulation – not less – to guarantee basic transparency,accountability and representation. The state has to oversee, regulate, and if necessarysanction local authorities so that poor people really benefit from political reform.Effective decentralisation also needs local authorities to which finance, functions andfunctionaries have been devolved. Responsibilities for delivering social services have tobe devolved to local authorities through legislative or constitutional means thatformally transfer control over functions as well as functionaries. But functionariescannot perform their functions without adequate finance. It remains a question as towhether local authorities should only receive grants from the centre, or also have alocal tax-raising authority, but in either case, the adequacy of finance will determinetheir success.Finally, when creating the local authorities, the central government should also createinstitutional mechanisms to ensure that the voice of the citizenry can



Page 50be heard through formal mechanisms by the local authorities. Of course, the level oforganization of the citizenry will be a crucial determinant of how those mechanismsare actually used. In the state of West Bengal, India, poverty declined sharply in the1980s where local authorities (panchayats) have been in existence long before theywere mandated by a constitutional amendment in every state in 1992 (Crook andSverisson 2001). Under Operation Barga, the panchayats helped improve agriculturaltechnology and reform land tenancy. They also helped register 1.4 millionsharecroppers, a near-revolutionary act in a country where in the majority of poornorthern states sharecroppers have few written contracts or rights, and are usually atthe mercy of powerful landlords.Similarly, since the late 1980s Mazdoor Kisan Shakti Sangathan (MKSS, or theOrganization for Workers and Peasants Power) in Rajasthan, India, has beencampaigning for the Right to Information Act. MKSS organizes public hearings toexamine official information – detailed accounts derived from official spending records– and evaluate its validity. It uses these “social audits” to promote democraticfunctioning at the level of the village. Such articulation of voice is made effectivewhen the local authority is empowered to remedy the wrongs done by governmentemployees.Section four: Elementary education: the synergy of collective voice and stateaction in two poor Indian statesIndia has a federal structure with two tiers of government, the centre and the states.However, since 1993 India has had a fairly successful decentralisation to localgovernments, which had been moribund since the 1950s. The political commitmentwith which decentralisation has been implemented in the different states has variedenormously. However, two states have enabled the “voice” of the people to bearticulated: Madhya Pradesh (MP) and Rajasthan. In sectors like school education,where functions have been effectively transferred to the local authority, the resultshave been remarkable. MP was the first state to put the newly resurrected system oflocal government, the panchayati raj institutions (PRI) into effect (after the enablingconstitutional amendment was passed by the Indian parliament in 1993).6 A workingPRI system was in place in MP in 1994. It is a facilitating structure for directcommunity action. The government converted selected programmes, two of whichwere primary education and basic health, into a mission. We discuss each briefly inturn.MP and Rajasthan are two of the most populous states in India, and along with fourother northern states7 account for three-fourths of the children who were out ofschool in the country during the mid-1990s. However, in the 2001 Census of India,MP showed an increase of 20 percentage points in its literacy rate (from 44.2 to 64percent), the highest increase of any state during the period of 1991–2001. Clearly,the two governments (MP and Rajasthan) were doing several things right. What werethey?Of the three prerequisites of successful democratic decentralisation – an effectivestate authority, an empowered local body to which functions, functionaries and



Page 51finance have been transferred by the state authority, and voice of the citizenry – notall were present in equal degree in these two states. In the federation of India,education is a concurrent (central and state) subject, as opposed to a federal (orcentral) government subject according to the Indian constitution. There is little doubtabout the competence of the state authorities, and the first prerequisite probablyexists in most of the state governments of the country. Despite the fact that theconstitutional amendment calls upon state governments to transfer functionaries tothe local bodies, most state governments have been loath to do so. Finance too hasnot in most cases been transferred for the school system to the local authority,8 butas we shall see below, for the purposes of creating new schools, funds were indeedtransferred to the local body to respond to community demand for a school. However,despite such flawed decentralisation, there has been remarkable change because ofthe three-way dynamic between the state government, transfer of functions to localauthorities, and the state creating mechanisms for the articulation of the voice of thecitizenry. In other words, even a weak form of deep democratic decentralisation canimprove governance, and produce tangible results in terms of service delivery in basicservices.In Madhya Pradesh, instead of academic institutions or the Central StatisticalOrganisation conducting a sample survey, the resulting democratic decentralisationopened an opportunity to undertake a door-to-door survey in 1996 by teachers andelected people’s representatives to discover the names of children in and out ofschool. Ironically, this survey was carried out as part of a centrally financed primaryeducation programme (DPEP)9 – which drives home the point about the differencebetween taking enabling action by the state, and empowering the community. Whilemost states in India took the enabling action of creating the PRI system, none took asmuch empowering action as Kerala and MP.The village council (panchayat) leadership was seen by the MP state government as akey player. There were three differences between past practice on data collectionregarding schooling and the Lok Sampark Abhiyaan (or Campaign for MassMobilization for Schooling). First, instead of using the schoolteacher for datacollection, the responsibility was widened to a local group including the local villagecouncil representatives and literacy activists. Second, the idea was not just to collectinformation on which children were in school (from school statistics), but whichchildren (5–14 year olds) from the village were not in school. Third, the objective insurveying households and children was not statistics collection, but to lead themotivational campaign to persuade parents to send their children to school(Gopalakrishnan and Sharma, 1999). It was, in other words, intended to consolidatecommunity management of the primary education system in the state.10A remarkable conclusion – with significant policy implications – emerged from theparticipatory survey. It is well known that government school-based statistics ofenrolment in India are grossly exaggerated, showing inflated enrolment. However,most “out of school” children contacted through the survey described themselves as“unenrolled” and not “dropped out.” The policy implication was that, in addition to theproblem of dropout (which, though not non-existent, might be much smaller thanbelieved hitherto), the major problem was that children had never gone to



Page 52school. The participatory survey revealed that only 70 percent of habitations in MPhad access to a primary school.In other words, access to schools itself is a problem, despite past claims that thenorm of the provision of one school within a one kilometre distance had been met for95 percent of India’s children. The policy response of the state government was tointroduce a programme (from January 1997) to guarantee primary schools to allhamlets – not just all villages (given that a village consists of a number of discretehamlets). Under the Education Guarantee Scheme (EGS), if the parents of 40 childrenin a locality (or 25 in a tribal area) seek a school for their children, routed through thevillage panchayat (elected council), the state government is committed to provide,within 90 days, a lower-paid teacher’s salary for the purpose.11 The village panchayatcan appoint the teacher from within the community, and it must also makearrangements for space where the children can organise into classes. The stategovernment will transfer funds to the village council to enable it to pay the salary ofthe new teacher recruited from within the community.The results have been remarkable. While 80,000 schools had opened in the first fiftyyears since independence in MP as part of the regular government primary schoolsystem, 30,000 new schools were created within three years of the scheme’sannouncement (after January 1997). The EGS school are formal schools, and part ofthe formal school system. What is particularly important is that it led to a hugeincrease in enrolment of tribal children – the very children who had some of thelowest enrolment rates among vulnerable groups. It also led to a larger than averageincrease in girls’ enrolment.The features of the EGS scheme offer profound lessons for other similar situationsaround the world. The expansion of schools and enrolment was the outcome of amutually dependent action by the state government, local government and thecommunity. The community’s demand (“collective voice”) for a school is the initialpremise of government action. Even the provision of a school is a reciprocal actionwhereby the community recommends the teacher from among its local people, thestate government remunerating and training her, the community providing space forthe centre, and the state government providing educational and other contingencymaterials.It is not suggested here that access to schooling alone should be a criterion forjudging the effectiveness of an education policy; however, reasons of space preventus from examining the challenge the increased enrolment level poses for thegovernment’s teacher training infrastructure (i.e. the District Institutes of Educationand Training, the Block and Cluster Resource Centres, on which see Mehrotra 2006;Mehrotra et al. 2005; Government of MP, HDR, 2002). But in the absence of theinterventions resulting from a policy of deep democratic decentralisation, millions ofchildren – especially tribal children and girls – did not even have access.Other actions in education (and health)12 have also been enabled by the democraticdecentralisation put in place. Following the success of EGS, a similar model was triedin adult literacy in MP starting in 2000. People who were nonliterate would cometogether as a Padhna-Badhna Samiti (or a study circle); they



Page 53would choose a local educated person to be their teacher; they would register at thenearest local council or government financed Cluster Resource Centre; the governmentwould undertake an external evaluation of the learners; and based on the number ofpeople who cleared the examination the teacher would be given a stipend. The bigdifference with the central government’s National Literacy Mission that had beenrunning for over a decade with precious little impact was the addition of a newstipend for the teacher based on the learners who passed the examination. This was amajor factor behind the remarkable increase in MP’s literacy rate recorded in theCensus of 2001.Like MP, another state that made remarkable strides during the 1990s is thenorthwestern state of Rajasthan. The processes that led to this achievement inRajasthan are rather similar to those in MP. All the measures involved an interactionbetween the state government, the local government and the community (Mehrotra2006).What Rajasthan and Madhya Pradesh demonstrate is that, despite being two of thepoorest and most educationally backward states of low-income India, and having theworst social indicators in the country, it is possible to transform schooling for thepoor, provided that the local government functionaries are mobilised in a participatorymanner, and government structures, both state and local-level ones, are made torespond to collective pressure from the people.Section five: The unfinished agenda of governance reform in India: Achievingdeep democratic decentralisationIndia resurrected its system of local governance and decentralisation in constitutionalamendments of 1993 (the 73rd and 74th Amendments), after a hiatus of nearly 40years. There have been many positive changes as a result – as the previous sectionsuggested. Until 1993 India effectively had essentially two levels of government: thecentre and the states, each with their parliamentary form of government. The thirdtier of government, at district level, was essentially an arm of the state government,and was not elected. The 1993 constitutional amendment made three levels of electedgovernment mandatory in every one of the 619 districts of India: at district, block andvillage level (and in urban areas in towns and cities it created elected municipalities).One-third of all seats at each level of local government was reserved for womencandidates – although such reservation does not exist for either the state legislativeassemblies or the national Parliament (the Parliament has endlessly debated in vainfor over five years such a reservation for women). This has brought thousands ofwomen into the political decision-making process within a decade throughout thecountry, a remarkable change in a country notorious for its gender discrimination atall levels of society.Unfortunately, however, most state governments have not been willing to transferfunds and functionaries to the local levels of government – undermining the wholepurpose of the reform. Twenty-nine subjects in the state list of the Indian Constitutionwere required to be transferred to the PRIs, many of which have indeed beentransferred by most state governments to the PRIs. But funds, and the



Page 54officials responsible for performing those functions, have not been transferred. Inother words, the accountability of officials for those subjects still remains with theirvertical line ministry in the state capital, rather than with their local clients. Teacherscan continue not to teach if they so wish, doctors and nurses can fail to turn up atprimary health centres without suffering any consequences, and the system ofcorruption survives with little change.PRIs accounted for barely 4 percent of total government expenditure of all levels ofgovernment (centre, state and PRIs) combined. There is barely any change since the1993 Constitutional Amendment and the constitution of the PRIs.A fundamental reason for the unwillingness of the state-level politicians andbureaucracy to transfer funds and functionaries derives from the fact that stategovernments have been bankrupt and face serious fiscal deficits, and they havelimited funds that they can transfer. However, on account of rapid economic growthand the buoyancy of state revenues, this situation is much better now than since theearlier part of the current decade. Nevertheless, in the poorer states, the fiscalconstraints still prevail.That by itself should be a sufficient reason not to transfer funds, and certainly shouldexplain why the functionaries have not yet been transferred to the PRIs. A moreimportant reason lies in political economy: state-level politicians do not want to losetheir powers of patronage to local-level politicians. And if funds are limited, onaccount of the fiscal deficit, the willingness to commit untied funds to the PRIs isfurther undermined. Meanwhile, the dysfunctional public health and public schoolsystem struggle along much as before.The tragedy is that the federal structure of the governance system in the countryprevents the central government from exercising any sanctions against stategovernments when they fail to fulfil their obligations under the constitutionalamendments. Even though the central government transfers significant developmentfunds to state governments under the Five Year Plans, this transfer is seen as anentitlement by state authorities, and cannot really be used to trigger reforms. Thefurther tragedy is that the system of transferring funds to state governments worksthrough the central government line ministries, which often have programmes of theirown in the health and education fields (as well as in agriculture, rural development,sanitation, water, etc.). Central line ministries are not exactly enthusiastic to transferfunds and control of these programmes to the PRIs either; so there is almost aconsonance of interests between state and central government line authorities to notfulfil the obligations under the constitutional amendments of 1993. The single mostimportant governance reform that could put ordinary people in villages around thecountry in control of their affairs remains in abeyance as a result. Unless this changeoccurs, the goal of inclusiveness in growth – the declared goal in the 11th Five YearPlan – will be undermined.Section six: Concluding remarksThis chapter has suggested that deep democratic decentralisation – characterized by afunctioning/strong central state, empowered local authorities, and “voice” – is an



Page 55ideal governance structure that states should aspire to. However, reasons of politicaleconomy may well lead to imperfections and distortions along the way. The Indianfederal structure, in particular, leaves enough leeway for state governments toundermine the decentralisation objective in practice. However, state governments thathave been bolder have shown that progress towards deep democratic decentralisationis possible, even under conditions of fiscal constraints.Notes1 Senior Adviser, Planning Commission, Government of India.2 See Bardhan and Mookerjee (1999, 2000) for a theoretical framework for appraisingthe trade-offs involved in delegating authority to a central bureaucracy – as opposedto an elected local government – for delivery of public services, from the viewpoint oftargeting and cost-effectiveness of public spending programmes in developingcountries.3 Thus, policy-making at the national level may represent greater compromise amongthe policy platforms of different parties.4 See Bates (1988); for a critique see Grindle and Thomas (1991) and Streeten(1993).5 The story is more complicated in much of East Asia (Khan 2001) and Latin America(Ames 2001).6 In contrast, Bihar (which has had the worst educational and social indicators in thecountry, comparable to those in much of sub-Saharan Africa) was the only statewhich had not had its PRI elections until 2001. Jharkhand, which emerged after thebifurcation of Bihar in 2000, has still not had any PRI elections – the only state in thecountry not to do so.7 Bihar, Uttar Pradesh, West Bengal and Andhra Pradesh are the other four. Thesestates (plus Assam and Tamil Nadu) are the subject of two studies on the financing ofelementary education (Mehrotra et al. 2005; Mehrotra 2006a).8 For instance, as Rao (2000) notes, barely 4 percent of total governmentexpenditure by central, state and local governments is undertaken in India by localgovernments.9 The District Primary Education Programme (DPEP) has been run on a decentralisedand participatory basis by state-level bodies, but with many of the top-down elementsbeing retained (Bashir and Ayyar 2001). MP also had the largest number of districtscovered by DPEP of all Indian states, and hence received a considerable proportion oftotal DPEP funds disbursed by the central government.10 One outcome of the survey was the development of a Village Education Register asa basic record of educational statistics of each village, to be maintained in two copiesat the village panchayat and the school. The survey was also used as a basis ofcohort monitoring for completion of primary schooling.11 One of the main reasons for the success of the EGS is its cost-effectiveness. Inregular (non-EGS) schools, teacher salaries account for over 90 percent of costs at theprimary level. Schoolteachers in regular schools are strongly unionised and animportant political force, and receive salaries which are high relative to per capitaincome (Kingdon 1994). However, EGS school teachers are paid only a third or less ofwhat regular schoolteachers are paid.12 The management of public hospitals has been transformed in MP – a remarkabledevelopment for one of the poorest states of India – by Rogi Kalyan Samitis (PatientsWelfare Committees), which have been helping hospital management since 2000.They are permitted to levy a modest user fee that can be retained at the hospitallevel to improve delivery of services. Over 770 public hospitals are managed by suchcommittees. Similarly, water management committees have helped since 2001 tocreate one million rainwater harvesting structures (see Government of MP 2002).



Page 56ReferencesAmes, B. 2001, ‘The Politics of Public Spending in Latin America.’ American Journal ofPolitical Science, vol. 21, issue 1 (February): 149–76.Bardhan, Pranab 1984, The Political Economy of Development in India. Oxford: BasilBlackwell.Bardhan, Pranab 2002, ‘Decentralization of Governance and Development.’ Journal ofEconomic Perspectives, 16(4): 185–205.Bardhan, Pranab and Mookerjee, D. 1999, Relative Capture of Local and CentralGovernments: An Essay in the Political Economy of Decentralization, Department ofEconomics, Berkeley: University of California.Bardhan, Pranab and Mookerjee, D. 2000, ‘Capture and governance at local andnational levels.’ Department of Economics, Berkeley: University of California. Bashir, S.and Ayyar, R. V. 2001, ‘District Primary Education Programme.’ Encyclopaedia ofIndian Education. New Delhi: National Council of Educational Research and Training.Bates, Robert H. 1981, Markets and States in Tropical Africa. Berkeley: University ofCalifornia Press.Bates, Robert H. 1988, ‘Contra Contractarianism: Some Reflections on the newInstitutionalism.’ Politics and Society, 16: 387–401.Crook, R. C. and Manor, J. 1998, Democracy and Decentralization in South Asia andWest Africa. Participation, Accountability and Performance. Cambridge: CambridgeUniversity Press.Crook, R. C. and A. S. Sverisson 2001, ‘Decentralization and Poverty Alleviation inDeveloping Countries: A Comparative Analysis, or, Is West Bengal Unique?’ IDSWorking Paper 130, Brighton: University of Sussex.Dutt, R. C. 1903, Economic History of India, Vol 2, New York: Burt Franklin.Gopalakrishnan, R. and Sharma, A. 1999, Education Guarantee Scheme. Bhopal:Government of Madhya Pradesh.Government of MP (Madhya Pradesh) 2002, Third Human Development Report. Usingthe Power of Democracy for Development. Bhopal: Government of Madhya Pradesh.Grindle, M. S. and J. W. Thomas, 1991, Public Choices and Policy Change: Thepolitical economy of reform in developing countries. Baltimore and London: JohnHopkins University Press.Hirschman, A. O. 1970, Exit, Voice and Loyalty: Responses to Decline in Firms,Organizations and States. Cambridge, MA: Harvard University Press.Khan, M. H. 2001, ‘The New Political Economy of Corruption’ in Fine, B., Lapavitsas,C., and Pincus, J., eds. Development Policy in the Twenty-first Century. Beyond thepost-Washington consensus. London and New York: Routledge.Kingdon, G. G. 1994, ‘An Economic Evaluation of School Management Types in UrbanIndia – A Case Study of Uttar Pradesh.’ D. Phil. Thesis, Oxford: University of Oxford.Mehrotra, S. 1998, ‘Education for All: Policy Lessons from High Achieving Countries.’International Review of Education, vol. 44, 5/6: 461–84.Mehrotra, S. 2006, ‘Governance and Access to Basic Social Services: EnsuringAccountability in Service Delivery Through Deep Democratic Decentralization’, Journalof International Development, 18/2.Mehrotra, S. 2006a ed. The Economics of Elementary Education in India, New Delhi:Sage Publications.Mehrotra, S., and Jarrett, S. 2002, ‘Improving Health Services in Low-IncomeCountries: Voice for the Poor.’ Social Science and Medicine, 54(11): 1685–90.Mehrotra, S., and E. Delamonica, 2007, Eliminating Human Poverty. Macro-economicand Social Policies for Equitable Growth. London: Zed Press.Mehrotra S., P. R. Panchamukhi, and R. Srivastava 2005, Financing ElementaryEducation in India, Uncaging the “Tiger” Economy. New Delhi: Oxford UniversityPress.



Page 57Paul, S. 1992, ‘Acccountability in Public Services: Exit, Voice and Control.’ WorldDevelopment, 20(7): 1047–60.Planning Commission 2006, An Approach to the Eleventh Five Year Plan, Governmentof India, New Delhi.Putnam, R. 1993, Making Democracy Work: Civic Traditions in Modern Italy.Princeton: Princeton University Press.Rao, Govinda 2000, Fiscal Decentralization in Indian Federalism. Mimeo, Bangalore:Institute for Social and Economic Change.Sender, J. and Smith, S. 1986, The Development of Capitalism in Africa. London:Taylor and Francis.Sen, A.K. 2000, Development as Freedom. Oxford: Oxford University Press.Streeten, P. 1993, ‘Markets and States: Against Minimalism.’ World Development,21(8): 1281–98.Tanzi, V. 1995, Taxation in an Integrating World. Washington, DC: BrookingsInstitution.Tendler, J. 1998, Good Governance in the Tropics. Baltimore: John Hopkins UniversityPress.Wade, R. 1985, ‘The Market for Public Office: Why the Indian State is Not Better atDevelopment.’ Word Development, 13(4): 467–97.——1989, ‘Politics and Graft: Recruitment, Appointment, and Promotion to PublicOffice in India’, in Ward, P. M., ed, Corruption, Development and Inequality: SoftTouch or Hard Graft? London: Routledge.World Bank 1997, World Development Report 1997. Oxford: Oxford University Press.World Bank 2006, World Bank Governance Indicators 2006,http://info.worldbank.org/governance/wgi2006/, accessed 1 March 2008.



Page 584 Is foreign aid compatible with good governance?Theory and evidence from the PhilippinesEduardo Araral, Jr.AbstractIs foreign aid compatible with good governance? Using the case of irrigation aid in thePhilippines, I hypothesize that aid is embedded in a perverse set of incentives thatcould undermine the application of good governance principles espoused by donors.My findings suggest that the moral hazard problem and aid fungibility are embeddedin irrigation aid and that these incentives also drive the vicious cycle problem inirrigation commonly found in many developing countries. These findings indeedsuggest that incentives structured into the relationship between donors and theirrigation agency may undermine the application of some principles of goodgovernance espoused by donors.IntroductionThe subject of governance has increasingly become a mantra in the developmentcommunity. Consequently, it has spawned a cottage industry of scholars attempting tomeasure its significance on development and the effectiveness of foreign aid as ameans to improve governance (Kaufmann 2005). The World Bank (2006), forinstance, has constructed and tracked down since 1996 aggregate governanceindicators that cover more than 200 countries using more than 350 variables obtainedfrom dozens of institutions worldwide.I use the term “governance” to refer to the traditions and institutions by whichauthority in a country is exercised for the common good (Kaufmann 2005). It includesthe processes by which those in authority are selected, monitored, and heldaccountable, the quality of public service provision, the competency of bureaucracy,credible commitment to policies and regulatory quality as well as the rule of law andthe control of corruption (World Bank 2006).The conventional wisdom holds that foreign aid works in a good policy environment.For instance, financial assistance leads to faster growth, poverty reduction and gainsin social indicators in developing countries with sound economic management, and theeffect is large (World Bank 1998). Yet, despite the volumes of research on aid, little isempirically known about whether or not incentives embedded in aid could underminethe application of good governance principles espoused by donors themselves. Inparticular, little is known about how incentives



Page 59embedded in foreign aid affect the incentives of bureaucracies in developing countriesto deliver public goods. This article attempts to explore this gap in the literature.Incentives in foreign aidSeveral incentive problems have been identified in the extant literature on aid, whichhave implications for governance, for instance Gibson et al. (2005), Martens (2002),Kanbur (2000), Collier (2000), Stiglitz (1997), Svensson (2000), van de Walle (2001)and World Bank (1998), among others.Gibson et al. (2005) and Martens (2002) suggest that institutions govern the complexrelationships between the main actors in the aid delivery system and often generate aseries of perverse incentives that promote inefficient and unsustainable outcomes.Feyzioglu et al. (1998), for instance, note the problem of fungibility in foreign aid.Using panel data, they suggest that donor financing of a project releases resources ofthe recipient government for other marginal projects. In effect, they argue, donors donot finance the project they appear to pay for but rather the one the recipientgovernment chooses to undertake with the resources freed up by their aid. In thispaper, I empirically demonstrate this problem of aid fungibility.Collier (2000) also argues that aid is embedded with another incentive problem: moralhazard. He suggests that the injection of aid alleviates the immediate fiscal crisis ofthe recipient government and hence the urgency for change. In poor policyenvironments, therefore, argues Collier, aid can delay reform. Given that aid is facedwith a time consistency problem, there is no incentive to maintain a promise forreform unless the recipient government is itself interested in reform.However, argues Collier, there is often no cost to defaulting on promises of policyreform. The reason for this is because of the moral hazard problem embedded in aid.Donor agencies that should enforce the terms of the contract also have an interest inrealizing some form of success. The enforcement of the terms of the aid contract,argues Collier, is often relaxed when the recipient shows some signs of promise butthese indications are sometimes withdrawn by the recipient after the aid has beendisbursed.This relationship between donors and recipients has been succinctly described as “theKenya-Fund dance” which plays out as follows:Over the past few years, Kenya has performed a curious mating ritual with its aiddonors. The steps are: one, Kenya wins its yearly pledges of foreign aid. Two, thegovernment misbehave, backtracking on reform and behaving in an authoritarianmanner. Three, a new meeting of donor countries looms with exasperated foreigngovernments preparing sharp rebukes. Four, Kenya pulls a placatory rabbit out of thehat. Five, the donors are mollified and the aid is pledged. The whole dance startsagain.There are three possible reasons for this sort of dynamics. The first is the Samaritan’sDilemma (Gibson et al. 2005): donors are in a dilemma when stopping aid,



Page 60particularly when there is so much dependency on aid, since stopping it sharply wouldcause a major chaos in the recipient country’s economy and only hurt the poor. If therecipient government didn’t spend enough on the poor and thus violated aidconditionality, donors are in a dilemma, since imposing sanctions might well mean adouble whammy for the poor.Second is the Patron’s Dilemma, which suggests that it is not in the donor’s interestto impose the sanction of aid withdrawal even when aid conditionality is violated forreasons of political clientilism (Kanbur 2000). When heavily indebted countries areinvolved, donors are understandably reluctant to cut aid inflows, as this would meaninterrupting debt servicing.Finally, there is the Career Dilemma, in that the incentive of agents within aidagencies is to keep aid flows going, and imposing sanctions stops these aid flows.This would jeopardize the preparation of aid programs on which depend the careersof agency staff, as well as the image of the agency in the eyes of its political masters.More generally and fundamentally, the problem of moral hazard and fungibility offoreign aid are variants of two basic incentive problems of governance – the problemof motivation and information (Campbell 1995; Weimer and Vining 1999). Motivationand information problems arise depending on three broad factors and the interactionamong these factors (Gibson et al. 2005; Araral, 2005):1 The nature of a good in terms of excludability, rivalry and congestibility (i.e.whether the good has the characteristics of a private good, public good, toll good,club good and common pool good) and the degree of uncertainty inherent in the good(i.e. whether it has the characteristics of a search good, experience good, post-experience good, etc.);2 the attributes of relevant players in the game (i.e. group size, history ofcooperation, social capital, etc.); and3 the institutions or the rules of the game (i.e. operational rules, collective choice orpolicy-level rules, or constitutional choice or rules about rule making) which structurethe patterns of interaction among actors and, along with the nature of the good andattributes of the players, affect the incentive structure of players.As an illustration, non-cooperation and free riding are dominant motivations androbust outcomes in situations involving public goods and common pool resourceswhen large groups of users are involved in the context of weak, perverse or non-existent institutions. However, the problem of overextraction or free riding expectedfrom a common pool resource can also be mitigated in the presence of informal andformal institutions (Ostrom 1990).Other scholars have examined variations of a similar problem of motivation andinformation, for instance Olson (1965) in the case of public goods and group size;Axelrod (1985) in the case of prisoner’s dilemma; Kreps et al. (1982) in the case ofthe finitely repeated prisoner’s dilemma; and Sandler and Tshirhart (1980) for clubgoods, among others. More recently, Schmid (2006) reconsiders these debates fromthe perspective of institutional and behavioural economics.



Page 61In the succeeding sections of this paper, I will empirically and analytically show thatthe incentives structured into the relationship between donors and the irrigationagency may undermine the application of the very principles of good governanceespoused by donors.Model and hypothesesIn this paper, I present an analytic case study focusing on the effects of foreign aidon the incentives of government agents to provide for a public good in one sector inone country. Specifically, I examine the effects of foreign aid in the governance ofpublic irrigation systems in the agriculture sector in the Philippines. I hypothesize thatcurrent approaches to irrigation aid are fraught with the problem of moral hazard andthe fungibility of aid, which drives the vicious cycle problem that is commonly foundin public irrigation systems in many developing countries and thus undermines thequality of public service provision (Figure 4.1).As Figure 4.1 shows, underinvestment in irrigation maintenance leads to unabateddeterioration of facilities, poor water service, lower cropping intensities and poorproductivity and income for farmers. Low income among farmers in turn leads topersistently low collection of irrigation fees, which further aggravates the problem ofchronic underinvestment in irrigation maintenance.My arguments proceed as follows: First, irrigation agencies in developing countries –which are often struggling financially – have strong incentives to underinvest in themaintenance of irrigation systems because this helps justify new loans from donors forcapital-intensive investments in rehabilitation. Donors need irrigation agencies asclients in order to grow their loan portfolio just as much as irrigation agencies needdonors to finance their capital expenditures.

Figure 4.1 The vicious cycle problem in irrigation and moral hazard problem in aid.Source: Araral, 2006.



Page 62

Figure 4.2 The effect of aid fungibility on bureaucratic incentives.Second, these new loans for irrigation rehabilitation provide direct and indirectsubsidies for financially strapped irrigation agencies, which is vital for keeping themafloat. However, these subsidies in their current form are highly fungible and thisproblem creates incentives for irrigation agencies to engage in a strategic behaviorthat leads to the chronic underinvestment in irrigation maintenance. How thefungibility of irrigation aid leads to strategic behavior by the irrigation agency ismodeled in Figure 4.2.In Figure 4.2, the horizontal axis shows the quantity of a good or service that theirrigation agency provides; for example, irrigation operation and maintenance (O&M)denoted by the variable X. The vertical axis shows the agency’s expenditure on allgoods other than X. The line that connects B/Px and B (i.e. BL1) represents theagency’s initial budget line without a subsidy. Given a total budget of B, the irrigationagency could do three things. First, it could spend nothing on X (irrigation O&M), andinstead spend its entire budget on all services other than X. Second, it could spendeverything on X and nothing on other services. Third, and more realistically, it couldchoose a budget allocation at any point in the budget line BL1 between theseextremes.Given this budget line, assume that the irrigation agency decides to provide X0 unitsof irrigation O&M denoted by X. The indifference curve labeled I0 gives all thecombinations of X and expenditures on other goods that would be equally assatisfying to the irrigation agency as X0 and b0 spending on other goods andservices.Suppose the national government borrows from donors to subsidize the irrigationagency the amount S for each unit of X (irrigation O&M) it will provide. This would bea matching grant because it matches the irrigation agency’s expenditures for O&M atsome fixed percentage. It is open-ended because there is no ceiling Expenditures onall other goods



Page 63on the total subsidy that the irrigation agency can receive from the nationalgovernment.With this subsidy, the irrigation agency’s new budget line now shifts to the right fromBL1 to BL2. The effective price that the irrigation agency sees for X (irrigation O&M)falls from Px to Px-S because of this subsidy. With this new budget line, the irrigationagency now procures X1 units of X reaching a higher level of satisfaction indicated byindifference curve I1. However, as a result of this subsidy, the irrigation agencyspends more on other goods and services unrelated to canal maintenance as some ofthe subsidy for X (irrigation O&M) spills over to goods and services other than X.The subsidy therefore becomes decategorized and the area bounded by the horizontallines from points b0 and b1 shows the extent to which the subsidy to O&M spills overto goods and services other than irrigation O&M. This also represents the extent ofthe fungibility of irrigation aid which, combined with the moral hazard problem,creates incentives that lead to the problem of chronic underinvestment in irrigationmaintenance, which in turn drives the vicious cycle problem of irrigation.The case of irrigation aid in the PhilippinesThe evolution of irrigation in the PhilippinesModern irrigation development in the Philippines, as in most developing countries,evolved through several phases. The first phase can be characterized as a capital-intensive, government-sponsored expansion phase that occurred during the period ofthe “green revolution” during the Marcos martial law years in the mid-1970s to themid-1980s. The second phase – from the mid-1980s to the late 1990s – consisted ofincremental improvements during which the role of the National IrrigationAdministration (NIA) in irrigation development was largely accepted and uncriticallyquestioned. The third phase, from the 1990s to the present, which is the focus of thisstudy, is characterized by the survival phase of the irrigation bureaucracy.Growth phaseThe National Irrigation Administration was established in 1964 for the purpose ofaddressing the problem of food security in the country, which had become a matter ofpolitical concern. The reason was that rice productivity during that time averaged only1.7 tons per hectare, one of the lowest in the world, but the country had a populationgrowth rate of 2.8, which was one of the highest in the world. With a persistentannual deficit, the national government was faced with the problem of meeting thebasic food requirement of its population. The solution then was to increase productionthrough the expansion of irrigated areas. NIA was thus created in 1964 as a semi-autonomous, government-owned and government-controlled corporation.



Page 64

Figure 4.3 Growth of irrigation aid in the Philippines, million dollars (2002 prices).Source: NIA 2003.In subsequent years, NIA’s charter was amended, which saw the increase in itscapitalization by more than 33 times and the size of its staff by nine times, adevelopment that has important consequences in the survival phase of NIA. Thisgrowth in the size of the irrigation bureaucracy was driven by the growth of irrigationaid in the Philippines (Figure 4.3).The growth phase of irrigation development was accompanied by reforms inoperational efficiencies, including the recognition of the vital role of irrigationassociations (IAs) in operation and maintenance of irrigation systems. During thisgrowth phase, as demand for water increased, property rights to water were definedand delineated to reduce uncertainties and potential conflicts and to encourage moreinvestments in irrigation infrastructure. During the growth period, irrigated areasincreased from 742,447 ha to 1,436,880 ha, an increase of 93 percent. This translatesto an annual average growth rate of 7.19 percent, or about 3.5 times faster than theinternational annual growth rate of two percent a year for the same period (NIA1990).This rapid growth in irrigation development in the Philippines also necessitated a moredecentralized approach to irrigation management, as the capacity of the irrigationagency could not cope with this growth. Thus, in the mid 1970s, NIA launched apioneering program to gradually decentralize the construction and O&M of small- andlarge-scale public irrigation systems as a means to improving their performance. Earlystudies have shown that these reforms have produced consistently positive results,and earned widespread international documentation and recognition as a role model.The decentralisation efforts of NIA not only caught the attention of researchers anddonors alike but also of irrigation authorities from India, Indonesia, Thailand, SriLanka and Nepal, who imported and adapted them to their countries. They alsoattracted widespread documentation from experts and scholars, arguably one of thewidest of its kind in the irrigation literature to date (Sabio and Mendoza 2000; Briscoe2000; Vermillion 2002; Panella 1999; Fujita et al. 1999; Raby 1997; Merrey 1994;



Page 65Oorthuizen and Kloezen 1995; Wijayaratna and Vermillion 1994; Bagadion 1994a;Meinzen-Dick et al. 1995; Korten and Siy 1987; NIACONSULT 1994a, 1994b). Thus,by the late 1980s, NIA had become the undisputed international leader in irrigationdecentralisation, so much so that the World Bank acknowledged NIA’s efforts as a“venerable tradition of reform” (Briscoe 2000:3).Incremental improvement phaseThe second phase of modern irrigation development in the Philippines was theincremental improvement phase during the mid-1980s to the late 1990s. During thisperiod, the role of NIA in irrigation development was largely accepted and uncriticallyquestioned by donors. The type of irrigation projects during this period took theexisting incentive structure and modus operandi of NIA as given and did notadequately examine alternative governance modes of providing irrigation to farmers.In fact, during this period, the role of NIA was reinforced by the national governmentand donors alike by building its capabilities through staff training, new irrigationtechnologies, equipment outlays, information and decision support systems, and othermanagerial and technical improvements.These capabilities were further augmented by organizing IAs to serve as NIA’scontractors in the collection of irrigation fees and O&M of irrigation systems. ThePhilippine model of decentralized irrigation management followed an incremental,voluntary approach. The transfer was first done at the tertiary canals for IAs withservice areas up to 200 ha and these could then federate at the secondary level. Forsome minor (1000 ha) and medium-scale systems (2000 ha), the IAs could federateup to the level of the entire system. In large irrigation systems, farmers’ control waseffectively limited in tertiary canals while NIA retained control of main and secondarycanals. Most of the irrigation systems turned over by NIA to farmers were those below1,000 ha. Thus, after more than 25 years since the so-called participatory irrigationmanagement program was launched, and for which NIA gained international acclaim,no more than 15 percent of the total irrigation service areas in the Philippines wereactually effectively transferred to farmers.NIA’s model of decentralized irrigation management in the large-scale irrigationsystems constituted two distinct types. One is administrative decentralisation,particularly the deconcentration of provision responsibilities – i.e. the transfer ofdecision-making authorities overseeing O&M functions to lower levels of the NIAhierarchy – and the privatization or contracting out of the production of O&M servicesto IAs. This is the case for 85 percent of the irrigation systems where IAs wereorganized and contracted by NIA to serve as its agents in O&M. Under this set-up,there is an upward flow of accountability as the IAs are accountable to NIA ascontractors, and regional field offices are accountable to the NIA central office. TheseIAs are not autonomous on matters of rule making and they have little control overthe irrigation infrastructure and finances, O&M, conflict resolution, the right towithdraw water, and, most importantly, the right to exclude non-members or non-payors from receiving irrigation services.The other model can be properly described as political or democratic



Page 66decentralisation. In this set-up, the authority, responsibility and accountability forirrigation O&M are fully transferred to the farmers through their IAs, who aredownwardly accountable to their members. These IAs gain autonomy and discretionfrom NIA in rule making and enforcement and they have full control over the irrigationinfrastructure and finances, O&M, conflict resolution, the right to withdraw water, and,most importantly, the right to exclude non-members or non-payors from receivingirrigation service. This is the case for approximately 15 percent of all IAs in publicirrigation systems in the Philippines.Consequences of the incremental phaseThe incremental phase of irrigation development in the Philippines has had seriousrepercussions for the subsequent phases. The first consequence of the contractingrelationship between NIA and the IAs is that 93 percent of all IAs have an averagenet worth of less than 59,000 pesos ($1,030 in 2005 prices), an amount that barelypays for routine maintenance and rehabilitation work. This is because the IAs wereprimarily organized by NIA to serve as its agents for O&M, and thus their financialbase is primarily dependent on labor contracts from NIA.Second, this patron-client relationship has stunted farmers’ abilities to govern theirirrigation systems. For instance, according to the Japan International CooperationAgency (JICA) study, about half of all IAs are poorly functioning or exist only onpaper, while only 21 percent, by NIA’s account, are considered functional (JICA2003:3–11). IA functionality is rated by NIA in terms of a composite set of criteria,including: 1) O&M (planning, implementation and performance); 2) organization interms of membership, record keeping, holding of meetings (general assemblies, boardof directors, service area meetings); 3) financial performance, and 4) organizationaldiscipline. Not surprisingly, only 8 to 25 percent of IAs bother to implement their O&Mplans (JICA 2003:3–13).Because of the poor functionality of IAs, and because the benefits of being a memberare unclear, farmers are, not surprisingly, reluctant to join them. Almost all of IAs (95percent) have memberships less than 50 percent of potential members (i.e. of all thefarmers in a given irrigation system and less than 50 percent bother to formally jointhe IA). Of those who do, less than 25 percent actively participate in IA activities.This patron-client relationship has bred a moral hazard problem that to a majorextent accounts for the free rider problem in irrigation in the Philippines. For instance,over a ten-year period from 1990 to 2000, some 80 percent of irrigation fees infarmers’ back accounts remain uncollected. NIA estimated that some 567,041 farmershave outstanding balances with the agency totaling 6.67 billion pesos ($125.9m).Farmers know that NIA would be unable to enforce the collection of unpaid irrigationfees, which further strengthens the incentive to free ride. This expectation by farmersis repeatedly proven correct – and the moral hazard problem reinforced again. In2003, NIA announced a program providing amnesty to delinquent payors in return forpartial payment of unpaid fees. The initial result of this latest effort,



Page 67reported in early 2004, was not encouraging. A year after launching the compromiseprogram, only 1.5 percent of all the farmers have settled their back accounts withNIA. Again, this indicates a major incentive problem in the relationship between NIAand farmers that can be attributed to the patron-client relationship.Thus 20 years after it was launched, an examination of the decentralisation programclearly indicates that irrigation in the Philippines is now faced with the vicious cycleproblem of irrigation as described in Figure 4.1. For example, I find that at least 80percent of all the 196 large-scale irrigation systems are in poor condition and requiremajor rehabilitation, which is a clear indication of the problem of chronicunderinvestment in maintenance.Consequently, when irrigation facilities are in poor shape, water service alsodeteriorates. For instance, cropping intensity, which is a measure of the quality ofirrigation service, averages only 68 percent over a ten-year period (1990–2000). Poorwater service in turn leads to lower aggregate farm productivity and lower incomesfor farmers, which in turn drive the problem of persistently poor collection of irrigationfees. For instance, from 1990 to 2000 the incidence of free riding among farmers inthe payment of irrigation fees has been reported at 66 percent. Also, less than 25percent of irrigation associations take responsibility for the maintenance of theirsystems even after 20 years of effort at decentralisation (JICA 2003).The role of irrigation aidSince 1969, NIA has contracted $2.2b in foreign loans to finance capital expenditure,O&M support and institutional development projects. Practically all of NIA’s capitalinvestment projects were financed with foreign funding, particularly from the WorldBank, Asian Development Bank (ADB) and the Japan Bank for InternationalCooperation (JBIC). Irrigation funding from donors increased substantially from 1971to 1983 at the height of the Marcos martial law period but conspicuously droppedfrom 1984 to 1989 (JICA 2003). The reason for this is that the World Bankrecommended that NIA scale down its investment program for future irrigationprojects because of the projected huge rice surpluses and declining commodity prices.The World Bank’s recommendation effectively signaled the end of the constructionphase of irrigation in the Philippines, and ushered in the second phase of irrigationlending in the mid-1980s – the emphasis on irrigation rehabilitation and improvement.This shift in emphasis in funding did not bring about fundamental changes in irrigationgovernance in the Philippines but in fact strengthened existing business-as-usualgovernance arrangements. As noted by Briscoe (2000), these types of projects takethe incentive structure and modus operandi of NIA as given and do not adequatelyexamine alternative governance modes of providing irrigation to farmers. The wayirrigation subsidies were distributed merely reinforced the central role played by thenational government in irrigation management.For example, the purpose of two major World Bank funded projects in the early 1990swas to provide direct subsidies to the O&M of irrigation systems nationwide



Page 68as well as subsidies to the general administration of the National IrrigationAdministration. Follow-up projects in the mid to late 1990s by the ADB and JBIC alsofocused on providing direct subsidies to NIA for the O&M and rehabilitation ofirrigation systems These types of projects constituted almost 90 percent of all projectsfrom 1990 to 2002 (JICA 2003).Why the preference for these rehabilitation projects? One consideration is theriskiness and the probability of failed implementation when introducing alternativeways of doing business (Araral 2005). During project design, project officers mustconsider project risks involved and how these might affect the quality and size of theirproject loan portfolio. These projects tend to involve straightforward engineeringdesign with familiar contracting and construction supervision mechanisms. These typesof loans tend to disburse quickly and are relatively easily monitored. They alsogenerate reliable bribe revenue for procurement officers. Also, aid project officers areoften trained as engineers, and are therefore comfortable with these projects.In contrast, projects with attached policy reform conditionalities are susceptible togreater risks and uncertainties (Araral 2005). Reform issues go beyond the control ofNIA and might involve other powerful government agencies or political actors. As thenumber of players and interests grow, and key issues spill into the political arena, theprobability of successful reforms within the limited project life cycle – typically fiveyears – diminishes. Unattractive risk/benefit ratios will tend to discourage most projectofficers from pursuing novel approaches. In addition, when the national governmentcannot credibly commit to pursuing needed reforms, risks grow still farther. Whenreform efforts fail, project disbursements are stalled. This impacts the loan portfolioand careers of bank officers, as disbursements are a typical indicator of staffperformance (for a discussion of aid incentives see also Gibson et al. 2005).The moral hazard problem and the fungibility of irrigation aidI have earlier argued that the relationship between donors and irrigation agencies isfraught with the moral hazard problem, i.e. donors need irrigation agencies as clientsto grow their loan portfolio, while irrigation agencies need donors to finance theircapital expenditures and subsidize their operating expenditures.Because of the moral hazard problem, donors do not have strong incentives toeffectively enforce loan provisions requiring borrowing countries to adequately investin O&M. Borrowers routinely promise to provide adequate funding for O&M but arefaced with a negligible cost for non-compliance. Regardless of their compliance record,irrigation agencies can correctly expect donors to continue financing rehabilitationprojects because it is in the donor’s interest to do so. Absent credible enforcement,and considering the negligible costs of non-compliance, the irrigation agency’sdominant incentive is to default on its responsibilities towards O&M.This behavior by NIA is reflected in the magnitude of the problem of deterioratingfacilities: 80 percent of all the large-scale irrigation systems in the Philippines are



Page 69in poor and deteriorating condition. These systems require major rehabilitation andthus would qualify as candidates for new loan projects from donors.The strategy of neglecting adequate irrigation O&M by NIA makes sense from thestandpoint of bureaucratic survival. From 1991 to 2000, NIA’s operating income wason average 73 percent below its operating expenses (i.e. what it collects fromirrigation service fees is not enough to cover its operating expenses) (JICA 2003). Tosurvive, NIA relies on direct and indirect subsidies from irrigation loans. For instance,17 percent of its total income over a ten-year period (1993–2002) is constituted bymanagement fees that it charges foreign-funded irrigation projects to pay forpersonnel and operating costs of its headquarters (JICA 2003). Every time anirrigation loan project is undertaken, NIA uses part of the loan to subsidize theoperation and maintenance of its headquarters.Thus, a portion of loans for irrigation rehabilitation, for example, spills over to otherservices not directly related to the actual O&M of irrigation systems. In this case, aportion of the subsidy for irrigation rehabilitation becomes “decategorized” or spillsover to other spending categories not directly related to irrigation O&M; for example,paying the salaries of staff in the headquarters who do not contribute to irrigationO&M. (See Figure 4.2 for the analytic model.) The fungible nature of irrigation loans inthe Philippines thus gives NIA a strong incentive to underinvest in maintenance toenable it to borrow irrigation loans because these management fees keep the NIAcentral office afloat.Second, foreign-funded projects also provide equipment assets to NIA, which in turngenerate equipment rental fees that are used to subsidize the operation of NIA’sregional offices. Equipment rental fees are fees collected by NIA when it rents outthese loan-funded equipments to contractors of irrigation projects or to othercontractors in the construction industry. While proceeds of irrigation loans arecategorically intended for specific projects, part of the loan again spills over to othercategories or they become decategorized. In the case of subsidies for equipmentassets, NIA rents out these equipments to private contractors – not necessarily to beused in projects where they were originally intended to be used – in order togenerate additional revenues to subsidize the salaries and O&M of regional offices.From 1993–2002, such fees accounted for 15 percent of its total income (JICA 2003).Third, rehabilitation projects bring in additional income from irrigation service fees tosupport the operations of NIA’s Irrigation System Offices (NISOs). These fees arecollected from farmers as payment for irrigation services. Newly rehabilitated irrigationsystems bring in additional sources of revenues for NIA that would not havematerialized when irrigation systems are non-functioning. Between 1993 and 2002,irrigation fees generated 38 percent of NIA’s total income. Ideally, the revenue fromirrigation fees should be plowed back from where it was originally collected. However,these irrigation fees revert to NIA’s General Fund, which is then used to subsidizeservices other than actual O&M of irrigation systems.Overall, the fungible nature of irrigation loans in the case of the Philippines hascreated a strong incentive for NIA to underinvest or postpone investments in



Page 70irrigation maintenance since irrigation systems that are poorly maintained soonbecome candidates for rehabilitation and therefore justify capital expenditure fromdonors, which, given their fungible characteristics, helps keep NIA financially afloat.The fungible portion of capital expenditures provides at least 40 percent of NIA’soperating budget, giving it a strong incentive to maintain the current subsidy scheme.Donors also have little incentive to alter this incentive structure because this is whatmotivates NIA to continue borrowing for irrigation aid, which is important for thegrowth of the donor’s irrigation loan portfolio.Conclusions and policy implicationsIs foreign aid compatible with good governance? Using the case of irrigation aid in thePhilippines, I hypothesized that aid is embedded in a perverse set of incentives thatundermine the application of good governance principles espoused by donors. Myfindings suggest that the moral hazard problem and aid fungibility are embedded inirrigation aid. These findings suggest that the incentives structured into therelationship between donors and the irrigation agency may undermine the applicationof some of the very principles of good governance espoused by donors.If indeed these conclusions are correct and commonplace among irrigationbureaucracies in developing countries – a pattern noted by Vermillion (2002) amongseveral countries in Asia – the question then is what can be done about it? This is animportant policy question, since irrigation plays a crucial role in developing countries interms of agricultural development, food security, poverty alleviation and livelihoodgeneration, as well as in stabilizing food prices and inflation levels in many developingcountries (Briscoe 2000). Irrigation is also the largest recipient of public agriculturalinvestment in the developing world and a major recipient of public operatingsubsidies.One way to deal with these problems is to use a form of aid conditionality such as amatching grant with a maintenance of effort (MOE) requirement. How the schemeworks is illustrated in Figure 4.4. Compared with the scheme earlier illustrated inFigure 4.2, under the MOE requirement in Figure 4.3, the irrigation agency is nowfaced with a new budget line, BL3. With this new budget line, only units beyond X0would be subsidized. The budget line with this new requirement follows the originalbudget line up to X0 (point A), and then rotates to the right so that it becomesparallel to the budget line BL2 for the subsidy without the maintenance of effortrequirement. At any point of the new budget line BL3 (with MOE), the agency nowresponds by providing more of X2 (higher effort at maintenance) with a newindifference curve (I2), and by spending less on other goods and services than itwould without the subsidy (X0) and by a subsidy without maintenance of effortrequirement (X1).The key policy lessons here – which can be generalized more broadly – are twofold.First, the MOE provisions may be useful for targeting subsidies at levels that have thehigher desired impacts. Second, MOE provisions can help ensure that subsidies arespent where they are actually needed instead of being spent



Page 71

Figure 4.4 Solution mechanism to the problems of moral hazard and aid fungibility.on the general operation of public bureaucracies with little or no direct impact ontargeted behavior. Thus, the MOE requirement can be used as one strategic policyinstrument that can be deployed to address the problem of moral hazard andfungibility of aid in general. It can also be used by national or federal governments inconjunction with the provision of matching grants to state or local governments, state-owned enterprises or public bureaus, or to other public entities where there is astrong possibility of moral hazard and fungibility of aid.BibliographyAraral, E. 2005, ‘Bureaucratic Incentives, Foreign Aid and Path Dependence.’ PolicySciences, 38(2): 131–57.Araral, E. 2006, Decentralization Puzzles: The Political Economy of Irrigation PolicyReform in the Philippines. Unpublished PhD Dissertation: Indiana University-Bloomington.Asaduzzaman, M. 1981, ‘The Emergence of Cooperation among Egoists.’ AmericanPolitical Science Review, 75: 306–18.Axelrod, R. 1985, ‘An Evolutionary Approach to Norms.’ American Political ScienceReview, 80(4): 1095–111.Bagadion, B. 1994a, Case Study on Farmers’ Participation in Philippines NationalIrrigation Systems, paper presented at the Workshop on Participatory Development,May 17–20, World Bank, Washington, D.C.——1994b, Joint Management of the Libmanan-Cabusao Pump Irrigation SystemBetween Farmers and the National Irrigation Administration in the Philippines, paperpresented at the International Conference on Irrigation Management Transfer,September 20–24, Wuhan, China.Briscoe, J. 2000, The World Bank’s Role in Water Resources Management in thePhilippines: Results of a Consultation. Washington, D.C.: The World Bank.



Page 72Campbell, D. 1995, Incentives: Motivations and the Economics of Information.Cambridge: Cambridge University Press.Collier, P. 2000, ‘Conditionality, Dependence and Coordination: Three Current Debatesin Aid Policy’, in C. L. Gilbert and D. Vines, eds. The World Bank: Structure andpolicies, Cambridge: Cambridge University Press.Feyzioglu, T., S. Vinaya, and Z. Min, 1998, ‘A Panel Data Analysis of the Fungibility ofForeign Aid.’ World Bank Economic Review, 12(1): 29–58.Fujita, M., Y. Hayami and O. Kikuchi 1999, The Conditions of Collective Action forLocal Commons Management: The Case of Irrigation in the Philippines, mimeo, paperpresented at the Workshop in Political Theory and Policy Analysis at IndianaUniversity, June 1999.Gibson, C. K., E. Andersson, Ostrom and S. Shivakumar 2005, The Samaritan’sDilemma: The Political Economy of Development Aid. New York: Oxford UniversityPress.Japan International Cooperation Agency (JICA) 2003, A Study on the Strengthening ofthe Organization and Management of NI., Manila.Kanbur, R. 2000, ‘Aid, conditionality and debt in Africa’, in Finn Tarp, ed. Foreign Aidand Development: Lessons Learnt and Directions for the Future. New York:Routledge.Kaufmann, D. 2005, ‘Back to Basics: 10 Myths About Governance and Corruption.’Finance and Development, 42(3). Available fromhttp://www.imf.org/external/pubs/ft/fandd/2005/09/basics.htm.Korten, F. and R. Siy, eds. 1987, Transforming a Bureaucracy: The Experience of thePhilippine National Irrigation Administration, Manila: Ateneo University Press.Kreps, D. M., P. Milgrom, J. Roberts and R. Wilson 1982, ‘Rational Cooperation in theFinitely Repeated Prisoners’ Dilemma.’ Journal of Economic Theory, 27(2): 245–52.Martens, B. 2002, The Institutional Economics of Aid, Massachusetts: CambridgeUniversity Press.Meijia, A.1999, ‘Farmers Participation in Irrigation Development and Management: TheNIA Experience’, in Abernethy, C. and F. Heim, eds. Irrigator’s Associations:Government Actions Toward Effective Irrigation Associations. Food and AgricultureDevelopment Center and Deutsche Stiftung für Internationale Enwinklung, Germany.Meinzen-Dick, R., R. Reidinger, and A. Manzardo 1995, Irrigation ParticipationSourcebook Technical Paper, Washington, D.C.: The World Bank.Merrey, D.J. 1994, Institutional Design Principles for Accountability on Large IrrigationSystems. Paper presented at International Conference on Irrigation ManagementTransfer, September, 20–24, Wuhan, China.National Irrigation Administration (NIA) 1990, The History of Irrigation in thePhilippines. Manila: Philippines.National Irrigation Administration (NIA) 2003, The Study on the Irrigator’s AssociationStrengthening Project in National Irrigation Systems. Manila: Philippines.NIACONSULT. 1994, Farmer’s Participation In National Irrigation Systems in thePhilippines: Lessons Learned. Washington, D.C.: The World Bank.Olson, M. 1965, The Logic of Collective Action. Public Goods and the Theory ofGroups, Cambridge, Mass.: Harvard University Press.Oorthuizen, Joost, and Wim H. Kloezen 1995, ‘The Other Side of the Coin: A CaseStudy on the Impact of Financial Autonomy on Irrigation Management Performance inthe Philippines.’ Irrigation and Drainage Systems, 9: 15–37.Ostrom, E. 1990, Governing the Commons: The Evolution of Institutions for CollectiveAction. New York: Cambridge University Press.Ostrom, E. 2002, ‘How Types of Goods and Property Rights Jointly Affect CollectiveAction’, Journal of Theoretical Politics, 15(3): 239–70.Panella, T. 1999, Irrigation Development and Management Reform in the PhilippinesStakeholder Interests and Implementation, paper presented at the InternationalResearcher’s Conference on IMT, December 11–14, Hyderabad, India.



Page 73Raby, N. 1997, Participatory Irrigation Management in the Philippines: The Learning ProcessApproach in the National Irrigation Systems, paper presented at the InternationalWorkshop on Participatory Irrigation Management, February 9–15, Cali, Colombia,World Bank Institute and the International Water Management Institute.Sabio, E. A. and A. D. Mendoza 2000, Organizational change for participatory irrigationmanagement. Tokyo: Asian Productivity Organization.Sandler, T., and J. T. Tshirhart 1980, ‘The Economic Theory of Clubs: an Evaluative Survey.’Journal of Economic Literature, XVIII: 1481–1521.Schmid, A. 2006, Conflict and Cooperation: Institutional and Behavioral Economics. Simon andBlackwell.Stiglitz, J. 1997, Can Aid Facilitate Development? A New Vision for Development Cooperation in the21st Century, paper presented at a public lecture, 17 September Tokyo, Japan.Svensson, J. 2000, ‘When is foreign aid policy credible? Aid dependence and conditionality.’ Journalof Development Economics, 61: 61–84.Swaroop, V. and S. Devarajan 1998, The Implications of Foreign Aid Fungibility for DevelopmentAssistance. World Bank Policy Research Working Paper No. 2022, Washington, D.C.: The WorldBank.Van de Walle, N. 2001, African Economies and the Politics of Permanent Crisis, 1979–1999. NewYork: Cambridge University Press.Vermillion, Doug. L. 2002, Irrigation Sector Reform in Asia: From Participation with Patronage toEmpowerment with Accountability, paper presented at the Asian Irrigation in Transition Workshop,April 22–23, Asian Institute of Technology, Bangkok.Weimer, D. and Vining, L. 1999, Policy Analysis: Concepts and Practice, 3rd Edition. New Jersey:Prentice Hall.Wijayaratna, C. M., and D. L. Vermillion 1994, Irrigation Management Turnover in the Philippines:Strategy of the National Irrigation Administration, Short Report Series on Locally ManagedIrrigation, No. 4. Colombo, Sri Lanka: International Irrigation Management Institute.World Bank 1998, ‘Assessing Aid: What Works, What Doesn’t and Why.’ Oxford: Oxford UniversityPress. Available athttp://siteresources.worldbank.org/INTASSAID/Resources/Overview_(ENGLISH).pdfWorld Bank 2006, ‘Governance Matter VI’, World Bank, Washington, D.C. Available athttp://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTPUBLICSECTORANDGOVERNANCE/EXTDS



Page 745 Public sector transparency and corporate accounting practices in AsiaXun WuIntroductionCorporate governance in Asia has been in the spotlight since the Asian financial crisisin 1997. Corporate accounting practices have in particular drawn heavy scrutiny, andthere is now a consensus among experts that poor accounting practices in Asian firmswere a leading factor contributing to the crisis (Gelos and Wei 2002; Johnson et al.2000a; Vishwanath and Kaufmann 1999). Poor accounting practices led to excessiveexposure to debt, weak protection of the shareholders’ interests, and distortions inresource allocation in the economy (ADB 2001; Mitton 2002).Corporate accounting practices also have significant impacts on public sectorgovernance. Because poor corporate accounting practices reduce the chance ofdetecting and exposing malfeasance, they open the door for shady exchanges withcorrupt public officials. The impacts of such dubious accounting practices on publiccorruption have been confirmed by some recent empirical studies. Kimbro (2002) hasshown that the quality of accounting standards is inversely associated with corruption,and Wu (2005) finds that prevalent corruption problems in many Asian countries areindeed due in part to the poor quality of corporate financial reporting in the samemilieu. In response to the Asian financial crisis, governments in many Asian countrieshave launched various reforms aimed at strengthening corporate accounting practices,in hopes of restoring investor confidence and enhancing public sector governance(Asian Development Bank [ADB] 2001). In Thailand, for example, companies listed inthe stock market must now submit quarterly and annual financial statements thatconform to International Accounting Standards (IAS). In South Korea, the governmenthas taken significant measures to enhance the enforcement of accounting andauditing standards to conform to international standards. Countries largely unaffectedby the crisis, such as China, have also recognized the financial risks associated withlow-quality accounting practices and have embarked on accounting reforms of variouskinds (Lin and Chen 2000).Despite these efforts, progress in strengthening corporate accounting practices in Asiahas been rather modest in the decade since the financial crisis (Morris et al. 2004);some have even perceived a decline in quality in recent years (Claessens and Fan2003). Decisions on financial reporting are ultimately the responsibility of individualfirms; yet such decisions are not made in a vacuum, and the external



Page 75environment plays a critical role in how these decisions are made. Choi (2002)concludes in his study of accounting reform in Korea that because corporate financialdisclosure is deeply imbedded in cultural norms, change in accounting practices maytake a long time to become fully implemented. Rosser (2003) observes that althoughadoption of international accounting standards among Indonesian firms may nowproject a positive image for foreign investors, little has changed in actuality, owing tothe absence of a political environment conducive to accounting reform.The impact of external environments on corporate accounting practices has also beenhighlighted in the literature on cross-country determinants of corporate financialreporting decisions. Studies have shown that corporate financial reporting isinfluenced by cultural values (Gray 1988; Zarzeski 1996), legal systems (LaPorta et al.1998; Ball et al. 2003; Jaggi and Low 2000), and politico-economic factors (Ball 2001;Archambault and Archambault 2003; Bushman et al. 2004). The importance ofexternal environmental influences suggests that corporate and public decision makerswould be prudent to develop reasonable expectations of what accounting reforms canachieve in view of such constraints, and to think strategically about implementation ofthose reforms.The present analysis seeks to extend the literature by focusing on the role of publicsector transparency in determining firms’ decisions on accounting practices.Government plays an important role in shaping the operating environment to whichcorporate accounting practices respond, and the degree of transparency ingovernment actions shapes the degree of economic, financial, and political risksperceived by the corporate sector. In conditions of low public sector transparency,firms might find it advantageous to adopt substandard or dubious accounting practicesin order to cope with various risks arising from asymmetry in information exchangebetween government and business. The analysis presented here used an internationaldata set drawn from individual corporations to explore characteristics of public sectortransparency as perceived by firms in Asia, and to test empirically the relationshipbetween public sector transparency and quality of corporate accounting practices.Understanding the linkage between public sector transparency and corporateaccounting practices has several important policy implications. First, policy makers inAsia should devise reform strategies that reflect on key features of the underlyingpolitical environment, such as the level of public sector transparency. Ignoring therelationship between public sector transparency and corporate accounting practicescould significantly reduce the effectiveness of the accounting reforms. Second,neglecting the characteristics of public sector governance (such as transparency) mayfuel unrealistic (high) expectations with regard to how quickly accounting reforms canachieve their intended goals. Changes in accounting practices may only slowly comeabout, as long as public sector transparency remains unchanged. Third, a betterunderstanding of the role of public sector transparency in determining firms’ financialdisclosure will enhance and expand the strategies and measures at the disposal ofgovernment. Targeted efforts to improve public sector transparency can in turngreatly improve prospects for successful accounting reforms.



Page 76The discussion below first reviews theoretical linkages between public sectortransparency and corporate accounting practices, and derives several testablehypotheses. Public sector transparency and corporate sector accounting practices inAsia are then considered in light of data from the World Business Environment Survey,and econometric models are tested. Concluding remarks focus on the policyimplications of these findings.Public sector transparency and corporate accounting practices: theoreticallinkagesThere is no commonly agreed-upon definition of public sector transparency (Bellverand Kaufman 2005). In the context of its impacts on the corporate operatingenvironment, public sector transparency can be defined as the assurance of firms’rights to certain types of information that help to prevent potential abuses arisingfrom asymmetry in information exchange between government and business.Information asymmetry between government and business may take several forms.First, firms may not be aware of the existence of laws, governmental policies, andregulations affecting the nature and conduct of their business. Information asymmetryis more pronounced in environments where changes in laws, policies, and regulationshave occurred recently or frequently. The asymmetry may have disproportionatelylarge effects on smaller firms, which may lack the resources to keep track of suchchanges. Second, firms may have inadequate access to information on theconventions and procedures whereby changes in laws and regulations are effected. Insuch circumstances, firms might undermine their own positions by conforming toexisting laws and regulations that will soon be modified or superseded. Third,information asymmetry may arise from discretionary administrative action ininterpreting laws and regulations, as applied to specific situations. Laws andregulations are often quite vague because of political compromises that were neededto ensure their passage, and this ambiguity leaves enormous discretionary leeway tothe agencies or officials that implement them.Corruption can further exacerbate the impacts of information asymmetry betweengovernment and business. Although it is important to distinguish lack of transparencyfrom outright corruption, the two phenomena are closely linked, especially in theirrelationship to firms’ operating environments. Corrupt officials can more easily extractbribe payments from businesses that are confronted with acute informationasymmetry problems; they can also intentionally increase the opaqueness of publicsector activities in order to secure more bribe payments from businesses.Firms’ responses to unpredictable operating environments that result from a low levelof public sector transparency can be quite predictable. First, as a counterbalancingstrategy to asymmetry arising from governmental sources, firms may choose toreduce informativeness in return, fashioning their accounting reporting so as toobstruct access to corporate financial information. Firms may also simply choose tolimit public access to actual corporate financial information in hopes of cushioningagainst risks induced by an unpredictable operating environment.



Page 77Second, lack of transparency in the public sector reduces firms’ incentives to improvethe quality of financial disclosure. Maintaining high-quality financial informationreporting requires considerable time and resources, including high long-term costs andthe deployment of highly educated human capital such as accountants and lawyers(Bushman and Smith 2001); the reward is presumed to be better prospects ofattracting investments. Lack of transparency in the public sector undermines this typeof effort by raising the variance of asset values and increasing the risk of investments,which can severely diminish investment flow to a particular country, sector, orindividual firm.Third, lack of transparency in the public sector increases the business costs byimposing high transaction costs. In such conditions, firms must expend criticalresources to monitor and cope with unpredictable changes in laws and regulationsand the procedural “rules of the game,” and senior corporate managers must spendsignificant amounts of time dealing with officials who hold discretionary powers ofinterpreting and applying these directives. Faced with such costly government-relatednecessities, firms may feel no compunction in evading tax responsibility in order tosurvive. The main argument here is that the costs of not hiding corporate informationcan be too high.Fourth, lack of transparency offers more leverage to predatory revenue officials whoexact excessive rents or accept bribes not to do so. Firms may choose to underreportprofits through fallacious accounting practices in order to avoid being targeted by suchschemes. Clarke and Xu (2004) and Svensson (2003) have found that the level ofbribe payments is positively related to “ability to pay”; that is, firms that are moreprofitable are expected to pay more in bribes. Firms may elect to divert activitiesunderground as a means of reducing vulnerability to extortion by government officials(Johnson et al. 2000b).Fifth, lack of transparency also reduces the costs of non-compliance withgovernmental strictures, especially with regard to corporate tax laws and regulations.Although tax evasion through flawed financial reporting is considered illegal in almostall countries, firms may find it easier to get away with such activities in circumstanceswhere interpretation and implementation of tax laws and regulations are lesstransparent.From these various considerations regarding low public sector transparency andasymmetries in private–public exchange of information, three hypotheses can beformed:Hypothesis One: Firms that find it difficult to obtain information on laws andregulations are more likely to adopt low standards for corporate accounting practices.Hypothesis Two: Firms that must cope with unexpected changes in rules, laws, andregulations are more likely to adopt low standards for corporate accounting practices.Hypothesis Three: Firms that perceive the interpretation of regulations asunpredictable are more likely to adopt low standards for corporate accountingpractices.



Page 78Empirical findingsDataTo test these hypotheses, data from the World Business Environment Survey (WBES)were used to supply measures for economic modeling of relationships between publictransparency and corporate accounting practices. Conducted by the World Bank in1999–2000 with the aim of understanding the constraints that businesses confronted,WBES covers 83 countries, including 12 in Asia. Table 5.1 shows WBES coveragerelating to Asia and the number of firms surveyed in each country.This data set differs in several respects from data used in other empirical studies onthe determinants of corporate information disclosure. First, it contains not only publiclylisted firms but also privately held firms, thus providing unique insights into thedeterminants of financial disclosure for privately held firms. This is especially useful ina study of Asian countries, where most firms are privately held.Second, whereas most studies acknowledge the importance of external factors ininfluencing corporate financial disclosure, and include measures of such factors in theiranalyses, the variables chosen are typically countrywide in nature and thus do notvary for firms within the same country; as a result, variations across firms withinindividual countries have not been well accounted for in terms of internal factors. Inthe WBES-based data set developed for the present analysis, external variablesprovide firm-specific information, and reflect the characteristics of externalenvironments as experienced by individual firms.Third, although the level of conformity to international standards – as reflected initems included in firms’ financial statements – has often been used to measureTable 5.1 World Business Environment Survey (2000) coverage in AsiaCountry Number of firmsAzerbaijan 128Bangladesh 50Cambodia 326China 101India 210Indonesia 100Kazakhstan 127Malaysia 100Pakistan 103Philippines 100Singapore 100Thailand 422Total 1867



Page 79the quality of financial disclosure, such a measure is intermediate in nature and maynot accurately reflect the actual quality of financial disclosure. By contrast, firmsparticipating in WBES were directly asked about the percentage of sales reported fortax purposes; the WBES-based data set used here for Asian countries thus canprovide a direct measurement of quality of financial disclosure.Measuring public sector transparency in AsiaTwo aspects of public sector transparency are of particular importance from firms’perspectives. The first is the predictability with which governments change the rules,laws, and regulations affecting firms (Bellver and Kaufmann 2005). The second is thepredictability with which these laws and regulations are interpreted in implementation.Several questions in the WBES were directly related to these same aspects of firms’perceptions of public sector transparency. One question assessed the level of difficultyfirms had experienced in obtaining information on laws and regulations affecting theirbusiness; another elicited firms’ impressions regarding predictability of changes insuch laws and regulations. Firms were also asked whether interpretations ofregulations affecting their business were predictable. Responses to these questionswere used to measure public transparency in the analysis presented here.Table 5.2 shows that the majority of firms surveyed in Asia (66 percent) found thelaws and regulations affecting their firms to be easy to obtain, at least to somedegree, although significant variations occurred across countries. For example, roughlyhalf of firms in Bangladesh, Indonesia, and Thailand rated access to laws andregulations affecting their business as very easy to somewhat easy; but in Singapore,the proportion of firms reporting that degree of accessibility was 98 percent, with theremaining two percent reporting that access was only somewhat difficult. Sizeablevariations can also be observed across firms within countries, suggesting thatperceived information asymmetry regarding the difficulty of obtaining relevant lawsand regulations may depend on a firm’s particular sector and geographical location.Table 5.2 also indicates a similar pattern in Asian firms’ impressions of thepredictability of interpretations of laws and regulations on implementation. Althoughthe majority of firms (64 percent) found these interpretations to some degreepredictable in application, significant variations can be seen both across countries andacross firms within countries.In comparison, a high percentage of firms expressed concern about the predictabilityof laws and regulations affecting their businesses: 52 percent believed that changes inlaws and regulations were to some extent unpredictable. This perception was verystrongly reported by firms in Thailand (60 percent), Bangladesh (61 percent),Indonesia (66 percent), and especially Kazakhstan (88 percent).



Page 80Table 5.2 Public sector transparency as perceived by firms across Asian countriesDifficulty in Obtaining the Laws and Regulations Affecting Firms Very difficult Difficult in

most casesSomewhat

difficultSomewhat

easyEasy inmostcases

Very easy

Azerbaijan 6% 5% 13% 20% 42% 15%Bangladesh 0% 12% 36% 6% 34% 12%Cambodia 4% 5% 21% 49% 9% 13%China 3% 10% 14% 22% 37% 14%India 4% 4% 18% 32% 31% 10%Indonesia 11% 12% 26% 40% 8% 3%Kazakhstan 6% 13% 11% 14% 27% 29%Malaysia 0% 1% 8% 39% 42% 10%Pakistan 12% 9% 16% 41% 16% 6%Philippines 1% 6% 16% 32% 29% 15%Singapore 0% 0% 2% 11% 49% 38%Thailand 3% 12% 34% 34% 16% 1%Subtotal 4% 8% 21% 32% 24% 12%Changes in Laws and Regulations Completely

unpredictableHighly

unpredictableFairly

unpredictableFairly

predicableHighly

predictableCompletelypredictable

Azerbaijan 5% 9% 12% 46% 6% 23%Bangladesh 2% 20% 39% 33% 4% 2%



Page 81Cambodia 14% 17% 28% 35% 4% 2%China 5% 12% 19% 36% 24% 3%India 4% 9% 40% 44% 1% 1%Indonesia 10% 35% 21% 22% 11% 0%Kazakhstan 41% 12% 34% 8% 2% 2%Malaysia 13% 8% 16% 46% 13% 3%Pakistan 9% 5% 19% 55% 12% 0%Philippines 6% 13% 28% 35% 12% 6%Singapore 2% 2% 9% 55% 27% 5%Thailand 8% 12% 40% 30% 9% 1%Subtotal 10% 13% 29% 36% 9% 3%Interpretations of Laws and Regulations Completely

unpredictableHighly

unpredictableFairly

unpredictableFairly

predicableHighly


Azerbaijan 4% 6% 12% 22% 48% 9%Bangladesh 0% 10% 27% 29% 23% 10%Cambodia 5% 4% 25% 49% 9% 7%China 1% 8% 21% 33% 28% 9%India 4% 12% 29% 46% 7% 1%Indonesia 13% 10% 35% 35% 5% 2%Kazakhstan 9% 20% 19% 16% 23% 13%



Page 82Table 5.2 Public sector transparency as perceived by firms across Asian countriesMalaysia 2% 0% 8% 38% 41% 11%Pakistan 5% 12% 21% 37% 23% 2%Interpretations of Laws and Regulations Completely

unpredictableHighly

unpredictableFairly

unpredictableFairly

predicableHighly


Philippines 3% 11% 20% 21% 35% 9%Singapore 1% 0% 3% 14% 48% 34%Thailand 1% 8% 32% 35% 21% 3%Subtotal 4% 8% 24% 35% 22% 8%Source: WBES (2000) and author’s calculations.



Page 83Corporate accounting practices in AsiaAs mentioned above, maintaining high-quality financial reporting is expensive in bothtime and resources. Many firms in developing countries in Asia may be reluctant tocommit the resources necessary for maintaining high-quality financial informationreporting. However, the Asian financial crisis in 1997 prompted changes in firms’perception of the importance of full and accurate financial disclosure. It is now widelybelieved that inadequate disclosure was a leading cause of the crisis. Choi (2002) hasargued that non-compliance of financial statements with international standards,deficiencies in disclosure, and lack of rigorous monitoring by external auditors wereamong the leading causes for the financial crisis in Korea. Rahman (1998), whoconducted a comparative study of five East Asian countries affected by the crisis –Indonesia, Malaysia, the Philippines, South Korea and Thailand – found that most ofthe companies in these countries did not follow international accounting standards.Since the 1997 crisis, many Asian countries have undertaken accounting reforms tostrengthen the quality of financial disclosure. Countries that largely escaped the crisis,such as China, have also recognized the financial risks associated with pooraccounting practices (Lin and Chen 2000). Typical reform measures in Asian countrieshave involved the adoption of international accounting standardsTable 5.3 Asian firms adopting International Accounting Standards (IAS) and externalauditing of annual financial reporting (AUDIT) Percent of firms that use

international accountingstandards (IAS)

Percent of firms that have annual financialstatements that have been reviewed by

external auditorAzerbaijan 18% 8%Bangladesh 76% 95%Cambodia 27% 22%China 12% 43%India 75% 97%Indonesia 45% 52%Kazakhstan 63% 37%Malaysia 20% 47%Pakistan 64% 52%Philippines 31% 81%Singapore 68% 95%Thailand 62% 83%Total 48% 58%Data source: WBES (2000) and author’s calculations.



Page 84and independent auditing practices. Table 5.3 shows that among firms, the effects ofthese reform initiatives have been quite impressive: roughly 50 percent of firms in theWBES sample now use international accounting standards, and about 60 percent hireexternal auditors to review their annual financial statements.Adoption of rules and regulations in accounting reforms, however, should not beinterpreted as equivalent to having good accounting practices. Accounting scandals inthe United States, involving noted firms such as Enron, WorldCom, and Tyco, suggestthat accounting flaws occur even in developed countries with good accounting rulesand highly competent financial professionals. Accounting standards are only as goodas the enforcement mechanisms driving them, and external auditors may align theirinterests with corrupt corporate boards and managers, turning a blind eye onirregularities in accounting reports.The difficulties of carrying out accounting reforms are confirmed by empirical resultsof the present analysis. Table 5.4 indicates that there is a sizeable disparity betweenaccounting standards and their actual implementation. Although 50 percent of firmsused international accounting standards and 60 percent hired external auditors toreview annual financial reports, only 42 percent reported 100 percent of their sales foraccounting purposes. This disparity was especially notable for firms in South Asia. Forexample, in Bangladesh, where 76 percent of firms used international accountingstandards and 95 percent used external auditors for annual financial reports, only 18percent reported 100 percent of their sales for tax purposes – in fact, more than halfreported less than 70 percent. It is clear that conforming to new accounting standardsdid not in itself guarantee good-quality financial disclosure.Table 5.4 Accounting practices in Asian firms: sales reported for tax purposes Less than 50% 50-59% 60-69% 70-79% 80-89% 90-99% All (100%)Azerbaijan 11% 16% 14% 8% 5% 9% 37%Bangladesh 0% 15% 36% 10% 10% 10% 18%Cambodia 0% 41% 9% 5% 6% 7% 31%China 8% 48% 10% 1% 3% 7% 24%India 0% 2% 3% 4% 1% 9% 81%Indonesia 15% 14% 5% 2% 14% 11% 39%Kazakhstan 3% 7% 10% 4% 8% 12% 56%Malaysia 44% 19% 3% 1% 1% 9% 22%Pakistan 16% 20% 18% 6% 4% 3% 33%Philippines 8% 12% 3% 4% 6% 17% 49%Singapore 8% 5% 1% 0% 1% 2% 82%Thailand 1% 18% 17% 9% 15% 10% 31%Total 7% 19% 11% 5% 7% 9% 42%Data source: WBES (2000) and author’s calculation.



Page 85Econometric modelsTable 5.5 summarizes the variables used in the models developed for the presentanalysis. Corporate accounting practices were measured by five variables, focusing ontwo related but distinct aspects – adoption of accounting standards and quality ofactual accounting practices. The first variable, IAS, indicates whether or not firmsadopted International Accounting Standards. The second variable, AUDIT, indicateswhether firms had their annual financial reports reviewed by external auditors. Thenext three variables measure the quality of firms’ actual accounting practices in termsof how they reported sales for tax purposes. The third variable, REPORT, indicates theextent to which firms reported their sales for tax purposes. The fourth and fifthvariables, REPORTLOW and REPORTUP, indicate what percentages of the sales firmsreported for tax purposes.Public sector transparency was measured by three variables: TRANSAVA, TRANSLAW,and TRANSINT, corresponding to the three hypotheses stated at the close oftheoretical discussion above. The first variable, TRANSAVA, measures the degree ofdifficulty firms perceived in obtaining information on laws and regulations affectingtheir business. The second, TRANSLAW, measures the extent to which the changes inlaws and regulations affecting firms’ business were perceived to be predictable. Thethird variable, TRANSINT, measures the extent to which interpretations of laws andregulations affecting firms were regarded as transparent.The variables just described are the main focus of interest here, but other variableswere included in the models to control for alternative interpretations of corporateaccounting practices. First among these are firm characteristics such as size andownership type. Large firms may have more resources with which to provide bettercorporate financial reporting than small firms do. Also, large firms may be moresecurely established and thus able to reveal comprehensive financial information withless fear that it will be misinterpreted (Jaggi and Low 2000).Second, ownership structure may also affect firms’ accounting practices. Enterpriseswith foreign shareholders are likely to adopt corporate accounting standards thatconform to international practices, in order to avoid trouble when sharing financialreportage. Governance structure can also affect firms’ decisions on accountingpractices. Firms established as single proprietorship or partnership may be under lesspressure to require comprehensive and detailed disclosures, as principle–agentproblems are not conspicuous. By contrast, firms with dispersed ownership are morelikely to adopt high standards for accounting practices, as managerial responsibilitiesare shared among board members and sharing of financial information is essential. Inmost countries, firms listed in the stock market have to adopt high standards foraccounting practices, in order to meet stringent requirements to be listed in stockexchange.Third, operating environment can also play a significant role in firms’ financialreporting decisions. For example, firms may choose poor accounting practices in orderto avoid paying taxes. Johnson et al. (2000b) have documented that in EasternEuropean countries, firms facing higher effective tax rates are indeed likely to hidetheir sales and profits. For present purposes, the extent to which high taxes



Page 86Table 5.5 Dependent and independent variables: description and descriptive statisticsVariable Description MeanStd.

Dev.Dependent VariablesIAS Dummy variable. 1=firm adopts international accounting

standards; 0=all others0.480.50

AUDIT Dummy variable. 1=Annual financial statements reviewed byexternal auditor; 0=all others

0.580.49

REPORT Categorical variable indicating percentage of total salesreported for tax purposes (1=less than 25%; 2=25 to 49%;3=50 to 59%; 4=60 to 69%; 5=70 to 79%; 6=80 to 89%;7=90 to 99%; 8=100%)

5.112.58

REPORTLOWThe percentage of the total sales reported for tax purposes(lower bound)

0.520.27

REPORTUP The percentage of the total sales reported for tax purposes(upper bound)

0.650.23

Test VariablesTRANSAVA The extent to which information on the laws and regulations

affecting the firms is easy to obtain. Scale from 1 to 6 (1=verydiffi cult; 2=diffi cult in most cases; 3=somewhat diffi cult;4=somewhat easy; 5=easy in most cases; 6=very easy)

3.991.26

TRANSLAW The extent to which the changes in rules, laws and regulationsare predictable. Scale from 1 to 6 (1=completely unpredictable;2=highly unpredictable; 3=fairly unpredictable; 4=fairlypredictable; 5=highly predictable; 6=completely predictable)

3.301.21

TRANSINT The extent to which the interpretations of rules, laws andregulations are predictable. Scale from 1 to 6 (1=completelyunpredictable; 2=highly unpredictable; 3=fairly unpredictable;4=fairly predictable; 5=highly predictable; 6=completelypredictable)

3.861.19

Control VariablesSMALL Dummy variable. 1=Small size firm (less than 50 employees);

0=all others0.470.50

MEDIUM Dummy variable. 1=Medium size firm (50 employees andabove but less than 500); 0=all others

0.350.48

FOREIGN Dummy variable. 1=Firm with some share of foreignownership; 0=all others

0.230.42



Page 87Variable Description MeanStd.

Dev.Control VariablesSINGLEPROPDummy variable. 1=Firm registered as single proprietorship;

0=all others0.280.45

PARTNER Dummy variable. 1=Firm registered as partnership; 0=allothers

0.160.36

PRIVATE Dummy variable. 1=Firm registered as privately held company;0=all others

0.350.48

LISTED Dummy variable. 1=Firm listed on a stock market; 0=all others 0.100.31TAXBURDEN The extent to which high taxes are problematic. Scale from 1

to 4 (1=no obstacle; 2=minor obstacle; 3=moderate obstacle;4=major obstacle)

2.971.06

were perceived by Asian firms as problematic (TAXBURDEN) has been used as a proxyfor the level of taxes firms encountered. The expectation is that the more problematictaxes are perceived to be, the more likely firms would be to choose low-qualityfinancial disclosure in order to avoid the problem.Given the nature of the dependent variables, three types of econometric model –probit, ordered probit, and interval regression – were used to test the threehypotheses regarding the role of public sector transparency in determining corporateaccounting practices in Asia. The probit model focuses on firms’ decisions to adoptInternational Accounting Standards, as well as on their decisions to use externalauditors. The ordered probit model and the interval model focus on the determinantsof the quality of corporate financial reporting, as measured by the percentage of salesfirms reported for tax purposes.In the probit model, it is assumed that the propensity for firms (yi*) to adopt goodaccounting practices is a function of a set of variables, including the test variables andcontrol variable. That is,

(1)where yi* is a ‘latent’ variable that cannot be observed directly. What is observed is adummy variable yi defined by

(2)In the present model, where yi is measured by IAS and AUDIT, the likelihood functioncan be written as



Page 88

(3)where F is the cumulative distribution function of u.In the ordered probit model, it is assumed that reported sales for tax purposes(yi*) isa function of a set of variables, including the test variables and control variable. Thatis,

(4)where yi* is a ‘latent’ variable that cannot be observed directly. What we observe is

The corresponding probabilities for each ordinal interval can be stated as follows:



Page 89And log-likelihood of the model can be specified as

(5)where and Zij is an indicator variable whichequals 1 if yi = j and 0 otherwise.The same as in the ordered probit model, the dependent variables in the intervalregression model – REPORTLOW and REPORTUP – are constructed from the firm’sresponses to the question on the percentage of the firm’s sales reported for taxpurposes. There are, however, two important distinctions. First, only firms reportingless than 100 percent of their sales are included in the estimation. Second, the actualpercentage terms, instead of categories, are used in the model. Seven brackets areconstructed, corresponding to firms reporting less than 25 percent, 25 to 49 percent,50 to 59 percent, 60 to 69 percent, 70 to 79 percent, 80 to 89 percent, and 90 to 99percent. They are, respectively, (0.0, 0.25), (0.25, 0.49), (0.5, 0.59), (0.6, 0.69),(0.7, 0.79), (0.8, 0.89) and (0.9, 0.99). The two numbers in each bracket indicate thelower (REPORTLOW) and upper (REPORTUP) bounds of the percentage of salesreported by the firm for tax purposes. The likelihood function for the intervalregression model can thus be expressed as

ResultsDescriptive statistics of variables in the model are shown in Table 5.5. Many firms hadadopted international accounting standards (48 percent) and had their annual financialstatements reviewed by external auditors (58 percent). The majority of these firmswere small and medium-sized enterprises (SMEs); about half being small (50employees) and one-third being medium-sized (51 to 200 employees). Close to aquarter of the firms in the sample involve foreign interests, an indication thatglobalization has made significant impacts on the landscape of firm ownership in Asia.About 40 percent of the firms in the sample were formed either as singleproprietorship or as partnership, and ten percent of the firms were listed on a stockexchange. This is not surprising given the dominance of SMEs in the sample.Table 5.6 shows the results of the probit model of firms’ adoption of accountingstandards. Columns 1–3 in Table 5.6 report the coefficients and standard errors ofprobit models for the adoption of International Accounting Standards; columns 4–6show the results for use of external auditors. The differences among Models 1, 2, and3 are in the measures of public sector transparency, and the same can be said aboutModels 4, 5, and 6. Because the measures of public sector transparency are highlycorrelated, as one might expect, we entered those measures into the models one at atime, to forestall multicollinearity problems.



Page 90Table 5.6 Probit models: adoption of accounting standards Dependent Variable: IAS Dependent Variable: Audit

(1) (2) (3) (4) (5) (6)SMALL –0.545*** –0.543*** –0.566*** –0.593*** –0.608*** –0.618*** (0.116) (0.116) (0.116) (0.139) (0.139) (0.139)MEDIUM –0.196* –0.222** –0.218** –0.442*** –0.466*** –0.471*** (0.108) (0.108) (0.107) (0.137) (0.136) (0.136)FOREIGN 0.730*** 0.744*** 0.719*** 0.742*** 0.699*** 0.743*** (0.094) (0.095) (0.094) (0.115) (0.115) (0.114)SINGLEPROP –0.763*** –0.792*** –0.803*** –0.985*** –1.015*** –0.965*** (0.145) (0.149) (0.146) (0.168) (0.172) (0.167)PARTNER 0.151 0.124 0.120 –0.104 –0.119 –0.085 (0.147) (0.149) (0.147) (0.164) (0.166) (0.163)PRIVATE 0.230 0.178 0.196 –0.011 –0.007 0.016 (0.141) (0.143) 0.141 0.165 0.166 0.163LISTED 0.455*** 0.374** 0.401** 0.823*** 0.730*** 0.808*** (0.170) (0.172) (0.170) (0.236) (0.238) (0.231)TAXBURDEN –0.072** –0.090** –0.074** –0.037 –0.044 –0.036 (0.036) (0.037) (0.037) (0.042) (0.042) (0.042)TRANSAVA 0.032 0.080** (0.030) (0.035) TRANSLAW 0.027 0.069*** (0.032) (0.038) TRANSINT 0.027 0.110** (0.032) (0.036)CONSTANT 0.591** 0.668** 0.636** 0.626* 0.670** 0.665* (0.296) (0.288) (0.300) (0.360) (0.342) (0.365)CountryDummies

Yes Yes Yes Yes Yes Yes

IndustryDummies

Yes Yes Yes Yes Yes Yes

Number ofObservations

1693 1682 1693 1593 1586 1593

Pseudo-R2 0.289 0.284 0.287 0.446 0.446 0.446Note: The table reports un-standardized coefficients, with standard errors inparentheses. *p < 0.1; **p < 0.05; ***p < 0.01.



Page 91Most control variables in these models are generally consistent with the priorpredictions as well as the findings of other empirical studies, although statisticalsignificance levels vary depending on model specifications. In the present analysis,variables on firm characteristics proved to be statistically significant in determining thefirms’ accounting practices. The coefficients on SMALL and MEDIUM show that biggerfirms were more likely to adopt accounting practices conforming to internationalnorms and expectations – a finding that is statistically significant for all models.Coefficients for FOREIGN indicate that firms with foreign ownership were more likelyto adopt International Accounting Standards and to use external auditors than firmsthat were domestically owned, a sign that globalization may have positive impacts onthe spread of better accounting practices. Expectations regarding the effects of firms’ownership structure are also confirmed in the results: firms formed as singleproprietorship were less likely to adopt International Accounting Standards or toemploy external auditors, and listed corporations were more likely to adopt highstandards of accounting practices. Lastly, although the negative coefficients onTAXBURDEN in all six models suggest that firms that reported resentment of hightaxes were less likely to adopt high standards in accounting practices, the perceptionof high taxes has greater impacts on the adoption of International AccountingStandards than on the use of external auditors.Public sector transparency, measured by predictability of policies, laws, andregulations, as well as their implementation, was shown to have the expected effectson both adoption of International Accounting Standards and use of external auditors,although statistical significance levels vary depending on the dependent variables inthe model. First, all three variables for public sector transparency registered positiveeffects on the adoption of International Accounting Standards and the use of externalauditors. Second, the coefficients for public sector transparency variables arestatistically significant for Models 4, 5, and 6 (for which AUDIT is the dependentvariable), but not for Models 1, 2, and 3 (for which IAS is the dependent variable).Third, the size of coefficients for public sector transparency is much larger in modelsfor AUDIT than those for IAS. Overall, effects of public sector transparency manifestedmore clearly in firms’ decisions to engage external auditors than in their decisions toadopt International Accounting Standards.As discussed earlier, the adoption of International Accounting Standards andinternational norms should not be interpreted as equivalent to having good-qualitycorporate accounting practices. Table 5.7 reports the results of the ordered probitmodel and interval regression model for firms’ actual performance in corporateaccounting practices: Models 7, 8, and 9 are ordered probit models for which thedependent variable is REPORT; Models 10, 11, and 12 are interval regression modelsfor which the dependent variables are REPORTLOW and REPORTUP. Firm size isshown to be inversely related to percentage of sales reported (except for coefficientson SMALL in Models 11 and 12), but the effects are not statistically significant for allmodels. Firm ownership matters: the coefficients on FOREIGN are statistically andeconomically significant in all six models, suggesting that foreign ownership hadpositive effects on the quality of firms’ actual accounting



Page 92practices. While the results show that the listed firms are more truthful in reportingtheir sales for tax purposes, the coefficients on SINGLEPROP and PARTNER show thatthe effects of ownership structure on the quality of actual accounting practices are notas clear-cut as on the adoption of International Accounting Standards and the use ofexternal auditors.Table 5.7 also shows that predictability of laws and regulations, and also transparencyof their interpretation, are positively correlated with sales reported for tax purposes.The coefficients of all three variables (TRANSAVA, TRANSLAW, and TRANSINT) notonly have expected signs on all models but are also statistically significant. It thusappears that firms had greater incentive to reveal accurate financial information wheninformation on laws and regulations was easier to obtain, when the change of lawsand regulations was predictable, and when the interpretation of these laws andregulations was transparent.DiscussionThe opaqueness of the public sector in Asia has been widely reported, and theempirical findings of the present analysis confirm that perception. Just over half (52percent) of the firms sampled for this analysis reported that changes in laws andregulations that affected their business activities ranged from “fairly unpredictable” to“completely unpredictable.” Although fewer firms reported difficulty in obtaining lawsand regulations – or unpredictability in interpretations applied to laws and regulations– those factors nevertheless comprised a sizeable share of the picture: one-third ofthe firms sampled held a negative opinion of these aspects of public sectortransparency.At the same time, many firms in the sample opted for low-qualitycorporate accounting practices. Only 42 percent reported 100 percent of their sales fortax purposes; in China, more than 50 percent of firms reported less than 60 percent.These findings strongly support other research that has recognized shady accountingpractices as a major determinant of the Asian financial crisis of 1997, and as a majorobstacle to sustaining the impressive growth of Asia’s corporate sector.The regression models presented here verify the three hypotheses that were statedwith regard to the effects of public sector transparency on firms’ decisions onaccounting practices, although strength of the relationship varies across models. Asexpected, the perceived difficulty of obtaining laws and regulations relevant tobusiness activities (TRANSAVA) had positive effects on adoption of InternationalAccounting Standards, the use of external auditors (AUDIT), and the percentage ofsales that firms reported for tax purposes (REPORT, REPORTLOW, and REPORTUP).These effects proved to be statistically significant for AUDIT, REPORT, REPORTLOW,and REPORTUP, but not for IAS. The same can be said about both the transparencyof laws and regulations (TRANSLAW) and the predictability of interpretation of lawsand regulations (TRANSINT).Overall, the effects of public sector transparency on the adoption of InternationalAccounting Standards turned out to be quite weak (none of the coefficients isstatistically significant), in contrast to results for use of external auditors (all three



Page 93Table 5.7 Ordered probit models and interval regression models: accounting practices Dependent Variable:

REPORTDependent Variable: REPORTUP andREPORTLOW

(7) (8) (9) (10) (11) (12)SMALL –0.073 –0.089 –0.091 0.006 0.004 0.006 (0.095) (0.094) (0.094) (0.024) (0.024) (0.024)MEDIUM –0.026 –0.050 –0.059 –0.002 –0.002 –0.003 (0.088) (0.087) (0.087) (0.023) (0.022) (0.022)FOREIGN 0.237***0.205***0.234*** 0.045** 0.037* 0.042** (0.074) (0.075) (0.074) (0.019) (0.019) (0.019)SINGLEPROP –0.023 0.058 0.026 –0.009 0.000 –0.008 (0.118) (0.118) (0.117) (0.029) (0.029) (0.028)PARTNER 0.216* 0.251** 0.245** 0.061** 0.058* 0.056* (0.123) (0.123) (0.122) (0.030) (0.030) (0.030)PRIVATE 0.182 0.222* 0.200* 0.054* 0.052* 0.051* (0.115) (0.115) (0.114) (0.028) (0.028) (0.028)LISTED 0.274* 0.358** 0.300** 0.083** 0.080** 0.077** (0.141) (0.141) (0.139) (0.036) (0.036) (0.036)TAXBURDEN –0.011 –0.030 –0.015 0.010 0.008 0.010 (0.029) (0.029) (0.029) (0.007) (0.007) (0.007)TRANSAVA 0.120*** 0.021*** (0.024) (0.006) TRANSLAW 0.055** 0.017*** (0.025) (0.006) TRANSINT 0.094*** 0.018*** (0.026) (0.006)CONSTANT 0.445*** 0.471*** 0.451*** (0.058) (0.057) (0.061)Country Dummies Yes Yes Yes Yes Yes YesIndustry Dummies Yes Yes Yes Yes Yes YesNumber ofObservations

1519 1513 1519 1032 1029 1036

Pseudo-R2 0.063 0.059 0.060 Note: The table reports un-standardized coefficients, with standard errors inparentheses. *p < 0.1; **p < 0.05; ***p < 0.01.



Page 94coefficients are both statistically and economically significant) and quality of actualaccounting practices measured as percentage of sales that firms reported for taxpurposes. The disparities across models suggest that the effects of public sectortransparency were more pronounced in corporate information disclosure. Both the useof external auditors and the percentage of sales reported for tax purposes involvedthe revelation of corporate financial information to outsiders, whereas adoption of IASdid not.Concluding remarksIn the aftermath of the Asian financial crisis of 1997, many Asian countries haveundertaken accounting reforms to strengthen the quality of accounting practices.Typical reform measures have involved the adoption of International AccountingStandards and independence in audit. The success of these accounting reforms willnot only determine the business competitiveness of Asia but will also influence thequality of public sector governance (Wu 2005).The evidence presented here demonstrates the mutual dependency between publicsector governance and corporate governance, and reveals mechanisms through whichsuch interdependency might be harnessed to enhance the effectiveness of governancereforms. Quality of corporate financial reporting was shown to be positively related topublic sector transparency, as measured by the predictability of policies, laws, andregulations, as well as their interpretations in implementation.Several implications arise from these findings. First, firms may choose to reduce theinformativeness of their financial reporting, creating information asymmetry as acounterbalancing strategy against perceived information asymmetry in public sectoractivity. Second, lack of transparency in the public sector reduces the incentives forfirms to improve the quality of accounting practices. Third, lack of transparency in thepublic sector increases business costs by imposing high transaction costs upon firms,to the extent that some may be forced to evade tax responsibility through fallaciousaccounting practices in order to survive. Fourth, lack of transparency offers moreleverage to predatory public officials who seek to extract rents from businesses; firmsmay choose to underreport profits through fallacious accounting practices in order toavoid being targeted. Fifth, lack of transparency also reduces the costs of firms’ non-compliance with public directives, especially with regard to tax laws and regulations.This analysis of the linkage between public sector transparency and corporateaccounting practices offers several important policy lessons for designing andimplementing accounting reforms in Asia. First, it confirms that quality of corporateaccounting practices is constrained by quality of public sector transparency. Policymakers in the region should take public sector transparency into consideration whendesigning and implementing accounting reforms. Different countries will face differentchallenges in accounting reforms, depending on how their corporate systems matchup with prevailing political environments. This variability demands varied solutions.Ignoring such interrelationships between public sector



Page 95transparency and corporate accounting practices may reduce the relevancy of reformpolicies.Second, governments should not commit to specific, comprehensive reform measuresprematurely. Much of the real progress that occurs might depend on what happens tofirms’ political environments. Firms’ corporate governance practices are largely shapedby forces outside their corporate boardrooms. There is little chance that adoption ofInternational Accounting Standards will lead to high-quality accounting practices aslong as firms and their political patrons both have vested interests in keeping thepublic sector opaque to protect existing rent-seeking schemes. For example,tightening standards for publicly listed firms prematurely may serve as a disincentivefor firms to become listed; they may become convinced that it is impossible to meetthe proposed standards under opaque public sector conditions.Third, the findings presented here do not at all imply a narrowing of available choicesfor measures of reform. In fact, they broaden the scope for effective measures forcorporate financial disclosure by placing a new set of instruments, focused oncorporate operating environments, at the disposal of those charged with projectingand implementing reforms. Pointed efforts can now be made to improve both theeffectiveness of accounting reforms and the success of anticorruption campaigns byenhancing public sector transparency.ReferencesAsian Development Bank (ADB) 2001, Corporate governance and finance in East Asia.Archambault, J. J., and Archambault, M. E. 2003, ‘A Multinational Test ofDeterminants of Corporate Disclosure.’ International Journal of Accounting, 38: 173–94.Ball, R. 2001, ‘Infrastructure Requirements for an Economically Efficient System ofPublic Financial Reporting and Disclosure.’ in R. Litan and R. Herring, eds. Brookings-Wharton Papers on Financial Services, Washington, DC: Brookings Institution Press:127–83.Ball, R., A. Robin and J. Wu 2003, ‘Incentives Versus Standards: Properties ofAccounting Income in Four East Asian Countries, and Implications for Acceptance ofIAS.’ Journal of Accounting and Economics, 36(1–3): 235–70.Bellver, A., and D. Kaufmann 2005, ‘Transparenting Transparency: Initial Empirics andPolicy Applications.’ World Bank Policy Research Working Paper.Bushman, R., and A. Smith 2001, ‘Financial Accounting Information and CorporateGovernance.’ Journal of Accounting and Economics, 32: 237–333.Bushman, R., J. Piotroski and A. Smith 2004, ‘What Determines CorporateTransparency?’ Journal of Accounting Research, 42(2): 207–252.Choi, J. 2002, ‘Financial Crisis and Accounting Reform: A Cultural Perspective.’ Journalof Accounting and Finance, 1: 77–93.Claessens, S., and J. Fan 2003, ‘Corporate Governance in Asia: A Survey.’International Review of Finance, 3(2): 71–104.Clarke, G., and L. Xu 2004, ‘Privatization, Competition and Corruption: HowCharacteristics of Bribe Takers and Payers Affect Bribe Payments to Utilities.’ Journalof Public Economics, 88(9–10): 2067–97.Gelos, R., and R. Wei 2003, ‘Transparency and International Investor Behavior.’Working paper, National Bureau of Economic Research.Gray, S. 1988, ‘Towards a Theory of Cultural Influences on the Development ofAccounting Systems Internationally.’ Abacus, 24(1): 1–15.



Page 96Jaggi, B., and P. Low 2000, ‘Impact of Culture, Market Forces, and Legal System onFinancial Disclosures.’ International Journal of Accounting, 35(4): 495–519.Johnson, S, P. Boone, A. Breach and E. Friedman 2000a, ‘Corporate Governance in theAsian Financial Crisis, 1997–1998.’ Journal of Financial Economics, 58: 141–86.Johnson, S., D. Kaufmann, J. McMillan and C. Woodruff 2000b, ‘Why Do Firms Hide?Bribes and Unofficial Activity After Communism.’ Journal of Public Economics, 76: 495–520.Kimbro, M. 2002, ‘A Cross-country Empirical Investigation of Corruption and itsRelationship to Economic, Cultural and Institutional Variables: An Examination of theRole of Accounting and Financial Statements Quality.’ Journal of Accounting Auditingand Finance, 17(4): 325–349.La Porta, R., F. Lopez-de-Silanes, A. Shleifer and R. Vishny 1998, ‘Law and Finance.’Journal of Political Economy, 106: 1113–55.Leuz, C. and F. Oberholzer-Gee 2005, ‘Political Relationships, Global Financing, andCorporate Transparency.’ Working paper, Wharton School of Business, University ofPennsylvania.Lin, Z. and F. Chen 2000, ‘Asian Financial Crisis and Accounting Reforms in China.’Managerial Finance, 26(5): 63–79.Liu, Q. and G. Xiao 2004, ‘Look Who’s Disguising Profits?’ mimeo, Hong KongUniversity.Low, P. 1998, ‘The Effects of Agency and Proprietary Cost on Corporate FinancialDisclosures.’ Working paper, City University of Hong Kong.Mitton, T. 2002, ‘A Cross-firm Analysis of the Impact of Corporate Governance on theEast Asian Financial Crisis.’ Journal of Financial Economics, 20: 293–315.Morris, R., B. Ho, T. Pham and S. Gray 2004, ‘Transparency of Financial ReportingBefore and After the Asian Financial Crisis: An Empirical Study of Indonesian CompanyPractices.’ Asia Pacific Journal of Accounting and Economics, 11(2): 1–26.Rahman, Z. 1998, ‘The Role of Accounting Disclosure in the East Asian Financial Crisis:Lessons Learned.’ Division on Investment, Technology and Enterprise Development,Enterprise Development Strategies, Finance and Accounting Section, Geneva: UNCTAD(United Nations Conference on Trade and Development).Rose-Ackerman, S. 2002, ‘“Grand” Corruption and the Ethics of Global Business.’Journal of Banking and Finance, 26: 1889–918.Rosser, A. 2003, ‘Globalisation, International Norms, and the Politics of AccountingReform in Indonesia.’ in G. Underhill and X. Zhang, eds. What Is To Be Done? GlobalEconomic Disorder and Policies for a New International Financial Architecture.Cambridge: Cambridge University Press, 263–82.Svensson, J. 2003, ‘Who Must Pay Bribes and How Much? Evidence from a Cross-section of Firms.’ Quarterly Journal of Economics, 118(1): 207–30.Vishwanath, T. and D. Kaufman 1999 ‘Towards Transparency in Finance andGovernance’, Working Paper, The World Bank.Vogl, F. 1998, ‘The Supply Side of Global Bribery.’ Finance and Development, 35(2):30–33.Werlin, H. 2002, ‘Politics Versus Economics: A Comparison of Ghana and South Korea.’Journal of Social, Political and Economic Studies, 25(4): 439–64.Wu, X. 2005, ‘Firm Accounting Practices, Accounting Reform and Corruption in Asia.’Policy and Society, 24(3): 53–78.Zarzeski, M.T. 1996, ‘Spontaneous Harmonization Effects of Culture and Market Forceson Accounting Disclosure Practices.’ Accounting Horizons, 10(1): 18–37.



Page 976 The Hara factorSome lessons from innovations in governance in JapanJorrit de Jong and Arre Zuurmond1Innovations: policies, projects and peopleWhy do innovations occur when they occur? Is it just because the people involved puttheir minds and willpower to it and make it happen? Or can they be successfulbecause they have certain social, institutional and technological resources available tothem? To what extent is a successful innovation the result of explicit efforts to make achange, and to what extent are there implicit forces at work? How do these explicitand implicit forces interrelate?There is a vast range of literature on innovations in organisations. Comparative andin-depth studies of innovations have yielded many insights; for example, in the wayinnovators work, the way systems react to innovations and the strategies to managethe environment. There is, however, not much knowledge about the relationshipbetween variables in the innovation’s environment and the innovation’s success –especially not when the unit of analysis is not the individual organisation but anetwork of organisations with a public task or responsibility.In this study2, based on extensive site visits and interviews with individuals involved,we analyze a network of service delivery related to electronic medical records (EMR)in Japan, in which innovations have been successfully developed and implemented.We ask the following questions: Which factors in the cultural, institutional,organisational and technological environment of the innovation can be distinguished?How does the success of the project relate to the general innovation policies orreform agendas? And what does that tell us about the possibilities of promotinginnovation through top-down initiatives? This study is explorative and presentsqualitative research from only one case study. However, since some of the results arenot concurrent with commonly held beliefs about innovation, the conclusions of thiscase study may stimulate the academic and political debate on innovation policy andmanagement.Hara’s innovations: HelloBaby and the K-Mix systemMedical professionals, especially in hospitals, are renowned for their desire forprofessional autonomy. They rarely use each other’s data or judgment. Their workingprocesses and administration are organized mostly around their own



Page 98expertise, not around the patient’s problem. Last but not least, every kind of doctorhas a particular view on the patient, with subsequent taxonomies and datacategorizations. To put it boldly, the radiologist thinks in X-rays, the pharmacist thinksin recipes, the surgeon thinks in incisions and stitches, and the nurse thinks in theamount of minutes she can spend on a patient. The intentions of all individual actorsare to act in the best interest of the patient, but they all tend to adopt a partial view.This often leads to a fragmented image of the patient, inefficient and ineffective datatransfers, time-consuming procedures and many mistakes, varying from minormiscommunications to fatal errors. The central challenge of this paper is to explorethe question of what it means to innovate in such a context. We first describe theinnovation associated with Professor Hara and his network.Professor Hara is a quiet, modest man. The 59-year-old gynaecologist from the islandof Shikoku, Japan, has had a long and successful career, in academia as well as inpractice. He was a pioneer in the field of medical informatics before it was even seenas a field in its own right. In recent years Hara has brought together hisgynaecological expertise and his ideas on medical informatics.Specifically, his HelloBaby system is a great success in Japan. It is a web-based portalthrough which general practitioners, physicians and nurses can access data onmothers and their unborn children. The availability of electronic medical records forperinatal care and delivery has increased efficiency and effectiveness ofcommunication between professionals across organizational boundaries. Moreimportantly, health and safety of mother and child benefit from the system. If womenexperience complications during delivery, they often need to be transported tohospitals with specialized care. Every minute counts in these situations. So if nurses,surgeons, anaesthesiologists and gynaecologists can start preparing surgery and carewhile the patient is still under way, chances of survival are increased. Although somewould think that cooperation and information exchange between medical professionalsjust makes sense, the HelloBaby system is a breakthrough innovation.Hara’s HelloBaby system is in fact part of a bigger network: the K-Mix system. Itconnects clinics and hospitals electronically. It was designed by Hara and his teamand further developed by a network of organizations, including vendors, hospitals andthe medical association.In his native prefecture of Kagawa, Shikoku, there are small islands in front of thecoast, and no easy access to health care. Kagawa is a small prefecture, but thepeople are very scattered. Most people live in Takamatsu, the big city, but many livein small islands that are hard to reach. Specialized care and expensive facilities arenot available on the small islands. Kagawa favors an equal distribution of health carefacilities. The K-Mix system is a major contribution because it enables nurses andphysicians to communicate over long distances. They can share X-ray files, consulteach other on diagnoses and keep digital medical records, such as HelloBaby. But theelectronic communication has many more advantages. It reduces errors, enhancescooperation between different medical professionals, and saves costs. The valueproposition was clear to the innovators. The authorizing environment, however, wasdivided. Some liked the long-distance aspects, some liked the reduction of medicalerrors, and others were interested in cost saving.



Page 99In Kagawa University Hospital, Takamatsu, an electronic patient record system hasbeen implemented. Nurses and doctors use nearly no more paper records to filemedical information about their patients. There are 600 beds and 600 laptops in thehospital. There is a wireless network that connects every laptop to both the internalserver and the internet. On the internal server, a software program called K-Mindenables nurses and physicians to enter, edit and view data about a patient’s past orpresent condition and about the drugs and treatments received. There are levels ofauthorization, but virtually all doctors can view each other’s information. Theintroduction of the technology has led to other work processes: the role divisionbetween doctors and nurses has changed, and facility management has changed aswell. The supply chain (pharmacies, laboratories, home care, etc.) has beenredesigned and streamlined.The department of medical informatics, headed by Professor Hara, is working onseveral projects concerning medical information exchange. The ultimate purpose is tomake K-mind web-enabled and to have a true electronic medical record, accessible byall medical institutes that have or need to have information about the patient. Inorder to make that possible, Professor Hara’s department has developed K-Mix withthe Prefecture Government of Kagawa, in association with Fujitsu, the software vendorthat created K-Mind and ST-Net, a data warehouse providing hosting and applicationservice providers (ASPs). K-Mix is an XML- and HL7- based exchange protocol thatdefines standards for the exchange of medical information, such as images, statistics,test results, etc. Even though K-Mind is not integrated with many systems yet, anumber of applications are already in use: HelloBaby (perinatal information exchangebetween nurses, general practitioners and gynaecologists), Remote Visual ConsultationSystem (for follow-up and second opinion) and a shared information system forinternal medicine.Hara operates with a grand vision, but with small steps at a time. By building small,useful applications for target groups (such as radiologists and gynaecologists), hegains critical mass based on positive experiences. He uses this as leverage to obtaininvestments in the infrastructure at large (e.g. K-Mix). With the infrastructure in placeit is easier to convince larger and more reluctant target groups and stakeholders tojoin the project.The introduction of information sharing systems has, according to the hospital directorand the medical informatics department, led to improved health care, both in terms ofeffectiveness and efficiency. Through remote consulting, specialization is enforced.Doctors can easily consult each other on difficult cases, so they don’t need to bespecialized in everything. Waiting lists have shortened because logistics have improvedand administrative burdens have diminished. Because work processes can beredesigned, nurses can do more of the intake and leave the diagnosis and treatmentto the doctors.Professor Hara and the four mythsThe innovations described above raise intriguing questions. Professor Hara is withoutdoubt an innovator. He is completely devoted to inventing new products,



Page 100procedures, treatments and technology. But he is more than an inventor in alaboratory. He is just as focussed on his environment as on his innovations. He useshis context as a resource for his actions. He does not change his project in order toobtain money from innovation budgets; but he obtains money from budgets that arenot at all labelled as innovation funds. He does not take risks with his innovations, buthe defines the status quo as risky and his innovations as risk-reducing solutions. Hedoes not feel paralyzed or limited by his rigid authorizing environment; rather, heactivates it by subtly but decidedly shifting the burden of proof to his superiors. Andby doing so, he seemingly unconsciously debunks four common beliefs aboutinnovation, myths that would appear to be contradicted by his example.Myth number one: innovation is stimulated mainly through strong financialmanagement and competitionThe advent of New Public Management (NPM) as a paradigm for governmentperformance has led politicians and managers all over the world to believe thatinnovation and improvement are driven by business-like approaches such ascompetition, financial incentives, planning and control, and benchmarks. “Rungovernment like a business” was the idea that influenced many governments in the1980s and 1990s. NPM is based on public choice and management theory, usingeconomic concepts and control-type interventions along with a strong focus on“entrepreneurship” (Osborne and Gaebler 1993). Its main hypothesis is that a strongmarket orientation enhances efficiency and effectiveness of the public sector.Consequently, New Public Management pre-supposes public managers who complywith the business directives and who are motivated by (financial) incentives.In our case, we witness something completely different. Professor Hara is insensitiveto financial incentives and top-down control. His drive is not to enhance the efficiencyof the present system, but to enhance effectiveness as defined from the perspectiveof the citizen-patient. Professional ethics and passion are his driving force. From thatperspective he operates very cautiously, since the innovation he envisions might beproblematic in terms of the existing financial control mechanisms. Furthermore, hedoes not behave like a leader, at least not an outspoken leader of the troops. Instead,he works informally, in networks, giving credit to others in his network. He operateslike a diplomat, under the radar.Myth number two: if there is a budget, there will be innovationThe introduction of digital network technology has made major improvements incommunication processes possible. Many hospitals and other medical institutions haveimplemented ICT systems to optimize their business processes. However, given theobservations above, it is not surprising that the automatization of insulated practiceshas deteriorated rather than improved cooperation and data exchange in health care.The fragmentation in the medical sector has been consolidated through incompatibleelectronic systems. This has been noticed by the Japanese



Page 101government and e-health was made a key issue in the prestigious E-Japan reformagenda of the Koizumi Administration.This agenda was coordinated by the Cabinet Office, headed by Prime Minister Koizumihimself. Hundreds of millions of yen were allocated to develop medical informatics andstimulate innovations in the sector. The Ministry of Health, however, was reluctant tospeed up the process. Officials there knew how complicated the task was, and werestill studying the best way to go about it. They did not want to invest randomly inspecific innovations, but rather in a well-coordinated effort to develop and implementcomprehensive medical record systems. As a result, criteria for subsidies weredeveloped at a system level.Professor Hara’s bottom-up approach did not quite match those criteria. That did notstop Hara or redirect his course. He just redefined his project. The Ministry ofAgriculture had a budget for rural development. It subsidized the development ofinfrastructure for people living in remote areas. Hara’s HelloBaby project could beseen as an infrastructural project. After all, the distances between a small clinic in avillage and the hospitals in the city are large. Emergency situations during childdeliveries in the countryside are therefore more serious. People in rural areas needHelloBaby more than people in cities do, reasoned Hara. He persuaded the Ministry ofAgriculture, and got funding for the development of his medical informaticsinnovation.“Yes”, says Hara, “money is important. But the idea that budgets for innovationsserve their genuine purposes is a myth. Practice is always ahead of policy and budgetdefinitions.” Therefore, innovators need be creative not only in their own field ofpractice, but also in their fundraising strategies.Myth number three: innovators are risk takersInnovation implies change. Change implies uncertainty. Uncertainty is often perceivedas a risk. Therefore, innovation is associated with risks. Another way people thinkabout innovation is in terms of experiments. One never knows in advance if anexperiment is going to work. The trial and error approach again requires theacceptance of risks. Risk-averse people, organizations or environments are thus notlikely to engage in innovations. Much literature on innovation in the public sectorargues that the public sector in general is more risk averse than the private sector(Borins 1998). This is partly because of political interests and partly because thepublic sector has a responsibility for the continuity, stability and reliability of publicservices that directly affect people’s lives.These bases for risk aversion may be particularly applicable in the medical sector,agrees Professor Hara. But that is why innovation to reduce risks is so important.Every improvement in treatments or procedures that reduces risks is encouraged. InJapan, people don’t forgive doctors who make medical mistakes. That is why doctorsare so interested in the HelloBaby system: it not only reduces costs, it reduces thechances of medication or treatment errors. Says Hara: “I never take risks. I test everyinnovation over and over until I am sure it works. My credibility as an innovatordepends on my risk-reducing capabilities vis-à-vis my colleagues.”



Page 102Myth number four: some institutional environments are better than othersJapan is known to be a society with a rigid, hierarchical social structure. The civilservice system is no different: every single official knows his place and bosses havevirtually unquestioned power over their subordinates. There is no way a subordinatewould or could bypass their boss. Apart from the official governance structures, thereis an informal pecking order based on age, status and reputation, which is known andrespected by everyone.This kind of social-institutional environment does not at first sight seem veryconducive to innovation. For example, to get a mandate for cross-boundaryinformation exchange, one needs to get permission from one’s superiors and, throughthese superiors, permission from the responsible officials in all other pillars ororganizations involved. The transaction costs add up with every required signature.Again, Professor Hara turns the problem around: according to him, the accountabilitystructure is an advantage for innovation. He simply persuaded his boss, the directorof the Takamatsu University Hospital, that he would not be able to perform optimallyif he could not innovate; if there is a way to perform better, medically speaking, andit is not implemented, one is not acting responsibly. Professor Hara put the burden ofproof on his authorizing environment, not on his innovation. The rigid institutionalframework now became a blessing: everyone who needed to account for theirperformance joined Hara’s innovation efforts, simply because they would not be ableto explain to their superiors why they did not use the innovative alternative.Mind the context: theories on public sector innovationOne might summarize the above by suggesting, albeit colloquially, that Professor Haradoes not get “played” by the context; he plays it. As we have shown in the textabove, some essence of innovation lies in the way the innovator interacts with hisenvironment. For now we will call this the Hara factor. We will build further on thislater.First, however, we want to take a closer and more analytic look at the context ofinnovations in general before applying them to HelloBaby and other innovations in themedical sector in Shikoku. What elements does the environment of an innovationconsist of? Analytically speaking, what variables are to be distinguished and how dothey interrelate? In the pages that follow, we lay out a number of definitions andpropositions about how innovations may relate to their broader environment, basedon our reading of the literature. We then return to see what light the single casestudy we have presented may shed on these propositions, knowing full well that theanswers in the framework of the current study are at best meant to provoke andstructure further discussion.InnovationMany lessons can be drawn from research in the private sector, where innovation haslong been an object of study. Borins (2002), among others, however, points out



Page 103that there are genuine differences in the circumstances under which innovations occurin the private and the public sector. But also, within the public sector innovation is nota uniform and unambiguous concept. Hartley (2005) sums up different possible loci ofinnovation: product level, process level, service level, position level, strategic level,governance level and rhetorical level. One could conclude from this, that public sectorinnovation in general is au fond a multilevel game. Moreover, depending on theconception of public administration and governance, innovation is interpreteddifferently, according to Hartley. In contrast to New Public Management or citizen-centered government, traditional public administration puts the task of innovation inother places and in the hands of other people. This means that the dominantconception or paradigm of public administration and governance determines on whatlevel and by what actor innovative action is to be expected.Pollit and Bouckaert (2004) have analysed how national governments have shapedtheir reform agendas. They argue that the politico-administrative system of a countryis an important determinant for the feasibility of change and reform, but they arereluctant to draw general conclusions. They note that most literature on innovationlacks a theory of context, which is a conceptual framework that makes innovation orthe lack of innovation in one context comparable to others. Moore (2005) drawsattention to another theoretical challenge: how to relate particular innovations to theorganizational environment in which they occur. He distinguishes the breakthroughmodel of analysis, which focusses on the actual occurrence of specific importantinnovations, and the continuous improvement model, which focusses on learningorganisations and their ability to continually innovate. Moore stipulates someinteresting open questions within both models, but does not solve the theoreticalincompatibility of the two models.It is too much of a modernist-positivist ambition to try to assemble one unifyinggrand theory of innovation. But it might scientifically be worthwhile to discover towhat extent the existing theories are complementary or contradictory to each other.Are innovations at different levels, in different paradigms, by different actors allexpressions of the same conscious desire for change and improvement? And if so,how are the practices of innovation at various levels interrelated?In this study we are interested in specific innovative action, in motivations and inmethods. We want to know how an evidently successful innovation such as thatpromoted by Professor Hara has occurred; that is, what the crucial factors were thatled to their occurrence. Was it the political or administrative system? Theorganisational culture? A crisis? The right people in the right place at the right time?Or was it a collusion of all those factors?Institutional arrangementsEven though developments in modern states cannot be seen separated from theglobal context in which they emerge and interrelate (Castells 1999), the nation stateis still the main governmental framework of action for most countries. Problems maybe caused by many external factors; they still have to be solved



Page 104primarily by national and local governments, markets and civil societies. Thegovernmental system of a country is therefore perhaps one of the most relevantpolitico-administrative contexts of the innovations we study in this project. Pollitt andBouckaert (2004) have distinguished several features of this context from a publicsector reform perspective: the state structure, the nature of executive government atthe central level, the relationships between political executives and top civil servants,and the administrative culture.The state structure reflects the dispersion of power in a country, both in verticaldimensions (relation between central and local government) and in horizontaldimensions (separation of fusion between powers). Characteristics include votingsystems, powers and responsibilities of the legislative and executive bodies and thejudiciary, etc. The degree of centralization might have an especially major influenceon the way the system as a whole is likely to handle innovations and need forchange. The more centralized a state structure is, the more likely changes will beimposed top-down, and the less likely it is that bottom-up innovations will emerge.Following Lijphart’s (1999) typology of majoritarian and consensual democracies, onecould argue that consensual regimes produce less rigorous management reforms thanthe majoritarian governments, because they have to negotiate and compromise onideas and ambitions. The more majoritarian a model of government is, the more likelyinnovations are to be implemented.A great deal of public management reform, especially in those countries that wereinfluenced by New Public Management (NPM), entails redistribution of responsibilitiesand governance arrangements across sectors. Especially in health care, services inmany countries are delivered by public, private and even voluntary sectororganisations. Motivations and reasons to allocate funds and responsibilities to onesector or the other may differ (e.g. expected efficiency gains, quality control, inducingchoice or setting standards). One thing is clear, however: if innovation is to take placein, for example, a network of health care providers, the more actors involved and themore heterogeneous they are, the more difficult it is to get them together to decideand implement. The degree of difficulty is influenced by the specific governancearrangement, consisting of the legal framework and the funding regime. But evenmore than those elements (legal framework often leaves room for interpretation andfunding regimes can be changed), the dominant conception of public value of thenetwork matters: the more citizen oriented a network of service delivery is, the morelikely organisations within the network succeed in achieving results in improvingservice delivery. We would like to advance the following propositions:• The more majoritarian a model of government is, the more likely innovations are tobe implemented• The more centralized a state structure is, the more likely changes will be imposedtop-down, and the less likely it is that bottom-up innovations will emerge• The more mandarins are directly accountable for their efforts to innovate orimplement innovations, the more adaptive the system as a whole is to change



Page 105• The more citizen-oriented a network of service delivery is, the more likelyorganisations within the network are to succeed in achieving results improving servicedelivery• Funding regimes are major bodies of resistance, and consequently crucial obstaclesor – when redesigned - vehicles for changeChange managementPublic management reform has to do with deliberate changes to the structures andprocesses of public sector organizations with the objective of getting them (in somesense) to perform better. These changes are informed by specific sets of ideas orideologies. And many times, they are induced by specific change events: a fiscalcrisis, a scandal, a disaster, etc. Many authors (Behn and Altshuler 1997; Hartley2005; Pollitt and Bouckaert 2004) have stressed the fact that a considerable amountof change is only existent in the realm of political rhetoric. In other words: much adoabout nothing. Politicians, especially when running for office, make promises,announce changes and propose innovations. Even if the politico-administrative systemthey are in provides them with the powers to actually turn their plans into action, it isnot guaranteed that they put their money where their mouth is. Also, civil servantsmay have considerable power to obstruct change. Those who know the TV series Yes,Minister will have recognized the change-reluctant and arch-conservativemachiavellian Sir Humphrey Appleby in many real-life cases where the civil servicesimply disguised disloyalty as impossibility. It is therefore important, when studyinginnovation, not only to look at the talks or plan, but rather at the way these wordsand plans are put into action. We then enter the broad domain of changemanagement.Moore (2005) has made a distinction between the innovation as a breakthrough andthe innovation as a logical expression of a learning organisation continually pursuingimprovement. Both models reflect different styles of change management. Thebreakthrough model could be interpreted as an innovation that occurred despite themanagerial regime, whereas the incremental model has a management style thatpromotes and fosters innovation. As mentioned before, the level or locus at which theinnovation is taking place could very much be related to the people who are theprincipal change agents. For example, when an innovation is only at the product level,organisational structure might not be altered and the principal change agents wouldbe the people in the research and development section. But when an innovation atgovernance level is at stake, it is most likely that a whole range of senior officials andeven politicians are involved in the change management process. The level (e.g.process level, product level, institutional level, etc.) at which the innovation takesplace determines its scope and significance. For instance, Professor Hara’s HelloBabysystem may look like an innovation at products level (a website), but when you take acloser look it is an innovation at process level (work processes have been changed)and governance level (a horizontal, inter-organizational network has been established)as well.Many countries, states, provinces, cities and sectors have explicit reform



Page 106agendas. A reform agenda is a change program with more or less specified goals,objectives and means. Depending on the dominant political and managerial ideas, atypical reform agenda might set out to decrease administrative burdens forentrepreneurs, improve service delivery performance, deregulate specific domains inthe public sector and the market, and achieve efficiency gains. In that sense, theyrepresent the recent interpretations of NPM.The management reform agendas in any particular country will almost certainly beshaped by the local preoccupations and priorities of the politicians and private actorsmost concerned. Even if the reform agenda is declared at a higher level ofgovernment, lower levels of government and private or voluntary sector parties will,most of the time, still be needed to actually make it happen. Reforms occur atdifferent levels and may be of a broader or lesser scope. And different governmentscan propose different reform plans at the same time. Given the fact that mostinnovations in the public sector tend to bubble up from below (Stone 1997; Borins2002), it is not at all clear what the actual relation between an agenda and the actualoccurrence of innovations is, let alone their successful implementation.Ways to put agendas into action and means to achieve the goals they contain arecalled trajectories (Pollitt and Bouckaert 2004). A trajectory is more than a trend;trajectories are intentional. They lead from a starting point to some desired place orstate of affairs in the future. A scenario consists of three basic elements: an initialstate, a trajectory, and a future state. A scenario is not always complete. Pollitt andBouckaert are of the view that the real world is less tidy; there are poorly specifiedvisions of the future, inadequate analyses of the status quo, and partial andsometimes conflicting or oscillating trajectories for different aspects of theadministrative apparatus. Four components of the substance of reform that make up atypical trajectory are to be distinguished: finance, personnel, organization, andperformance measurement. For the process of reform there are three more features:top-down/bottom-up, legal dimensions and organizational processes. Since the wholeuse and effect of a trajectory depends on its consistency and managerial feasibility,one could say that the more elaborated and focussed a reform trajectory, the moreimpact a reform agenda has on specific innovations.Borins (1998, 2002) has analyzed over 150 innovations in government, focussing onthe role innovators play as principal change agents. He concentrated on the buildingblocks for innovation and looked for patterns in strategies of innovators. Given thefact that innovations always cause confrontations and conflicts with existing structuresand practices, innovators often meet obstacles that they have to overcome. Therepertoire of intervention and conflict resolution techniques is therefore an essentialvariable in the study of change management. We would like to advance the followingpropositions:• Breakthrough innovations typically occur in rigid organisational structures• Incremental innovations typically occur in adaptive, innovative and learningorganisations• National reform agendas, seen as recent interpretations of the NPM movement,



Page 107have very little impact on the successful implementation of innovations in servicedelivery networksSocial capitalThere are many ways to look at organizations, and the way you look at themdetermines what you see (Morgan 1994). Looking from an institutional perspective,you see structure. Looking from a management perspective, you see strategy. Lookingfrom a social capital perspective, you see human relations. The explanation of socialcapital by Lin (2001) is consistent with various renditions by scholars who havecontributed to the discussion (Lin and Marsden 1982; Coleman 1988; Flap 1989,1994; Burt 1992; Putnam 1993, 1995; Erickson 1995, 1996); namely, “investment insocial relations with expected returns in the marketplace”.Social capital, as a relational asset, must however be distinguished from collectiveassets and goods such as culture, norms, trust, and so on. In his interpretation ofsocial capital, Lin (2001:29) distinguishes three critical components for analysis: (1)the resources, (2) being embedded in a social structure, and (3) action.In the conceptualization of this research, social capital may be defined operationally asthe resources embedded in a social structure that are accessed and/or mobilized byactors in purposive actions. Resources are possessed by actors that together form asocial structure. The pivotal proposition that several social capital theorists advanceconcerns the expected return on investment of social capital. Better access to socialcapital and better use of it yields better outcomes for those who possess or use it (Linet al. 1981; De Graaf and Flap 1988; Granovetter 1995). Several scientistsinvestigated the effect of social capital on the obtainment of a job. The studies showthat social capital enhances the likelihood of getting better jobs. For example, in ananalysis of the National Longitudinal Survey of Youth Labor Market Experience,Wielgosz and Carpenter (1987) found that informal social networks reduced theamount of time applicants needed to find a job. The weak ties of a social network(acquaintances, friends of friends, colleagues you don’t see frequently, or occasionalbusiness contacts) appear to be especially important for getting and spreadinginformation about resources and opportunities. The strong ties (close friends, long-time colleagues, etc.) have less information to add to one’s knowledge, but providemore substantial resources such as support, money, facilities, labor, etc. Anotherimportant aspect of social networks is redundancy or overlap: the more diverse one’snetwork, the more competitive one’s social capital. People who have friends in morethan one “scene” or professional circle have more resources available to them and arekept in more than one information loop. Most of the social capital studies examine thereturn on an individual or a societal level. The focus of this research is the network oforganizations involved in the innovation process. In other words, what are the resultsof social capital of a network in terms of effectiveness in achieving the goals of thenetwork? With respect to the central question of this research (factors that influencesuccessful innovation), we would like to advance the following propositions:



Page 108• There is a positive correlation between the amount of social capital in a network andthe effectiveness of innovation by the network• The stronger the ties in the social network, the more resources are available andthe more likely the network is able to innovate in a successful way• The less redundant the social capital of the network, the more resources areavailable and the more likely the network is able to innovate in an effective wayDiscussion: electronic medical records in ShikokuWhat light can the above propositions shed on Hara’s innovations, and vice versa?Institutional framework• The more majoritarian a model of government is, the more likely innovations are tobe implementedAlthough the Liberal Democrat Party (LDP) has been the ruling majority party inJapan since 1996, it is not likely at all that this has influenced the innovations inShikoku. Of course, former Prime Minister Koizumi was a relatively strong figure withambitious action plans (E-Japan), but the way these were administered has not yethad any impact on the work floor in Kagawa University Hospital. Typical is VanWolferen’s description of “power” in Japan (1989:25):The Japanese prime minister is not expected to show much leadership; labour unionsorganise strikes to be held during lunch breaks; the legislature does not in factlegislate; stockholders never demand dividends; consumer interest groups advocateprotectionism; laws are enforced only if they don’t conflict too much with the interestsof the powerful; and the ruling Liberal Democratic Party is, if anything, conservativeand authoritarian, is not really a party, and does not in fact rule.In this context, we find no evidence to support this proposition based on our casestudy.• The more centralized a state structure is, the more likely changes will be imposedtop-down, and the less likely it is that bottom-up innovations will emergePrefectures in Japan have a relatively large amount of autonomy when it comes topolicy and implementation. They can levy taxes and spend the money they collect.They also get additional funding from the central government for specific purposes.On the other hand, the national government decides on virtually all legislation; theprefectures mainly decide on allocation of funds, operations and regionaldevelopment. Health care is administered at the prefectural level, but almost all



Page 109health care legislation is national. Against this complex backdrop, the occurrence ofHara’s grassroots innovation does not seem to have much to do with Japan’sparticular state structure.• The more mandarins are directly accountable for their efforts to innovate orimplement innovations, the more adaptive the system as a whole is to changeThe high-ranking medical professionals and civil servants that we have met who wereinvolved in the innovation process did not have specific personal targets. Nobody toldProfessor Hara to develop HelloBaby or K-Mix. Once he got started, nobody gave himindicators, targets or whatever. The innovators we met were never held accountablefor the success or failure of innovations in the medical sector. Instead, mainly theywere held accountable for the quality and budget control of health care. Our findingsdon’t support this proposition. But then again, we have spoken with intrinsicallymotivated innovators. We can’t say if people without that kind of drive would startinnovating when ordered to.• The more citizen-oriented a network of service delivery is, the more likelyorganisations within the network succeed in achieving results improving servicedeliveryThis proposition is fully supported by our case. Hara managed to get across themessage that the patient is the ultimate constituency for change. The more peoplewho shared that value, the more successful they were in overcoming obstacles andspreading their innovation.• Funding regimes are major bodies of resistance, and consequently crucial obstaclesor – when redesigned – vehicles for changeFour ministries are actively funding e-health projects in one way or another: theMinistry of Health, the Ministry of Economy, Trade and Industry (METI), the Ministryof Agriculture (Rural development) and the Prime Minister’s Office (E-Japan strategy).Respondents told us that the involvement of so many ministries was far from helpful.In fact, due to lack of policy coordination, the national government was sometimesseen as an obstacle, as Professor Hara noted to us:Another hurdle was the lack of coordination and cooperation of the different involvedministries. There was no cooperation between the three ministries. Health andWellness was focused on the existing health system and the risks of innovation. METIhad the tendency to take everything as a matter of industry and economic innovation.They took the medical sector as “service industry”. That’s why they invested themoney in the branch. But that framed the project as an economic effort, notnecessarily an improvement in health care. This is still the case. METI is now moreaware of the sensitivities, but Health is still not very happy with the entrepreneurialspirit of the program.



Page 110Now e-health is one of the main goals of the E-Japan strategy of the Prime Minister’soffice. That could be a good thing, but it is again another ministry, with a differentangle. My own experience was that the Ministry of Health was rather conservative inthe beginning. They had a medical informatics division. I visited them when I startedmy program. There was no jurisprudence and I was told I would never succeed in myplans. I admit it is not an easy job, but we are heading in the right direction. Ministryof Health is held responsible for the national care system. They can’t afford to makemistakes and are conservative for that reason. METI has a different approach and cansupport a pilot project.The result of this was that the innovators in Japan did not want to waste energy onfundraising. They just started a project themselves and kept an eye open to see whensome money became available. Thus the proposition is supported by our findings:funding regimes matter. They can be beneficial if they align with the dynamics ofinnovation processes. They can be quite adversarial when they take the institutionalor policy frames as a starting point. All innovators need money and all innovators haveproblems finding money. Hara succeeded not by doing exactly what the money wasintended for, but by redefining what he found important in terms of the objectives ofthe fund provider.Change management• Breakthrough innovations typically occur in rigid organisational structures• Incremental innovations typically occur in adaptive, innovative and learningorganisationsJapan has a rigid institutional framework and hierarchical organisations. None of theinnovations we found in Shikoku were breakthrough innovations. They were long-termincremental strategies, led by careful visionaries. The innovators were facilitatingevolution rather than leading revolutions. Professor Hara told us:We began very small and with a subject that I knew very well. You know I am agynecologist. We focused on the fetus. We thought: what kind of information isrelevant for all possible care around birth. It was very difficult to map all those data.We thought: if we do this right, then the rest will follow. So we started working onHelloBaby. Every following step was thought out by myself. I had a big picture inmind about sharing information for all medical purposes. But I knew I would fail if Istarted working on a grand design. So I moved step by step. After we had one or twoapplications, we needed infrastructure and standards, then we needed money, afterthat we needed more users, after that we needed more applications, after that weneeded legislation, and so on and so forth.



Page 111• Thus, organizational structure was not a crucial factor for innovation to occur in thiscase studyNeither national reform agendas, nor the NPM theories behind the same, showed anystrong influence on the successful implementation of innovation in this service deliverynetwork. All we can say about the reform agenda in Japan is that it was noticed, if atall, by Hara only after he had already started his innovations. He used the broadagenda to get funding or support for things he believed in and projects that he hadalready started. In other words, reform agendas seem to facilitate those who alreadyinnovate rather than trigger or stimulate those who don’t feel that urge. This suggeststhat innovators do not necessarily share the incentive structures as depicted by NPM.Their desire to innovate does not stem from the competitive force that comes fromserving one’s self-interests, as the public choice literature suggests. For the innovatorswe’ve seen, their professional ethics demanded the innovation, not any efficiencygain, let alone personal financial gain or other financial, competition-enhancingincentive.Social capital• There is a positive correlation between the amount of social capital in a network andthe effectiveness of innovation by the networkJapanese culture is characterised by “harmony”. Harmony means, in fact, powerbalance and mutual understanding. There is a covert power structure based onintricate ties between business and government: the Keiretsu. Deals are made alongthe lines of that covert power structure. For outsiders this system is hard to grasp.One does not usually speak about it. Many respondents, however, when talking offthe record, indicated that this covert power structure can make and break innovations.As long as the innovation is not disruptive to the power balance, it has a chance. If itis disruptive, you can forget it. The “harmony” is highly valued in Japanese societyand it is of a higher value then practical problems.There is, however another side. If the power balance is not directly affected, theJapanese become very practical and pragmatic. One respondent told us:Concerning the culture in the medical sector or the social network, you can easily sayit is rather conservative. When innovation works it is mostly thanks to the effort of asingle person who is willing to make the idea succeed. Leadership is very important inmedical innovation. For a leader in the medical field, connections to politics have highvalue.• The stronger the ties in the social network, the more resources are available andthe more likely the network is able to innovate in a successful way• The less redundant the social capital of the network, the more resources areavailable and the more likely the network is able to innovate in an effective way



Page 112The innovators we spoke with had many strong as well as weak ties. The lattercontributed highly to the diffusion of the innovation in the network. The creation ofthe innovation depended more on strong ties – ties that became even stronger duringthe innovation process. One of the respondents told us about the importance of hispersonal network:It is very, very important! Everything I achieve is through my network. I don’t haveanything to be proud of, but I have many friends. What I have achieved is thanks tothose friends and despite the government. I have three groups of contacts: people atthe grass root level (nurses, doctors, technicians, etc.), CEOs, and fellow surgeons.Through these contacts, their expertise, and their respective networks I can achievealmost anything I need. It is important that I have a diverse network. When youmerely talk to doctors, you don’t get very far. Doctors have their own professionaltunnel vision on reality. They don’t understand the world outside. I am not like that, Iam curious by nature.We believe that innovators such as Hara may have at least one thing in common:they enjoy a very heterogeneous personal network, consisting of politicians, medicalprofessionals, technicians, etc. All resources needed for their innovations are“sourced” through their social networks. This is certainly consistent with our presentcase study, which therefore supports both of the above propositions.ConclusionsWhich socio-cultural, institutional and management variables influence the occurrenceof successful innovations in networks of service delivery? Literature from disciplinesranging from political science to organizational sociology has led us to propositionsabout relations between variables in the context of an innovation. Looking at theJapanese case, we have to conclude that many of those variables seemed hardlyrelevant. Other variables, however, seemed to deserve much more attention.Institutional factors hardly mattered in the case. True innovators are more thaninventors. Their capabilities exceed their talent for envisioning and delivering publicvalue: they are creative in managing their authorizing environments. Reform agendaswith budgets attached may come in handy, but the practice of innovation is almostalways ahead of policy definitions.Management does matter, but not so much the management by the superiors ofinnovators. Innovators act on their own behalf, with their own professional sense ofdirection and without incentives or sanctions attached to their mission. Innovatorsthemselves seem to have a quite distinct idea about their own management; theystress the importance of developing a clear image of the ideal situation; creating asense of direction among stakeholders, taking into account the varied interests ofstakeholders while producing tangible benefits to keep them engaged in the effort;and going step-by-step, forgoing grand designs for the sake of strategic andpragmatic incrementalism.We also found that social capital is vital to the creation, implementation and



Page 113dissemination of innovations. Strong ties are needed to obtain the required resourcesand weak ties to spread the innovation and to scale up. All the innovators we spokewith manage their social networks carefully.Apart from answering the research questions, interviewees indicated what, accordingto them, were additional influential success factors. One is human capital. Innovatorslike Hara share certain characteristics which distinguish them from others, such asgreat social competencies, a broad and heterogeneous social network, an above-average professional drive and multiple expertise (political-administrative,management, technical and medical skills). The people who started and disseminatedthe innovation were not mainstream professionals. They were – more so than theirpeers – driven by the will to improve.Another influential success factor is business process redesign. Technology is oftenmentioned as a push for innovation. We found that technology is indeed perceived asan enabler, but no more than that. The availability of technology did not lead toinnovations right away. Innovators basically made old and new systems interoperable.The innovations were made in the process of redesign rather than in the technology.A third factor is serendipity. None of the interviewees said: “this or that was thefactor that made the innovation happen.” They all said things like: “it was a perfectset of circumstances” or “it was a nice accident of timing” or “there were just theright circumstances.” Some factors like money, dedicated people, and strongleadership came together. There was a policy window. Thus, we have to concludethat there is not one distinctive factor but a set of factors that makes innovationhappen.It may be the case that nothing is more disappointing for policymakers and academicsthan having to conclude that innovation boils down to having the right person, at theright time in the right place. What we have called the “Hara factor” is the coincidenceof talent, vision, resources and circumstances. Common notions about innovation inthe public sector were easily debunked as myths. Innovation is not about top-downbusiness-like management based on (financial) incentives. Innovation is not aboutthrowing money at something. Innovation is not about taking risks or allowing peopleto take risks. And innovations occur to a large degree irrespective of the institutionalenvironment and reform agendas.However, the Hara factor implies more than negative recommendations. When weabandon the idea that innovation is something that we can steer and administer likeany policy issue, there might be some interesting lessons to be drawn from the cases.If we really want to understand and promote the Hara factor in general, scholars andpolicy-makers may want to take into account the factors that actually did influencethe innovations: the professional, political, managerial and social skills of theinnovator, his particular strategies and tactics, and his motivation and professionalethics. Studying and creating the factors that influence the likeliness of innovation,much attention has been paid to the external institutional, cultural and managerialfactors. Further research will need to pay just as much attention to factors related tothe innovators and how they do what they do, often against the stream.



Page 114Notes1 Jorrit de Jong is Research Fellow at the Ash Institute for Democratic Governanceand Innovation at Harvard University’s Kennedy School of Government. Arre Zuurmondis Professor of Public Management at Delft University in the Netherlands. They aremost grateful to The Netherlands-Japan Institute in Tokyo, Noor Huijboom, Albert JanKruiter, Floor van Dijk, Bas Valckx, Sebastiaan Rust, Rik Schurmann, Femke Smuldersand Arco Strop for their advice and/or assistance during the research. They wouldalso very much like to thank all the innovators that were so kind to share theirstories, opinions, insights and dilemmas.2 Site visits and interviews were conducted in the summer of 2005 in Tokyo, Kyoto,Takamatsu, and several medical centres across Kagawa Prefecture on the island ofShikoku, Japan. This study was part of a larger research project, involving casestudies on similar projects in Canada and Denmark. The conclusions andrecommendations in this paper on innovation in Japan were corroborated in the othertwo case studies.ReferencesBehn, Robert and Alan A. Altshuler 1997, Innovation in American Government:Challenges, Opportunities, and Dilemmas, Washington, DC: Brookings InstitutionPress.Borins, Sandford 1998, Innovating with Integrity: How Local Heroes Are TransformingAmerican Government, Washington, DC: Georgetown University Press.Borins, Sandford 2002, ‘The Challenge of Innovating in Government’, in Innovation,Abramson, M.A. and Littman, I. D., eds. Maryland: Rowman & Littlefield Publishers.Burt, Ronald S. 1992, Structural holes: the social structure of competition. Cambridge,MA: Harvard University Press.Coleman J. S. 1988, ‘Social capital in the creation of human capital.’ American Journalof Sociology, 94: S95–S121.Castells, Manuel 1999, The rise of the network society: The information age, Vol. 1.Oxford: Blackwell.De Graaf, N. D. and Flap, H. D. 1988, ‘With a little help from my friends.’ SocialForces. 67(2): 452–72.Erickson, B. H. 1995, ‘Networks, success, and class structure: a total view.’ SunbeltSocial Networks Conference, February, Charleston, SC.Erickson, B. H. 1996, ‘Culture, class and connections.’ American Journal of Sociology,102(1): 217–51.Flap, Henk and Marco H. D. van Leeuwen, eds. 1994, Op lange term n: verklaringenvan trends in de geschiedenis van samenlevingen. Verloren.Ganzeboom, Harry B. G. and Henk Flap, eds. 1989, New Social Movements and ValueChange: Theoretical Developments and Empirical Analyses. Amsterdam: SISWO.Granovetter, Mark 1995, Getting a Job: A Study of Contacts and Careers. Chicago:The University of Chicago Press.Hartley, Jean 2005, ‘Innovation in Governance and Public Services: Past and Present.’Public Money and Management, 25(1): 27–34.Lijphart, Arend 1999, Patterns of Democracy: Government Forms and Performance inThirty-six Countries. New Haven, CT: Yale University Press.Lin, Nan 2001, Social capital: A Theory of Social Structure and Action. Cambridge:Cambridge University Press.Lin, Nan, Walter M. Ensel and John C. Vaughn 1981, ‘Social Resources and Strengthof Ties: Structural Factors in Occupational Status Attainment.’ American SociologicalReview, vol. 46, no. 4: 393–405.Marsden, Peter V. and Nan Lin, eds. 1982, Social Structure and Network Analysis.Beverly Hills: Sage Publications.



Page 115Moore, Mark H. 1995, Creating Public Value: Strategic Management in Government.Cambridge, MA: Harvard University Press.Moore, Mark. H. 2005, ‘Break-through Innovations and Continuous Improvement: TwoDifferent Models of Innovative Processes in the Public Sector.’ Public Money andManagement, 25(1): 43–50.Morgan, Gareth 1997, Images of Organization, 2nd ed. Thousand Oaks: SagePublications.Osborne, David and Ted Gaebler 1993, Reinventing Government: How theEntrepreneurial Spirit is Transforming the Public Sector. New York: Penguin.Pollitt, Christopher and Geert Bouckaert 2004, Public Management Reform: AComparative Analysis, 2nd ed. Oxford: Oxford University Press.Putnam, Robert D., Robert Leonardi and Raffaella Y. Nanetti 1993, Making DemocracyWork: Civic Traditions in Modern Italy. Princeton, NJ: Princeton University Press.Stone, Deborah 1997, Policy Paradox: The Art of Political Decision Making. New York:Norton.Van Wolferen, Karel 1989, The Enigma of Japanese Power: People and Politics in aStateless Nation. London: Macmillan.Wielgosz, J. B. and Carpenter, S. 1987, ‘The Effectiveness of Alternative Methods ofSearching for Jobs and Finding Them.’ American Journal of Economics and Sociology,46: 151–64.



Page 1167 “Good governance” and the constraints of local conditionsThaksin and the breakdown of Thailand’s public integrity systemAlex M. MutebiIntroductionCorruption in the public sector is one of the greatest challenges that many developingcountries face today. The corrosive effects of corruption are well documented(Bardhan 1997; Khan 1996; Rose-Ackerman 1999; Tanzi and Davoodi 1998).Particularly insidious forms of corruption involve powerful non-state actors taking thereigns of power and using their newly acquired leverage and influence to shape acountry’s legal and regulatory framework to their own advantage, thereby distortingthe beneficial effects of competition, private-sector-led growth, and the state itself.For the most part, that kind of corruption – different from the common image ofcorruption as involving fraudulent street-level bureaucrats – is more ominous, as itoften implies that a powerful elite commandeers key aspects of state power.This paper explores the old paradox of why such corrupt behaviour can thrive despitethe existence of rather comprehensive anticorruption mechanisms, themselves oftenpart of a larger and more extensive web of “good governance” institutionalarrangements. Some easy explanations for the prevalence of corruption highlight suchfactors as authoritarian regime types, inherently endemic corrupt cultures,beleaguered civil society and media, skewed donor/government relations in favour ofthe latter, and so on. However, this paper points to something more fundamental.Using the case of Thailand during its former Prime Minister Thaksin Shinawatra’s timein office, this paper suggests that it was the country’s anticorruption repertoire, set upbefore Thaksin’s premiership, which proved woefully inadequate, primarily because thetype and levels of corruption that predominated during his leadership simplyoverwhelmed the existing anticorruption measures. In particular, by not focussing onthe problems of state and regulatory capture and the effective implementation of thevarious administrative corruption mechanisms that were adopted, the reformersdiminished Thailand’s potential for greater progress in the fight against corruption asThaksin cynically and expertly exploited loopholes in what, until then, was consideredThailand’s best written and “good governance” constitution.A controversial but very intelligent and shrewd businessman, Thaksin is one of EastAsia’s most colourful politicians in recent years. Having initially made



Page 117his fortune in the 1980s from several concessions he was granted during military rulein Thailand, Thaksin had a political reincarnation in the mid-1990s, when he firstserved briefly as the Information Minister. Following the Asian financial crisis, duringwhich he had abstained from direct participation in a political party, Thaksin re-entered politics in 1998, started a new party, Thai Rak Thai (Thais Love Thais), andlaunched the first of his bids for the premiership. He then made history by leading hisTRT party to overwhelming victories in two back-to-back elections, first in 2001, andthen in 2005 (Kuhonta and Mutebi 2006; Nelson 2002). Under Thailand’sparliamentary system, Thaksin’s overwhelming majority virtually ensured that the self-made telecommunications billionaire would remain prime minister for a long time, andhold an even firmer grip on power than many of his 30 or so predecessors (Chambers2005; Ganesan 2004; Hicken 2006; Ockey 2003).Thaksin’s electoral appeal was largely due to his populist economic platform, dubbed“Thaksinomics,” which was very popular with large sections of the country. Primarilydesigned to support farming and cottage industries, and to boost the incomes of thecountry’s downtrodden, the policies quickly turned him into something of a popularhero. Not only was he hailed by his admirers as a decisive, no-nonsense leader whohad restored Thai pride by lifting his country-folk from the despondency of the Asianfinancial crisis that began in 1997, his leadership was also seen as critical in theimpressive economic mini-boom that occurred during his tenure as prime minister. Tohis detractors, however, Thaksin was not only a semi-autocrat who ran roughshodover opponents and relentlessly extended his own power, he was also a leader whopresided over an administration that many believe to have been “captured” by special(mostly business) interests close to him (Kurlantzick 2003; McCargo 2002; Ockey2004).Ironically, pre-Thaksin Thailand had seen promising trends in which corruption,though not necessarily lessened, had been brought more into the open. This wasfueled largely by various awareness-raising campaigns organized by local andinternational civic organizations, as well as the country’s increasingly dynamic media,which thrived in the wake of an abortive coup in 1991. When Thailand adopted areformist constitution in 1997, the document encompassed not only most of thestandard provisions said to underpin “good governance”, and which one finds in theconstitutions of most Western liberal democracies, but it specifically includedprovisions for increasing transparency and probity in public officialdom. In particular,the constitution included requirements for the establishment of a number ofindependent integrity agencies to provide checks and balances in the political system.This was primarily meant to lead the country towards a less corrupt and better-governed society.However, under Thaksin – who had ridden to power on the coattails of that 1997“good governance” constitution – Thailand’s new set of anticorruption and “goodgovernance” institutions, laws, and regulations turned out be rather spectaculardisappointments. Not only did many of the fledgling independent institutionsmandated by the constitution struggle to establish their authority and credibility(Callahan 2005a; Mutebi 2006; Ockey 2003; Ungpakorn 2002), but Thaksin and



Page 118his close interests were also constantly plagued by accusations of insidious forms ofcorruption (Bidhya Bowornwathana 2004; McCargo and Ukrist Patthamanan 2005;Pasuk Phongpaichit and Baker 2004).Whereas space considerations preclude any detailed examination of the variouscorruption scandals that dogged Thaksin and other leading public figures close to him,I briefly highlight one case early in his premiership that symbolized the kind of politicshe represented. As Deputy Prime Minister in 1997, Thaksin had been charged withfailing to declare assets after officials discovered a large fortune’s worth of hiscorporation’s shares in his maid’s and driver’s bank accounts. A guilty verdict from thecountry’s Constitutional Court would carry the real possibility that his tenure as primeminister would be cut short. Thaksin defended himself by blaming his wife andclaiming that his actions were commonplace anyway (Pasuk Phongpaichit and Baker2004:1–7). In the weeks leading up to the verdict, Thaksin essentially ran against thecountry’s highest court, crisscross ing the country and telling his party’s rural basethat he and he alone could make Thailand prosperous. The implication was essentiallythat, were the court to rule against him, his voters should not stand for it (Greenfield2001). Indeed, some in his TRT party even warned of mob violence in the event of aguilty verdict. The verdict was predictable: the seemingly invincible new PrimeMinister beat the charges by a slim majority of one vote (7 out of 13). His criticsallege that rather than being shaken by the close call, Thaksin was insteademboldened, and he then set out to pursue not just his populist and strong-maneconomics and politics, but also policies that specifically benefited his own companiesas well as the business interests of those around him (Ukrist Pathmanand, 1998).Before examining some of the factors that may explain the alleged systemic corruptionthat characterized Thaksin’s Thailand, this paper includes several sections thatcontextualize the main argument. Section two provides a brief discussion ofanticorruption programs in much of the developing world today, specifically linkingthem to the “good government” paradigm and briefly mentioning why most of themfail to curb state and regulatory capture, the more insidious forms of corruption.Section three briefly summarizes Thailand’s anticorruption institutional framework, asinstated in the late 1990s. Section four discusses four plausible but incompleteenvironmental explanations of the contradiction of why corruption thrived underThaksin even in the presence of a comprehensive anticorruption framework. Sectionfive advances the main argument of the paper; that is, that Thailand’s post-1997integrity and anticorruption framework was incomplete from the very beginning.Finally, Section six concludes the article.Administrative corruption, state “capture” and “good governance”Compelled by the fact that the term “corruption” is a catch-all for many kinds ofbehaviour with different causes and impacts, scholars routinely draw distinctionsbetween different kinds of corrupt activity and have even created entire typologies tomake better sense of the phenomenon. Although the various typologies are impossibleto enumerate here, common distinctions in the literature include



Page 119“systemic” versus “individual” or “opportunistic” corruption; “grand” versus “petty”corruption; “good” versus “bad” corruption; “centralized” versus “decentralized”corruption; “state capture” versus “administrative corruption” (Thomas and Meagher2004).The latter typology is obviously not intended to define absolute and unambiguousdivisions among countries and is used mostly as a heuristic device to highlightanalytically, and at times empirically, the differences in corruption patterns acrosscountries. However, the typology is quite useful for the purposes of this paper. Ingeneral, administrative corruption refers to the use of illicit favours to distort theimplementation of existing policies. State capture refers to collusion between privateand public agents, with the former providing illicit favours to the latter in order toinfluence the formulation of laws, regulations and policies, and secure specialadvantages (Fogel 2006; Hellman et al. 2003; Hellman and Schankerman 2000; Philp2001). Variants of corruption in the form of state capture are widely recognized today,sometimes arising through subtle, quasi-legal forms; for example, legal lobbyingcontributions from business interests in exchange for particular legislation predisposedin favour of those interests, or the allocation of lucrative contracts and concessionsinvolving the interaction of private and public sector agents where the latter takeadvantage of the power of their positions at the expense of broader public welfare.Among other things, the administrative corruption/state capture typology allows for asimple division of countries into four spheres to determine, roughly, the relative levelsof state capture and administrative corruption (World Bank 2000). Table 7.1 is apartial representation of such a division, including only four spheres. Countries inquadrant A are considered to have both types of corruption, but in manageable levels.Countries where the central problem is administrative corruption, while the state isless subject to capture by the private sector than in comparable countries, fall intoquadrant B. Likewise, countries that have largely been able to contain the level ofadministrative corruption within the context of high state capture fall into quadrant C.Finally, quadrant D represents thoseTable 7.1 A simple typology of corruption State Capture MEDIUM HIGHAdministrativeCorruption

A B

MEDIUMMedium administrativecorruption

High administrativecorruption

Medium state capture Medium state capture C D HIGH Medium administrative

corruptionHigh administrativecorruption

High state capture High state capture



Page 120countries where administrative corruption is a serious problem, perhaps due to weakstate capacity, and which also exhibit conditions of extreme state capture.Obviously, the typology is not dynamic and can thus only represent corruptionpatterns within or across countries statically at a given time. A more dynamic typologywould allow for different patterns of development concerning corruption, as well asvarious evolutionary paths across those patterns. If we take Southeast Asia as anexample, a country may register progress in one dimension of corruption while laggingor falling behind in another, or may even move ahead on both fronts concurrently.Indeed, the rapid economic development in certain parts of Southeast Asia over thelast several decades has engendered excellent conditions for state capture andadministrative corruption (Khan and Jomo 2000; Lim and Stern 2002; Marsh et al.1999; Quah 1999). Of course, the level and extent of such corruption has variedacross countries, and even within countries over time, based on a complex set ofinteractions and lines of causality at different stages of the countries” development.For example, countries where national wealth has generally been concentrated in afew key productive sectors have faced greater risks of powerful interests seeking togain control over them, even while those interests have also “invested” some of theirgains in a bid to sustain and entrench their positions. Likewise, the sophistication ofthe public administrative machineries that countries have inherited from earlier eras,as well as the depth and strength of social networks and civil society in thosecountries, have all had an impact on the level of corruption in the institutionalarrangements that have ensued.As attention has increasingly been directed towards the corrosive effects of bothadministrative corruption and state capture in recent years, many countries haveenacted anticorruption policies, often part of “good government” or “goodgovernance” reforms. These have been regarded as pivotal to the developmentprocess. This “good government” framework – mostly normative – is normallyexpected to include several key components: (i) political legitimacy for the statethrough democratic elections and the transfer of power, and an effective politicalopposition and representative government; (ii) accountability through transparencyand the provision of information; (iii) separation of powers; (iv) effective internal andexternal audits; (v) effective means of combating corruption and nepotism; (vi) officialcompetency, such as trained public servants; (vii) realistic policies and low defenseexpenditure; (viii) human rights as indicated by freedom of religion and movement;(ix) impartial and accessible criminal justice systems; and (x) the absence of arbitrarygovernment power (Doig and Riley 1998; Kaufmann 2001). Good government, theargument goes, is not an end in itself. Rather, it is an essential condition toward thewider goal of “good governance,” the other anchor of much of the anticorruptionliterature and one seen as concerning not just the organization and activity ofgovernment but also the ends to which both those are put in terms of achievinglevels of economic, human and institutional development (World Bank 1991).For its part, “good governance” – an also mostly normative framework – involvesgovernmental arrangements accepted by participatory publics as legitimate,



Page 121responsive to the preferences of the population and dedicated to improving itswelfare, capable of managing law and order and delivering public services, andproviding effective policy environments and open-handedness in conduct (Ginther etal. 1995). In addition, such a system of government would want to disengage itselffrom direct involvement in other areas, particularly economic production, and insteadconcentrate on societal priorities. That argument also usually suggests that oncepolitical reforms are in progress and centrally controlled economies dismantled, liberalmarket principles can then begin to advance economic development, which in turnshould promote participation in political processes (Doig and Riley 1998:47). Theculmination of this argument entails economic and political objectives complementingeach other to work toward an educated populace with both political knowledge andthe will to act, coupled with modern industrial economies and deep-rooted democraticpolitical values.1At any rate, whereas the “good government” and “good governance” literature tendsto suggest that anticorruption strategies should be situational, with variation based onthe weaknesses of specific environments, most standard advice coming out of boththese frameworks has mostly focussed on measures geared toward addressingadministrative corruption, at least in practice. Thus, anticorruption strategies based oneither of these frameworks typically focusses on one or several of the following: (i)formation of anticorruption agencies; (ii) reliance on public opinion surveys; (iii)raising of public sector wages; (iv) reduction in the size of the public sector; (v)strengthening of financial accountability; (vi) increases in media freedom; (vii)strengthening in judicial independence; (viii) boosting citizen participation; (ix)decentralisation; or (x) changes in bureaucratic culture.However, anticorruption policies and mechanisms often fail miserably, particularlywhen they fail to target broader structural relationships, including the internalorganization of countries” political systems, relationships among core state institutions,interactions between state organs and private firms, and relationships between thestate and civil society (Maor 2004; Quah 2006; Schacter and Shah 2000).The next section briefly outlines the key elements of Thailand’s extensive web ofintegrity, anticorruption and “good governance” institutional arrangementsencapsulated in its 1997 constitution.Key “good governance” arrangements in post-1997 ThailandWhereas it is impossible in a paper of this scope to disentangle the intricateinteractions and lines of causality that have shaped the development of Thailand’sintegrity and anticorruption institutions over the last several decades, a few key factsare clear, beginning from the 1970s. Prior to 1975, investigating and fightingcorruption had mostly been the responsibility of regular government agencies (mainlythe police) who relied on the criminal law and an assortment of regulations applicableto public officials who, if found guilty, would face disciplinary reprimand by theirrespective government agencies. After the public uprising against the government of14 October 1973, there was a crisis of public



Page 122confidence in Thailand’s public institutions. This compelled the promulgators of thecountry’s subsequent Constitution (1974) to target the issue of corruption for the firsttime. The constitution stated that “the State should organize efficient systems ofgovernment […] and should take all steps to prevent and suppress the quest forbenefits by corrupt means.’ In 1975, a Counter Corruption Act was promulgated and aspecialized agency, the Office of the Counter Corruption Commission (OCCC), wasestablished to carry out the legislation.Between 1975 and 1999, the OCCC operated as a special government unit within theOffice of the Prime Minister. However, the levels of corruption remained high, in partbecause of the absence of a code of ethics for public officials, a generally tolerantenvironment for corruption, ineffectual enforcement, inadequate coordination ofanticorruption efforts, and the close relationship between the OCCC and the executivebranch. Furthermore, there was general public mistrust concerning the OCCC andother integrity agencies, in particular the criminal justice system (United NationsDevelopment Program 2005).As stated previously, Thailand’s contemporary anticorruption repertoire is largelyanchored in the country’s 1997 Constitution, written by a generally non-partisanassembly selected following the political upheavals of the early 1990s. TheConstitution Drafting Assembly (CDA) was comprised of various elite factions, some ofwhom had only reluctantly been compelled onto the reform bandwagon (Freedman2006). The CDA was led by two distinguished Thai statesmen, Uthai Pimchaichon andAnand Panyarachun, and included some 99 representatives: 76 delegates from each ofThailand’s provinces, eight public law experts, eight political and public administrationexperts, and seven officials with experience either in drafting constitutions and laws,or in bureaucratic regulation. The constitution, which included 33 articles, containedprovisions similar to those in constitutions of Western democracies. It also greatlyrestructured both the public and non-public sectors, including the legislature, electoralsystem, judiciary, cabinet, bureaucracy, and so on.To ensure transparency and accountability, the Constitution established several newoversight institutions that were “independent” of executive interference, orrestructured existing ones to make them so. Among the most notable were thoseaimed at balancing and controlling administrative power (including a ConstitutionalCourt; Administrative Courts; a National Committee on Human Rights; a State AuditCommission; and an Ombudsman). In addition, other institutions were aimed atbalancing and controlling political power (including a National Election Commission(EC) whose rulings were beyond appeal and a National Counter CorruptionCommission (NCCC) composed of various administrative bureaus).2 Table 7.2summarizes the key functions of each of these institutions. In addition, the newconstitution mandated, for the first time in Thailand’s history, not only that the 200-member Senate (Wuthisapha) would be directly elected, but that senators would benon-partisan. This was aimed at increasing the Senate’s oversight role of both theExecutive and the House of Representatives (Sapha Phuthaen Ratsadon).Further measures to improve governance in Thailand were encapsulated in new set ofguidelines, the Regulations of the Prime Minister’s Office on Good



Page 123Table 7.2 Key “good governance” provisions in Thailand’s 1997 constitutionInstitution/agencyKey functionsElectionCommission (EC)

Conduct elections and regulate political parties

Eliminate rampant vote buying at election time Investigate all alleged electoral fraud and disputes - where

necessary, order new elections or referendums in any or all pollingstations

Order re-elections, ban cheats from running again, or dissolvepolitical parties that did not follow regulations

National CounterCorruptionCommission(NCCC)

Oversee mandatory annual disclosures of assets and liabilities ofmost politicians and senior bureaucrats

Inquire and decide whether public officials are “unusually wealthy”,demonstrate corruption, or make false or incomplete declarations(as stipulated by the Organic

Law on Anticorruption (1999) and authorized by the Constitution) Bar offenders from office for up to five years, subject to the

approval of the Constitutional Court.State AuditCommission(SAC)

Examine state expenditures for evidence of misappropriation offunds under the leadership of an independent and impartial AuditorGeneral

Audit all public receipts and payments each fiscal year Audit country’s currency reserve account Examine fees and other income of audited public agencies in

connection with the collection of taxes Search and seize suspects and offenders in the exercise of the

above dutiesConstitutionalCourt (CC)

Oversee and decide on all organic laws, bills, decrees, andordinances under the new Constitution

Establish the Administrative Court systemAdministrativeCourt (AC)

Rule on cases between government departments or officials andprivate organizations or ordinary people, and between governmentdepartments or public offi cials.

Office of theOmbudsman

Oversee administrative problems in cases where state organizationsare accused of either failing to comply with the law, exercisingpowers beyond their authority, or failing to perform their duties.These transgressions would trigger independent recommendations tothe Constitutional

Court or the Administrative Court for further actionNational HumanRights

Safeguard constitutional human rights, primarily through

Commission(NHRC)

examining allegations of human rights violations and reportingfindings to the National Assembly

Governance, in 2001, In particular, these guidelines aimed to promote transparency ingovernment, improve the quality of public services, and strengthen integrity in publiclife. This ambitious set of regulations aimed to prevent, inter alia, corruption,misconduct, and malpractice for personal benefit and gain in both the public andprivate sectors, as well as to create a sense of mutual responsibility



Page 124towards society. Also listed in that directive were the “Rule of Integrity” and the “Ruleof Value for Money’, as key good governance goals. Many of these measures werethemselves linked to much larger civil service reform efforts enacted in the aftermathof the 1997 Asian Financial Crisis. Among other things, the civil service reformsinvolved various external actors including the two Bretton Woods organizations as wellas other multilateral and bilateral partners providing technical assistance withThailand’s anticorruption strategies (Painter 2006).In sum, by the late 1990s, Thailand had in place – at least on paper – some verystrong institutional arrangements for reducing certain forms of corruption and forpromoting “good governance’. In particular, the new integrity framework would notonly improve transparency in the public sector and the integrity of public servants, aswell as reduce corrupt practices; it would also build and empower domesticconstituencies that would, in turn, generate sustained demand for anticorruptionmeasures and good governance. In essence, the 1997 Constitution was seen asproviding critical tools for public monitoring and accountability that would be essentialfor placing constraints on politicians and bureaucrats. These tools would also underpina degree of state capacity that enabled politicians to assert greater control withinpublic bureaucracies if they chose to do so.Why then is the Thaksin administration believed to have presided over widespreadcorruption despite the existence of the comprehensive anticorruption mechanismsabove, themselves often part of a larger and more extensive web of “goodgovernance” institutional arrangements? The next section presents several possibleexplanations for that contradiction – variables which, though consistent, overlook themost important explanation: a misdiagnosis by Thai reformers in the 1990s of thelevel and type of corruption that would pose the greatest challenges to the countryafter 1997.Credible but insufficient explanations for widespread corruption underThaksinCorruption in Thailand has long been recognized as deep-rooted, diverse and complexin both size and form, ever adapting to the rapidly changing political, economic andsocial environment (Doner and Ramsay 1997; Nattakrit Thewphaingarm 1996; Neher1976; Nishizaki 2005; Pasuk Phongpaichit and Sungsidh Phiriyarangsan 1996; Rock2000). Among others, Arghiros (2001), Callahan (2005b), Ockey (1994), PasukPhongpaichit et al. (1998), and Pasuk Phongpaichit and Sungsidh Phiriyarangsan(1996), have all detailed more various discreet, though no less illicit, forms ofcorruption, often involving intricate business networks and even illegal business.Others, including Neher (1977), Thinapan Nakata (1977), and Waterbury (1973), havediscussed corruption in Thailand’s public service, as well as in the legislation regulatingthis area, which has at times been poorly developed and weakly enforced thanks to abureaucracy with considerable discretionary powers to intervene in the country’seconomy.A simple argument for the failure of the Thailand’s integrity institutions during theThaksin era would simply be to point to the country’s historic legacy



Page 125of corruption, which, presumably, would have overwhelmed any leader. Thus, whereasThaksin may have both inherited or even accelerated various technocratic reforms thatwere useful points of departure into tackling endemic corruption, all such efforts wereno match for a kleptocratic polity. Such an argument, however, is clearly inadequate.Another such explanation of the failure of Thailand’s integrity institutions underThaksin, despite what seemed like a far-reaching integrity and anti-corruptionframework, is to link to it the well-documented authoritarian tendencies of his regime(McCargo 2002; Pasuk Phongpaichit and Baker 2005). While there is no definitive linkbetween regime type and reform policy outcomes, there is some evidence, notably inthe study of the implementation of reforms, that suggests that more democratically“open” regimes provide more opportunities to a wider range of actors to pursue policychange than do authoritarian regimes dominated by a closed circle of elites(Brinkerhoff 1996). Whichever label one uses to describe Thaksin’s regime (“semi-authoritarian”, “soft-authoritarian”, etc.), there is little doubt that his administrationdemonstrated greater authoritarian inclinations than those of his immediatepredecessors (Ganesan 2006; Simpson 2006). In particular, violations of Thailand’sdemocratic institutions during Thaksin’s tenure as prime minister were so frequentthat Thailand was often portrayed as failing to meet conventional minimum standardsfor a true democracy. Levitsky and Way (2002:51–4) have coined what may be amore apt term to describe Thaksin’s type of regime: “competitive authoritarianism’. Inthis type of regime, violations of democratic criteria are both frequent and seriousenough to give the incumbent an unfair advantage over the opposition. Theincumbent routinely abuses state resources, harasses and denies media coverage forthe opposition, and spies on, threatens, harasses, or even arrests government critics.Consequently, one could rightly argue that, in general, regimes with authoritariantendencies such as Thaksin’s are not only more likely to have leaders less willing totackle corruption seriously, but will also offer few motivations to citizens to articulateconcerns about corrupt behaviour. On the other hand, more open regimes usuallyinstitute the basic elements of rule of law, accountability, transparency, and access, atleast to the extent that they have effective democratic governance systems. Not onlydo active competition and accepted rules of accountability make it difficult for anyindividual, faction, or interest group to dominate politics or the economy in suchregimes, but clearer distinctions between those two areas check the worst types ofabuses. Regimes that are more open are more likely not only to have leaders morewilling to address corruption, but also to generate the circumstances conducive to theinitiation and support of anticorruption activities by other societal groups.Similarly, another plausible but incomplete explanation of the failure of Thailand’sintegrity institutions under Thaksin points to the nature of the relationship betweenthe state on the one hand, and the media and civil society on the other. In general,an unfettered civil society and media can criticize any credible anticorruption efforts,acting as either countervailing forces against government excesses or as sources of



Page 126encouragement to the efforts of reformers (Johnston 2005; Norris 1999). Indeed,partnerships with the media and civil society, as well as alliances with interestedanticorruption partners within the state, are always a strong foundation for anycredible comprehensive anticorruption strategy. Such “interventions” from below, inthe civic republican tradition, tend to be longer lasting and also help build socialcapital necessary for the consolidation of democracy.To be sure, efforts to promote collective action among anticorruption constituenciesand competing interest groups have varied widely over time in Thailand. Throughmuch of the 1990s, for example, the media and Thai civic groups such asTransparency Thailand, the Anticorruption Network, and the Foundation for Clean andTransparent Thailand served as watchdogs that tried hard to enforce theaccountability and transparency of public officials. However, during the Thaksin era,the media and civil society’s potential as agents of change was greatly constrained, atleast in comparison to earlier democratic administrations. Civic groups and sections ofthe media were either overtly harassed or they were routinely pressured by variousstate organs (Kurlantzick 2003; Mutebi 2003:104–6, 2004:78–9). In part because ofthis, one could argue that the political space in which Thailand’s civil society, mediaand other non-state actors could organize and express their views, including oncorruption and anticorruption, was significantly constrained.Yet another reasonable but partial explanation of the failure of Thailand’s integrityinstitutions under Thaksin despite the country’s comprehensive integrity andanticorruption framework could be the altered nature of the relationship betweenThailand and the international donor community during the Thaksin era. In general,international donor involvement can play a major role in pushing countries towardgreater progress on their anticorruption agendas, particularly by preventing politicalcomplacency and generating political will among reformers.Although Thailand had long ceased to be a major aid recipient by the late 1980s, theAsian financial crisis of 1997 temporarily thrust it back into this situation, particularlywith funds from the two Bretton Woods institutions, the World Bank and theInternational Monetary Fund. In particular, the IMF reached an agreement with thegovernment to adopt a comprehensive reform program that included, among otherthings, radically cutting expenditures, decreeing many corporations bankrupt,liberalizing foreign investment laws, and privatizing state enterprises in exchange for amulti-billion dollar rescue fund. Much of the assistance of the other significant sourceof aid, the World Bank, included programs requiring administrative reforms to controlcorrupt practices and move Thailand towards good governance (World Bank 1999). Inaddition, bilateral donors became involved in supporting various anticorruptioninitiatives, thus providing additional knowledge and partnerships to move forward insome of the most difficult areas of governance.However, in part because Thaksin’s government was able to pay off the bulk ofmultilateral debt by the middle of his first term in office, one could argue that theinternational donor community lost much of its leverage for buttressing the efforts oflocal counterparts in tackling corruption. If anything, Thaksin’s ability to stave



Page 127off pressures from international donors enhanced his managerial credibility and fedinto his populist appeal.Lastly, another intuitive but incomplete explanation for the failure of Thailand’sintegrity and anticorruption institutions under Thaksin was the endemic nature ofcorruption in sections of Thai society. In general, the extent of the stakes involved inany anticorruption programs – themselves dependent on the extent and nature ofcorruption – undoubtedly influence both the degree of political determination requiredfor any such undertaking and the design and implementation of anticorruption reformprograms. Within the administrative corruption/state capture typology presentedearlier, it is generally agreed that state capture is more difficult to deal witheffectively.In sum, parsimonious explanations of the failure of Thailand’s integrity andanticorruption institutions under Thaksin, though compelling, not only tend towardssome circularity, but are rather unpersuasive. The next section builds on thatargument and presents the paper’s main proposition, that is, that Thailand’s post-1997 anticorruption framework was incomplete – something, I argue, which is linkedto the sheer breadth and extent of corruption that ensued in the years following thepromulgation of that constitution, particularly during Thaksin’s term as prime minister.More basic explanations for the ineffectiveness of Thailand’s “goodgovernance” provisionsAlthough there are a wide range of diverse models for structuring anticorruptionefforts, most such efforts share key goals: enhancing state capacity and public sectormanagement, strengthening political accountability, enabling civil society, andincreasing economic competition. Yet many anticorruption efforts fail because they arenarrowly focussed on the weaknesses in the capacity of government, particularly inpublic administration and public finance management. Laudable as it is, such a focusfails to target broader structural relationships, including the internal organization ofcountries” political systems, relationships among core state institutions, interactionsbetween state organs and private firms, and relationships between the state and civilsociety (Huther and Shah 2000; Larmour and Wolanin 2001). Indeed, a well-recognized factor in the success or failure of anticorruption reforms, or any reformsfor that matter, is the range of relationships between the various interest groups infavour of, or opposed to, such measures, and other state and non-state actors.Powerful interests that face significant losses from reform are invariably opposed tochange and often deploy their resources to contest any changes to the status quo(Grindle and Thomas 1991; Waterbury 1973). In weak democracies such as Thailand,reformers are often politically weak and ineffectual, in part because they rest onbroad and unstable coalitions dependent on the goodwill of numerous state and non-state actors, many of whom are wary of change. Sustaining a supportive coalition insuch conditions where political will is unpredictable and transient, and reformers areweak and can be neutralized, can call for significant trade-offs and concessions, andcan sometimes feed factionalism and anxiety (Theobald 1990).



Page 128Some of the most powerful vested interests in Thailand, as in other parts of the world,feature the interaction of businessmen and politicians, with the former “purchasing”opportunities and favours that only the latter can provide (Doner and Ramsay 1997;Pasuk Phongpaichit et al. 1998). In fact, in many parts of Southeast Asia that nexus ofbig business and public servants is often linked to well-connected family dynastieswho, through their power and influence, control politics (Hutchcroft 1997; Kang 2002;Kidd and Richter 2003; Lim and Stern 2002). Such a phenomenon has indeed longbeen part of the Southeast Asian landscape and Thaksin Shinawatra was simply themost recent addition to the list. Although not a main thrust of this paper, we shouldnote that several scholars have argued that in high-growth East Asia, such closeinteractions between business interests and politicians may indeed have the positiveeffect of enhancing policy coordination, and ensuring that market regulatory reformsand other economic measures meet the needs of business. Doner and Laothamatas(1994), for example, have examined the joint public-private sector consultativecommittees in Thailand in the 1980s, where businesses could raise concerns on creditlimits, government effectiveness, taxes, and export incentives. By providing accessand meeting some business demands, this helped in return to get cooperation forbuilding support for other reforms including painful stabilization. Ironically, thesignificant GDP growth recorded under Thaksin’s first term may very well have beenfacilitated by such interaction – growth that not only had clear benefits for thecountry, but also, the sort of growth that indicated a recovery which few peopleexpected. At what point did positive, business-government consultation cross the linein Thailand to become state capture? What measures should 1997 constitution writershave put in place to better prevent state capture, while encouraging the positive sideof business/state interaction? These are questions that lack quick and easy answers.However, what is clear is that the main obstacle to further progress on the anti-corruption reform front during the late 1990s in Thailand lay less in the weakness ofstate institutions than in the power of vested economic interests and the privateinterests of powerful politicians, at the apex of which was Thaksin himself (Imai 2006;Ukrist Pathmanand 1998). Although Thailand’s basic political institutions and civilsociety were much more developed than at earlier times in its history, many of thecountry’s mechanisms of accountability were attenuated at the onset of the Thaksinera. Not only was political access for some societal groups often blocked, andinstitutional restraints within the state generally weaker, but political contestability wasalso less robust than at earlier times when the country was under democraticgovernments (McCargo 2002). Almost inevitably, conflicts of interest grew as Thaksinand the various interests around him regularly blurred the boundaries between privateinterest and the wellbeing of the state.Following the reforms inaugurated in the wake of the Financial Crisis of 1997, thecapacity of these powerful vested interests to influence the country’s fast-changinglegal, legislative, and regulatory frameworks appears to have outpaced the constraintsimposed by competing interest groups, civil society, and Thailand’s reasonably strongpublic administrative system (Wingfield 2002). Not put in place to check the powerand practices of old and emerging commercial, financial, and



Page 129industrial groups, were additional measures essential for both the enhancement ofpolitical transparency and accountability, and the strengthening of the country’seconomic competition (Chutatong Charumilind et al. 2006; Doner and Ramsay 2000;Imai 2006; Sakkarin Niyomsilpa 2000). However, to be effective, any national effortsto reduce corruption must be recognized to move beyond the narrow response toimmediate manifestations of corruption to a broader approach which attends to itsunderlying causes, themselves largely influenced by a complex mix of historicallegacies, institutional factors, and policy choices.The contention here is that the anticorruption institutional framework set up in theyears preceding Thaksin’s election failed largely because the extent of state capturethat ensued during the Thaksin era simply overwhelmed them. Specifically, I arguethat Thailand’s integrity and anticorruption institutional structure was based on“medium state capture/high administrative corruption” assumptions (quadrant C ofTable 7.1) instead of the “high state capture/medium administrative corruption”assumptions (quadrant B), or perhaps even the “high state capture/ highadministrative corruption” (quadrant D), the two quadrants I believe are best suitedto the circumstances during Thaksin’s government. In general, effective anticorruptionstrategies for countries in quadrant A should seek to capitalize on favourableconditions for strengthening political accountability and transparency through furtherinstitutional reforms. On the other hand, countries in quadrant D have highlyconcentrated economic interests capable of blocking reforms, as well as the generallyrestricting channels of access for countervailing interests, and ought to focus onbreaking the hold of vested interests on the process of reform (World Bank 2000).As mentioned in an earlier section, corruption in Thailand not only predates Thaksin,but also has long been recognized as deep-rooted, diverse and complex in both sizeand form. Yet, although Thailand during the period leading up to the early to mid-1990s had considerable unresolved corruption problems in the public sector, it did notexhibit any considerably higher levels of state capture or administrative corruptionthan other countries in the region at similar levels of socioeconomic development(Quah 1999). Indeed, what distinguished pre-Thaksin Thailand from most of itsSoutheast Asian neighbours may not have been the relatively modest level of thecorruption challenges the country faced, but the level of development of many of itskey institutions and tools necessary to confront these challenges. These not onlyincluded a civil society that was beginning to grow in sophistication and strength(particularly following the abortive 1991 coup) a relatively dynamic independentmedia, an evolving system of institutional restraints within the state, fairly well-developed public administration institutions, more robust political competition thanbefore, and an increasingly sophisticated electorate. One could even argue that theproblem of corruption in Thailand at the time had greater exposure and hence greaterpublic recognition in the 1990s.However, even with many of the key constituents as well as some of the institutionsand tools to combat corruption, there was little political will among most of Thailand’ssuccessive, mostly coalition, governments in the late 1980s and much of the 1990s.That lack of political will, coupled with the inability to create enduring



Page 130anticorruption coalitions, particularly at the highest levels of Thailand’s political system,proved to be among the most serious risks to an effective anticorruption strategy, atleast until the promulgation of the 1997 constitution. Unwilling to launch seriousanticorruption efforts by tackling corruption at the very top, Thai political leaders bredfurther complacency at all levels of the state as well as cynicism among the public. Infact, some scholars, for example Pasuk Phongpaichit and Sungsidh Phiriyarangsan(1996:131–66) found that the challenges posed by corruption in the Thai public sectorwere indeed compounded by widespread cynicism among the population and aprofound lack of trust in the state. Whether or not these traits were cultural legaciesor offshoots of the many years of military rule and uncertain democracy, theyundermined the pressure that civil society could potentially exert on politicians andbureaucrats. Thus, the government had less accountability in addressing bothadministrative corruption and state capture, and monitoring and accountabilitymechanisms vital for containing both problems were stymied.Among the various “reform” factions that wrote Thailand’s 1997 Constitution weresome that set out to tackle the problem of corruption. Starting with what one couldrefer to as a more “traditional” approach, the foundation was laid for the eventualintroduction of “best practice” and “good governance” technocratic reforms andknowledge transfers, including those meant to tackle corruption, particularly in thearea of public administration as briefly outlined in an earlier section.As Prime Minister, Thaksin and influential interests close to him both within andoutside government used their influence to shape the country’s legal and regulatoryframework to their own advantage (i.e. capture), thereby exacerbating the corruptionproblem. In addition, administrative corruption thrived. Just to list a few examplesfrom Table 7.2: the Shin Corporation Plc share sale exhibited tax avoidance; Thaksin’swife, Pojaman Shinawatra’s Ratchadapisek land purchase violated the NCC Act; thescandal involving the procurement of the baggage handling system for SuvarnabhumiAirport was an alleged bribery case; the Airport Rail Link had a faulty tender process,and so on – all examples of administrative corruption anticipated in the Constitution,with appropriate processes in place to investigate, prosecute and convict. However,these processes did not work effectively because of the measures cited in Sectionfour, inter alia, the Thaksin regime intimidated the country’s constitutional officerbearers as well as their supporters in the media and civil society. All of this points tothe incompleteness of the country’s 1997 constitutional measures. Safeguards againstadministrative corruption were needed, but in addition, more robust measures shouldhave been added to prevent state capture.ConclusionDespite the presence of strong anticorruption policies, state and regulatory capturecorruption often persists and thrives in the highest echelons of some countries. Thispaper explored this contradiction based on the case of Thailand under former Prime



Page 131Minister Thaksin Shinawatra’s administration, and using a common typology thatdifferentiates between administrative corruption and state capture. Specifically, thepaper argued that the primary reason for this contradiction in Thailand lies in thecountry’s post-1997 anticorruption framework, which, because it was overwhelmed bythe corruption during the Thaksin era, proved to be incomplete. Effective control ofnot only administrative corruption, but also state and regulatory capture – the latter asignificant form of corruption that characterized the Thaksin era – called for a strategythat should have extended far beyond the technocratic approaches focused on byThai reformers in the mid to late 1990s. In the end, whereas the 1997 “goodgovernance” constitution introduced various instruments to curb administrativecorruption, in particular, conflict of interest mechanisms, the Thaksin era –demonstrated the ineffectiveness of the implementation of many of thosemechanisms.Even though we can not and should not generalize from this one case, we can at thevery least appreciate the challenges of addressing both administrative corruption andthe various forms of “capture” in any integrity, anticorruption and “good governance”reforms.Notes1 To be sure, this is a disputed contention (see, for example, Bukovansky 2006;Nanda 2006).2 In addition to administrative units, NCCC comprised seven main bureaus: CorruptionPrevention Measures; Public Relations and Ethics Promotion; Policy and Planning;Corruption Suppression in Local Government and the Social Sector; CorruptionSuppression in other sectors; Asset Inspection; and Legal Affairs.ReferencesArghiros, D. 2001, Democracy, Development and Decentralization in ProvincialThailand, Democracy in Asia, no. 8, Richmond, Curzon.Bardhan, P. 1997, ‘Corruption and Development: A Review of the Issues’, Journal ofEconomic Literature, 35 (September): 1320–46.Bidhya Bowornwathana 2004, ‘Thaksin’s Model of Government Reform: PrimeMinisterialisation Through “A Country Is My Company” Approach’, Asian Journal ofPolitical Science, 12(1): 133–51.Brinkerhoff, D. 1996, ‘Perspectives on Participation in Economic Policy Reform inAfrica,’ Studies in Comparative International Development, 31(3): 123–51.Bukovansky, M. 2006, ‘The Hollowness of Anti-Corruption Discourse’, Review ofInternational Political Economy, 13(2): 181–209.Callahan, W. 2005a, ‘The Discourse of Vote Buying and Political Reform in Thailand’,Pacific Affairs, 78(1): 95.——2005b, ‘Social Capital and Corruption: Vote Buying and the Politics of Reform inThailand’, Perspectives on Politics, 3(3): 495–508.Chambers, P. 2005, ‘Evolving toward What? Parties, Factions, and Coalition Behaviorin Thailand Today’, Journal of East Asian Studies, 5(3): 495–520.Chutatong Charumilind, R. Kali, and Yupana Wiwattanakantang 2006, ‘ConnectedLending: Thailand before the Financial Crisis’, Journal of Business, 79(1): 181–217.Doig, A., and S. Riley 1998, ‘Corruption and Anti-Corruption Strategies: Issues andCase Studies from Developing Countries’, in S. J. Kpundeh and I. Hors eds. Corruptionand



Page 132Integrity Improvement Initiatives in Developing Countries, New York, NY, UnitedNations Development Program.Doner, R. F., and Anek Laothamatas 1994, ‘Thailand: Economic and PoliticalGradualism’, in S. Haggard and S. B. Webb, eds. Voting for Reform: Democracy,Political Liberalization, and Economic Adjustment, New York, Oxford University Press.Doner, R. F., and A. Ramsay 1997, Competitive Clientelism and Economic Governance:The Case of Thailand, in S. Maxfield and B. R. Schneider, eds. Business and the Statein Developing Countries, Ithaca, Cornell University Press.——2000, Rent-Seeking and Economic Development in Thailand in M. H. Khan and J.K. Sundaram, eds. Rents, Rent-Seeking and Economic Development: Theory andEvidence in Asia, New York, Cambridge University Press.Fogel, K. 2006, ‘Oligarchic Family Control, Social Economic Outcomes, and the Qualityof Government’, Journal of International Business Studies, 37(5): 603–22.Freedman, A. 2006, ‘Thailand’s Missed Opportunity for Democratic Consolidation’,Japanese Journal of Political Science, 7(2): 175–93.Ganesan, N. 2004 ‘Thaksin and the Politics of Domestic and Regional Consolidation inThailand’, Contemporary Southeast Asia, 26(1): 26–44.——2006, ‘Appraising Democratic Consolidation in Thailand under Thaksin’s Thai RakThai Government,’ Japanese Journal of Political Science, 7(2): 153–74.Ginther, K., E. Denters, and P. J. I. M. d. Waart 1995, Sustainable Development andGood Governance, Boston, MA, Kluwer Academic Publishers.Greenfield, K. T. 2001, ‘In the Clear’, TIME Asia, 14.Grindle, M. S., and J. W. Thomas 1991, Public Choices and Policy Change: ThePolitical Economy of Reform in Developing Countries, Baltimore, Johns HopkinsUniversity Press.Hellman, J. S., G. Jones, and D. Kaufmann 2003, ‘Seize the State, Seize the Day:State Capture and Influence in Transition Economies’, Journal of ComparativeEconomics, 31(4): 751–73.Hellman, J. S., and M. A. Schankerman 2000, ‘Intervention, Corruption and Capture –the Nexus between Enterprises and the State’, Economics of Transition, 8(3): 545–76.Hicken, A. 2006, ‘Party Fabrication: Constitutional Reform and the Rise of Thai RakThai’, Journal of East Asian Studies, 6(3): 381–407.Hutchcroft, P. 1997, ‘The Politics of Privilege: Assessing the Impact of Rents,Corruption, and Clientelism on Third World Development’, Political Studies, 45(3):639–58.Huther, J., and A. Shah 2000, Anti-Corruption Policies and Programs: A Framework forEvaluation, Washington, DC, World Bank, Operations Evaluation Department, CountryEvaluation and Regional Relations Division.Imai, M. 2006, ‘Mixing Family Business with Politics in Thailand’, Asian EconomicJournal, 20(3): 241–56.Johnston, M., ed. 2005, Civil Society and Corruption: Mobilizing for Reform, Lanham,MD, University Press of America.Kang, D. C. 2002, Crony Capitalism: Corruption and Development in South Korea andthe Philippines, Cambridge Studies in Comparative Politics, New York, CambridgeUniversity Press.Kaufmann, D. 2001, ‘Improving Governance and Controlling Corruption: New EmpiricalFrontiers and the Case for Collective Action’, in L. M. Gabriel, R. Stapenhurst and M.Thomas, eds. The Role of International Donors in Fighting Corruption, Maastricht, TheNetherlands, Dutch Ministry of Foreign Affairs and the World Bank Institute.Khan, M. H. 1996, ‘A Typology of Corrupt Transactions in Developing Countries’, IdsBulletin-Institute of Development Studies, 27(2): 12.Khan, M. H., and K. S. Jomo 2000, Rents, Rent-Seeking and Economic Development:Theory and Evidence in Asia, Cambridge, New York, Cambridge University Press.Kidd, J., and F. J. Richter 2003, Corruption and Governance in Asia, Houndmills, UK,New York, Palgrave Macmillan.



Page 133Kuhonta, E., and A. M. Mutebi 2006, ‘Thaksin Triumphant: The Implications of One-Party Dominance’, Asian Affairs: An American Survey, 33(1): 39–51.Kurlantzick, J. 2003, ‘Democracy Endangered: Thailand’s Thaksin Flirts withDictatorship’, Current History, 102(665): 285–90.Larmour, P., and N. Wolanin 2001, Corruption and Anti-Corruption, Canberra, AsiaPacific Press.Levitsky, S., and L. A. Way 2002, ‘The Rise of Competitive Authoritarianism’, Journalof Democracy, 13(2): 51–65.Lim, L. Y. C., and A. Stern 2002, ‘State Power and Private Profit: The PoliticalEconomy of Corruption in Southeast Asia’, Asian-Pacific Economic Literature, 16(2):18–52.Maor, M. 2004, ‘Feeling the Heat? Anticorruption Mechanisms in ComparativePerspective’, Governance: An International Journal of Policy and Administration, 17(1):1–28.Marsh, I., J. Blondel, and T. Inoguchi 1999, Democracy, Governance, and EconomicPerformance: East and Southeast Asia, Changing Nature of Democracy, Tokyo, NewYork, United Nations University Press.McCargo, D. 2002, ‘Democracy under Stress in Thaksin’s Thailand’, Journal ofDemocracy, 13(4): 112–26.McCargo, D., and Ukrist Patthamanan 2005, The Thaksinization of Thailand, Studies inContemporary Asian History, Copenhagen, NIAS.Mutebi, A. M. 2003, ‘Thailand in 2002 – Political Consolidation Amid EconomicUncertainties’, Asian Survey, 43(1): 101–12.——2004, ‘Thailand in 2003 – Riding High Again’, Asian Survey, 44(1): 78–86.——2006, ‘Thailand’s Independent Agencies under Thaksin: Relentless Gridlock andUncertainty’, Southeast Asian Affairs, 2006: 303–21.Nanda, V. P. 2006, ‘The “Good Governance” Concept Revisited’, The ANNALS of theAmerican Academy of Political and Social Science, 603(1): 269–83.Nattakrit Thewphaingarm 1996, ‘Corruption and Thailand’s Economic and PoliticalDevelopment’, Thesis (LL M), School of Law, University of Wisconsin-Madison,Madison, WI.Neher, C. 1976, Modern Thai Politics: From Village to Nation, Cambridge, Mass,Schenkman Pub. Co.——1977, ‘Political Corruption in a Thai Province’, Journal of Developing Areas, 11(4):484–9.Nelson, M. H. 2002, ‘Thailand’s House Elections of 6 January 2001: Thaksin’sLandslide Victory and Subsequent Narrow Escape’, in M. H. Nelson, ed. Thailand’sNew Politics: KPI Yearbook 2001, Bangkok, Thailand, King Prajadhipok Institute.Nishizaki, Y. 2005, ‘The Moral Origin of Thailand’s Provincial Strongman: The Case ofBanharn Silpa-Archa’, South East Asia Research, 13(2): 184–234.Norris, P. 1999, Critical Citizens: Global Support for Democratic Government, Oxford,New York, Oxford University Press.Ockey, J. 1994, ‘Political Parties, Factions, and Corruption in Thailand’, Modern AsianStudies, 28: 251.——2003, ‘Change and Continuity in the Thai Political Party System’, Asian Survey,43(4): 663.——2004, ‘State, Bureaucracy and Polity in Modern Thai Politics’, Journal ofContemporary Asia, 34(2): 143.Painter, M. 2006, ‘Thaksinisation or Managerialism? Reforming the Thai Bureaucracy’,Journal of Contemporary Asia, 36(1): 26.Pasuk Phongpaichit, and C. Baker 2004, Thaksin: The Business of Politics in Thailand,Chiang Mai, Thailand, Silkworm Books.——2005, ‘Thaksin Dismantles the Opposition’, Far Eastern Economic Review, 168(3).Pasuk Phongpaichit, and Sungsidh Phiriyarangsan 1996, Corruption and Democracy inThailand, Chiang Mai, Thailand, Silkworm Books.Pasuk Phongpaichit, Sungsidh Phiriyarangsan, and Nualnoi Treerat 1998, Guns, Girls,



Page 134Gambling, Ganja: Thailand’s Illegal Economy and Public Policy, Chiang Mai, SilkwormBooks.Philp, M. 2001, ‘Corruption and State Capture: An Analytical Framework’, in Paperpresented at the 10th International Anti-Corruption Conferences, 1–10 October 2001,“Designing Strategies, Assessing Impact, and Reforming Corrupt Institutions”, Prague,Czech Republic.Quah, J. S. T. 1999, ‘Corruption in Asian Countries: Can It Be Minimized?’ PublicAdministration Review, 59(6): 483–94.——2006, ‘Curbing Asian Corruption: An Impossible Dream?’ Current History,105(690): 176–9.Rock, M. T. 2000, ‘Thailand’s Old Bureaucratic Polity and Its New Semi-Democracy’, inM. H. Khan and J. K. Sundaram, eds., Rents, Rent-Seeking and EconomicDevelopment: Theory and Evidence in Asia, New York, Cambridge University Press.Rose-Ackerman, S. 1999, Corruption and Government: Causes, Consequences, andReform, New York, Cambridge University Press.Sakkarin Niyomsilpa 2000, The Political Economy of Telecommunications Reforms inThailand, New York, Pinter.Schacter, M., and A. Shah 2000, ‘Anti-Corruption Programs: Looking Before You Leap’,in Paper presented at the ADB-OECD Conference on Fighting Corruption in Asian andPacific Economies: Fighting Corruption in the New Millennium, Seoul, South Korea.Simpson, A. 2006, ‘Downfall? Capitalism, Dissent and the Media in Thaksin’s Thailand’,in Paper presented at the Second Oceanic Conference on International Studies, 5–7July 2006, University of Melbourne.Tanzi, V., and H. Davoodi 1998, Roads to Nowhere: How Corruption in PublicInvestment Hurts Growth, Washington, DC, International Monetary Fund.Theobald, R. 1990, Corruption, Development, and Underdevelopment, Durham, DukeUniversity Press.Thinapan Nakata 1977, Bureaucratic Corruption in Thailand: Incongruities betweenLegal Codes and Social Norms, Bangkok, School of Public Administration.Thomas, M. A., and P. Meagher 2004, A Corruption Primer: An Overview of Conceptsin the Corruption Literature, Baltimore, MD, Center for Institutional Reform and theInformal Sector: University of Maryland.Ukrist Pathmanand 1998, ‘The Thaksin Shinawatra Group: A Study of the Relationshipbetween Money and Politics in Thailand’, Copenhagen Journal of Asian Studies, 13:60–81.United Nations Development Program 2005, International Arrangements to CombatCorruption: A Comparative Study, Bangkok.Ungpakorn, J. G. 2002, ‘From Tragedy to Comedy: Political Reform in Thailand’,Journal of Contemporary Asia, 32(2): 191.Waterbury, J. 1973, ‘Endemic and Planned Corruption in a Monarchical Regime’, WorldPolitics, 25(4): 533–55.Wingfield, T. 2002, ‘Democratization and Economic Crisis in Thailand: Political Businessand the Changing Dynamic of the Thai State’, in E. T. Gomez, ed., Political Business inEast Asia, London, Routledge.World Bank 1991, The Governance Dimension, Washington, DC, The World Bank.——1999, Report and Recommendation of the President of the IBRD to the ExecutiveDirectors on a Proposed Public Sector Reform Loan, Washington, DC, World Bank.——2000, Anticorruption in Transition: A Contribution to the Policy Debate, A WorldFree of Poverty, Washington, DC, World Bank.



Page 1358 The judiciary’s role in good governance in KoreaJoongi KimIntroductionIn Korea, transparency and accountability throughout the public sector havesignificantly improved. Among recent achievements, the general parliamentary andpresidential elections, held in 2002 and 2004 respectively, have been credited as thecleanest in modern history. Public trust toward elected officials and bureaucracy hasimproved due to progress in better governance not only through legal and regulatoryreform, but more importantly through stricter enforcement. This paper will focus onthe role that the judiciary branch has played in improving public sector governance inKorea.Overall, the role of the judiciary in achieving good governance has thus far beenunderappreciated. The emphasis of this paper will be on the past decade since thefinancial crisis, during which the judiciary has become appreciably more active in legalenforcement. It will argue that the increased level of enforcement has become thecornerstone for the consolidation of rule of law and good governance. The paper willconcentrate particularly on enforcement through the judicial sentencing against seniorpublic officials for violations in two areas: campaign financing and election laws, andanticorruption laws. The level of enforcement in these two sectors serves as abarometer in the progress toward good governance because, as they are part of theestablishment that tends to maintain a concentration of power, they are among themost protected. The focus will therefore be the judicial treatment that senior publicofficials such as National Assemblypersons or senior bureaucrats have received underthe legal system.While Korean courts were haphazard and inconsistent in the past, they have becomemarkedly stricter in applying the law and levying countermeasures. Ultimately, thepotential for credible legal enforcement has had a significant impact in establishingdiscipline necessary to accomplish good governance in the public sector. This paperwill hypothesize on the factors that led judicial enforcement to be “soft” in prior years.Institutional dynamics that led to the transformation of the judiciary will then bediscussed. Reforms such as sentencing guidelines emerge as being necessary formore uniformity in court judgments in white-collar crime cases. Finally, this paper willtry to distil some comparative lessons that can be learned from the Koreanexperience.



Page 136The judiciary’s role in good governanceEmpirical research shows that governance plays a crucial role in a country’sdevelopment (Kaufman et al. 2006). Good governance of a country is a multifacetedtask that requires an infrastructure that must operate across a broad spectrum ofsectors. The World Bank, of the most closely watched measures, has traditionallyassessed governance in terms of six interconnected indicators: (1) rule of law; (2)control of corruption; (3) voice and accountability; (4) political stability; (5)government effectiveness; and, (6) regulatory quality. The judicial branch of mostgovernments has an essential function in the operation of these factors, withparticular impact upon the first three factors.As outlined under the World Bank’s indices, the rule of law includes such criticalfactors as the fairness of the judicial process, the independence of the judiciary, thestrength and impartiality of the legal system, and the honesty, enforceability andquickness of the court system. Trust in the legal system as implemented through thejudiciary therefore must exist. Yet, despite the importance of the rule of law ingovernance, Korea remains one of the few countries in the Organization for EconomicCooperation and Development (OECD) in Asia that maintains a weaker rule of lawrelative to their respective income categories (Graph 8.1). Korea must redouble itsefforts to overcome the negative perception toward its rule of law to improve itsoverall governance rating. The judiciary must function to render effectiveenforcement.

Graph 8.1 Rule of law (World, 2005).(Source: Worldwide Governance Indicators 1996–2007).



Page 137The control of corruption indicators covers the public’s trust in the financial honesty ofpoliticians and the degree to which elected leaders are not associated with graft. Thelegal system must remain impartial and stern when confronted with cases involvingsenior political officials. Otherwise, public credibility toward governance will beundermined. Once again, among Asian countries, Korea remains notably laggingbehind in this category relative to its income category (Graph 8.2).The voice and accountability role of governance is evaluated through civil and politicalrights, of which free and fair elections for public officials comprise an integralcomponent. Elections in turn can only operate effectively if relevant laws andregulations are implemented through proper enforcement. The failure of electoralcampaign infractions to be properly disciplined has been a continuous problem. Aswith the categories of rule of law and control of corruption, Korea still remains behindthe average level of other OECD countries, but notably Korea fares relatively betterthan many other comparable Asian countries. This progress can be attributed to thejudiciary’s transformation in taking an active stance in establishing legal disciplineagainst electoral violations.Despite the theoretical imperatives, many reasons can be found for the judiciary’slenient posture toward legal enforcement, particularly toward political and governmentofficials. First, for public officials, the reputational damage and public shame that theyexperience serve as sufficient deterrent and penalty. Because most senior officialstend to be first-time offenders, the public shame has such a powerful effect thatnumerous public officials facing investigations have even committed

Graph 8.2 Control of corruption (World, 2005).(Source: Worldwide Governance Indicators 1996–2007).



Page 138

Graph 8.3 Voice and accountability (World, 2005).(Source: Worldwide Governance Indicators 1996–2007).

Graph 8.4 Korea, South (2005).(Source: Worldwide Governance Indicators 1996–2007).



Page 139suicide. Contributions made to the country while acting as public servants during theirtenure are cited as another mitigating reason. From a socio-historical perspective,Confucian tradition also emphasizes respect for the authority held by senior publicofficials. Under this thinking, judges believed that harsh sentences would underminepublic trust in the government and would lead to negative consequences for thecountry (Chan Jin Kim 2000).A practical factor that cannot be dismissed is that senior officials can hire better legalrepresentation. This increases the likelihood that they can take full advantage of thelegal protections afforded to criminal defendants. To another degree, a bias againststrict punishment of white-collar crime exists. The economic damages were dismissedas being far less serious than other violence-ridden crimes. Suspicions also existedthat prosecutions in this area were politically motivated. Finally, courts face intrinsiclimitations in the administration of justice, such as the fact that they cannot initiatelitigation themselves. Only prosecutors have the authority to pursue criminal chargeson behalf of the state, and they themselves often rely upon regulators or victims torefer criminal cases or press charges. Hence, the lack of prosecution should be faultedfirst, not the judiciary. Taking into consideration these multifarious factors, the court’srole in establishing proper governance might appear confined within a narrow, legalframework.On the other hand, numerous negative factors such as graft or cronyism can be citedfor the judiciary’s leniency. The effectiveness of legal counsel, for example, especiallyin criminal cases, has been traditionally measured not by one’s performance record,but by whether one was a former judge in the relevant court and when one left suchposition. Public perception prevails that former judges can exercise influence oversitting judges, particularly within the same district, who were once their colleagues orjunior members. Given the rigid hierarchy within the judiciary coupled with Confuciantradition, junior justices, for example, face enormous social burdens to withstandrequests made by former superiors who also acted as their mentors. Newly retiredjudges have allegedly reaped more income in the first two years of retirement thanduring their entire previous career, because of their ability to wield influence overtheir former colleagues. Major law firms have been known to target judges that areretiring from courts where they have significant cases pending. Suspicions of sittingjudges providing ex parte communication and generous rulings to former colleaguesor superiors have plagued the judiciary. The practice of partiality has even led to thecoining of a pejorative term, “jeongwanyeu (respect to former judicial officials).”Other demographic and practical problems also complicated the situation in thejudiciary. For example, an inordinately large number of judges attended the same highschools and universities as many senior public officials. For example, until recently, 90percent of the judiciary came from only five universities, with 64 percent coming froma single university, and most were concentrated in the law departments (Kim 2006).Predominantly male as well, this exclusivity and these commonalities sowed the seeds,in the eyes of many, for communal, protectionist cronyism. From another perspective,inadequate compensation for judges has also been cited as a reason for theirsusceptibility to improper influence. In 1998 and



Page 1401999, two major scandals involving improper payments and relationships betweenattorneys and judges generated public uproar. Private practitioners were convicted forfinancially supporting and bribing judges.Several factors contributed to recent attitudinal shifts in the judiciary. First, theadvances in the democratic political process have helped spur judicial independence.With the consolidation of democratic, peaceful transitions of power, particularlybetween ruling parties and opposition parties, the judiciary has now been able tofocus more on the strict application of the law instead of being exposed to themanipulation of political forces. This process has helped create judicial accountabilityand scrutiny. During more authoritarian single-party rule, the executive branch couldwield inordinate influence over the judiciary through appointments and outrightthreats. A multi-partisan legislature now acts as a check and balance, and conductsrigorous screening of Supreme Court justices, a process that allows the Chief Justiceto appoint judges within the judiciary from an impartial position. Most recently,legislative branch scrutiny has even forced the withdrawal of judges nominated tosenior positions. Second, from a more practical standpoint, attempts have been madeto correct conflicts of interest. The courts have established ethics codes and retiringjudges now ostensibly face post-employment restrictions; as a general matter, publicofficials cannot do work related to their former public functions. This provision,however, has not been strictly applied in the case of former judges.Finally, the judicial system has undergone a demographic transformation. The size ofthe judiciary almost doubled, from 1,185 in 1994 to 2,157 in 2005. Not only did thenumber of judges increase, but also the diversity. The change in judicial culture canbe attributed to the broader range of judges with different backgrounds andexperiences, the jump in the number of female judges, who hold generally fairer andimpartial reputations, being among the most notable. The portion of women amongthe judiciary in particular has risen from 4.64 percent in 1994 to 14.75 percent in2005 and is expected to reach 30 percent by 2015, with the first female SupremeCourt judge being appointed in 2004. Diversity has therefore helped decentralize theconcentration of power that revolved around a limited number of judicial elites.Good governance can be achieved only through effective judicial discipline toward thepolitical and bureaucratic branches of government. Credibility in the judicial systemestablishes public confidence in a country’s governance platform. The Korean judiciaryfaced numerous obstacles, but a broad range of factors have contributed to enhanceits independence and effectiveness. As a newly industrialized country that joined theranks of high-income countries in the OECD, Korea offers an important case study ofthe potential and limitations that effective judicial enforcement can provide.Election and campaign financing reformOne of the most dramatic transformations in Korea’s governance involves theenhanced accountability and transparency of the electoral process. From a system



Page 141riddled with illicit solicitations, vote buying and other forms of graft, electoralcandidates now adhere to strict procedures when engaging in campaign activities. Akey driver behind these changes has been the judiciary’s role in establishing credibleenforcement discipline. One of the most significant measurements of change remainsthe number of newly elected senior public officials who were forced to forfeit theirpositions due to strict judicial sentences based upon violations of election-relatedlaws. Similarly, the delays associated with election trials that eviscerated the authorityof the judicial process have been effectively corrected.The risk of forfeiting one’s seat due to electoral violations has finally become a seriousthreat in Korea. An analysis of the fate of National Assemblypersons that served in themost recent National Assembly elections held in 2000 and 2004 proves revealing. First,in the 16th National Assembly election held in 2000, the National Election Commission(NEC) underwent a significant shift toward active enforcement. They requestedcriminal prosecution or investigation of a record 656 persons (Table 8.1). The subjectsincluded not only parliamentarians themselves, but also their accountants, staffmembers and direct relatives. At the next stage, the prosecutor’s office maintained asimilar commitment by indicting 283 persons. For example, out of 308 NationalAssemblypersons elected, 57 were indicted for election law violations. Ultimately, thecourts completed the cycle by sentencing 11 National Assemblypersons to prison andfined 69 of them more than a million won (US$1,000), the benchmark that mandatesforfeiture of an electoral position. Twelve National Assemblypersons therefore endedup relinquishing their seats due to the strict sentencing.In the following 17th National Assembly that was elected in 2004, the processcontinued. The National Election Commission intensified their efforts and referred 854persons for criminal prosecution or investigation. The prosecutor’s office in turnremained firm and indicted 535 of these persons. Ultimately, the courts rendered eightprison sentences and 158 fines of over a million won each. Out of the 46parliamentarians indicted for election law violations, 12 were eventually forced torelinquish their seats due to strict prosecutions and court sentences. At the sametime, critics cite that 34 parliamentarians still managed to receive fines just below amillion won and maintain their seats. This progress contrasts with the treatmenttoward the members of the 15th National Assembly, who served from 1996 until2000. For those cases, 18 National Assemblypersons were indicted and 14 receivedfines over a million won at the trial level, but on appeal, seven had their finesreduced below the benchmark level so that only seven eventually lost their seats.Some experts advocate that the minimum monetary benchmark of a million won itselfshould be abolished so that anyone sentenced to a fine of any amount wouldautomatically lose their elected office. In total, out of 103 National Assemblypersonsindicted for electoral violations between 1993 and 2005, 20 received prison sentencesor fines that forced them to give up their positions.Another factor that has contributed to the establishment of judicial discipline has beenthe speed at which courts now review election-related violations. Historically, theadministration of justice suffered due to chronic delays of judicial review. Back in1969, Korea first adopted a provision that stipulated that election-related



Page 142Table 8.1 Election-law-relatedElection National

electioncommitteeaction

Prosecutorial decisionTotalIndictment Non-

indictment

Subtotal SuspensionTermination Nocharges

Investigationended

16thParliament(2000-2004)

Total 656 283 373 56 6 67 244

Requestprosecution

248 184 64 37 0 23 4

Investigationreferral

408 99 309 19 6 44 240

16thPresident(2002-2007)

Total 350 168 182 12 1 19 150

Requestprosecution

124 111 13 7 0 6 0


226 57 169 5 1 13 150


Total 854 535 319 14 4 93 208

Requestprosecution

453 419 34 8 1 25 0


401 116 285 6 3 68 208

Source: State of Prosecutions for Each Election, 2006, National Election Commission.



Page 143Table 8.2 Election-law-related court judgmentsElection National

electioncommitteeaction

JudgmentTotalPendingImprisonmentProbationFine Conditional

releaseInnocent

Over1 M\

Under1 M\


Total 283 0 11 49 142 69 12 0

Requestprosecution

184 0 6 32 86 51 9 0


99 0 5 17 56 18 3 0

16thPresident(2002-2007)

Total 168 0 1 21 75 66 5 0

Requestprosecution

111 0 12 46 50 3 0


57 0 1 9 29 16 2 0


Total 535 1 8 78 271 158 11 8

Requestprosecution

419 0 7 63 215 121 9 4


116 1 1 15 56 37 2 4

Note (M\=million won).(Source: State of Trial Proceedings for Each Election, 2006, National ElectionCommission).



Page 144Table 8.3 Sectors affected by corruption (2005)Countries Legal system/judiciary Political parties Parliament/legislatureIndia 4.3 4.7 4.4Indonesia 3.8 4.2 4.0Korea 3.7 4.4 4.4Philippines 3.4 4.2 4.2Taiwan 3.4 4.1 4.3Japan 3.0 4.2 3.7Malaysia 2.9 3.7 3.1Thailand 2.8 3.9 2.8HK 2.4 3.1 3.5Singapore 2.1 2.2 1.8Average 3.18 3.87 3.62(Source: Transparency International, Global Corruption Barometer 2005; 1 = not atall; 5 = extremely).trials should be completed within one year. The provision, however, remaineddeclaratory at best and did not mention anything about factoring in the length ofappeals. In 1991, 22 years later, the time period was reduced to six months, and thenin 1994, specific deadlines were imposed for appeals. Judicial review at the High Courtand Supreme Court levels was limited to three months so that the total timeexpended would not exceed the maximum one year. The deadlines then became“mandatory provisions” in February 2000.Despite the legislative developments, among the cases against parliamentarians andtheir staff of the 16th National Assembly, none were completed within the new legallystipulated time period of one year, including appeals. In fact, only 55.9 percent weredecided within six months at the trial level, while only 15.5 percent were decidedwithin three months at the first appeal level. The statistics still represent animprovement from the 15th National Assembly, when 28.6 percent were completedwithin six months at the trial level and none were completed within three months atthe first appeal level. In a notorious case, one indicted National Assemblymanmanaged to complete all but five months of his four-year tenure because his criminaltrial lasted for over 38 months due to delays and continuous appeals, which went allthe way up to the Supreme Court. In practice, the deadlines still remained hortatorywithout adherence by the courts because nothing could force them to comply. Thisundermined the deterrent effect of enforcement. Nothing altered the belief amongaspirants to public office that they should just seek to get elected at all costs becausethey could delay future trials for extended periods.In 2004, the judiciary finally decided to take a firmer stance toward meeting



Page 145these timelines. The Supreme Court announced that it would designate for expeditedreview serious electoral violations that could jeopardize an elected official’s status.This marked a significant shift compared to the past. The Supreme Court alsodeclared that they would follow the timelines as provided under the law so that thetotal time expended including appeals would not exceed the maximum one year. Thisdramatically changed the electoral landscape. For the 17th National Assembly, 94.6percent of the cases were completed within six months at the trial level and 51.4percent were decided within three months. In May 2006, the courts decided to furtherexpedite the process so that the time for judicial review would be within two monthsfor each stage, so that the total time from trial level to the Supreme Court would notexceed six months. At present, only nine cases have taken more than a year to reachtheir final judgment.Finally, another legislative measure that added significant momentum to theenforcement process was the increase in whistleblower rewards that was adopted inMarch 2004. According to new provisions, whistleblowers who notify authorities ofelectoral violations could receive rewards as large as 500 million won (US$500,000).This led to a considerable increase in notifications. In the 16th National Assemblyelection, whistleblowers only received 4.86 million won (US$4,860) for 78 cases,whereas in the 17th National Assembly election they received a substantially largersum of 770 million (US$770,000) for 349 cases. This provided prosecutors, andultimately courts, more evidence with which to convict violators.As testimony to the overall confidence in the enforcement involving election-relatedviolations, the number of legal challenges brought by losing candidates has declined.Losing candidates have the right to raise claims of election law violations or requestprosecutions against their winning opponents. The number of such instancesincreased from nine cases in the 15th National Assembly election to 28 cases in the16th National Assembly election, but substantially declined to three cases in therecent 17th National Assembly election. This demonstrates that not only arecandidates complying with regulations to a higher degree, but that losing candidateshave reached higher levels of satisfaction toward the level of monitoring, action andenforcement carried out by the legal system.As witnessed in the realm of electoral process, the judiciary therefore has played asignificant part in creating a more effective compliance framework. This process tooka combination of legislative changes, stricter sentencing, faster review, greaterincentives for whistleblowers and, ultimately, judicial recognition of the importance ofestablishing enforcement discipline. The result has been historic progress.Corruption-related judicial decisionsCritical governance indicators such as the rule of law, control of corruption,government effectiveness, and regulatory quality cannot be fulfilled properly unlessKorea can shed the perception that senior public officials in elected office and thebureaucracy receive inadequate penalties for bribery and other corruption-relatedcases. Convincing judicial action plays a crucial role in overcoming this perception.



Page 146

Graph 8.5. Rule of law (World, 2005).(Source: Worldwide Governance Indicators 1996–2007).While steady progress has been achieved, Korea’s standing in corruption indices incomparative rankings, for example, has remained relatively constant in recent years.The 2006 Transparency International Corruption Perceptions Index gives Korea a 5.1rating that amounts to 42nd out of 163 countries, but Korea significantly trails theOECD average of 7.2, and has been unable to close its gap with other leading Asiancountries such as Japan or Singapore. The lack of substantive progress can be tracedto the lack of effectiveness of courts in establishing credible enforcement discipline forcorruption involving public officials, unlike their success in electoral violations.The weakness in enforcement discipline in corruption-related offenses can bemeasured through the way courts have issued suspended sentences for convictions.For instance, in general graft-related cases, the courts have issued a high degree ofsuspended sentences against defendants even after levying prison terms. From 1981until 1988, the portion of suspended sentences against senior public officials forcorruption-related offenses was 65.63 percent (Table 8.4). Even after the full-fledgedtransition to democracy (other than a temporary decline to 41.03 percent between1988 and 1992), from the entire period from 1992 until 2004, spanning threepresidencies, the percentage of suspensions still stood at 65.35 percent. As a generalmatter, from 1983 until 2003, the commutation rate for all crimes as a whole hasranged three to four percentage points lower (between 61 to 62 percent). Thisdifferential serves as an indicator of the lenient posture taken by the courts towardcorruption cases involving senior public officials,



Page 147Table 8.4 Corruption-related court judgments Court sentence

Timeperiod

Guiltyjudgment

Servedactualprison

sentence

Suspendedsentence

Averageprison

sentences

% ofsuspendedsentences

All crimes 1983–1992*

896,914 342,103 554,811 – 61.96

1993–2003

1,695,551 667,806 1,027,745 – 60.61

1981–1988.2

64 22 42 3 yrs 65.63

Crimesinvolvingseniorofficials

8 mo

1988.3–1993.2

39 23 16 2 yrs 41.03

11 mo 1993–

2004Corruptioncases

131 41 90 – 68.70

Aggravatedcrimes under50 MW

62 13 49 – 79.03

Aggravatedcrimes over50 MW

35 17 18 – 51.43

Aggravatedcrimes total

97 30 67 – 69.07

Total cases 202 70 132 – 65.35Note * Only trial level statistics.(Source: Combination of Donga Ilbo and Jang Joon-Oh, p. 62.)



Page 148In a similar manner, although Korea has a specific law that stipulates aggravatedpunishment for certain corruption-related offenses, not only were stiffer sentences notrendered, but they also became futile due to suspensions. An analysis of thecorruption cases involving grand corruption where public officials were sentencedunder aggravated punishment provisions highlights the problems. Until recently,wrongdoers involved in bribes greater than 50 million won (US$50,000) were subjectto prison sentences longer than 10 years, whereas amounts between 10 million wonand 50 million won (US$10,000–50,000) required sentences greater than five years.Instead of the stricter sentences as mandated under the spirit and letter of the law,courts have been unusually lenient.Specifically, from 1993 through 2004, 97 officials were convicted for violating theaggravated provisions related to bribery and peddling influence (Table 8.4). Amongthe convicted, however, only 30 (or 30.9 percent) served actual prison sentences,whereas the remaining 67 (or 69.1 percent) received commutations. Instead of beinglower, the suspension rate was 4 percent higher than the average 65 percent forgeneral corruption cases, and almost 8 percent higher than the rate for all crimestaken together. Among the 35 defendants convicted for receiving more than 50million won in bribes (who should have received the stiffest sentences of ten years ormore), only 17 (or 48.6 percent) received actual prison sentences, while 18 (or 51.4percent) received suspended sentences (Table 8.4). The suspension rate was lower inthese cases, but the portion of suspensions still remains inappropriately high given thegrander scale of corruption involved. In terms of actual sentencing against NationalAssemblypersons, over the past 12 years only 17 have been indicted for non-electionlaw violations and only five received prison sentences that led to forfeiture of theirpositions. Courts have thus exercised maximum judicial discretion to reduce sentencesby up to 50 percent of mandated terms. In the end, over half of the defendants foundguilty under the aggravated punishment ultimately did not serve actual prisonsentences.One criticism that levied toward the aggravated punishment law has been that itsprovisions were too stringent. This particularly holds some weight when one considersthat the benchmark amounts of 10 million and 50 million won were established in1990 and were only amended in 2006 to reflect inflation. Many judges also believedthat 10 years’ imprisonment for serious crimes, for example, was too harsh, especiallywhen compared to sentences meted out to violent criminals. Nevertheless, theseremain legislative considerations and the courts technically should strictly apply thelaw.Given that senior public officials are prosecuted on a more selective basis, the higherrates of suspensions are significant. Prosecutions against senior officials generallyoccur under more extraordinary circumstances, such as when overwhelming evidenceexists or the scale of wrongdoing or sums involved are inordinately large. For publicpolicy reasons, courts should have been tougher toward public officials, particularly ifthey wanted to send a message, under media scrutiny, that corruption would not betolerated for those that breached public trust. They should have established thatsenior public officials faced serious consequences if they compromised public trust,particularly given the enormous reputational



Page 149and pecuniary benefits they derived. Technically, public officials found guilty forcorruption-related crimes would lose their positions even if they received a suspendedcriminal sentence and, when combined with reputational damage, this might offerconsiderable deterrence. Nevertheless, the high portion of suspensions added to theimpression that when society’s trust-bearers committed grand corruption they weretreated more leniently.As with election violations in the past, court delays for more serious crimes such asbribery and political financing have plagued the system. Public officials have beenallowed to remain in their positions for extended periods until all appeals areexhausted. In the 17th National Assembly, for example, eight parliamentarianspromptly received prison sentences at the trial and appeal level for corruption-relatedcharges, but the Supreme Court’s final decision required over a year to be completed.This contrasts with the recent changes in the court’s position toward electoralviolation cases. To a certain extent, the relatively longer delays for corruption-relatedcases could reflect the complexity of facts because they involve more serious crimesthat require more time to review for a judgment to be reached. Nevertheless, anydelays add to the perception of judicial ineffectiveness.Finally, a critical factor that has undermined judicial enforcement in corruption cases isa long history of granting presidential pardons and reinstatements on a liberal basis.Pardons remain a common measure available to the head of state of most modern,democratic societies. Reinstatement in turn allows those that have completed prisonsentences but have been restricted from certain official positions to regain their statusor qualifications. The problem in Korea arises because these pardons andreinstatements have been issued to an excessive degree. Korea maintains the highestnumber of presidential pardons compared to any major economy. The practiceundercuts the authority of the courts and public trust in the legal system. Koreanpresidents issue special pardons on a regular basis when they first gain office, andthen again on special holidays such as Buddha’s Birthday in May

Graph 8.6 Original prison sentence and actual sentences served for senior publicofficials.(Source: Special Investigation Team 2005, ‘[Investigative Report] Released AfterServing One-Third of Actual Sentences’).



Page 150Table 8.5 Corruption-related pardons and re-instatementsGovernment Surveys Prison sentence

without suspensionPardons

(%)Reinstatements

(%)Chun Doo Hwan1980.9–1988.2

A 39 8 (20.5) 10 (25.6)

Roh Tae Woo1988.2–1993.2

A 64 32(50.0)

41 (64.0)

B – 6 Kim Young Sam1993.2–1998.2

A 54 28(51.8)

12 (22.2)

B – 15 (49) Kim Daejung1998.2–2003.2

A 7 0 (0) 0 (0)

1998.2–1999.12 B - 8 (42) Roh Moo-hyun2003.2–2008.2

2003.2–2006.11 - 19

C 42 / 22 / 1993–2004 /

Over 50 MW17 less

than 7

C 70 22(31.4)

1993–2004 /Under 50 MW

Total MW = Million Won(A: Jang Joon-Oh 2001; B: Segye Ilbo 2005; C: Donga Ilbo 2005).and National Liberation Day in August. Hence, even when courts have sentencedpublic officials to imprisonment, a substantial portion of them avoid serving most oftheir terms because they receive presidential pardons and reinstatement (Table 8.5).To stem the excesses of the executive branch, some advocate the establishment of aPardons Committee composed of unbiased experts to review and recommendapplicants for pardons. Another option proposes that presidential pardons should notbe allowed for election violators, corrupt officials or those that commit heinouscrimes. Yet another proposal seeks to require National Assembly approval for allspecial presidential pardons. Various legislative proposals have been made to try toreform the process, but they continue to remain in limbo.Overall, Korean courts need to follow the lessons learnt from election-relatedviolations and become more stringent in corruption-related cases. In 2003, theSupreme Court even declared that in principle all public officials who received bribesof more than five million won (U$5,000) should be subject to actual prison sentences.One potential solution is the adoption of stricter sentencing guidelines to establishmore uniformity in court judgments. In fact, judicial precedent has already beenestablished by two district courts adopting sentencing guidelines that focused



Page 151specifically on white-collar crimes in March 2006. The results and impact of theseinitiatives potentially have great promise in solidifying balanced and consistent judicialsentencing against corruption-related offenses.ConclusionsAs with many Asian countries, Korea must continue its efforts to shed the perceptionthat looms over it that enforcement related to governance remains soft, unpredictableand inconsistent. In particular, Korea has made great strides to improve enforcementdiscipline in the judicial sector. A bona fide democracy with peaceful transitions ofpower, a vibrant multi-party system, stable economic growth and critical oversightthrough the media have all helped the judiciary make this transition. Nevertheless,despite the progress, relative to its income category, Korea’s rule of law still lagsbehind.This paper analyzed the transformation of the judiciary through its application andsentencing of electoral law violations and corruption-related cases. These areasremain significant because they are the most exposed to political influences. Acomprehensive analysis confirms that judicial enforcement has improved, particularlywith regard to electoral violations, but must strengthen further in corruption cases.The courts must follow the lead taken in electoral cases to overcome obstacles toengaging in stricter sentencing and shorter judicial review. Recently adoptedsentencing guidelines, although of a limited scope, provide hope that more uniformand effective court judgments can be rendered. Ultimately, more judicious usage ofpresidential pardons by the executive branch will also help the judiciary to enhancethe efficacy of enforcement discipline.The judiciary plays an essential role in establishing the rule of law through its abilityto render criminal judgments. As the final arbiter, they instill public trust in thegovernance system through effective enforcement of the laws against elected officialsand the bureaucracy. The judicial sector therefore must embrace such full-fledgedreforms if the country is to reach levels of good governance achieved in the mostadvanced countries.ReferencesJudiciary Annul Reports, Office of Court Administration.Ministry of Justice Annual Reports, Ministry of Justice.National Election Commission Annual Reports, National Election Commission.2004. ‘Public Surveys’, Korean Institute for Criminal Justice Policy.2006. ‘Public Surveys’, Changwon and Jeonju District Court. March.Public Official Election Act.Political Funds Act Art.Special Crimes Aggravated Punishment Act.2000–2006. Law Times2004. Hankyoreh, 15 May.Buscaglia, Edgardo 2001, ‘An Analysis of Judicial Corruption and Its Causes: AnObjective Governing-Based Approach.’ International Review of Law and Economics,vol. 21, no. 233.



Page 152Buscaglia, Edgardo, and Dakolias, Maria 1999, ‘The Role of Legal Institutions in theEconomic Development of the Americas: An Analysis of the Causes of Corruption inthe Judiciary’, Law and Policy in International Business, vol. 30, no. 95, Winter.Carothers, Thomas 1998, ‘The Rule of Law Revival.’ Foreign Affairs, (March–April): 95.Chae, H. C., Lee, C. J., Kim, G. S., Kim, T. H., Kang, G. Y., Shin, M. Y., and Oh, S. J.2005, ‘Investigative Report: Major Dissection of Special Pardons.’ Segye Ilbo, 15 May.Corruption Perceptions Index 2006. Transparency International. Available online athttp:// www.transparency.orgDam, Kenneth W. 2006, The Judiciary and Economic Development. Univ. Chi. John M.Olin Law & Econ Working Paper No. 287, 2nd Series. Available online at http://ssrn.com/abstract=892030.Daniels, Ronald J. and Trebilcock, Michael 2004, ‘The Political Economy of Rule of LawReform in Developing Countries.’ Michigan Journal of International Law, vol. 26, no.99.Davis, Kevin E. 2004, ‘What Can the Rule of Law Variable Tell Us About Rule of LawReforms?’ Michigan Journal of International Law, vol. 26, no. 141.Ehrlich, Craig P., and Kang, Dae Seob 2002, ‘Independence and Corruption in Korea.’Columbia Journal of Asian Law, vol.16.Esty, Daniel C. 2006, ‘Good Governance At The Supranational Scale: GlobalizingAdministrative Law.’ Yale Law Journal, vol. 115, no. 1490.Global Corruption Barometer 2005. Transparency International.Hoggard, Shiloh 2004, ‘Preventing Corruption in Colombia: The Need for An EnhancedState-Level Approach.’ Arizona Journal of International and Comparative Law, vol. 21no 577.Hwang, Jeong Geun. 1999. ‘Direction for Amending the Election Laws’, Legal Times,vol. 6.Jang, Joon Oh 2001, ‘Corruption 1980–1990 Political and Economic Collusion’, KoreanInstitute of Criminology, vol. 1–13, p. 62.Jeong, Suk Yong 2005, ‘14 Times Increase Number of Female Prosecutors within Past10 years.’ Naeil Shinmun, 18 December.Kaufmann, D., Kraay, A. and Mastruzzi, M. 2006, Governance Matters V: GovernanceIndicators for 1996–2005. Available at: http://info.worldbank.org/governance/kkz2005[15 September 2006]Kim, Chan Jin 2000, ‘Korean Attitudes Towards Law’ Pacific Rim Law & Policy, vol.10,no. 6.Kim, Do Hyun 2006, ‘Career Patterns of the Korean Judges’, Law & Society, vol. 31.Min, Suk Gi 2006, ‘Completing Electoral Violations Trials within 6 Months.’ MaekyungEconomy, 2 May.Moyer, Homer 2004, ‘The Role of Law in Combating Official Corruption’, AmericanSociety International Law Proceedings, vol. 98, no. 169. People’s Solidarity forParticipatory Democracy (PSPD) Report, 28 Jan. 2005. Available online athttp://www.peoplepower21.orgSandgren, Claes 2005, ‘Combating Corruption: The Misunderstood Role of Law.’International Lawyer, vol. 39, no. 717.Shihata, Ibrahim F. I. 1998, ‘Role of the Judiciary in the Prevention and Control ofCorruption.’ Paper prepared for the Conference At Avvocatura Dello Stato, April 1,Rome.Special Investigation Team 2005, ‘[Investigative Report] Special Crimes AggravatedPunishment Act = Particularly Light Punishment Act’, Donga Ilbo, 12 January.Special Investigation Team 2005, ‘[Investigative Report] Special Crimes AggravatedPunishment Act = Particularly Light Punishment Act’, Donga Ilbo, 12 November.Special Investigation Team 2005, ‘[Investigative Report] 159 Indicted for ReceivingBribes only 41 Receive Actual Sentences’, Donga Ilbo, 12 January.Special Investigation Team 2005, ‘[Investigative Report] “Corrupt Heavyweights”Caught Through Investigations, How Were Their Cases Treated?’, Donga Ilbo, 12January.



Page 153Special Investigation Team 2005, ‘[Investigative Report] Even After Actual Sentences,Released Through Pardons or Parole’, Donga Ilbo, 13 January.Special Investigation Team 2005, ‘Major Dissection of Special Pardons’, Segye Ilbo, 15May.Special Investigation Team 2005, ‘[Investigative Report] Released After Serving One-Third of Actual Sentences’, Donga Ilbo, 13 January.Special Investigation Team 2006, ‘[Investigative Report] ‘Special Crimes AggravatedPunishment Act = Particularly light Punishment Act’, Donga Ilbo, 13 September.State of Prosecutions for Each Election 2006, National Election Commission (on filewith author).State of Trial Proceedings for Each Election 2006, National Election Commission (onfile with author).Tan, Eugene Kheng-Boon 2000, ‘Law and Values in Governance: The Singapore Way.’Hong Kong Law Journal, vol. 30, no. 91.Thachuk, Kim and Dakolias, Maria 2000, ‘Attacking Corruption in the Judiciary: ACritical Process in Judicial Reform.’ Wisconsin International Law, vol. 18, no. 353.Wallace, J. Clifford 1998, ‘Resolving Judicial Corruption While Preserving JudicialIndependence: Comparative Perspectives.’ California Western International LawJournal, vol. 28, no. 341.‘Worldwide Governance Indicators 1996–2007’, World Bank. Available online at http://info.worldbank.org/governance/wgi/index.aspYoo, Hee Youn 2004, ‘Special: Bribe Recipients, Actual Imprisonment only to 11%.’Munhwa Ilbo, 16 January.Yoon, Jong Gu 2006, ‘48 Election Law Violators, 34 Receive “Indulgence” Judgmentsthat Allow Maintenance of National Assembly Positions.’ Donga Ilbo, 13 September.



Page 1549 Healthcare reforms in ThailandRethinking conventional wisdomM. RameshBy the norms of contemporary mainstream public management thinking, the Thaihealthcare system is a policy disaster. Its already flawed healthcare system,characterized by the dominance of public hospitals, deteriorated, according to thismainstream view, with the adoption of a tax-funded financing scheme for publicsector workers in 1980 and health insurance for private sector workers in 1991. Themost severe lapse, however, was the adoption of the tax-funded Universal Coverage(UC) for the general population in 2001. As it turned out, healthcare expenditures didnot explode, and indeed declined with successive expansion of the state’s role in thesector. The Thai experience casts serious doubts on public sector reform thinking thatis sceptical of state intervention to address pressing public problems. The case studyshows that there is scope for plenty of innovations by policymakers daring to defyconventional thinking.Public ownership of service delivery agencies such as hospitals is an anathema incontemporary health economics and new public management because it is believed tobe inherently inefficient, notwithstanding contradictory empirical evidence1. Thesupposed differences between public and private hospitals is problematic becausethere is no reason to believe that ownership on its own makes a difference. Publichospitals can behave very much like their private counterparts in their competition forrevenues and surplus when allowed the opportunity, and private hospitals can bemade to serve public purposes under appropriate conditions. What is critical are theincentives confronting hospitals (OECD 1992) and it is possible for governments tostructure the incentives in ways that stymie hospitals’ propensity to generate surplusfor their managers and shareholders.Public financing of healthcare for all but the poor is another taboo topic incontemporary health reform debates. It is argued that public financing inflatesdemand, which is not only economically inefficient but also fiscally ruinous. Privatepayments, on the other hand, are believed to curb unnecessary consumption and thuspromote efficiency and prudence. There is no empirical evidence showing that privatehealthcare payments translate into low overall expenditures, but it is widely believednevertheless. Indeed healthcare expenditures have grown exponentially in China andIndia over the last few decades, for instance, despite the very heavy burden of out-of-pocket payments (OOP). In Thailand itself, expenditures grew rapidly in the 1980sdespite the prevalence of high OOP payments and the growth



Page 155slowed down after the launch of the tax-financed UC scheme. Any third-partyfinancing, public or private, is likely to lead to expenditure escalation unlessaccompanied by policy measures designed to curb the behaviour of providers andinsurers. In the case of Thailand, public ownership of hospitals and capitation formulasfor paying providers have served as effective mechanisms for containing expenditureswhile expanding access. In the following discussion, I will describe how, by swimmingagainst the tide of mainstream thinking on healthcare reforms, Thailand has achievedenviable outcomes.ProvisionThe machinery for providing healthcare in Thailand consists of a dominant publicsector and a relatively small private sector; though in rural areas, public facilities areall that are usually available. The public healthcare system is hierarchical, with theMinistry of Public Health (MPH) at the apex, trailed by provincial and regional hospitalsin urban cities of each province and in Bangkok, followed by community hospitals atdistrict level, and health centres at the sub-district level. However, Thailand has a lowratio of physicians and hospital beds to population compared to the developed, oreven many developing, countries. In 2000, there were 0.29 physicians per onethousand in population, compared to the OECD average of 2.60 and Middle Incomecountries’ average of 1.5 (World Development Indicators Online, visited 22 May 2006).Total numbers of hospital beds have grown consistently in Thailand, rising from under80,000 in 1987 to just under 140,000 in 2002. The proportion of beds in the publicsector has declined in recent years (from 88 percent in 1990 to 80 percent in 2005),though it is still larger than in most countries in the developing world. Of all publichospitals, more than four-fifths are community hospitals, located mostly in rural areas.The dominance of the public sector is also reflected in the proportion of doctors in thepublic sector: 79 percent in 2002, though this was lower than the 93 percent in 1971and higher than the 76 percent in 1995. Similarly, the proportion of professionalnurses working in the public sector dropped from 93 percent in 1971 to 86 percent in1995, but then rose to 90 percent in 1997.The government owns and operates public hospitals through the Ministry of PublicHealth (MPH), the largest owner-operator of hospitals in the country. The ProvincialHealth Office in each province and Bangkok is responsible for planning, coordinating,regulating and administering health services provided by the public sector at theregional level. The Public Organisation Act of 1999 grants each public hospital anindependent status with its own Chief Operating Officer, Board of Directors (whichincludes local residents), and personnel. In practice, however, autonomy means little,due to meagre funds and lack of local management expertise, which makes themheavily dependent on MPH. But the situation is changing and it is expected that a lotmore functions will be devolved to local governments in the future.The Office of Hospital Accreditation was set up in 1995, and the 1998 Hospital Actencourages hospitals to secure accreditation on a voluntary basis. Since 2002,



Page 156all hospitals registering for participation in UC have had to receive at least minimalaccreditation. As of June 2003, 51 hospitals had already received accreditation andanother 488 were at some stage in the accreditation process (Viroj andChanwongpaisarn 2005).The Thai government places few formal restrictions on private healthcare providers.The number of private hospitals increased from 23 in 1970 to 473 in 1998, butdecreased amidst economic crisis, and numbered only 346 in 2003. Nearly half of allprivate hospitals (43 percent) have 50 beds or fewer, and nearly half (44 percent) ofall private beds in the country are concentrated in the Bangkok area. However, exceptfor a handful of well-known hospitals catering to the rich Thai and foreign patients,private hospitals have a dubious reputation of overcharging, and providing services ofinconsistent quality.In Thailand, there is little competition among hospitals outside major cities. While thefinancing system, which will be discussed shortly, does allow consumers choiceregarding providers, they have little choice in reality because they must choose fromamong the few available within their regional network. The limitation is particularlyacute in rural areas and smaller towns where there is usually only one medical facility.It is estimated that only about 10 percent of all Thais have any choice of facility.(Interview with Dr Suwit, 16 May 2006.) Competition among hospitals is thus limitedto patients in urban areas who are covered by public or private health insuranceplans. Competition is difficult to achieve, due to a shortage of medical resources andthe dominance of MPH as a provider, and the situation is unlikely to changesubstantially until hospital resources expand substantially.FinancingThailand has a range of public healthcare financing schemes that developed overthree decades. The first public scheme to be established is the WorkmenCompensation Scheme (WCS), which was launched in 1974. It is an employer liabilityscheme for work-related injuries, illness, disability and death.The next major expansion was the launch of the Civil Servant Medical Benefit Scheme(CSMBS) in 1980, providing comprehensive healthcare to all current and retiredgovernment employees (including state enterprise employees) and their dependents(parents, spouses and children). It covers around seven million civil servants and theirdependents, forming nine percent of the population (in 2003). It is a non-contributoryprogram funded entirely by the government, though limited user charges haverecently been introduced. It pays providers on a fee-for-service basis which,unsurprisingly, makes it an expensive program (Viroj and Chanwongpaisarn 1997).Following various studies in the aftermath of the 1997 economic crisis, thegovernment reformed the CSMBS to include prospective capitation for ambulatorycare and diagnosis-related groups (DRG) payment for certain inpatient care.Public financing was extended to private sector workers with the launch of SocialHealth Insurance (SHI) under the Social Security Act 1991. The scheme



Page 157started with mandatory coverage of employees (but not dependents) of privatecompanies with more than 20 employees, which was successively reduced to tenemployees, then five, and eventually to all employees in all firms in 2003. Participationis voluntary for the self-employed and informally employed, but few have taken it up.SHI covers around six million employees, forming ten percent of the population (in2003). It provides healthcare for non-work-related illness and injuries and reimbursesproviders on a contract capitation basis. Members have the choice of registering withany participating provider within their provider network. It is financed by acontribution of 4.5 percent of wages, shared equally among employee, employer, andthe government. The fact that the premium is tied to income, but the benefitsreceived are the same, makes SHI a progressive scheme.Thailand, similar to other countries, has found it difficult to develop a comprehensivefinancing program for the poor and informally employed, and it was not until recentlythat it succeeded. In 1975, the government initiated a scheme (called the MedicalWelfare Scheme) providing free medical care to poor workers. It was renamed theLow Income Card scheme (LIC) in 1994 and extended to everyone under 12 yearsand over 60 years of age, secondary school students, the disabled, veterans andmonks. The service package included free medical care at public facilities forambulatory as well as inpatient care. By 2001, approximately 32 percent of thepopulation in 2001 was covered under this scheme (Rural Health Division and Facultyof Public Health 1994; Human Resource Development Project 1996). Severalassessments of the program have indicated that it inadequately targeted the real poor(Kongsawat 2000). The government started reforming the LIC in 1998 bystrengthening regulations, improving accountability, decentralising funds managementand strengthening the primary care network before it was disbanded with the launchof UC.The Voluntary Health Card (VHC) project was launched in 1983 with the objective ofexpanding health insurance coverage to the non-poor and the uninsured. It was avoluntary health-insurance program with an affordable premium for rural householdsnot covered by the LIC. In 1994, the scheme evolved into a government-subsidisedvoluntary health-insurance scheme financed equally from household contribution andgovernment budget. The VHC covered around 21 percent of the population in 2001. Itwas an ill-designed program characterized by adverse selection, inadequate publicfunding, and widespread flouting of the referral system which was central to cost-containing measures. Several efforts were made to reform the scheme but itultimately collapsed under the weight of the 1997 fiscal crisis.The climate for healthcare financing reforms changed rapidly with the establishmentof the National Health System Reform Committee (NHSRC) under the chairmanship ofthe Prime Minister in 2000. A Health System Reform Office (HSRO) was establishedwithin Health System Research Institute (HSRI) with a three-year mandate to leadand coordinate reforms. Criticized in 2001, the HSRO rejected emulating the CSMBSand the Free For Service (FSS) payment and instead proposed a government-financeduniversal scheme based on capitation payment.



Page 158The ruling Democratic Party was favorably disposed to the proposal, but it was theThai Rak Thai party (TRT) that embraced it enthusiastically. Its popular leader,Thaksin, made universal healthcare for a modest flat rate of 30 baht the centrepieceof his election campaign. The promise turned out to be enormously popular and theTRT went on to win an unprecedented landslide victory. The proposal was branded as“populist” and “fiscally irresponsible” by business interests and critics of Thaksin, butthe government moved resolutely and implemented the scheme as promised.The Universal Health Coverage (UC) scheme, popularly known as the “30 BahtScheme”, was launched on a pilot basis in late 2001 and extended nationwide withthe promulgation of the National Health Security Act in November 2002. Immediatelyupon launch, it covered nearly three-quarters of the Thai population, comprisingeveryone except those covered by SHI and the CSMBS. Its launch was mostsignificant for the 30 percent of the population that had remained outside the net ofthe existing healthcare financing programs (LIC, NHI, CSMBS). The scheme providescomprehensive service coverage, including treatment for catastrophic illnesses andprescribed medication. Only certain expensive services such as renal replacementtherapy and anti-retroviral drugs are excluded. However, the quality of serviceprovided is very basic, offering only essential services and generic drugs whenavailable.UC is financed almost entirely from the government’s general tax revenues. Until2006, hospitals charged 30 baht at the time of each visit, which generated revenuesthat added up to less than 3 percent of hospitals’ total revenues. Even then, morethan half of the UC subscribers did not pay anything, due to various exemptions (forpoverty, age, disability). In late 2006, the newly appointed military governmentabolished the 30 baht contribution on the grounds that the costs of administering thecollection were higher than the amount collected.Residents are required to register with one provider network of their choice. Whilepeople have a choice among competing networks, in reality there are few or noalternative providers outside major cities. Private clinics at district level are oftenunable to participate in the UC scheme, due to their inability to meet requirementssuch as full-time operation and provision of preventive care. A typical district healthsystem serves as a primary care contractor network consisting of health centres and adistrict hospital serving about 50,000 people. A typical province will have five to sevendistrict primary care networks. In urban areas, the network consists of provincial ortertiary hospitals as well as several urban health centres. Referral from the primarycare provider is required to access tertiary and hospital care.Following practices established under SHI, participating providers are paidprospectively on a capitation basis (Mills et al. 2000; Viroj and Chanwongpaisarn2005). The capitation amount per subscriber was 1,202 baht at the time of the launchin 2001, but has gradually risen to 2,089 baht (approximately USD 65). The paymentamount is split into two parts: prospective capitation payment for ambulatory andpreventive care and retrospective payment for inpatient services. A Diagnosis RelatedGroup (DRG), also known as a Case Mix Index, is extensively



Page 159employed for retrospective payments. Prospective payment for outpatient care ismeant to encourage hospitals to invest in preventive care, whereas retrospectivepayment on a DRG basis is meant to prevent hospitals from under-servicing while stillremaining cost-conscious.The National Health Security Office (NHSO) functions as the purchaser of medicalservices on the behalf of both UC and SHI schemes. As the largest purchaser ofmedical services in the country, it is in a position to impose prices and serviceconditions on providers. However, its ability to use its dominant purchaser positionwas limited until 2006 because funding of public hospitals was through the MPH,which was difficult to negotiate with due to its large size and its more establishedposition within the government. This has changed, as NHSO now negotiates directlywith individual contractors, including MPH-owned hospitals.Contrary to worldwide trends, total healthcare expenditures’ share of the GDP hasdeclined over the past two decades in Thailand, from five percent in 1990 to 3.8percent in 1996 to 3.5 percent in 2005 (Murray et al. 1994; World Health Organization[WHO] 2006). Another notable feature is that the public sector’s share of total healthexpenditures increased substantially, from 22 percent in 1990 to 65 percent in 2005.The share accounted for by OOP declined correspondingly, as private insuranceaccounted for a negligible share in Thailand. Prior to the introduction of UC, Thaipublic hospitals received 20–50 percent of their income from user fees (Towse et al.2004). The public sector’s current share is similar to the 59 percent in OECD andconsiderably higher than 48 percent in Middle Income countries (WHO 2006). Again,this is contrary to the experience of many other countries, where the public sector’sshare has shrunk since the 1980s.The declining expenditures on healthcare have not been at the expense of reducedusage of services. In fact, hospital admissions on average across the countryincreased significantly in the year following the launch of UC, from 6.3 percent of thepopulation in 2003 to 6.9 percent in 2004; though by 2006 it had returned to the2001 level. What is remarkable, as evident in the table below, is the steady decline inadmissions rate in the Bangkok and Central areas and the corresponding increase inNorth and Northeast regions, confirming the regional disparities that characterizedThai healthcare before the launch of UC.Table 9.1 Hospital admissions rate, percentage of total population 2001 2003 2004 2005 2006Bangkok 4.8 4.0 4.4 4.0 3.6North 6.6 7.6 8.7 8.0 7.8Northeast 6.8 6.9 7.6 7.6 6.5South 6.5 6.0 6.2 6.7 6.8Central 6.3 6.1 6.3 5.8 6.0Overall 6.4 6.3 6.9 6.7 6.4Source: National Statistical Office 2001-2006, Health and Welfare Survey, Bangkok:NSO.Note: Before 2001, the survey was conducted every five years. There was no surveyin 2002.



Page 160Healthcare reforms: assessmentThe current healthcare system in Thailand is characterized by many remarkable andsome highly salutary features. Since the launch of UC, comprehensive healthcareprotection has been available to the entire Thai population, which is no small feat fora developing country. The program has been most beneficial to the third of thepopulation that had remained uncovered despite numerous attempts to cover themduring the 1990s. The increased access to healthcare benefits is reflected inexpanded utilization rates: between 2002 and 2005, the ambulatory utilization rateincreased by 4.3 percent annually while the hospital admission rate increased by 2.2percent. And this was achieved without corresponding increase in expenditures, as wesaw earlier.The introduction of UC also streamlined the allocation of public funding for healthcareand made it more equitable. Earlier, budget was allocated according to historicalprecedence, which was in turn based on the size of facilities – larger facilities tend tobe older and located in prosperous urban areas – resulting in more affluent areas withhigher population/bed ratios and more professional staff continuing to get more publicfunds. By tying funding to the number of people registered with a provider, areas withpoorer population began to receive their fair share of public funds (Viroj andChanwongpaisarn 2005). However, to prevent financial collapse of the larger hospitalswhose budget allocation would decline under the new formula, the governmentgradually introduced the changes.Moreover, the establishment of UC put an end to extensive application of usercharges – which provided nearly half of public hospitals’ revenues in the early 1990s –that were not only inequitable but also inefficient. The hospitals’ ability to impose usercharges about which consumers could do little was a major reason for rising healthcare costs and worsening equity in the healthcare sector in Thailand during the 1980sand 1990s. The light regulations that governed public and private hospitals until thelate 1990s allowed, and indeed encouraged, hospitals to acquire unnecessarytechnology and provide unneeded services that were subsequently recovered throughuser charges.The system’s efficiency is further enhanced by the capitation system by whichproviders are paid under UC and SHI (but not the CSMBS). Prospective paymentimposes a global cap on expenditures with no incentive to supply services in excess ofa pre-set amount. It also pressures providers to provide primary rather thansecondary care when appropriate, because the latter is more expensive. Theenhanced emphasis on primary care is evident in the rising proportion of visits toprimary care providers: the proportion of outpatient care (OP) at primary careincreased from 67 percent in 2001 to 85 percent in 2004.The decline in hospital expenditures since the introduction of UC in Thailand isremarkable. This is a result of the global budget limits and increased emphasis onprimary care mentioned earlier, as well as the referral system in place. It is a well-established empirical fact that restraining hospitalization is a vital tool for containingexpenditures because of the higher costs it involves (Pauly 1980). Since this verycostly activity is controlled almost exclusively by physicians, the UC requires patientsto first visit a primary care provider and get a referral before



Page 161approaching the hospital system. Following UC’s introduction, a significant proportionof users, who in the past went directly and often unnecessarily to hospitals andspecialists, now have had their access curtailed. System-wide efficiency is furtherenhanced by the payment system, whereby UC pays the primary care unit based onregistered population, which has led secondary and tertiary healthcare facilities tooffer primary care in order to attract patients, thus resulting in economy of scale(Pongpisut Jongudonsuk 2002).DRG payment for hospital care also promotes efficiency because it mitigates thecapitation system’s incentives to undersupply services. DRGs pose less financial risk tothe hospitals than capitation because of the payment certainties they offer. With a setpayment schedule for each procedure, hospitals have less incentive to turn awaypatients requiring expensive treatment.The position of NHSO as the dominant purchaser in the country offers opportunitiesfor raising service standards because of the requirement for providers to meet setservice standards before they can participate in the UC scheme. However, NHSO hashad only modest success in this area due to inexperience, resource shortage, andMPH’s opposition to supervision by an outside body. The shortage of medical facilitiesand personnel in Thailand also constrains the NHSO’s ability to employ its dominantpurchaser position to raise standards. Currently there are simply not enough providersfor the NHSO to be able to reject participants. As a result, it must spend considerableresources on the education and training of hospital management and medicalpersonnel. It also has a long term contract with the Hospital AccreditationDevelopment Institute to help both public and private hospitals to develop theiradministration system. The NHSO also randomly audits hospitals books for clinical andfinancial lapses – but again, resources in the area are thin, though expanding(Personal email from Charay, Head of Policy and Planning Department, NHSO, 18 June2007).The greatest improvement as a result of the healthcare reforms is in the area ofequity. The World Health Report 2000 ranked Thailand 128 out of 130 countries inFairness of Financial Contribution, based on 1997 data (Limwattananon andPrakongsai 2005). The inequity was to a significant extent a result of hefty usercharges that had proliferated through public hospitals during the 1980s and early1990s. By removing nearly all user charges, the reforms made healthcare accessibleto everyone. Equity is further enhanced by the dominance of the public sector: thepoorest two-fifths of the Thai population used public inpatient services approximately95 percent of the time compared to the richest one-tenth, who used it 54 percent ofthe time (Supon et al. 2005). Similarly, public outpatient facilities are used by thepoorest two-fifths about 50 percent of the time while the richest one-tenth use it 25percent of the time. That a significant number of rich patients also use public facilitiesindicates the inclusiveness of the reforms: they are available to everyone, eventhough the poor need them and use them more.Impoverishment due to catastrophic out-of-pocket healthcare payments declinedrapidly in Thailand after the introduction of UC. The incidence of catastrophic healthexpenditure (defined as OOP healthcare payments that form more than 10 percent oftotal household consumption expenditure) reduced from 5.4 percent



Page 162in 2000 to 2.8 percent in 2004 for all households, and from 4.7 percent to 2.6 forformer LIC and VHC members (the latter are a proxy for poor). Over the sameperiod, the number of Thai households that fell below the national poverty line due toOOP payments declined from 4.4 to 1.8 percent for the whole population, and from18.3 percent to 8.0 percent for the poorest quintile. The enhanced equity of thereforms is further evidenced in the fact that health expenditures for the lowest decilesdecreased from 87 baht in 2000 to 47 baht in 2002, while it increased from 836 bahtto 1,189 baht for the highest deciles (Phusit and Viroj 2006). This indicates that UCwas assisting the poor while making the rich pay more out of pocket.The basic level of services provided under the UC in a peculiar way promotesefficiency because it restrains the well-off from using UC, thus reducing overalldemand on public resources. The Health and Welfare Survey 2003 shows that onlyabout 57 percent of UC members used outpatient services, whereas 81 percent usedinpatient service. The take-up rate is, unsurprisingly, significantly higher in the lowerincome groups, but about one third of the population in higher income groups tendsto use the UC scheme as a fall-back scheme. (International Labour Organization [ILO]2004). By paying for UC through taxes but not actually using it, high income groups ineffect make a transfer to lower income groups, thus improving the overall distributionof income.The current healthcare system in Thailand is not without flaws, however, even afterdiscounting the fact that UC is still a new scheme going through teething problems.The most chronic of UC’s problems is the low level of overall funding compared to theambitious goals of the project. In one swoop the government expanded thegovernment’s healthcare responsibilities in 2002 without a commensurate increase infunding. The meagre capitation amount set by UC as a result of low funding imposestremendous constraints on providers. There was a 50 percent increase in hospitalutilization after the launch of the UC, for which hospitals were not prepared, leadingto extensive delays in providing needed treatment (Thai News Service 8 May 2006).At the same time, it caused deep anguish at hospitals among management andmedical personnel, who were often stretched to the breaking point – instances widelymentioned in the local media. There is anecdotal evidence of widespread disaffectionand low morale among medical personnel who feel the government is improvinghealthcare for the population at their expense.Another downside of the meagre funding is the draconian cost-cutting measures thatproviders have had to resort to in order to live within the allotted funding. There isanecdotal evidence that providers are under-providing needed services, evidencewhich is masked by data on increasing utilization rates. It is possible, though notconfirmed, that hospitals are denying complicated and expensive procedures due toshortage of funds. An audit in 2005 found that one-fifth of drugs provided to patientsunder the UC scheme were substandard (Bangkok Post 16 June 2005). Appropriateregulations and enforcement, in addition to adequate funding, are vital, and thegovernment has been lacking in these respects.The lack of meaningful competition among providers outside major urban



Page 163centres may also be undermining quality. NHSO tries to make virtue out of necessityby calling for co-operation rather than competition among hospitals. Having said that,there is some merit to its argument that competition in the hospital sector createssub-optimal use of resources because of heavy investment requirements which mustsomehow be recovered.ConclusionWe are said to live in a world in which privatization and marketization of publicservices are sweeping the world, and only poorly governed countries are resisting thetide. Yet Thailand went against the trend and adopted a publicly-financed healthcaresystem to expand the public sector’s existing dominance of healthcare provision. Theexpansion has not meant doom for the country, as many critics of the reforms hadpredicted. Indeed, early evidence suggests a massive improvement: overall healthcarespending has gone down while access has improved.The Thai experience should not come as a complete surprise, however, because thepublic sector’s domination of both provision and financing of healthcare does offersome unique advantages that are often overlooked in healthcare reform debates.Healthcare is rather unique in that it suffers from more market failures in the form ofinformation asymmetries and moral hazards than most sectors.In the area of provision, the public sector’s overwhelming ownership of hospitals givesthe government a powerful instrument for intervening in the healthcare sector. Whilemore competition is certainly conducive to improved services, in the healthcare sectorits absence is not without advantages, especially in a developing country with limitedresources. Wasteful capital investment, duplication of equipments and services, andemphasis on frills that are endemic to hospitals in a competitive environment are cutout when hospitals provide only what they need to, rather than trying to get an edgeover the competition. Unlike other services where such wastefulness is whittled awayby competition, in the healthcare sector hospitals are able to recover their costs byover-servicing unsuspecting customers. The basic level of service provided by publichospitals in Thailand also helps restrain government expenditure. It is arguable thatthe bureaucratic, impersonal services provided by public hospitals are an effectiverationing mechanism, in that they keep out those who value frills over medicalservices. The Thai government’s overwhelming dominance of the hospital system alsomeans that there are only a limited number of employment opportunities in theprivate sector, which helps stem the exodus of public sector doctors to the privatesector.The public sector’s dominance of the hospital system also offers political benefits, inthat it would have been nearly impossible to adopt the reforms if the system weredominated by private providers. In fact, were it not for the government’s dominanceof ownership, it is doubtful the government would have been able to adopt UC andthe capitation fee model due to imminent opposition from the medical profession. Thelow capitation rate that the government offers would have been especially hard toimpose on private providers. And without low capitation rates, it is unlikely thegovernment would have been able to



Page 164introduce UC because of the enormous costs of universal FSS payment, especiallyconsidering that the Thai economy was yet to recover from the 1997 economic crisis.These scenarios are supported by evidence from other Middle Income countries thathave adopted universal healthcare financing programs based on FSS: theirexpenditures have ballooned rapidly, and governments have been helpless in the faceof the medical profession’s resistance to any measure that might reduce theirautonomy or income (Viroj and Chanwongpaisarn 2004). In Thailand, in contrast, thegovernment was able to impose its decisions without difficulty, due to its ownership ofan overwhelming majority of hospitals in the country.The provision and financing measures in place have been highly effective inrestraining healthcare expenditures in Thailand. Not only have the total expendituresgone down, projections show that they will remain under 5 percent of GDP over thenext ten years (Interview with Pongpisut, NHSO, 16 May 2006). Far too often,policymakers think in terms of cost to the government rather than cost to society as awhole. Greater private spending may mean lower costs for the government, but thecosts for the whole society may be higher, due to the various inefficiencies of privatehealthcare markets.The publicly provided and funded healthcare in Thailand is no doubt currently verybasic, but this is a vast improvement over the situation prior to UC, which exposedthe poor to great financial uncertainties and stress. As we saw earlier, there has beena noticeable increase in utilization rates and a significant decline in incidence ofpoverty due to healthcare expenditures since the adoption of UC.Another attractive feature of the Thai system is that it is scalable and can beexpanded as necessary. The frugal services provided under UC are entirely a functionof the low capitation rate that the government pays the providers. As incomes riseand there is demand for higher-quality services, the current system has enoughflexibility for additional government expenditures and/or private contributions to meetthe extra costs.The Thai healthcare reforms also defy the aphorism that good policies make badpolitics, because UC is attractive in both economic and political terms. Thaksin’spledge to provide comprehensive healthcare coverage for 30 baht played a vital rolein securing TRT an unprecedented victory at the elections. Indeed, its popularity hascontinued to grow, so much so that the military government that toppled the Thaksingovernment not only retained it, but expanded it by eliminating the 30 bahtcontribution requirement. The program’s popularity lies in the fact that it addressesthe population’s understandable concerns about being able to afford healthcare.The reforms may have, however, gone too far in requiring universal coverage withoutcommensurate increase in funding. Indeed, it is imperative that government increasethe capitation rate further, given that costs have been cut to the point of affectingmedical professionals’ morale. There is also a need to increase healthcare facilitiesand personnel, because the current levels are too low, as the government itselfadmits. Expansion in the number of staff and facilities would not only relieve thestress experienced in many public hospitals, it would also inject some competitionamong providers, which would create pressures for improving services.



Page 165Table 9.2 Summary of the development of major health policy reform measures1954Social Security Act launched but not implemented1974Workers Compensation System established1975MCS was initiated1978CSMBS launched1980Royal Decree for CSMBS promulgated1981Low Income Cards were given to 10.9 million people with monthly family income

below 2,800 baht1983The Voluntary Health Card scheme launched1990The Social Security Act promulgated as a mandatory scheme for private sector

enterprises with more than 20 employees1992Reform of CSMBS Free medical care for elderly1993Traffic Accident Victim Protection Insurance1994Workmen’s Compensation Act promulgated, which transferred the responsibility

to the Social Security Office SHI coverage extended to private enterprises with more than 10 employees Medical Welfare Scheme expanded the free medical care for the poor to cover

elderly, children under 12 years and all disabled persons1996Attempt to reform health systems including financing1998New financial regulation for the MWS – management by national and provincial

committees, per capita budget allocation to provinces, reinsurance for high costcases; DRG; global budget.

The Health Promoting Hospital Master Plan (1999–2001) developed Co-payments by CSMBS beneficiaries introduced, only drugs quoted as essential

drugs to be reimbursed, limits on hospital stays in private rooms1999HSRI Board submits a plan to draft a National Health Security Act and proposes a

process of reform to cabinet2000Cabinet approves rationale of health system reform; NHSRC established The Social Security Scheme expanding to cover old age pension and child Benefits2001Launch of UHC on pilot basis soon after new government led by Thaksin took

Office Draft National Health Security Act introduced to the National Assembly SHI coverage extended to enterprises with more than 5 employees2002SHI coverage extended to enterprises with more than 1 employee National Health Security Act implementing UHC promulgatedAdapted from Viroj and Chanwongpaisarn et al. 2005, Viroj et al. 2002; PongpisutJongudonsuk 2002; WHO 2004.



Page 166\ The Thai experience in healthcare reforms shows that public provision and financingare an effective and relatively inexpensive way to provide healthcare on a universalbasis in developing countries. It also shows that the current widespread faith incompetition and individual responsibilities as tools for improving services andincreasing efficiency may be misplaced, especially in developing countries whose mainconcern is adequate quantity rather than high quality.Note1 For studies showing that private hospitals are more efficient, see Altman andShactman 1997; Clarkson 1972; Coles and Hesterly 1998; Ferrier and Valdmanis1996; Hoerger 1991; Hrebiniak and Alutto 1973; Lindsay 1976. For empirical studiesshowing that it is public hospitals which are more efficient, see Pattison and Katz1983; Robinson and Luft 1988; Silverman et al. 1999; Woolhandler and Himmelstein1997.ReferencesAltman, S. H., and Shactman, D. 1997, ‘Should We Worry About Hospitals’ HighAdministrative Costs?’ New England Journal of Medicine, 336: 798–99.Clarkson, K. 1972, ‘Some Implications of Property Rights in Hospital Management.’Journal of Law and Economics, 15: 363–84.Coles, J. W., and Hesterly, W. S. 1998, ‘The Impact of Firm-specific Assets and theInteraction of Uncertainty: An Examination of Make or Buy Decisions in Public andPrivate Hospitals.’ Journal of Economic Behavior and Organization, 36: 383–409.Ferrier, G. D., and Valdmanis, V. 1996, ‘Rural Hospital Performance and itsCorrelates.’ Journal of Productivity Analysis, 7: 63–80.Hoerger, T. J. 1991, ‘Profit Variability in For-profit and Not-for-profit Hospitals.’Journal of Health Economics, 10: 259–89.Hrebiniak, L. G., and Alutto, J. A. 1973, ‘A Comparative Organizational Study ofPerformance and Size Correlates in Inpatient Psychiatric Departments.’ AdministrativeScience Quarterly, 18: 365–82.Human Resource Development Project 1996, Evaluation of the 1990 Low Income Cardissued, Bangkok, National Institute for Development and Administration (NIDA) (inThai).International Labour Organization 2004, A Technical Note to the Government:Financing Universal Health Care in Thailand, TN/Thailand/TN.2, ILO. Available onlineat: http://www.ilo.org/public/english/protection/secsoc/downloads/publ/995sp1.pdfKongsawat, S. 2000, The Evaluation of the Sixth Round Implementation of Low-Income Card for the Poor Households. Nonthaburi: Ministry of Public Health.Limwattananon S. Tangcharoensathien and V. Prakongsai, P. 2005, Equity in FinancingHealthcare: Impact of Universal Access to Healthcare in Thailand. Working Paper #16, EQUITAP Project.Lindsay, C. M. 1976, ‘A Theory of Government Enterprise.’ Journal of PoliticalEconomy, 84: 1061–77.Mills, Anne, Sara Bennett, Porntep Siriwanarangsun, and Viroj Tangcharoensathien,2000. ‘The response of providers to capitation payment: A case-study from Thailand’Health Policy, vol. 51: 163–80.Murray, C. J. L., R. Govindraj and P. Musgrave 1994, ‘National Health Expenditures: AGlobal Analysis.’ in Murray, C. J. L. and A. D. Lopez, eds. Global ComparativeAssessments in the Health Sector, Geneva: World Health Organization.National Statistical Office 2001–2006, Health and Welfare Survey, Bangkok: NSO.OECD 1992, The Reform of Health Care: A Comparative Analysis of Seven OECDCountries, Paris: OECD.



Page 167Pattison, R. V., and Katz, H. M. 1983, ‘Investor-owned and Not-for-profit Hospitals: A Comparison Based on California Data.’ New England Journal of Medicine,309: 347–53.Pauly, Mark 1980, Doctors and Their Workshops: Economic Models of Physician Behavior. Chicago: University of Chicago Press for NBER.Phusit Prakongsai and Viroj Tangcharoensathien 2006, ‘Achieving Universal Coverage of Health Care: Lessons from Thailand.’ Unpublished paper presented at theNational Forum on Social Health Protection, December 5–6, Phnom Penh.Pongpisut Jongudonsuk 2002, ‘Achieving Universal Coverage of Health Care in Thailand through 30 Baht Scheme.’ Unpublished paper presented at SEAMICConference, January 14–17, Chaing Mai, Thailand.Robinson, J. C., and Luft, H. S. 1988, ‘Competition, Regulation, and Hospital Costs, 1982 to 1986.’ Journal of the American Medical Association, 260: 2676–81.Rural Health Division and Faculty of Public Health 1994, A Study of Coverage of Low Income Card. Bangkok: Ministry of Public Health (in Thai).Silverman, E. M., Skinner, J. S., and Fisher, E. S. 1999, ‘The Association Between For-profit Hospital Ownership and Increased Medicare Spending.’ New EnglandJournal of Medicine, 341: 420–25.Supon Limwattananon, Viroj Tangcharoensathien, and Phusit Prakongsai 2005, Equity in Financing Healthcare: Impact of Universal Access to Healthcare inThailand. EQUITAP Project, Working Paper # 16.Towse, Adrian, Anne, Mills, and Viroj Tangcharoensathien 2004, ‘Learning from Thailand’s Health Reforms.’ British Medical Journal, vol. 328: 103–5.Viroj Tangcharoensathien 1997, Reforming the Civil Servant Medical Benefit Scheme: A Case Study in B.E. 2538. Nonthaburi: Health Systems Research Institute,MPH (in Thai).Viroj Tangcharoensathein et al. 2002, Overview of Health Insurance Systems in Thailand, Nonthaburi: Health Systems Research Institute.Viroj Tangcharoensathien, Suwit Wibulpholprasert, and Sanguan Nitayaramphong 2004, ‘Knowledge-based Changes to Health Systems: The Thai Experience inPolicy Development.’ Bulletin of the World Health Organization: the International Journal of Public Health, 82(10): 750–6.Viroj Tangcharoensathien and Lalita Chanwongpaisarn 2005, ‘Challenges of Implementing Universal Health Care in Thailand.’ in Huck-ju Kwon, ed. Transformingthe Developmental Welfare State in East Asia, New York: Palgrave for UNRISD.Woolhandler, S., and Himmelstein, D. U. 1997, ‘Costs of Care and Administration at For-profit and Other Hospitals in the United States.’ New England Journal ofMedicine, 336: 769–74.World Bank 2006, World Development Indicators Online, Washington DC: World Bank. Available athttp://web.worldbank.org/WBSITE/EXTERNAL/DATASTATISTICS/0,contentMDK:20519297~isCURL:Y~pagePK:64133150~piPK:64133175~theSitePK:239419,00.html[22 May 2006]World Health Organization 2004, Regional overview of social health insurance in SouthEast Asia. New Delhi: WHO.——2006, World Health Statistics 2006, Geneva: WHO. Available at http://www.who.int/healthinfo/en/index.html [April 13, 2007].



Page 16810 Restructuring governmentAn empirical analysis of agencification in the public health system of RajasthanAvantika SinghAbstractAgencification, or the carving out of independent agencies from the government, is aNew Public Management (NPM) approach for restructuring government throughseparating the policy formulation and policy implementation functions. This researchempirically investigates agencification in the public health system of the State ofRajasthan in India, with reference to the rationale for agencification, composition ofagencies, autonomy and control of agencies, and service delivery by agencies. Thisresearch employs organisation theory perspectives such as interorganisationalrelations to explicate mechanisms for achieving functional integration and coordinationbetween agencies and parent departments.Agencies studied in this research are legally independent bodies with not-for-profitstatus, which undertake public functions. Though the agencies have operationalautonomy, they function within the policy framework provided by the parentministry/department. However, there is no explicit performance contracting betweenagencies and parent ministry/department. Though evidence on the rationale foragencification is not conclusive, the emergence of agencies is linked to the reformsprocess. Because of resource dependence, agencies achieve functional integration withthe parent ministry/department. Most agencies deliver services through a network ofprivate providers and voluntary organisations. These actors establish coordinationthrough the exchange of resources and information. Coordination is more formalisedand dependent on Memorandums of Understanding (MoUs), contracts and authority-based relationships than on trust, norms, reciprocity and informal communication.Introduction‘Agencification’, or the disaggregation of the government through the carving out ofindependent ‘agencies’, has gained currency among national governments undertakingreforms based on the New Public Management1 (NPM) principles. Throughagencification, the policy implementation function – along with autonomy andaccountability for the achievement of results – is hived off to an agency, while thepolicy formulation function is retained in the parent ministry/department. The



Page 169agency either delivers public services directly or outsources the delivery of publicservices to providers from the private and voluntary sectors. Agencification is lookedupon to ‘reinvent’ government by making it more effective, efficient, accountable andperformance-oriented. The central argument of this paper is that despite their legalindependence and structural disaggregation, agencies are dependent for resources onthe parent department, and that coordination is established in service deliverynetworks through exchange of resources and information, and that the actors in thenetwork are held together by formal mechanisms of coordination rather than byinformal ones.During the early and mid-1990s, health reforms were initiated in several states acrossIndia. Through changes in policy and institutional arrangements, these reforms aim toenhance the performance of the health system and thereby result in better healthoutcomes (WHO SEARO 1997). This paper investigates one such institutional reform,agencification, in the public health system of Rajasthan, a state in north-westernIndia. The empirical analysis is done with respect to the rationale for agencification,autonomy and control of agencies, composition of agencies mechanisms to achievefunctional integration between agencies and their parent ministry/department, servicedelivery by agencies, and coordination among agencies, parent ministry/department,and service providers. Perspectives from organisation theory such asinterorganisational relations have been employed to explicate coordination in thedelivery of services through networks.The remainder of the chapter is structured as follows. Section two provides a briefbackground on the performance of the public health system in India and Rajasthan.Section three presents a review of the existing literature on agencification, andexamines the relevance of the literature on interorganisational relations in the contextof agencification. In section four, I discuss the findings of the empirical analysis ofagencification in the public health system of Rajasthan. In the concluding section, Imove from practice to theory by drawing lessons from the agencification process inRajasthan. I discuss the similarities and points of departure between agencification inRajasthan and Western formulations of agencification, and derive how the presentanalysis of agencification advances our understanding of NPM.The public health system in India and RajasthanIndia spends 5.2 per cent of its Gross Domestic Product (GDP) on health, of which 83per cent is private and 17 per cent is public health expenditure.2 Since health is astate subject3 as specified by the Constitution of India, a large proportion (85 percent) of the total public health expenditure is contributed by all the state governmentscombined, and 15 per cent by the Union government (Ministry of Health and FamilyWelfare [MoHFW] 2005a). Traditionally, however, the union ministry – Ministry ofHealth and Family Welfare, Government of India (GoI) – has been carrying out thepolicy formulation role, and the state governments have been responsible forprovision of health services. In Rajasthan, apart from the policy guidance receivedfrom MoHFW, the ministry of health at the state



Page 170level, with assistance from the Secretariat (staffed by bureaucrats), formulates somepolicies, projects, and schemes. Policy implementation is the responsibility of the linedepartment, known as the Directorate of Medical and Health Services (DMHS), whichis largely staffed by health professionals.Located in the northwest of the country, Rajasthan is the largest state in India interms of area. Rajasthan is one of the states in India with low performance on healthand developmental indicators. The birth rate, total fertility rate (TFR) and infantmortality rate (IMR) in Rajasthan are higher than the country averages.4 While thetopography of Rajasthan compounds the problem of physical access to health care(almost two-thirds of the state is either desert or hilly), a number of factors such as ahigh proportion of marginalised groups (such as Scheduled Castes and ScheduledTribes), low female literacy, and unfavourable sex ratio5 add to the complexity ofsocial access to healthcare.AgencificationWhat are agencies?Agencification involves transfer of activities hitherto done by the government toagencies (Pollitt and Talbot 2004), through hiving off existing departments or forming‘new’ organisations. Agencies have been described as an “administrative fashionaccessory” (Pollitt et al. 2001:286), and agencification as ‘boutiquing’ of government(Schick 2002), and “internal structural devolution” (Christensen and Lægreid2004:129). Most literature on agencification has emerged in OECD countries and iswritten in a comparative perspective (Pollitt and Talbot 2004; Talbot et al. 2000).The terms ‘agencification’ and ‘agencies’ have come to mean different things tovarious national governments and scholars (see Allix and Van Thiel 2005; Aucoin1990; Boston et al. 1996; Halligan 2004; Roberts 1997). A large segment of theliterature deals with “signposting the zoo” (Gill 2002), i.e. the terminology to describe‘agencies’ and the definitional aspects of agencies (Greve et al. 1999; Pollitt 2003;Thynne 2004; Van Thiel 2000). In order to understand what an ‘agency’ is exactly, itis important to understand what is not an agency. An agency is neither a division ordirectorate within a ministry or department, nor a corporate entity with commercialpurpose. While Thynne (2004) classifies agencies on the basis of legal existence,Pollitt and Talbot (2004) emphasise structural separation from the core, irrespective oflegal independence. Thus, an agency is an entity with some measure of structuraldisaggregation from the parent ministry/department (Halligan 2004; Pollitt et al. 2001;Pollitt and Talbot 2004). Autonomy (Common et al. 1992; du Gay 2004; Thynne2004; Pollitt and Talbot 2004; Verhoest et al. 2003) coupled with performancecontracting and accountability for results (Christensen and Lægreid 2004; Pollitt et al.2001; Roberts 1997) is regarded as a defining characteristic of agencies. However, anagency is not a statutorily fully independent body free from ministerial instruction.Though the ultimate control of an agency lies with the parent ministry (du Gay 2004;Pollitt and Talbot



Page 1712004), the form of control shifts from ex ante to ex post (Verhoest et al. 2004). Anagency performs public tasks such as service provision, regulation, adjudication andcertification (Pollitt et al. 2001), and is therefore a public organisation (Pollitt andTalbot 2004).In the decision to agencify, there is an implicit assumption that agencification canimprove the effectiveness of government (OECD 1995). Being disaggregated from theministerial hierarchy, agencies are expected to function in a more business-likemanner (Halligan 2004). Yesilkagit (2004) has examined the separation of functionsthrough agencification from the point of view of the principal-agent problem andtransaction costs. In the case of public hospitals, it is expected that agencification canhelp overcome problems such as chronic under-funding, poor human resourcepractices and a fragmented approach to policy implementation (Yeoh 2003). In thepresent-day environment, with the reduced ability of the government to fund welfarethrough taxes (Pollitt and Bouckaert 2004), the bureaucratic structure no longersuffices for government to fulfil its emergent role (Osborne and Gaebler 1992), andthe agency structure is considered more appropriate for service delivery and financing(Schick 2002).Though citizens’ expectations of the quality of public services are rising (Pollitt andBouckaert 2004), they have lost faith in the government (Gill 2002; Nye et al. 1997).Several scholars (Common 1998; Dolowitz and Marsh 1998; Gill 2002) suggest thatagencification is, thus, an opportunity for government to regain its legitimacy. LaGuerre (1994) and Levitt (1991) have described the decision to agencify as theoutcome of coercion, Christensen and Lægreid (2004) have described it as an‘ideological import’, Van Thiel (2000) has described it as ad hoc, and Gill (2002) hasdescribed it as a means to ‘depoliticise’ government. Pollitt and Talbot (2004) proposethe task-specific path dependency (TSPD) model, according to which two sets offactors, namely path (the nature of cultural and institutional norms in a particularjurisdiction) and task specificity (the nature of the actual work to be done), influencethe spread and application of agencies. Pollitt et al. (2001) suggest that the behaviourof agencies is context-specific because of the different political systems. To sum up,there emerge varying perspectives on the rationale for agencification in the literature,and the question on rationale for agencification is not answered firmly.Though the idea of agencification as a means to restructuring and reforminggovernment sounds appealing, it has been criticised on several counts. The generaltendency in the literature has been to overemphasise operational autonomy ofagencies while underemphasising the role of departments in providing policy guidance(Schick 2002). While several scholars agree that the vertical separation of policy-making and policy implementation caused by agencification (du Gay 2004; Norman2003; Parker 1993; Pollitte et al. 1998) erodes the trust between politicians andadministrators (Christensen and Lægreid 2004; Rhodes 1996), others argue thatagencification may not necessarily cause the separation of the policy andimplementation functions (Halligan 2004; Pollitt 2002; Schick 1996; Shand 1996;Trosa 1994).Empirical evidence does not point out that effectiveness and efficiency have



Page 172improved with the formation of agencies (Boston et al. 1996; Christensen and Lægreid2004; Van Thiel 2000). Roberts (1997) argues that in the US, even thoughagencification has improved performance, claims have been overstated. Norman(2003) has questioned the appropriateness of such a results-oriented business modelfor the public sector, as the application of market-like principles alone cannot meetwelfare and equity goals.Perspectives on interorganisational relations and agenciesIn some cases, agencies directly deliver services to citizens, whereas in others, theyperform a planning, financing and coordinating role while outsourcing the servicedelivery function to private providers (PPs) and voluntary organisations (VOs).6Consequently the emergent institutional landscape is increasingly being defined by anetwork of government, autonomous agencies, PPs and VOs, and the linkages amongthem. The actors in the network engage in exchange of resources and informationand establish coordination, in order to accomplish mutual goals. Thus, it would be ofscholarly interest to study the interorganisational relations and coordination amongactors in such networks.Complexity and interdependence of public management problems necessitatescollaboration among various actors (Mason 1993), thereby making networks thepreferred mode of governance in contrast to markets and hierachies. Networks areformal and informal relationships (Jones et al. 1997) comprising relatively enduringresource transactions, flows, linkages (Oliver 1990), and information exchangesbetween two or more organisations, such that that their individual autonomy remainsintact (Kapucu 2005) while fulfilling mutually beneficial ends. The study ofinterorganisational relations (IOR) has been partially applied to the delivery of socialservices and human services (Goes and Park 1997; Hall 2002; Provan and Milward1995).The political economy perspective of networks focusses not on the actors per se buton the interorganisational linkages7 (Benson 1975; Milward and Provan 1998). Therelative position of an organisation in a network is determined by the centrality of itsfunctions or its strategic location in the network or its ability to influence resourceflow (Benson 1975). Therefore, resource control is an important issue in IOR (Baker1990; Ring and Van de Ven and Walker 1984; Whetten 1981). The interorganisational network may be described as a political economy with respect to thedistribution of resources, such as, money and authority (Benson 1975), information,social support, facilities, personnel (Hall 2002) and even clients (Boje and Whetten1981). However, resource transfer creates dependence (Baker 1990; Whetten 1981),reduces autonomy, and necessitates coordination (Van de Ven and Walker 1984).Several organisation theorists have described coordination in networks (Chisholm1989; Jennings 1994; O’Toole 1997; Whetten 1981). In network organisations,coordination is increasingly maintained through organic, informal social systems(Eccles and Nohria 1992; Gerlach 1992), open-ended contracts, and social capital(Axelrod and Cohen 2000; Coleman 1988; Fukuyama 1995,



Page 1732001; Granovetter 1992; Ostrom 1991; Putnam 1993) rather than through formalcontracts and authority-based relationships.Gaps in the literature on agencificationMost literature on agencification has emerged with reference to OECD countries and islimited to the analysis of prominent agencies. Though perspectives from economics,management and governance have been employed to describe agencification, theagencification literature has not reached a stage of maturity as yet, and there is needfor greater conceptual as well as empirical research in this area. The scholarship onthe experience of developing countries with agencification is weak. Further,agencification has not been studied from an interorganisational relations andinterorganisational coordination perspective. A key aim of this paper is to make amodest contribution to the agencification literature in the context of the public healthsystem in Rajasthan. This paper aims to empirically verify the existing literature onagencification, as well as apply organisation theory perspectives to understandinterorganisational linkages and coordination among agencies, parentministry/department and service providers.Agencification of the public health system in Rajasthan: an empirical analysisDefinitional aspects of agencies in the public health system of RajasthanGiven that agencification is a relatively less-researched area, an exploratory study wasconducted in Rajasthan at state level, and in one district. The study reveals that priorto agencification, National Health Programmes (NHPs) such as those for control ofblindness, vector-borne diseases, leprosy and tuberculosis (TB), and promotion ofReproductive and Child Health (RCH), were implemented by the line department – theDirectorate of Medical and Health Services (DMHS) located in the state capital, Jaipur.In the district, the Chief Medical and Health Officer (CM&HO) was responsible forimplementing these health programmes. With agencification, the implementationfunction with regard to these NHPs has been hived off from DMHS to separateagencies at state level as well as at district level.8 On the other hand, the NationalAIDS Control Programme is implemented through new agencies created at nationaland state level.9 All the agencies created at state and district level for theimplementation of NHPs are referred to as category A agencies for the purpose of thispaper.With state governments hard-pressed for resources, there has been a need to look atalternative options of financing. The idea of forming independent hospital boards wasfloated on the assumption that part-financing by the users would help improve thequality of services. Rajasthan Medicare Relief Societies (RMRS), which are agenciesattached to individual health facilities (which could be a hospital attached to a medicalcollege, a district hospital, a Community Health Centre [CHC], or a Primary HealthCentre [PHC]), were formed in a phased manner by



Page 174Table 10.1 Agencies selected for the study on agencification of the public healthsystem in RajasthanCategory A Category BAgencies formed for the management of healthprogrammes

Agencies at health facility level formanagement of public hospitals

Agencies at state level One at District HospitalRajasthan State AIDS Control Society (RSACS) One at Community Health CentreState Blindness Control Society (SBCS) One at Block Primary Health CentreState Leprosy Elimination Society (SLES) One at Primary Health CentreState Vector-Borne Disease Control Society(formerly State Malaria Control Society) (SVBDCS)

State Tuberculosis Control Society (STCS) State Health and Family Welfare Society forVoluntary Sector (SHFWSVS) - SCOVA

Agencies at district level District RCH Society (DRCHS) District Leprosy Elimination Society (DLES) District Tuberculosis Control Society (DTCS) District Blindness Control Society (DBCS) the Government of Rajasthan (GoR), in order to manage public hospitals.10 In thecontext of this paper, RMRS are referred to as category B agencies. The agenciesselected for the study are listed in Table 10.1.In Rajasthan, both categories of agencies are neither ministerial departments norcorporate entities with for-profit motive. All of them are registered under theRajasthan Societies Registration Act 1958, making them legally independent,structurally disaggregated bodies, with not-for-profit status. Both categories ofagencies undertake public functions, i.e. category A agencies are responsible forimplementation of NHPs, while category B agencies are responsible for managementof public health facilities and making them financially sustainable.Rationale for agencificationVarious rationales for agencification (in the case of category A agencies) emerge fromthe empirical analysis, such as to divorce policy formulation from policyimplementation, and to bring in a greater measure of autonomy and responsiveness inorder to overcome “bureaupathologies” (Khandwalla 1999); namely, procedural andfinancial delays associated with government, excessive rule-boundedness, unutilisedbudgetary allocation, utilisation of funds for purposes other than those they weremeant for, duplication of activities, and sub-optimal resource allocations



Page 175and output. Donor-supported programmes are implemented through the agency modebecause agencification helps donors circumvent the pitfalls of directly working withgovernment departments, and is expected to fulfil their concerns for outcomes. Therationale for agencification through the creation of RMRS (category B agencies) was todecentralise management of health facilities, grant autonomy at local level, ensurefinancial sustainability of the health facility by generating, retaining and usingresources at local level for improving availability and quality of services, and makeavailable modern diagnostic and treatment facilities at affordable cost to the citizens.Through the creation of agencies and development of service delivery networksinvolving VOs and PPs, the government has expanded public participation in healthcare, such that the responsibility for achieving the health goals of the country isshared with other actors in society, and accountability towards the society isstrengthened.Beginning early 1990s, agencification has occurred for implementing disease-specificprogrammes supported by the World Bank. Subsequently, agency structure hasincreasingly been adopted for programme implementation. The newly launchedNational Rural Health Mission (2005–2012) is also being implemented through theagency mode (agencies have been formed at state and district-level for implementingNRHM) (MoHFW 2005a). Agencification has come to be established as a best practice,and the government is increasingly adopting agencification. Thus, institutional change(in the form of agencification) is found to be path dependent in the sense that theadoption of agency structure has been deepening over the years.Composition of agenciesAs per the requirements of the Rajasthan Societies Registration Act 1958, eachregistered Society has to have a Governing Body and an Executive Committee. TheGoverning Body is responsible for strategic planning and steering, and the ExecutiveCommittee is entrusted with day-to-day management of the affairs of the agency. Inthe case of agencies in the public health system of Rajasthan, though the compositionof the Governing Body varies from one agency to another, we can generalise it tosome extent. The Health Secretary is the chairperson, and the State ProgrammeOfficer (RCH/AIDS/ Blindness/Malaria/TB/Leprosy) is the Member Secretary of thecategory A agency at state level. The members of the Governing Body include officialsfrom DMHS as well as other departments (Finance, Women and Child Development,Education, Social Welfare, etc.), eminent professionals, representatives from privatesector, VOs and GoI. The District Collector is the chairperson and the CM&HO is theMember Secretary of the category A agency at district level. The members of theGoverning Body include the Zila Pramukh,11 officials from various departments atdistrict level, representatives of VOs and PPs, eminent professionals, philanthropistsand social workers. In the case of category B agencies (RMRS), the DistrictCollector/Sub-Divisional Magistrate (SDM)/ Project Director (PD) or any other officialdirected by Divisional Commissioner/ Health Secretary serves as the chairperson ofthe Governing Board. The CM&HO/ Hospital Superintendent/any other health officialserves as the Member Secretary.



Page 176The members of the Governing Body include other health officials and medicalspecialists in the district, officials from other departments, representatives from VOs,PPs and Panchayats (local self-government institutions), businessmen, social workersand philanthropists.The members of the Executive Committee (of both categories of agencies) are mostlydrawn from the Governing Body. The Chairperson and Member Secretary are commonto the Governing Body and Executive Committee. The Chairperson presides overmeetings of the Governing Body and Executive Committee. The Member Secretary isvested with all executive and financial powers, and is responsible for planning,implementation and monitoring of the agency’s activities.Autonomy of agenciesBecause they are legally independent personalities, all agencies operate separate bankaccounts, have separate financial rules and procedures, and maintain separate booksof accounts. Moreover, these agencies have easier and faster billing and paymentsystems as compared to the government department. In the earlier case, paymentshad to be routed through the Treasury of the state government (GoR), making theprocess cumbersome and time-consuming. However, now payments are made throughcheques and demand drafts, thus improving the availability of funds for carrying outvarious interventions. Earlier, files would move up to the level of the Health Secretaryfor approval. Agency heads are now empowered to make certain decisions; this isexpected to reduce the red tape in getting approvals. These agencies are lessprocedure-oriented and have greater operational flexibility. Though most officials aredeputed from government departments to agencies, the agencies can hireprofessional as well as administrative staff on short-term contracts as required. Theagencies do not need to adhere to compensation and other benefits as offered by thestate government. Contractual workers may be paid high salaries, but there may beno social security measures such as pension and gratuity.The category B agency (RMRS) raises its own resources and spends them at locallevel. Apart from the seed money given by the state government at start-up, RMRSdoes not receive any grants from the state government. Its sources of funds includeuser charges, income from sale of medicines, and grants and donations from othersources. Prior to formation of RMRS, hospital income (if any) had to be deposited withthe Treasury of the state government. However, because of its independent legalstatus, RMRS can retain its income and plough it back into the hospital for variouspurposes such as maintenance, repairs and renovation of hospital premises, purchaseof medical equipment, purchase of drugs, provision of free services to the poor andthe disadvantaged, contracting out hospital support services, and other functions suchas computerisation.Control of agenciesIn accordance with the provisions of the Rajasthan Societies Registration Act 1958,the objectives and bylaws of the agency may be altered, or the agency



Page 177dissolved with the consent of the union/state government. Though these agencies areoperationally autonomous, they do not have policy autonomy – they function withinthe policy framework laid down by government – and in accordance with severaladministrative law procedures. Most members of the Governing Bodies of theseagencies are drawn from government. The ultimate control of these agencies lies withthe government, as state-level agencies are headed by senior officials of theSecretariat/DMHS, and agencies at district level and health facility level are headed bythe District Collector and relevant health officials in the district. Category A agenciesare primarily financed by donors such as the World Bank through the Uniongovernment. However, they are entitled to receive grants and donations from othersources as well. In contrast, category B agencies are less dependent on financingfrom the government, as they have a number of sources of funds, particularly usercharges.Control systems in both categories of agencies are ex post as the agencies arerequired to submit performance reports and audited financial statements to the parentministry/department. However, there is no explicit performance contracting. Thoughthese agencies have performance targets – such as number of cataract surgeries tobe conducted or number of new TB patients to be identified – these are not,however, linked to incentives or the continuance of funding. There is a condition onRMRS to spend 25 per cent of its income on providing free services (includingmedicines, investigations and other hospital services) to patients possessing MedicareRelief Cards. These include patients from Below Poverty Line (BPL) families, widows,freedom fighters, orphans, senior citizens (over 70 years of age), prisoners andaccident emergency cases.Functional integration of agencies with parent ministry/departmentApplying the interorganisational relations perspective, dyadic linkages of the agencieswith the parent ministry/department have been traced with respect to the exchangeof resources (including human resources, funds and items in kind such as equipment,drugs and supplies) and information (bylaws, policy guidelines, rules, procedures,performance reports, audits and evaluations). The flow of resources and informationmay be one-way or two-way, more frequent or less frequent, strong or weak, andhierarchical or horizontal. These interdependencies in terms of exchange of resourcesand information between the agency and the parent department necessitatecoordination to achieve the common objective of performance. Figure 10.1 illustratesa generic representation of the linkages of category A agencies with the parentministry/department and service providers. For prevention and control of HIV/AIDS,the World Bank routes funds and policy guidance through MoHFW to a national levelagency named National AIDS Control Organisation (NACO). Most states in India havestate-level agencies that receive funds and policy guidance from NACO (in Rajasthan,this is the Rajasthan State AIDS Control Society, RSACS). NACO disburses fundsdirectly to these state-level agencies, and does not channelise any funds through thestate governments. In the case of other health programmes (blindness, vector-bornedisease, leprosy, TB and RCH),



Page 178

Figure 10.1 Linkages of agencies with the public health system and their servicedelivery network.the state-level agencies receive policies, strategy documents, implementationstrategies, procedural rules, financial and organisational resources from the WorldBank (routed through the concerned department in MoHFW) as well as DMHS in thestate. The state-level agencies send performance reports, audit reports and evaluationreports to NACO, MoHFW, and DMHS. Because resource dependence on MoHFW ishigher, the state-level agencies are tightly integrated with the concerned departmentin MoHFW in comparison to DMHS (with the exception of RSACS, which is tightlyintegrated with NACO and not MoHFW).In the case of National AIDS Control Programme, there are no district-level agencies.Interestingly, the district-level agencies for leprosy, blindness and TB were inexistence for several years before the state-level agencies were formed.12 At thetime when the state-level agencies were not in existence, each district-level agencywould receive the budgetary allocation directly from GoI (under its CentrallySponsored Schemes), and send reports to the concerned department in MoHFW. Now,the district-level agencies for blindness, TB, leprosy and RCH work under theguidance of the respective state-level agencies, and receive policy guidance andresources from them. These district-level agencies send performance reports, auditreports and evaluation reports to the district administration (District Collector andCM&HO) and the concerned agency at state level. Appointing officials from the districthealth administration as agency heads helps achieve functional integration. However,due to resource dependence on the state-level



Page 179agency, the district-level agencies have stronger linkages with the state-level agencyas compared to those with the district administration.The appointment of the District Collector as Chairperson and a health official asMember Secretary of category B agencies (RMRS) helps achieve functional integrationwith the local hospital/district health administration. Though RMRS has a considerableamount of autonomy with regard to the generation and utilization of funds, it isdependent on the local hospital for other organisational resources. RMRS are requiredto work within the broader policy framework set by and as per directives issued bythe state government/hospital administration.Service delivery by agenciesNeither category of agencies engage in service provision. Category A agenciesoutsource services such as information, education and communication (IEC) for NHPs,voluntary counselling and testing for HIV, providing care and support to people livingwith HIV/AIDS (PLWHA), cataract surgery, identification of leprosy patients, dispensingdrugs, providing laboratory services and inpatient care to TB patients, DDT spray anddistribution of insecticide-treated bed-nets to prevent malaria, and RCH services suchas immunisation, distribution of contraceptives, IUD insertion and sterilisationoperations. The agencies enter into MoUs with private providers and voluntaryorganisations to deliver these services. Thus, a service delivery network emerges withthe agency as the focal organisation. Two such networks are illustrated in Figure10.2. In the diagram on the left, RSACS is depicted as the focal organisation havingcontracted out various activities, such as blood banks, IEC, targeted interventions forPLWHA, etc. to NGOs. In the diagram on the right, State Tuberculosis Control Society(STCS) is depicted as having contracted out the task of dispensing drugs for TB toNGOs, PPs and individuals. Few of these NGOs act as intermediaries and further sub-contract these activities to Unani providers.Category B agencies (RMRS) outsource hospital support services, such as kitchen,cleaning, laundry, computer operations, etc., and diagnostic services such as CTscanning, through a competitive bidding process. RMRS has led to strengthening ofinstitutions of curative care without creating additional financial burden on the alreadyconstrained state government. Secondly, it has given a considerable amount ofautonomy to the Governing Body and created space for the participation of theprivate sector in managing hospital services. It has led to the creation of markets andcompetition through contracting out of various hospital services. The RMRS model hasdemonstrated scalability, and is being replicated at PHC level now.The areas of concern in the performance of RMRS include low utilisation of healthcare by BPL patients, and the inability of RMRS to sustain themselves when located inareas with larger proportion of poorer families. However, there is no mechanism forcross-subsidising low-performing RMRS from high-performing RMRS because eachRMRS is an independent body. The revenue generation of RMRS has increased fromINR13 30.2 million in 1997–98 to INR 240.9 million in



Page 180

Figure 10.2 Service delivery networks in the public health system of Rajasthan withagencies as the focal organisation.2002–03, as depicted in Figure 10.3. The cumulative number of BPL beneficiaries hasincreased from 0.36 million in 2001–02 to 1.06 million in 2003–04. However, the totalnumber of BPL families identified through a survey and provided cards was2,301,058.14Mechanisms of coordinationLinkages of state and district-level agencies with the Union ministry (MoHFW), healthdepartment (DMHS) of the state government, district public health



Page 181

Figure 10.3 Revenue generation of RMRS over the period 1997–98 to 2002–03.administration, PPs and VOs are established through formal mechanisms such asorders issued by the government, MoUs and contracts. In the process of exchangingresources and information, the actors in the network establish coordination. There areseveral formal mechanisms of coordination, such as constitution of task forces andcommittees, issuance of guidelines and strategies to guide implementation, deputationof officials from government departments, appointment of government officials asagency heads, and information sharing through reporting, monitoring, auditing,independent evaluations and stakeholder meetings. Inter-agency coordination isestablished by having the Health Secretary head various agencies at the state leveland the District Collector head various agencies at district level. The representation ofofficials from other departments on the Governing Body is meant to achieve inter-departmental coordination. The nomination of PPs, VOs and professions on theGoverning Body is meant to involve them in policy-making. Though the practice ofdeputing officials from the parent ministry/department to agencies can help ensurecoordination, the downside is that the organisational culture of government maypercolate into these agencies.Several gaps in coordination exist, such as limited involvement of externalstakeholders – as well as internal stakeholders – in policy-making. Internalstakeholders, particularly DMHS officials (who are medical professionals), are involvedin a limited way in policy formulation, and they have expressed the desire for greaterparticipation in policy formulation. Category A agencies are highly specialised, andeach agency focusses on a specific disease control programme. This reinforces theexisting segmented and “vertical” approach to implementation of health programmes,which has come under sharp criticism by public health scholars in India (Antia 2001;Banerji 1994). The direct linkages between the Union ministry and various agencies atstate level, and those between state-level agencies and district-level agencies, havestrengthened the control of the Union ministry over the agencies, and areaccompanied by a corresponding weakening of



Page 182the state government in the transfer of resources, information exchange and policyimplementation. This could result in alienation and resistance from existing powerconstituencies. It could lead to agencification being perceived as an externallyimposed structure or centralization of types.The selection of PPs and VOs for delivering various services is done by the concernedagency at state/district level. The agency is responsible for selecting PPs and VOs,allocating them projects and budgets, and monitoring their activities. This in turnleads to a reduced role of the parent department (DMHS or district healthadministration) in project and budget allocation, monitoring, supervision or evaluation.In cases where NGOs are allocated projects and budgets by an agency at national(NACO) or state level (RSACS, SCOVA), district-level health officials have littleinformation about the NGOs working in their respective districts. Though thisarrangement places the NGOs at arm’s length from the government department, thegovernment loses direct control on the NGOs’ activities.Government officials are largely sceptical about the authenticity, performance andaccountability of NGOs. Similarly NGOs too look at the government with skcepticism,particularly with regard to the prevalence of corruption, and delay in receivingpayments. The study reveals that informal mechanisms of coordination in the networkof agencies, government departments, PPs and VOs seem to be weak. In cases wherewell-reputed NGOs are involved in service delivery, there exists an amount of trustbetween the NGO and the agency. The government, by virtue of its nature, reliesheavily on ex ante control through rules, procedures and documentation. Thus,coordination in the networks with the agencies as focal organisations, is morehierarchical than horizontal; it is more formalised and dependent on MoUs, contractsand authority-based relationships than on trust, norms, reciprocity and informalcommunication.ConclusionThe agencies studied in this research are legally independent, structurallydisaggregated bodies with not-for-profit status, which undertake public functions suchas implementation of National Health Programmes and management of publichospitals. Though they have operational autonomy, they function within the broadpolicy framework and guidelines provided by the parent ministry/ department.However, unlike the “original” formulation of agencification based on the NPM,agencification in Rajasthan has not been accompanied by explicit performancecontracting, lateral entry of agency heads, accountability for results, and performance-based rewards. Moreover, the idea of agencies in Rajasthan has not generated from“within” – it is path dependent and linked to the reforms process. Agencification hasbeen “imported” only as an organisational structure, mostly at the behest of donors.Agency structures serve as a channel for resource flow and implementation of healthprogrammes.The study of the functional integration of agencies with the parent ministry/department reveals that the agencies have stronger linkages with the Union ministry,MoHFW, as compared to that with the State health department (DMHS)



Page 183primarily because resource dependence on the former is higher. Similarly, due togreater resource dependence on the state-level agency, the linkages of district-levelagencies with the state-level agency are stronger as compared to that with thedistrict administration. Agencies outsource the delivery of services to PPs and VOs,thus, leading to the emergence of a service delivery network with the agency as thefocal organisation. The actors in the network establish coordination through theexchange of resources and information. Several gaps in coordination exist, such aslimited involvement of external and internal stakeholders in policy-making, and loss ofdirect control of the health department (DMHS or district health administration) overthe selection of PPs and VOs, allocation of projects and budgets, the provision ofpolicy guidance, monitoring, supervision and evaluation. Coordination is moreformalised and dependent on MoUs, contracts and authority-based relationships thanon trust, norms, reciprocity and informal communication. The underlying reasonscould be the procedure-oriented nature of government work, and scepticism and lackof trust of government in NGOs and vice versa.For success of NPM in the public health system of Rajasthan, it is important forcomprehensive and simultaneous implementation of NPM reforms. Agencification hasto be accompanied by policy changes that bring about lateral entry, greateraccountability, performance measurement, results orientation and incentive schemesin the public sector. For this to happen, political commitment needs to be high.Though this research has made a modest attempt to analyse agencification from anorganisation theory perspective, it reveals the potential for applying“interorganisational relations” to the study of emerging service delivery networks as aresult of disaggregation of government. Other issues pertaining to agencification thatcould be addressed through future research include organisational culture of agencies,institutionalisation of agencies, how to address equity issues through public hospitalswith autonomous boards, and how to improve coordination in networks through thedevelopment of social capital.AcknowledgementsThe author is grateful to Prof. M. Ramesh, University of Hong Kong, Hong Kong, forhis comments on earlier versions of this paper.The author was recipient of Sir Ratan Tata Trust (SRTT) Fellowship at Institute ofRural Management (IRMA), Anand, India. She acknowledges the financial supportprovided by SRTT, Mumbai, India for carrying out this research.Notes1 New Public Management (NPM) originated in UK during the late 1970s under theleadership of Margaret Thatcher, and was soon embraced by New Zealand, Australiaand other OECD countries. Its variants later spread to other countries including India.The two basic principles of NPM include managerialism and marketisation. NPM marksa departure from the Old Public Administration (Dunleavy and Hood 1994), and hasbeen described as the “reinvention” of government through change in its role from



Page 184‘rowing’ to ‘steering’ (Osborne and Gaebler 1992) or the shift from ‘government’ to‘governance’. Significant contributions to the scholarship on NPM have been made byAucoin (1990), Barzelay (2001), Ferlie et al. (1996) Hood (1991), Kettl (1997), Kickert(1997), Minogue et al. (1998), Pollitt (1993), Pollitt and Bouckaert (2004) and Walsh(1995).2 Public health expenditure as a percentage of total expenditure on health in India isthird lowest in the world. The only countries that spend less than India are Myanmarand Georgia (World Bank 2003).3 India has a federal polity, and the legislative powers are distributed between theUnion and the States, as mandated by the Constitution of India. Public health andsanitation, hospitals and dispensaries are defined and enlisted in List – II (State List)of the Seventh Schedule of the Constitution. Population control and family planning,medical education and medical profession are defined and enlisted under the List – III(Concurrent List) of the Seventh Schedule of the Constitution. Thus, while only theState Legislatures can legislate on the subjects in the State list, both the Parliamentand the State Legislatures can legislate on those in the Concurrent list, and thelegislations made by the Parliament supersede those made by the State Legislatures(Ministry of Law and Justice).4 The birth rate in Rajasthan was 28.6 per 1000 as compared to 23.8 for all Indiaduring 2005 (Indiastat 2006), TFR was 3.7 per woman in Rajasthan as compared to2.9 for all India during 2004 (Indiastat 2005 and 2006), and IMR was 68 infant deathsper 1000 live births in Rajasthan as compared to 58 for all India during 2005 (MoHFW2005b).5 According to the 2001 Census, Scheduled Castes and Scheduled Tribes togethermake up 29.8 per cent of the population of the State as compared to 24.4 per centfor all India. Just 43.9 per cent of the female population is literate in Rajasthan, ascompared to 53.7 per cent in the whole of India, and the sex ratio is 922 females per1000 males in Rajasthan, as compared to 933 for the whole of India (RegistrarGeneral of India [RGI] 2001).6 In this chapter, the term “voluntary organisations” has been employed as a broaderterm to include nongovernmental organisations (NGOs) as well as voluntaryassociations such as Lions Club or Rotary Club.7 Whetten (1981) has described four forms of interorganisational linkages; namely,dyadic linkage, organisation set, action set and network. Dyadic linkage refers to themutually beneficial interactions between two otherwise independent organisations.Organisation set is the sum total of the interorganisational linkages established by a“focal organisation”. In this case, the linkages between other interacting organisationsare ignored. Actions sets are clusters of organisations working together to accomplisha specific purpose. A network consists of all the interorganisational linkages in apopulation, and can be visualised as a system comprising several interconnected sub-systems.8 These include State-level agencies such as State Blindness Control Society (SBCS),State Leprosy Elimination Society (SLES), State Vector Borne Disease Control Society(formerly State Malaria Control Society) (SVBDCS), State Tuberculosis Control Society(STCS), and State Health and Family Welfare Society for Voluntary Sector (SHFWSVS)popularly known as SCOVA. The district-level agencies in the study district includeDistrict RCH Society (DRCHS), District Leprosy Elimination Society (DLES), DistrictTuberculosis Control Society (DTCS), and District Blindness Control Society (DBCS).There is no district-level agency for vector-borne disease control in the study district.9 The national-level agency is known as National AIDS Control Organisation (NACO),and the state-level agency in Rajasthan is known as Rajasthan State AIDS ControlSociety (RSACS).10 At the initiative of the then Health Secretary, 55 Rajasthan Medicare ReliefSocieties (RMRS) were formed through an order issued by the GoR in 1995 in allhospitals affiliated with medical colleges and District Hospitals. The success of theseagencies



Page 185motivated the GoR to issue another order in 1998, thereby paving the way forestablishing these agencies in all government hospitals with fewer than 100 beds andall Community Health Centres. Inspired by the success of RMRS, GoR issued anotherorder in 2005 for setting up RMRS in each Primary Health Centre in the state.11 Zila Pramukh is the chairperson of the Zila Parishad (District Council). ThePanchayati Raj System of local self-government comprises a three-tier architecture ofGram Panchayats (village councils) in the villages, Panchayat Samitis (intermediatebodies) at sub-district level, and Zila Parishad at district level. All these bodies consistof elected representatives.12 State Health and Family Welfare Society for Voluntary Action was registered in1997, and District Reproductive and Child Health Society, Tonk in 1999. RajasthanState AIDS Control Society was registered in 1998, and there are no correspondingdistrict-level societies. State Leprosy Society was registered in 2000, while DistrictLeprosy Eradication Society, Tonk was registered in 1994. State Tuberculosis ControlSociety was established in 2001, while District Tuberculosis Control Society wasestablished in 1999. Rajasthan State Malaria Control Society was registered in 2001,and renamed State Vector Borne Disease Control Society in 2004. Often there hasbeen a lag between the issue of guidelines by GoI and the establishment of theagency at State level. For instance, GoI issued guidelines for formation of StateBlindness Control Society (SBCS) in 1998, and it was only in 2002 that SBCS wasregistered. District Blindness Control Society has been in existence in Tonk since 1994.13 1 US$ = ~40 INR (Indian National Rupees).14 Bhatnagar, S. 2005, Health sector reforms in Rajasthan (Presentation shared bythe author).ReferencesAllix, M., and S. Van Thiel 2005, ‘Mapping the Field of Quasi-autonomousOrganizations in France and Italy.’ International Public Management Journal, 8(1): 39–55.Antia, N. H. 2001, ‘Health for All Through a Decentralized People’s Movement.’ Healthfor the Millions, 27(4): 10–11.Aucoin, P. 1990, ‘Administrative Reform in Public Management: Paradigms, Principles,Paradoxes and Pendulums.’ Governance, 3(2): 115–37.Axelrod, R., and M. D. Cohen 2000, Harnessing Complexity: OrganizationalImplications of a Scientific Frontier. New York: The Free Press.Baker, W. E. 1990, ‘Market Networks and Corporate Behavior.’ American Journal ofSociology, 96(3): 589–625.Banerji, D. 1994, ‘A Simplistic Approach to Health Policy Analysis: the World BankTeam on the Indian Health Sector.’ International Journal of Health Services, 24(1):151–59.Barzelay, M. 2001, The New Public Management: Improving Research and PolicyDialogue. Berkeley, CA: University of California Press.Benson, J. K. 1975, ‘The Interorganizational Network as a Political Economy.’Administrative Science Quarterly, 20(2): 229–49.Boje, D. M., and D. A. Whetten 1981, ‘Effects of Organizational Strategies andContextual Constraints on Centrality and Attributions of Influence inInterorganizational Networks.’ Administrative Science Quarterly, 26(3): 378–95.Boston, J., J. Martin, J. Pallot, and P. Walsh 1996, Public Management: The NewZealand Model. Oxford: Oxford University Press.Chisholm, D. 1989, Coordination Without Hierarchy: Informal Structures inMultiorganizational Systems. Berkeley, CA: University of California Press.Christensen, T. and P. Lægreid 2004, ‘Governmental Autonomisation and Control: TheNorwegian Way.’ Public Administration and Development, 24(2): 129–35.Coleman, J. C. 1988, ‘Social Capital in the Creation of Human Capital.’ AmericanJournal of Sociology, 94(Supplement): S95–S120.



Page 186Common, R. 1998, ‘Convergence and Transfer: A Review of the Globalisation of NewPublic Management.’ International Journal of Public Sector Management, 11(6): 440–50.Common, R., N. Flynn, and E. Mellon 1992, Managing Public Services: Competitionand Decentralisation. Oxford: Oxford University Press.Dolowitz, D. P., and D. Marsh 1998, ‘Policy Transfer: A Framework for ComparativeAnalysis.’ in M. Minogue, C. Polidano, and D. Hulme, eds. Beyond the New PublicManagement: Changing Ideas and Practices in Governance. Cheltenham: EdwardElgar.Du Gay, P. 2004, ‘Against “Enterprise” But Not Against “Enterprise”, for That WouldMake No Sense.’ Organization, 11(1): 37–57.Dunleavy, P., and C. Hood 1994, ‘From Old Public Administration to New PublicManagement.’ Public Money and Management, 14(3): 9–16.Eccles, R. G., and N. Nohria 1992, Beyond the Hype: Rediscovering the Essence ofManagement. Boston, MA: Harvard Business School Press.Ferlie, E., L. Asburner, L. Fitzgerald, and A. Pettigrew 1996, The New PublicManagement in Action. Oxford: Oxford University Press.Fukuyama, F. 1995, ‘Social Capital and the Global Economy: A Redrawn Map of theWorld.’ Foreign Affairs, 74(5): 89–103.——2001, ‘Social Capital, Civil Society and Development.’ Third World Quarterly,22(1): 7–20.Gerlach, M. L. 1992, ‘The Japanese Corporate Network: A Blockmodel Analysis.’Administrative Science Quarterly, 37(1): 105–39.Gill, D. 2002, ‘Signposting the Zoo – From Agencification to a More Principled Choiceof Government Organisational Forms.’ OECD Journal on Budgeting, 2(1): 27–80.Goes, J. B., and S. H. Park 1997, ‘Interorganizational Linkages and Innovation: TheCase of Hospital Services.’ Academy of Management Journal, 40(3): 673–96.Granovetter, M. S. 1992, ‘Problems of Explanation in Economic Sociology.’ in N. Nohriaand R. G. Eccles, eds. Networks and Organizations: Structure, Form and Action. 25–56, Boston: Harvard Business School Press.Greve, C., M. Flinders, and S. Van Thiel 1999, ‘Quangos – What’s in a Name? DefiningQuangos from a Comparative Perspective.’ Governance, 12(2): 129–46.Hall, R. H. 2002, Organizations: Structures, Processes and Outcomes, 8th ed. NewDelhi: Prentice Hall of India.Halligan, J. 2004, ‘The Quasi-autonomous Agency in an Ambiguous Environment. TheCentrelink Case.’ Public Administration and Development, 24(2): 147–56.Hood, C. 1991, ‘A Public Management for All Seasons?’ Public Administration, 69(1):3–19. Indiastat 2005 and 2006, ‘Selected State-wise Total Fertility Rate in India (1988to 2004).’Rajya Sabha Unstarred Question No. 4745, dated 06.05.2005., and Lok SabhaUnstarred Question No. 161, dated 22.11.2006. Viewed 15 June, 2007.<http://www.indiastat.com/india/ShowDataSec.asp?secid=369063&ptid=17795>.Indiastat 2006, ‘State-wise Birth Rates in India (2003 to 2005).’ Ministry of Health andFamily Welfare, Govt. of India, and Rajya Sabha Unstarred Question No. 1258, dated04.08.2006. Viewed 15 June, 2007.<http://www.indiastat.com/india/ShowDataSec.asp?secid=390929&ptid=17791>.Jennings, E. T., Jr. 1994, ‘Building Bridges in the Intergovernmental Arena:Coordinating Employment and Training Programs in the American States.’ PublicAdministration Review, 54(1): 52–60.Jones, C., W. S. Hesterly, and S. P. Borgatti 1997, ‘A General Theory of NetworkGovernance: Exchange Conditions and Social Mechanisms.’ Academy of ManagementReview, 22(4): 911–45.Kapucu, N. 2005, ‘Interorganizational Coordination in Dynamic Context: Networks inEmergency Response Management.’ Connections, 26(2): 33–48.Khandwalla, P. N. 1999, Revitalizing the State: A Menu of Options, New Delhi: Sage.Kettl, D. F. 1997, ‘The Global Revolution in Public Management: Driving Themes,Missing Links.’ Journal of Policy Analysis and Management, 16(3): 446–62.



Page 187Kickert, W. ed. 1997, Public Management and Administrative Reform in WesternEurope. Northampton: Edward Elgar.La Guerre, J. 1994, Structural Adjustment, Public Policy and Administration in theCaribbean. St. Augustine, Trinidad: School of Continuing Studies.Levitt, K. P. 1991, The Origins and Consequences of Jamaica’s Debt Crisis 1970–1990.Kingston: Consortium Graduate School of Social Sciences.Mason, J. C. 1993, ‘Strategic Alliances: Partnering for Success.’ Management Review,82(5): 10–15.Milward, H. B., and K. G. Provan 1998, ‘Principles for Controlling Agents: The PoliticalEconomy of Network Structure.’ Journal of Public Administration Research and Theory,8(2): 203–21.Minogue, M., C. Polidano, and D. Hulme, eds. 1998, Beyond the New PublicManagement: Changing Ideas and Practices in Governance. Cheltenham: EdwardElgar.Ministry of Health and Family Welfare (MoHFW) 2005a, National Rural Health Mission(2005–2012) – Mission Document, New Delhi: MoHFW.——2005b, ‘State-wise Infant Mortality Rate by Sex and Residence in India (2005).’Viewed 15 June, 2007. <http://www.indiastat.com/india/ShowDataSec.asp?secid=409099&ptid=17794>.——2006, Annual Report 2005–06. New Delhi: MoHFW.Ministry of Law and Justice 2007, ‘Seventh Schedule’, Constitution of India. Viewed 14May, 2007. <http://lawmin.nic.in/coi/SEVENTH-SCHEDULE.pdf>.Norman, R. 2003, Obedient Servants? Management Freedoms and Accountabilities inthe New Zealand Public Sector. Wellington: Victoria University Press.Nye, J. S., P. D. Zelikow, and D. C. King, eds. 1997, Why People Don’t TrustGovernment. Cambridge, MA: Harvard University Press.OECD 1995, Governance in Transition: Public Management Reforms in OECDCountries, Paris: OECD.Oliver, C. 1990, ‘Determinants of Interorganizational Relationships: Integration andFuture Directions.’ Academy of Management Review, 15(2): 241–65.Osborne, D., and T. Gaebler 1992, Reinventing Government: How the EntrepreneurialSpirit is Transforming the Public Sector. Reading, MA: Addison-Wesley.Ostrom, E. 1991, ‘Rational Choice Theory and Institutional Analysis: TowardComplementarity.’ American Political Science Review, 85(1): 237–43.O’Toole, L. J., Jr. 1997, ‘Treating Networks Seriously: Practical and Research-basedAgendas in Public Administration.’ Public Administration Review, 57(1): 45–52.Parker, R. 1993, The Administrative Vocation. Sydney: Ironmonger.Pollitt, C. 1993, Managerialism and the Public Services. Cambridge: Blackwell.——2002, Ministries and Agencies: Performance Measurement or Poor Parenting? ADiscussion Paper for NOB 2002, Rotterdam: Erasmus University.——2003, Agencies, Apples and Pears: Mapping the Agency Debate, Paper for the 7thInternational Research Symposium on Public Management, 2–4 October, Hong Kong.Pollitt, C., K. Bathgate, J. Caulfield, A. Smullen, and C. Talbot 2001, ‘Agency fever?Analysis of an International Policy Fashion.’ Journal of Comparative Policy Analysis:Research and Practice, 3(3): 271–90.Pollitt, C., J. Birchall, and K. Putnam 1998, Decentralising Public Service Management,London: Macmillan.Pollitt, C., and G. Bouckaert 2004, Public Management Reform: A ComparativeAnalysis, 2nd ed, Oxford: Oxford University Press.Pollitt, C., and C. Talbot 2004, Unbundled Government: A Critical Analysis of theGlobal Trend to Agencies, Quangos and Contractualisation, London: Routledge.Provan, K. G. and H. B. Milward 1995, ‘A Preliminary Theory of Network Effectiveness:A Comparative Study of Four Community Mental Health Systems.’ AdministrativeScience Quarterly, 40(1): 1–33.



Page 188Putnam, R. D. 1993, Making Democracy Work: Civic Traditions in Modern Italy.Princeton, NJ: Princeton University Press.Registrar General of India 2001, Primary Census Abstract: Census of India 2001.Accessed 26 January 2007. <http://www.censusindia.net/results/pca_main.html>.Rhodes, R. A. W. 1996, ‘The New Governance: Governance Without Government.’Political Studies, 44(4): 652–67.Ring, P. S., and A. H. Van de Ven 1994, ‘Developmental Processes of CorporativeInterorganizational Relationships.’ Academy of Management Review, 19(1): 90–118.Roberts, A. 1997, ‘Performance-based Organizations: Assessing the Gore plan.’ PublicAdministration Review, 57(6): 465–78.Schick, A. 1996, The Spirit of Reform: Managing the New Zealand State Sector in aTime of Change. Wellington: A Report Prepared for the State Services Commissionand the Treasury.Schick, A. 2002, ‘Agencies in Search of Principles.’ OECD Journal on Budgeting, 2(1):7–26.Schmidt, S. M., and T. A. Kochan 1977, ‘Interorganizational Relationships: Patternsand Motivations.’ Administrative Science Quarterly, 22(2): 220–34.Shand, D. 1996, ‘The New Public Management: An International Perspective.’ Paperfor the Public Services Management 2000 Conference, 11 October 2000, University ofGlamorgan.Talbot, C., C. Pollitt, K. Bathgate, J. Caulfield, A. Reilly, and A. Smullen 2000, ‘TheIdea of Agency: Researching the Agencification of the (Public Service) World.’American Political Studies Association Conference, August, Washington DC.Thynne, I. 2004, ‘State Organisations as Agencies: An Identifiable and MeaningfulFocus of Research?’ Public Administration and Development, 24(2): 91–9.Trosa, S. 1994, Next Steps: Moving On. London: Cabinet Office.Ulrich, D., and J. B. Barney 1984, ‘Perspectives in Organizations: ResourceDependence, Efficiency, and Population.’ Academy of Management Review, 9(3): 471–81.Van de Ven, A. H. and G. Walker 1984 ‘The Dynamics of InterorganizationalCoordination.’ Administrative Science Quarterly, 29(4): 598–621.Van Thiel, S. 2000, Quangocratization: Trends, Causes and Consequences. Utrecht:Universiteit Utrecht.Verhoest, K., B. G. Peters, G. Bouckaert, and B. Verschuere 2003, The Study ofOrganizational Autonomy: A Conceptual and Methodological Review. Working Paper,Leuven: Catholic University of Leuven.Verhoest, K., B. Verschuere, B. G. Peters, and G. Bouckaert 2004, ‘ControllingAutonomous Public Agencies as an Indicator of New Public Management.’Management International, 9(1): 25–35.Walsh, K. 1995, Public Services and Market Mechanisms: Competition, Contracting andthe New Public Management. London: Macmillan Press.Whetten, D. A. 1981, ‘Interorganizational Relations: A Review of the Field.’ TheJournal of Higher Education, 52(1): 1–28.WHO SEARO 1997, Public Private Sector Partnerships for Health: Role ofGovernments, New Delhi: WHO SEARO.World Bank 2003, World Development Indicators 2003, Washington, DC: The WorldBank.Yeoh, E. K. 2003, ‘Experiences in Hospital Sector Reform, the Performance ofHospitals Under Changing Socio-economic Conditions. WHO Global Study Undertakenin Collaboration with the International Hospital Federation.’ World Hospitals and HealthServices, 39(2): 2.Yesilkagit, K. 2004, ‘The Design of Public Agencies: Overcoming Agency Costs andCommitment Problems.’ Public Administration and Development, 24(2): 119–27.



Page 18911 The structural transformation of public bureaucracy towards goodgovernanceThe case of team-based management in the Korean governmentChang Kil LeeIntroductionThe global movement towards better governance is reshaping the internal structure ofbureaucracy. Even though the provision process for public service has alreadyundergone a transformation, traditional bureaucratic structures remain. For example,citizens increasingly participate in the process of policy-making. However,governments have not been able to reform their internal structures to adapt to thisdynamic. New paradigms and reforms related to governance must consider thetransformation in policy-making that has taken place in the government. Ingraham(2004:229) states that better public governance is accomplished by not only closerinteraction with external policy actors, but also by internal restructuring withinbureaucracy. Osborne and Gaebler (1992:12–18) also anticipate the emergence ofentrepreneurial government and the decline of traditional bureaucracy as trendsdealing with new management and governance. Thus, the burgeoning of newparadigms in government made the traditional hierarchical structure of bureaucracyoutdated and a new horizontal structure required.In spite of a variety of alternatives for internal and structural changes, two distinctivemethods are pervasive across Korea: agencification and empowerment. Agencificationcreates arm’s length agencies that are contracted from outside the department. Theseagencies are given managerial autonomy and flexibility in budgeting and humanresource management. By the 1980s, the United Kingdom had already delegatedresponsibilities to these “executive agencies” for improving their performance. Theagencies were able to sell their services for profit based on competitive marketmechanisms. Empowerment, on the other hand, delegates both authority andresponsibility to lower levels of the hierarchy. As a result, empowerment motivatesemployees and leads to better performance. These trends show that flat and flexiblestructures are emerging within government.Of the many alternatives for structural empowerment, the team-based approach hasbeen a popular management tool in the private sector since the 1990s. Jones andLindley (1998) explain that the team-based approach represents work sharing,decision-making involvement, team performance monitoring and feedback, highinterdependence, recognition of expertise, and a high discretion in decision-making.



Page 190In spite of this success in the private sector, the government has given little thoughtto its possible application. As a system of empowerment, the team system is animportant alternative for reducing rule-based hierarchies and establishing a new high-performing structure in the public sector.This paper examines the emergence of team-based management systems as modelsof structural change in the Korean government. First, it highlights how the team-based approach arose as an example of good governance. Second, it analyzes theorganizational features of the team system. Third, the paper investigates the threedrivers of the team system. These include innovative leadership, coercive imitation,and vicarious learning. Fourth, it discusses the relationship between theaforementioned drivers and the time needed for adoption of the team system. Finally,the issues and challenges for implementing reform are reviewed, with the hope ofproviding guidance for policy makers.Good governance and the team-based approachFrom hierarchical control to team-based collaborationThe global movement towards good governance calls for “deep and comprehensivechange” in the existing public management system. It stresses efficiency, flexibility,and responsibility as ways of attaining better performance. These values advocatetransforming the vertical control system of authority into a horizontal network(Heinrich et al. 2004). The vertical structure is “composed of classic command-and-control relationships within organizational units or programs,” dominated by rules andlaws (Heinrich et al. 2004:10). In this structure, traditional policy managementapproaches depend on institutional arrangements and legislations (Abramson et al.2006).“Good governance” rejects vertical structures by putting emphasis on the value ofhorizontal collaboration rather than control and command. This signifies a shift from“hierarchy” to “network” (Heinrich et al. 2004:10), exemplified by the decentralisationof central and local governments, as well as the delegation and devolution of authoritywithin the bureaucratic structure. In particular, delegation or devolution representflexibility and responsibility for executive bodies at the bottom of the hierarchy.The team-based approach in Korea was also initiated as a tool for structural changesand good governance. By definition, a team is “a small number of employees withcomplementary competencies who are committed to common performance goals andworking relationships for which they hold themselves mutually accountable” (Hellriegelet al. 2001:226). This definition presumes different perspectives and expertise outsideof a given department, which may also ease power struggles between departments. Itis advocated as ”a way of replacing the current inflexible, dehumanizing, bureaucraticsystem with flexible, decentralized, humanistic networks” (Hellriegel et al. 2001:226).Osborne and Gaebler (1996:252) argue for a shift from a hierarchical form to oneinvolving participation and teamwork because decentralized institutions are



Page 191more flexible, more effective, more innovative, and generate higher morale, morecommitment, and greater productivity. They state:Centralized, hierarchical organizations also divide themselves up into many layers andboxes. People begin to identify with their unit – their turf. Communication across unitand between layers becomes difficult. This explains why innovative organizations sooften use teams … In contrast, innovative organizations foster constantcommunication, so information flows quickly through their ranks. To do this, theyregularly create new teams and new configurations, so nearly everyone comes intocontact with nearly everyone else (Osborne and Gaebler 1996:269).Abramson, Breul, and Kamensky (2006) summarize managerial flexibility by sayingthat one can “let” managers manage and “make” managers manage. The first methodliberates managers from ex ante controls on inputs and operating procedures, whilethe second is premised on “setting clear and reachable targets while holdingmanagers personally and organizationally responsible for improved performance”(Abramson et al. 2006:3). Teams provide the capacity to innovate, accomplish tasks,and respond rapidly to change (Osborne and Gaebler 1992:268). Thus, the traditional,hierarchical form of bureaucracy based on vertical control must be eliminated in favorof horizontal cooperation.The initiation of the team-based approach in KoreaSince the beginning of 2005, around 22 of the 49 ministries in the Korean governmenthave adopted team-based management systems as a means of organizational reform.The Ministry Of Government Administration and Home Affairs (MOGAHA) adoptedteam-based management systems for the first time in March 2005. Though the timeperiod needed for adoption has varied, as of 2006 almost half of the ministries hadadopted team-based management policies. The remaining ministries, though reluctant,are increasingly adopting team-based management systems to accomplishorganizational reforms.MOGAHA’s reason for initiating reform can be applied to other Korean ministries.According to their code, MOGAHA has two types of responsibilities. One is to initiateand lead government reforms for the entire government, and the other is tocoordinate and support local governments in terms of organizational, financial, andelectoral management.The first motivation of initiating the new system in MOGAHA came from thedisadvantages of the hierarchical structure as well as the culture embedded across theministries. Even though hierarchical structure provides speedy and consistent decision-making from the top, it ignores employee creativity and provokes vague responsibilitybetween high- and low-level employees. In particular, MOGAHA was regarded as aministry which maintained a strong hierarchical and bureaucratic culture. In the past,authoritative regimes tended to control local governments by delivering central policyto lower-level employees. Ideas were stymied under the



Page 192multi-level reporting process of this system, which also led to long periods of wastedtime while awaiting the minister’s authoritative decision.The “deep and comprehensive” change of the administrative environment brings aboutmismatches with those embedded in the culture of MOGAHA. In spite of demands onhigher expertise in policy-making, and increased decentralisation of its top-levelofficials, MOGAHA attempted to maintain its existing authority and control over localgovernments. Though MOGAHA delegated responsibility, major functions or issueswere decided by the high-level authorities. Now, it appears to have changed. Strongpressure from the outside pushed the ministry towards organizational reforms basedon a horizontal and flexible culture. Minister Oh Young-Kyo (2005) also emphasizedthat the vertical framework constrained efficiency during the work process, flexibilityin decision making, and responsibility for improving organizational performance. Forthese reasons, the only way to achieve better performance was to destruct thecurrent hierarchical structure of decision-making and establish a new horizontalsystem such as “team-based” management.The second motivation for reform at MOGAHA was found from a feeling of crisisrelated to the identity of the ministry and organization. Employees were concernedwith the future of the organization and feared its abolition under future governments.By introducing the team system, MOGAHA was able to clarify its vision for the future,thus ending uncertainty. This rationale played a decisive role in introducing the newsystem.In addition, the two major functions of government administration and home affairs,which merged in 1998, maintained separate personnel, work processes, and cultures.As a result, organizational commitment and loyalty were impeded. The clear visionand strategy brought about by team-based management strengthened employeesolidarity, confidence, and commitment to the organization.Though both motivations were directed towards good governance in Korea, the firstmotivation represented the functional viewpoint, while the latter one was orientedtowards the political point of view. On the one hand, the movement from hierarchy toteam-based approaches was aimed at improving organizational performance throughstrengthened flexibility and responsibility. Also, successful political reform projectsdemonstrate the legitimacy of the organization’s existence to external actors. Byleading government reforms, the organization can protect itself, while providinginternal actors with a clear vision and goals.Organizational features of the team-based approachThe team-based management system in Korea has several distinct features that aredifferent from the private sector. First, it emphasizes interdependence andcollaboration within a team rather than across teams. It focuses on “within” dynamicsrather than “between” dynamics. Hellriegel, Slocum and Woodman (2001) classifyteams into four common types: functional, problem solving, cross-functional, and self-managed teams. Of the four team types, the Korean government resembles functionalteams rather than self-managing or cross-functional because



Page 193they have been formed from current divisions rather than selected team membersacross divisions.Second, teams in the Korean government are established before setting goals. Afterteam members are selected, they create a vision, goals, and strategies forimplementation. According to Tuckman and Jensen’s (1977) model, we can say that inKorea’s case the process of constituting a team starts with the “forming” stage, butproceeds directly to the “performing” stage without going through the “storming” and“norming” stages. These missing stages often produce conflict among team members.In addition, it takes a considerable amount of time to agree upon a vision, goals, andoperating procedures.Third, the team system in Korea is more of a comprehensive tool for innovation thana change in the administrative structure. It is closely related to the political motivationdescribed in the previous chapter. Team-based management includes a variety ofreforms. Some may cover performance-based management that aims to improve teamperformance. Also, the division of team functions may require the redesigning of workprocesses. This is why it is usually referred to as “team-based management” ratherthan simply “teams.” The team-based management system in Korea denotes fouraspects of reforms in an administrative system: structure, evaluation, humanresources, and communication.Structural changeThe first characteristic of team-based management in Korea is structural change toreduce the number of vertical layers of a bureaucratic hierarchy. Before team-basedmanagement was adopted, hierarchical trees of decision-making in the Koreangovernment consisted of seven layers from the top to the bottom: minister, viceminister, assistant minister, director general, director, assistant director, andemployees. The adoption of team-based management reduced this to five layers:minister, vice minister, headquarters, teams, and team members. Assistant ministerand director generals were merged, while assistant directors and employees mergedvertically into teams. Directors of a division were renamed “team chiefs,” with theirroles and functions significantly enlarged. Decentralizing authority is a key componentof team-based management, thus vertical allocation of decision-making wasextensively moved from the bureau directors” or assistant ministers” office down toteam leaders. The chief of a team was given autonomy and flexibility in policy-makingas well as human resource management within the division. They were also givenmore accountability with respect to team performance.The size of a team has become smaller than that of a division under the previoussystem. Since vertical layers were reduced, a team is now composed of about 10members on average, and ranges from as few as three to as many as 28. Table 11.1shows more details on the size of a team before and after the team system wasadopted. The number of team members in the Ministry of Planning and Budget (MPB)decreased from 8.65 members to 6.05 members, while the number of teamsincreased. After the MPB adopted the team system, the number went up from 40 to



Page 194Table 11.1 Change of team size and team number before and after the team systemMinistries/Team adoption Before team After team

Team sizeNo. of Teams Team size No. of TeamsMinistry of Planning & Budget (MPB) 5 to 16

(8.15)40 4 to 9

(6.05)45

Ministry of Labor (MOL) 7 to 14 (10.6)

31 4 to 17 (9.5)

36

Public Procurement Service (PPS) 6 to 26 (14.9)

24 3 to 28 (12.26)

31

Note: The numbers in parenthesis represent the average size of a team.Source: Ministry of Government Administration and Home Affairs, 2006.45 teams. In the Ministry of Labor (MOL), teams rose from 31 to 36. The PublicProcurement Service (PPS) increased from 24 to 31 teams.Evaluative changeThe second characteristic of team-based management in Korea is that teams, as wellas individuals, are regularly evaluated on work performance and customer satisfaction.The entire team’s performance depends on the collective efforts of individual teammembers. Conversely, the individual performance of team members is also dependenton the collective performance of the teams in which they are involved. In mostministries that adopted team-based management, individual performance wasdetermined 70 percent of the time by team-based scores and 30 percent byindividually-based scores (Kim 2005:473). Under this system, teams or team leadershave been given autonomy and flexibility, as a team is responsible for itsperformance.A team’s first objective must be to set goals. Then teams develop key performanceindicators for evaluating team performance. Based on a Balanced Scorecard (BSC),the Ministry of Government Administration and Home Affairs launched the “integratedadministration innovation system” (IAIS) for the first time in government agencies(MOGAHA 2006:45). The system categorized key performance indicators into fourelements: customers, work performance, operating processes, and learning/growth. Itautomates service workflows, as well as measures customer satisfaction in real-timebased on the services team members provide.More importantly, collective and individual work performance is closely connected withthe compensation system. Various performance-based incentives, such asdifferentiated allocation of personnel and budget, are given to high performers(MOGAHA 2005:45). In the case of MOGAHA, the team system worked as a usefultool for performance-based management.



Page 195Table 11.2 Performance indicators by balanced scorecard in MOGAHACategories Weight Key performance IndicatorsCustomers 30 Customer satisfactionWork 50 Goal achievement 50% Work diffi culties 20% Cooperation among members 30%Operations 10 Work process system score 40% Customer/expert meetings 30% Suggestions/Ideas 30%Learning/Growth 10 Innovation mileage scoreSource: Ministry of Government Administration and Home Affairs, 2006.Human resource management changesThird, the team-based management system created a radical change in humanresources management. After the introduction of team-based management, humanresource management entered a new phase of job posting for most positions in theagency (MOGAHA 2005:44). All of the team leaders are now chosen competitivelythrough job postings. Candidates who want to be team leaders apply for the position.Qualifications have also been radically revised to allow any capable and eligiblecandidate an opportunity to be team leader. Team leaders are given the authority toselect their own team members. Team members who want to work with a teamleader can apply for the team, while team leaders submit a written priority list of thecandidates they want to work with as a team. If the application of a candidate andthe priority of the team leader match, the candidate is appointed a member. If not,the candidate is forced into leave or out of the office. In this sense, the introductionof team-based management activates an internal market of labor in the government.The other change in human resource management under the team system is relatedto compensation and promotion. That is, work performance has become the mostimportant factor for promotion. Performance-based promotion and pay havesupplanted the seniority-based system.Communication changeUnlike the conventional bureaucratic system, the team-based system by definitionpresumes horizontal cooperation among team members, rather than vertical authority.Beyond the boundaries of each division, team members must cooperate to achievethe team goals. A team normally chooses its members from functional departments,and shares distinctive competencies and perspectives. In the Korean case, however,the team system improved horizontal as well as vertical communication. Also,horizontal cooperation among team members was enhanced because the position ofassistant director within the division was moved to the level of team member. Thisincreased the span of control of team leaders. Vertical communication was weakened,but horizontal communications grew stronger.



Page 196Now, teams operate cooperatively because common goals are set, collectiveperformance is evaluated, and compensation is allocated to the whole team. Thesevariables have led to overall performance improvements.Vertical communication within the Korean team system has significantly improved.There is less distance between minister and team leaders. Team leaders, becausethey have been given more authority, more frequently communicate with the ministeror vice minister to enhance team performance. The closer and more agile relationshipbetween high ranks and team leaders has come from the use of informationtechnology. All reports are replaced by electronic reports, to enable prompt decision-making that reduces time wasted during face-to-face reporting (MOGAHA 2005:45).By using personal computers, ministers may check any report by a team leader andapprove it while in the office or at home.Initiating drivers of team-based managementInnovative leadershipLeadership by administrators or ministers has been one of the most important driversof team-based management. Kane (2005:21) argues that the most critical ingredientin any effort towards change is leadership. Thus, leaders must create a change-friendly environment and prepare executive teams for change. Wilson (1989:227)indicates that whether changes are externally imposed or internally generated,understanding why they occur at all requires one to understand the behavior of theagency executive. More innovative agency leaders are more likely to adopt team-based management. Thus MOGAHA, which is mainly responsible for leading andcoordinating government reforms, pioneered the reform of the conventional verticalhierarchy found throughout the Korean governmental structure. As Rogers (2003)describes, the adoption of the system at MOGAHA can be attributed to the leadershipby the incumbent minister as an innovator.Government executives are particularly important in any changes in publicbureaucracy. In its fundamental character, public bureaucracy is vertically structured.Top executives usually have authority to make a final decision. Their importantresponsibilities are to maintain the agency, as well as to change it. Ministers in Koreaare also given authority and autonomy to implement organizational change. Whetheradopted or not, change is affected by the ministers’ innovative capabilities. The moreinnovative that ministers are, the more likely they will adopt the team-basedmanagement system. Not all ministries adopted the team structure simultaneously. Ittook nearly two years for half of the ministries to adopt the system. The willingnessand confidence of a minister were the key factors that determined if an agency wasable to adopt reform. The adoption of team-based management brings strongresistance from working level employees because it reshuffles the current structure.Therefore, in order to be successful, reform requires strong leadership to overcomeresistance.



Page 197Coercive imitationImplicit and explicit coercive pressures may affect the adoption of the team system.DiMaggio and Powell (1991) identify three mechanisms of institutional isomorphicchange as coercive, mimetic and normative. They explain that coercive isomorphismcomes from a more direct and formal force, while mimetic isomorphism is encouragedfrom uncertainty and implicitness. This paper does not differentiate between the twoterminologies, but instead integrates them into the category of coercive imitation. Thisis because implicit and uncertain coerciveness and formal and direct imitation can’t beseparated.Implicit but coercive pressure for the adoption of the team system seems to permeatethroughout government. Without any explicit authority, these pressures might forceministries to adopt the system. As an example, the President of Korea, Roh Moo-Hyun, did not express explicit support for the team structure in the government, buthe implemented structural changes necessary for government innovation. PresidentRoh’s emphasis on government innovation may be an implicit force since hisinauguration in 2003. He strongly encouraged restructuring work flow andtransforming management for better performance. But the ministries are free tochoose their own path of reform. Thus, the adoption of the team-based managementsystem gave the ministries flexibility with respect to the pace of innovation. Thepresident and the public recognized it as a symbol of actively implementinginnovation. The ministers, politically appointed by the president, may have simplyimitated the team system to show support for the president’s innovation drive. Theyaccepted those implicit pressures by adopting the system.Without other specific management techniques to improve performance, the teamsystem is a convenient tool for innovation as other ministries have already introducedit. As DiMaggio and Powell (1993:69) point out, “uncertainty is a powerful source thatencourages imitation.” Combined with coercive pressure, it forces ministries to adoptthe team system. In spite of the negative effects found in other ministries, ministrieshave had little choice other than to introduce the team system. The positive effects ofthis system were overestimated while the negative effects were underestimated. Someministries were in conflict over measuring the costs. In the midst of this uncertainty,team-based management was “theorized” as a standard model for innovation (Strangand Soule 1998). A “theorized” model can be a convenient source of practice thatministries use.Direct pressure also contributes to the selection of team-based management inministries. MOGAHA is responsible for leading and coordinating government-wideinnovations, and reviews and evaluates ministries” innovativeness and innovationinitiatives at the end of the year. Therefore the ranking of a ministry’s innovativenessis reported to the president and is open to the public. The evaluation displays thenumber of initiatives a ministry introduced, performance pay, and job postings. In thisway, ministries compete for higher ranking in terms of innovativeness. As a result,ministries must implement the team system to receive higher evaluations.



Page 198Vicarious learningSetting aside leadership or coercive imitation, the third driver comes from ministriesthat have started to see how others innovate. They read about the benefits and costsof the system. Given the freedom of choice, ministries calculate positives andnegatives to determine if they should adopt a policy. As Lee and Strang (2006) pointout, the drivers are not a process of blind imitation, but a professionally drivendynamism where policy experts select and codify “best practices.” In order to reducehierarchical levels and improve productivity within the government, ministries start to“theorize” or model the team-based management system.The calculation of economic or non-economic benefits forces some ministries to adoptthe team-based system. The resistance of employees, criticism from labor unions,derisive responses from congressional members, and the responsibility of failure areweighted against performance enhancement, fair competition, and enhanced customersatisfaction. Ministers are forced to ponder whether or not goals can be achievedunder the system. Analyzing the international diffusion of public sector downsizing,Lee and Strang (2006) demonstrate shifts in government employment in the UnitedStates and mutual influence among nations that are geographically close, and thattrade extensively. They show that “downsizing is contagious while upsizing is not:proximate downsizers but not upsizers are imitated, and states act on evidence thatdownsizing is economically beneficial while ignoring evidence that it is harmful.” Whileadministrative reform suggests growing convergence in the qualitative strategies usedto achieve downsizing, models of expanding the public sector are likely to be lessstandardized and more nation and context specific.Seen as breaking up bureaucracy, ministries are asked whether they should adopt thesystem or not. The response to the question varies. A lot of pros and cons wereissued before MOGAHA adopted the team-based system as an innovation model.Initiating drivers and adopting timeSince the initiation of the team system by MOGAHA, it took one and a half years foralmost half of the ministries to participate in the new system: 21 ministries during 18months from March 2005 to September 2006. One or two ministries introduce it everymonth. Based on the time of adoption, three periods taking five to six months can benoted since the adoption by MOGAHA: early adopters, mid-time adopters and lateadopters. As Table 11.3 shows, the first period for early adopters ranges from March2005 to July 2005. The second period for mid-time adopters was from August 2005 toDecember 2005. The third period, the late adopters, ranges from January 2006 toJune 2006. Why do ministries differ in this way? Rogers (2003:282–285) classifiedadopter categories into five: innovators, early adopters, early majority, late majority,and laggards. The time period of adopters classified above depends upon the rate ofadoption rather than absolute time. Rogers (2003) divided time periods into fivecategories by the percentage of adoption members to total members.



Page 199Table 11.3 Classification of three periods by adoption timePeriod Adoption Ministry Initiating forces2005 March-2005 July Early adopters 4 ministries Initiative leadership2005 August-2005 December Middle adopters 12 ministries Coercive imitation2006 January-2006 June Late adopters 6 ministries Vicarious learningStudies on team management and its effectiveness have remained a static method.Most articles have focused on organizational characteristics as explanatory variables ofteam effectiveness, called “feature-based models” by Erickson and Dyer (2004). Also,Jones and Lindley (1998) examined the effects of preparation and the building ofteam structures with participants. During the transition to teams, they argue, teamperformance correlated with four major concerns of the participants over time:managerial support for independent decision-making, role clarity, workload distributionpressures, and team social support. Erickson and Dyer (2004) examined the effects ofa team’s mobilization and launch activities on subsequent team progress andperformance by studying teams from start to finish. They argued that high performingteams mobilized relatively quickly, used comprehensive rather than limitedmobilization strategies, and conducted participatory rather than programmed launchedmeetings. In the same manner, the main drivers of the team system may differ byadoption periods.Early adopters: leadershipFirst of all, the initiating force for adopting the team system in the early stage wasleadership. Early adopters can serve as role models for many other members (Rogers2003:283). In addition, Rogers (2003:290) pointed out that early adopters are morehighly interconnected through interpersonal networks in their social system than arelater adopters. In the Korean case, ministries coordinating between ministries tend toadopt the team system earlier. The Ministry of Government Administration and HomeAffairs adopted the system early. It coordinates the organizational size and structureof each ministry, while the Ministry of Planning and Budgeting coordinates an annualbudget between ministries. The Public Procurement Service (PPS), an agency moreconnected with other ministries on the provision of materials and contract, alsoadopted the system earlier than others. These ministries have a higher degree ofopinion leadership within the government because their function is one of internalcoordination.A minister’s leadership plays an important role in implementing the team system. Theministers who have adopted it have shown a higher degree of innovativeness and awillingness to introduce new ideas. They have strongly implemented innovativeinitiatives such as performance management or quality control of policy. Minister OhYoung-Kyo of MOGAHA, an innovator of the team system



Page 200in the government, has had success as an innovative CEO at the Korean TradeOrganization (KOTRA), a public enterprise responsible for facilitating internationaltrade by bringing together Korean exporters and overseas buyers. Due to hisleadership, KOTRA became the best enterprise in work performance and customersatisfaction in 2003–2004. In the Ministry of Planning and Budgeting, the minister wasalso positive about innovation. Due to his leadership, the minister was ranked first onthe evaluation of innovation in 2004. According to evaluations, PPS was also the bestinnovator among the under-ministerial agencies in 2004. Leadership was a strongdriver for introducing the team system in the earlier stages.Proposition 1: Innovative leadership will be a more influencing driver to initiate earlyadopters to the team system than coercive imitation and vicarious learningMid adopters: coercive imitationRogers (2003:283) points out that the early majority interacts frequently with theirpeers, but seldom holds positions that allow them to express opinions. Mid adopters,Roger’s early majority, seem to have less power with respect to opinion within thegovernment. Though the ministers or administrators ultimately decide to introduce theteam system, the role of leaders is not as strong as with the early adopters. Coerciveimitation works as a more influencing driver for introducing the team system thaninnovative leadership. The adoption of innovative measures by pioneers in thegovernment makes other ministries fear possible repercussions if they do not acceptchange. Ministers tend to fear the possibility of having their ministry identified aspassive. Thus, a coercive climate takes place throughout the government. Someradical ministries often publicize a positive position on innovation itself, but will notintroduce the team system. Others suspend the implementation of the team system.In other words, ministers are forced to determine whether they should introduce it ornot. In this process, some ministries start to introduce the team system without anycalculation of the benefits and costs associated with the system.Though most ministers have a strong willingness to innovate, it is difficult to find amodel to improve organizational performance. The vague and uncertain picture ofinnovation makes ministries define the team system as an appropriate model ofinnovation. They usually agree upon the need for creating innovation in theirorganization, but cannot decide what they should do for innovation. To reduce thisuncertainty, some ministers follow the early adopters’ decisions to free themselves ofthe coercive climate. A ministry’s weakness in capacity and network accelerates theirpotential fear and makes them adopt the model earlier. Only five out of a total of 12mid-time adopters are ministries, while the other seven are sub-units withinministries. By imitating the early adopters, under-ministerial agencies are able toavoid uncertainty related to innovation, and to show that they are willing to innovate.The introduction of the team system was given a great amount of attention at the



Page 201Annual Innovation Evaluation (AIE). The ministries showed concern with respect totheir AIE score, which ranks each ministry based on their level of innovation.Proposition 2: Coercive imitation will be a more influencing driver to initiate the teamsystem for the mid-time adopters than innovative leadership and vicarious learningLast adopter: vicarious learningDuring 2005, the ministries faced a dilemma regarding the application of the teamsystem. While ministries that were weak in capacity or network quickly responded tothe uncertainty in 2005, more confident ministries suspended their decision orcalculated the positive and negative effects of the model. Rogers (2003:284)mentioned that the adoption by late majorities may be due to economic necessity andthe result of increasing peer pressure. As the year 2006 began, ministries receivedless coercive pressures. The strong push for the team system gradually cooled from2005, while innovation itself was still in progress. Ministries did not need their peers”approval to legitimize innovation. The possibility of defining them as an anti-innovation member decreased. In addition, the introduction of the team system didnot mean innovative leadership.As a result, more autonomy of choice prevailed. External forces pushing the teamsystem significantly decreased. In February 2006, Minister Oh of MOGAHA steppeddown from his office. Without the adoption of the team system, a few ministriesemployed a performance management system in its place. For example, the CivilService Commission implemented a performance agreement system in which theminister and bureau directors establish specific performance goals to be achieved forthe year, along with a written agreement outlining these goals. Thus, ministriesviewed benefits from a more calculative perspective. They also considered alternativesfor creating more innovative policies.During this period, uncertainty significantly decreased. Ministries witnessed the earlyand mid-time adopter progress through introduction and implementation for about ayear. Sixteen early and mid-time adopters taught possible followers the positive andnegative effects of its introduction. Ministries still in conflict evaluated organizationalcharacteristics and the environment they were facing. In spite of many complaintsabout the negative effects, five remaining ministries adopted the team system, withthe hope that it would be an effective tool for improving performance. Also, the teamsystem did not bring any invincible resistance from inside or outside the adoptingorganizations.Based on these changes, the team system started to be theorized from anexperimental model to a tested one, which drove its diffusion. However, the teamsystem may fail to theorize because the number of adopters has decreased. There stillseems to be a lot of skepticism on the application of this new system in thegovernment. This is why some ministries have adopted the system and others havenot. Ministries will adopt the system if they are convinced that they will receive morepositive than negative effects.



Page 202Proposition 3: Vicarious learning will be a more influencing driver to initiate the lateadopters to the team system than innovative leadership and coercive imitationInnovative leadership will be a more influencing driver to initiate the team system forearly adopters than coercive imitation or vicarious learning. While the most influentialdriver of early adopters is innovative leadership, coercive imitation will be the mostinfluential driver for initiating the team system for the mid-time adopters thaninnovative leadership or vicarious learning. For the late adopters, vicarious learningwill be the most influential driver for initiating the team system than innovativeleadership or coercive imitation. Table 11.4 summarizes the degree of three influentialdrivers over the three different adoption periods.Conclusion: challenges and sustainabilityTeam-based management bore pros and cons with the advent of its initiation as wellas its overall sustainability. On the one hand, proponents argue its positive effects onthe system. They lay emphasis on the functional motivations issued during the birthstages. Oh Young-Kyo (2005) reported that team-based management brought anefficient work process, responsive customer service, and budget savings. Sevenmonths after the team system was adopted, MOGAHA conducted a survey to check itsperformance. Customer satisfaction dealing with services increased from 56.3 percentto 61.7 percent in the public, and from 51.9 to 72.3 among the other ministry’s publicemployees. It was also reported in his article that the time spent for service wasreduced from an average of 9.3 days to 2.3 days. Over 70 percent of all work wasdelegated from the minister, vice minister or bureau directors to team leaders or teammembers. The long line of managers reporting to the minister decreased during thistime. Performance-oriented culture became spread through the government. Inanalyzing a team performance survey, given to team members of MOGAHA, Kim andAhn (2006) showed that structural change brought about by the team system createdefficiency, while the evaluative and human resource changes were increasedeffectiveness and responsiveness.The opponents did not accept the positive impact of team-based management.Instead, they stressed its limitations and dysfunctions. In particular, the sustainabilityof the positive effects of the team system would be reduced when the three driversare not involved. It would be unlikely to be successful without innovative leadershipduring the first stage. It would also not be easy to observe the sustainability of thesystem for the future without vicarious learning.Table 11.4 Three influential drivers, by adoption periodsInitiating forces Early Adopters Mid Adopters Later AdoptersInnovative leadership High Medium LowCoercive imitation Medium High MediumVicarious learning Low Low High



Page 203Opponents put emphasis on the positive aspects of the existing bureaucratic system.Ingraham (2004:229) pointed out the attractiveness of the so-called old black boxes:“rigid boundaries closed off external influence, decision-making was neutral and notintended to shape policy, individual motivations and incentives were not dysfunctionalto the organization, and leadership unquestionably had the skills and ability to lead.”Though he did not target MOGAHA, Yu (2001) examined team performance in severalpublic enterprises in Korea by comparing team and non-team departments. He foundno significant difference between the two divisions in terms of team performanceaccording to employees” perception of effectiveness and employee satisfaction.The Korean government also faced strong resistance from internal employees as wellas external actors. Labor unions representing public servants in Korea argued that theteam-based system is established on distrust and esteem between the ministers orhigh level officials and the employees. They note that the fundamental meaning ofteam-based systems is to encourage collaborations among employees, rather than tostimulate competitive instincts in the work place. On the basis of the marketmechanism, they add, it is intended to instigate strong competition between teams,even though government policies should be closely collaborative among differentpolicy divisions. An incumbent member of National Assembly in Korea interestinglyargues that the “customer” is an object that should be distinguished from “people” asa subject for better governance. Similarly, Lencioni (2002) also described thedysfunctions of a team as follows: (1) absence of trust, (2) fear of conflict, (3) lack ofcommitment, (4) avoidance of accountability, (5) inattention to results. He alsosuggested a team diagnostic instrument that team members can use to assess theirteam’s susceptibility to each of the five dysfunctions. This includes “team memberscalling out others’ deficiencies or unproductive behavior,” and “team membersengaging in discussions that result in clear and specific calls to action.”In spite of a lot of criticism, the team-based system is one of the importantalternatives for changing the internal structure and leading it towards goodgovernance. The success of the system may depend upon how the new system canmake a smooth transition from traditional hierarchal structure. As Che and Yoo (2001)state, the success of teams may rely on long-term interaction, decentralized authority,peer monitoring, and collective incentives. The question as to how the system can bemade sustainable becomes more important than its initiation itself. The findings ofErickson and Dyer (2004) on high performing teams give us a reference as to the“how” question. According to their analysis, successful teams tend to mobilizerelatively quickly, to use comprehensive rather than limited mobilization strategies, aswell as to conduct participatory rather than programmed launch meetings. Hackmanand Wageman (2005) propose that the motivational coaching, rather than theconsultative or educational coaching, of teams is most helpful when provided at thebeginning of a performance period.This study shows that the emergence of the team-based management systeminitiated by the Korean government is an alternative for change in the internalstructure of the government. The paper also examines the relationship between



Page 204204 C.K. Lee the temporal period of adoption and its key drivers. Early adoptersbetter utilize the ministers’ leadership, mid-time adopters illustrate coercive imitation,and later adopters use vicarious learning as a key driver for adopting the teamsystem. Different ministries initiate team systems at different times with differentrationales. As described above, a horizontally collaborating system is not an easyalternative for replacing the hierarchical structure and culture within the government.As Ezzamel and Willmott (1998) indicate, it extended rather than supplantedtraditional, hierarchical systems of management control, because it established a senseof self-identity as “machinist”, and had the unintended effect of fomenting hostilitytowards the managerial goals. When combined with supportive policies of recruitmentand selection, the team system might strengthen managerial control and intensifywork activity towards good governance (Ezzamel and Willmott, 1998). Accordingly,the team-based management system should be cautiously adopted and well preparedfor it to be successful and sustainable. This may help explain why half of theministries in Korea still remain undecided on the adoption of the team system.However, despite drawbacks and limitations, the horizontal team structure is asuccessful tool for administrative reforms that establish a hierarchical bureaucracy.ReferencesAbramson, M., Breul, J. and Kamensky, J. 2007, Public Manager, Spring 2007, vol. 36,issue 1: 3–11.Che, Y.K. and Yoo, S.W. 2001, ‘Optimal Incentives for Teams’, American EconomicReview, 91(3): pp. 525–41.DiMaggio, Paul J. and Walter W. Powell 1983, ‘The Iron Cage Revisited: InstitutionalIsomorphism and Collective Rationality in Organizational Fields.’ American SociologicalReview, 48: 147–60.DiMaggio, P.J. and Powell, W.W. 1991, The New Institutionalism in OrganizationalAnalysis. Chicago: The University of Chicago Press.Erickson, Jeff and Dyer, Lee 2004, ‘Right From the Start: Exploring the Effects of EarlyTeam Events on Subsequent Project Team Development and Performance.’Administrative Science Quarterly, vol. 49: 438–71.Ezzamel, Mahmoud and Willmott, Hugh 1998, ‘Accounting for Teamwork: A CriticalStudy of Group-based System of Organizational Control.’ Administrative ScienceQuarterly, vol. 43: 358–96.Heinrich, C.J., Hill, C.J. and Lynn, L.E. Jnr. 2004, ‘Governance as an organizing themefor empirical research, In P.W. Ingraham and L.E. Lynn, Jr. (eds.), The art ofgovernance: Analysing management and administration (pp. 3–19) Washington, DC:Georgetown University Press.Hellriegel, D., Slocum, J.W. Jnr., Woodman, R.W. 2001, Organisational Behaviour (9thed), South-Western College Publishing.Ingraham, Patricia W. 2004, ‘Analyzing Management Structure and Systems in aGovernance Framework: What Have We Learned?’ in Ingraham, Patricia W. and Lynn,Laurence E., (eds.) The Art of Governance: Analyzing Management andAdministration, Washington D.C., Georgetown University Press.Jones, Robert G. and Lindley, Warren D. 1998, ‘Issues in the Transition to Teams.’Journal of Business and Psychology, vol. 13, no. 1, Fall: 31–40.



Page 205Hackman, Rechard J. and Wageman, Ruth 2005, ‘A Theory of Team Coaching.’Academy of Management Review, vol. 30, no. 2: 269–87.Kim, Byeng Seob and Oh, Si-Young 2005, ‘Teams in Government K: Myth andRealities.’ Korean Journal of Public Administration, 43(4): 411–36.Kim, Tae-Ryong and Ahn, Hui-Jung 2006, ‘Effect Analysis on Team System.’ KoreanPublic Administration, 2006 Summer Conference: 345–61.Kim, Sang Hun 2005, ‘Team System and Performance Management in Public Sector.’Korean Public Administration, 43(4): 459–78.Lee, Chang Kil and Strang, David 2006, ‘The International Diffusion of Public SectorDownsizing: Network Emulation and Theory-Driven Learning.’ InternationalOrganization (IO), 60(4): 883–910.Lencioni, Patrick M. 2002, The Five Dysfunctions of a Team: A Leadership Fable. SanFrancisco: Jossey-Bass.Ministry of Government Administration and Home Affairs 2006, Introduction andOperations of Government Team Organizations, Research Report: Seoul, Korea.Organizations for Economic Cooperation and Development (OECD) 2004, ModernizingGovernment: The Way Forward. OECD: Paris.Oh, Si-Young 2005, ‘A Study on Team Effectiveness.’ Korean Association of PublicAdministration, 2006 Fall Conference: 1–17.Oh, Young-Kyo 2005, ‘The “Team” System for the First Time in Government Agency:Vision and Change Management.’ Korean Public Administration, 43(4): 379–410.Osborne, David and Ted Gaebler 1992, Reinventing Government: How theEntrepreneurial Spirit is Transforming the Public Sector. New York: Plume.Rogers, Everett M. 2003, Diffusion of Innovations, 5th edition, Free Press, New York.Strang, David and Sarah A. Soule 1998, ‘Diffusion in Organizations and SocialMovements: From Hybrid Corn to Poison Pills.’ Annual Review of Sociology, 24: 265–90.Tuckman, B.W. and Jensen, M.A.C. 1977 ‘Stages of small-group developmentrevisited’, Groups and Organization Studies, 2: 419–420Wilson, James Q. 1989, Bureaucracy: What Government Agencies Do and Why TheyDo It. Jackson, TN: Basic Books, United States.Wing, Linda 2005, ‘Leadership in High Performing Teams: A Model for Superior TeamPerformance.’ Team Performance Management, vol. 11, no. 1/2: 4–11.Yu, Min Bong 2001, ‘Realities and Effectiveness on Team Organization: Focus onPublic Enterprises.’ Korean Journal of Public Administration, 35(4): 157–77.



Page 20612 Technocratic solutions versus political realitiesImplementing governance reforms in the Balochistan province of PakistanRaza Ahmad1 and Syed A.A. AkifAbstractAlmost immediately after taking power in 1999, Pakistan’s fourth military regimearticulated a seven-point agenda that in large measure entailed governance policyshifts – the key being decentralisation of powers and democratic local governance.Pakistan’s Poverty Reduction Strategy Paper (PRSP) articulated these goals in the pro-poor growth framework. The international aid community favored this reform climateand the Asian Development Bank (ADB) took a lead in supporting public sectorreforms, especially by way of improving sub-national governance in a federal context.This paper focuses on the reform agenda articulated through ADB-assisted BalochistanPublic Resource Management, and the complementary devolved social servicesprograms. While it is still early to assess the impact of these policy shifts, weconclude that improved service delivery through better fiscal management is a long-term goal. The ongoing reforms are located within and impacted by historicallydifficult federal-province relations and a highly contested political climate. There is aninherent danger of institutional reforms being appropriated by entrenched elites. Inconclusion, sub-national reforms cannot be isolated from the imperatives of federalismand local patrimonial structures, i.e. the political economy context. Developmentagendas should overtly address politics in designing and implementing governancereform programs.Background – the emergence of governance discoursePakistan’s military regimes, which have spanned at least 30 of the country’s 59 years,have always justified their advent and continuation on the plea of misgovernance bypreceding elected governments. Soon after assuming power, General PervezMusharraf, who initially took on the rather unusual title of “Chief Executive”, lamentedthe misdoings of the former elected government and announced an ambitious seven-point agenda, the core of which related to governance reforms – getting the state,markets and citizens right.2 The strategy to accomplish the goals of the seven-pointagenda articulated a wide range of objectives: economic revival, poverty alleviation,political restructuring, and devolution of power to grass roots.



Page 207On the symbolic occasion of the nation’s 54th Independence Day – August 14, 2000 –the transformational “Devolution Plan 2000” was introduced. This plan radically alteredlocal and provincial governance arrangements that had been in place for over acentury and a half. The reforms were articulated through the Local GovernmentOrdinances3 (LGO) of 2001, which provided the policy, legal, and regulatoryframeworks for provincial and local government functions, roles, and accountabilities.The domestic reform climate and the imperatives of Pakistan’s strategic role in thepost-9/11 world drew immediate attention to the international developmentcommunity. Agencies such as ADB, among others, took the lead in technical andfinancial support to the governance reforms. This was a time when the emergingconvergence of liberal governance principles and poverty reduction was taking place ata global level.Notwithstanding the criticism of a military regime undertaking governance reform,engagement was preferred over political polemics, not just by international donors,but the local civil society as well. In fact, commentators have highlighted an alignmentbetween the socially advanced classes and the military rule (Zaidi 2005a). Littlewonder that there was wide support for General Musharraf’s takeover and subsequent“reforms”.That the devolution came about in the wake of the failure of a multi-donor SocialAction Program (SAP) was even more instructive. The failure of SAP to convertresources into results had established that issues of poor service delivery and humandevelopment indicators were largely embedded in governance failures. In all, over $9billion was invested towards improving basic services. While education was a keypriority, the net enrollment rates declined. In effect, the seven-point agenda wasnothing less than a direct domestic response to ailments already identified throughresearch and prognosis by the international community.Within two years, the devolution process was well into implementation. A transition ofthis scale, given the complexity of Pakistan’s federal structure and entrenchedpatterns of governance, was far from smooth. However, the regime showed itsresolution in staying the course. This resolve was undermined to some extent by thepost-2002 election scenario.Focus on subnational governanceWhile the thrust of 2003 PRSP-led reforms largely focused on economic growth andimproved governance, the new “devolution” reforms articulated improvements inservice delivery and on addressing the “social gap”4 that had resulted from years ofmisgovernance leading to abysmal social indicators.The favourable response of the international community was illustrated through theADB’s 2002–06 Country Partnership Strategy (CSP) which emphasized two things: (i)governance shifts through devolution and access to justice, and (ii) reforming marketsthrough an economic globalization strategy, together, leading to poverty reduction(ADB 2002). Other donors articulated similar views. From 2001, positive response andassistance related to Pakistan’s reforms steadily



Page 208increased. The major donor support to devolution came from ADB in terms of theUSD $350 million Decentralization Support Program (DSP).5However, devolution reforms were not easy to manage. The transfer of powers fromthe province to the local level was not complemented by a matching devolution ofpowers and functions from the federal to the provincial. More importantly, Pakistan’stenuous federal system, notably the provincial, territorial and patrimonial domains ofgovernance, stood challenged by the devolution reforms. Other than the politicaltension, another difficult feature to negotiate was that the devolution required theprovincial governments” policy processes to respond to the new systems of fiscal andadministrative decentralisation. The LGOs, for instance, required the Province FinanceCommissions (PFC) to make formula-based transfers to the local governments forimproved development outcomes; and the provincial departments – the erstwhilepolicy makers and implementers – underwent a shift in their role by being declaredmerely policy setters and monitors of local performance. The transfer of provincialtechnical staff to local governments and the strengthening of capacities at the locallevel emerged as key challenges to implementing devolution. Thus the focus ofreforms and development assistance shifted from the national to subnationalstructures and systems of governance (ADB 2004a).It soon became apparent that the pre-devolution fiscal management patterns had leftthe provinces indebted to Federal Government and locked in their recurrent budgets.In particular, Balochistan lacked the fiscal space to ensure that sufficient transfers tolocal governments could take place for improved service delivery.Having presented an overview of reform progress, we now turn to the province ofBalochistan to indicate how the wide-ranging donor support since decentralisationresponded to Balochistan’s poverty, isolation and constraints in implementingdecentralisation for improved services and local governance.Balochistan: economy, poverty and the need for governance reformThe province of Balochistan comprises 44 percent of Pakistan’s land area, has acoastline extending over 700 kilometers, produces 40 percent of Pakistan’s energyneeds,6 and is rich in mineral resources.7 Its very low population density means apoor return on infrastructure, and its great expanses of difficult, inhospitable terraindefy economic feasibilities. Thus development in all areas, particularly economic andsocial, remains below potential. It is not surprising that Balochistan is Pakistan’spoorest region, with 21 of its 29 districts being placed among the lowest 30 ofPakistan’s 112 districts (Government of Balochistan 2005). According to theBalochistan PRS, poverty levels are estimated to be as high as 47 percent(Government of Balochistan 2003). Social sector indicators of Balochistan are lowest inSouth Asia; female literacy, for example, is just 18 percent compared to the nationalaverage of 38 percent. Likewise, village electrification is at 25 percent – one third ofthe national figure. Balochsitan also ranks lowest of all Pakistan’s provinces on UnitedNations Human Development Index (HDI) (UNDP 2003).Balochistan is a predominantly agro-pastoral economy that faced severe droughts



Page 209in the last decade. Drought-related internal migration increased the vulnerability of thepoor, especially women. Over-exploitation of groundwater and the rapiddisappearance of traditional surface water sources have further aggravated waterscarcity. Added to this was the human capital deficit, the overcoming of which isessential for Balochistan to tap internal and external markets and capitalize onmarket-driven economic growth.Historically, relations between the Federal Government and Balochistan have beentenuous, characterized by provincial dependence on federal transfers as the mainstayof its budgetary and development needs. This situation is compounded by the popularperception that Balochistan does not receive its due from the energy revenues thatare managed by the Federal Government under the 1973 Constitution.8The ADB emerged as a major development partner in the province, extending itsfinancing through the 2004 Balochistan Resource Management Program (BRMP) (ADB2004b), approved for $130 million, and the 2005 Balochistan Devolved Social ServicesProgram (BDSSP), approved for $200 million. These programs share the objectives ofimproving the fiscal space and enhancing provincial and local governmentperformance by strengthening decentralisation and policy making in the provincial andlocal contexts. The aim is to generate improved citizen services through creating fiscalspace for a province that has the highest poverty rates and a budgetary (currentaccount) deficit of nearly 50 percent, and an inward monetary flow accounting foralmost 95 percent of the budget (ADB 2004b). The modality of assistance is policy-based lending (PBL) – a shift away from the structural adjustment that emphasizesand rewards domestic policy reform and avoids the use of the term “conditionality.”Rationale for subnational lending against policy change targetsIn 2004, provincial fiscal management was severely constrained due to a mix of thefollowing conditions: traditional economy with limited tax base, dependence on federaltransfers, and endemic debt with limited capacity for debt management. For instance,95 percent of provincial revenues consisted of federal transfers and grants, and 60percent of expenditures came under the head of “current expenditures” – the non-development part of the budget – including subsidies on tube-wells and wheat flour.The total debt liability of the government of Balochistan at the end of FY2003 was PRs38.1 billion (approximately US $635 million), and debt servicing has veered between11–15 percent of total expenditures since 2001. Domestic loans were predominantlycash development loans (CDLs) (PRs 12 billion or US $200 million in FY2003) from theFederal Government.9 The BRMP assistance was designed to restructure PRs 6 billion(US $100 million) of CDLs, leaving the province, by its own estimates, with aroundPRs 8.5 billion in high-interest debt. This was the underlying rationale for externalassistance through BRMP and, later, DSSP.However, the more important factor for these program loans was to enhance the



Page 210resource envelope for local development. It was argued that revenue expenditureitems of the government were provincial allocable,10 general administration, and debtservicing and this left little for development resources for local transfers. Since thelaunch of the Balochistan LGO in 2001, more than two-thirds of the revenueexpenditure was allocated to provincial allocable, most of which went to localgovernments as grants. Although funds transferred to the local government increased,these could barely cover the salaries – let alone basic operations and maintenanceexpenses. The social gap in Balochistan required increased resources.The political economy of BalochistanNo single model of political and social theory can adequately help us analyze thepolitical economy and governance context of Balochistan. Centuries of traditional socialrelations, the colonial and post-colonial periods and, now, the imperatives ofglobalization, have had profound impacts on the state, society and economy ofBalochistan. Whereas a full-scale analysis of such variables may not be feasible here,we will use the clientelist and (neo-) patrimonialist patterns of governance – and theirvariations – to understand the complexity of Balochistan’s political economy. We addto this the post-colonial, especially the recent wave of modernization attempts, drivenby imperatives of globalization and liberal market integration.National imperativesAmong the areas that were under the direct or indirect administration of the BritishRaj, Balochistan was the least “colonized”. Instead of replicating the colonizationexperiments implemented in northern and parts of southern India, the British chose touse a carrot-and-stick policy in Balochistan in order to gain the allegiance of itsdifferent tribes while leaving the medieval social organisation intact. Hence,Balochistan did not gain from the transformational processes of colonization that oftenbrought the pulls of modernity in their wake. This historical incongruity was inheritedby an independent Pakistan in 1947.11The process of “national” integration has been far from smooth; recognized as apolitical reality, the strong influence of federal pulls has resulted in bargains betweentribal elites and the federal power centres. Balochistan elites and the emergent middleclasses have continued to articulate their concerns regarding inadequate safeguardsfor provincial autonomy in Pakistan’s federal structure. Given the small indigenouspopulation (which is now under an increasing threat of outside economic interestgroups in areas of new opportunity), and consequent small share in the nationalrevenues, a general perception of discrimination and denial of “rights” over theprovince’s natural resources – especially natural gas, and revenues from theirexploitation – is increasingly popular with local citizens. In recent years these tensionshave come to the fore and become a key political question requiring a solution.Balochistan’s administration system is a hybrid of formal and informal systems.



Page 211The tribal system, with its archaic social structures and a justice system controlled bythe tribal elite,12 had long made formal rule-of-law framework irrelevant for most ofthe population. Such a system obviously perpetuates a high degree of conflict andinsecurity – thereby leaving the poor communities distant from the “formal” state.Fortunately, however, change in this area is in progress. Until very recently, less thanfive percent of the provincial territory was under regular police jurisdiction.Increasingly, the police are taking over law and order duties from the “levies”.The national context also has repercussions on provincial political stability. Politicalinstability at the centre has its spillover into the subnational political processes. Ingeneral, the perception is common that provincial governments are extensions ofcentral patrimonies. The issues of natural resource royalties and fiscal transfers arecited as illustrations of this trend. However, this is not as simplistic as the tribal elites”populist rhetoric maintains. The beneficiaries of central largesse, over the decades,have been the tribal chieftains who have continued to enter into patron-clientrelationships with most of the central governments. Continuous change in executivepower, studies have shown, results in under-investment in public goods and retardsgrowth (Alesina et al. 1996). Political instability results in frequent policy reversals,sends uncertain signals to the implementers, and weakens the accountabilities forperformance and policy outcomes. Pakistan and Balochistan suffer from this syndrome.For instance, prior to the military takeover during the decade of democracy (1988–99), there were nine prime ministers and federal cabinets and their counterpartdispensations at the provincial levels.Rent-seeking and corruptionRent-seeking occurs where individuals or groups influence policy and decision-makingto enhance their personal incomes and assets. Such behaviour is not uncommon tothe political process but in the context of patron-client systems, it assumes aparticularly negative effect by becoming unmanageable due to limited agents ofrestraint. (Collier and Pattillo 2000 cited in Brinkerhoff and Goldsmith 2002). Informalexchanges, rents and such payments may actually be completion of informal contractsbetween patrons and clients. This becomes even more relevant in the context ofBalochistan, where tribal loyalties and informal systems of give and take areembedded in social relations.Rent-seeking trends also help us understand the issues of subsidies (noted in sectionII) that cause an endemic fiscal crisis in Balochistan government. Similarly, thepersistent indebtedness can be located in the framework of clientelism as well. Alsooperative here is Edward C. Banfield’s construct, “amoral familism”, which refers to asystem in which each group focuses on maximizing its own interests and formstemporary coalitions to further its respective interests in its quest for a monopoly overthe state to gain personal power, at the cost of weakening the state. Examplesinclude cases from Liberia and Sierra Leone, where predatory elites and statecriminality have defined instability; parallels are found in Balochistan. Here, thegangster-politician, as the patrimonial chief, with the help of violence



Page 212and terror “controls commercial activities and is an effective source of governance insome territory or locality where the central state has withered or disappeared”(Brinkerhoff and Goldsmith 2002).Elite captureA key manifestation of clientelism has been the elite capture syndrome, generallyconsidered as a cause for dismal public service delivery. It has been argued that, in“oligarchic” societies, elites will oppose mass education because the more educatedthe population, the greater the pressures for democratization and the greater thethreat to the power of the privileged groups (Bourguignon and Verdier 2000). Othershave held that rural elites have blocked education reform (Haq 1998). Balochistan isno exception to this trend, and the age-old structures of tribal loyalty and traditionreinforce this situation. Public services demonstrate the worse forms of patronage.With recruitment to public office being the ultimate act of patronage, “relaxation ofrules” (an official term used in government files) has meant that quality of recruitmenthas been poor, leading to staff absenteeism in service delivery institutions. Thisexplains why policy makers are keen to hire new staff or start new projects ratherthan maintain the standards of existing services and public goods. In such anenvironment, electoral politics is nothing but a means to channel largesse to clientsbased on local influence and ethnicity; such largesse can even be appropriated bysmaller groups of individuals close to centres of patronage and patrimony (Wilder1999). All elected governments have continued to institutionalize patronage throughspecial development packages for the elected national and provincial legislatures.Ethnic polarization and factionalismIn yet another dimension of the governance patterns noted above, clientelism hasbeen linked to ethnicity, as the leaders use ethnic criteria for exclusion. Ethnic identityinfluences political discourse in Balochistan (Titus 1998). Ethnic politics in Balochistantend to be worked out through established processes and traditions of negotiation andmutual respect. Provincial-level politics provide intricate mechanisms for formal andinformal negotiation, conflict avoidance and conflict resolution. Tribal norms andcustoms provide the parameters of conflict resolution, and tribal leadership becomessynonymous with political leadership. The tribe acts as the institution for organizingintra-group collective action, on the one hand, while simultaneously defining theboundaries of the group, on the other. Social inclusion and social exclusion, therefore,are delineated along the lines of tribal affiliation. The tribal systems sustain strongpatriarchal norms concerning the rights of women, their access to resources andspaces, and their participation in social and political life. There is close inter-linkagebetween patriarchy and tribalism.Balochistan’s ethnic complexion lends additional complexity to this quandary. Thebasis of ethnic distinction in Pakistan is language. As per the census of 1998,Balochistan is inhabited by communities, identified on the basis of the mother



Page 213tongue, as follows: Balochi (55 percent), Pushto (30 percent), Sindhi (5.6 percent),Seraiki (2.6 percent), and others (Punjabi 2.5 percent and Urdu 1 percent). Further,these communities are concentrated in different regions: On a north-south divide,those in the south are predominantly Balochi-speaking, while almost all of theprovince’s Pushto speakers are concentrated in northern and north-eastern districts.These patterns are also reflected in the composition of the provincial legislature, thecivil service and the allocations for development. Competition over resources andpolarization at the provincial level is a well-established trend. A further compoundingelement in this complex situation is the rise of factional political organizations; forinstance, the quasi-religious political party Jamiat-i-Ulama-i-Islam has traditionallybeen an all-Pashtun party, while the Balochi speakers have continued to befragmented amongst the nationalist Baloch parties; in the middle is the “official”centrist Muslim League that has always enjoyed the support of the military, andbrings together a host of pro-establishment individuals.Empirical linkages between ethnicity and provision of public goods have beenestablished by researchers (Alisina et al. 1999). Spending on public goods is lower inethnically diverse contexts. Dilution of accountability also results from ethnicpolarization, as the elected leaders are returned to office on the basis of caste, tribeor ethnicity and not in terms of their performance (Keefer and Khemani 2003). In thecontext of Pakistan, low levels of human capital investment compared to the percapita income levels are linked to advanced degrees of polarization due to ethnicityand class (Easterly, 2003).Partial reform syndromeLastly, “partial reform syndrome” is a result of Balochistan’s governance system. It hasbeen said that institutional and political factors are more important than capacities inshaping incentives for policy makers to implement reforms (Brinkerhoff and Crosby2002). Gaps between stated and actual commitments to reform differ. Donor-ledreforms and the measures formally agreed upon at bilateral or multilateral foraminimize opportunities for rent-seeking and discretion to reward clients. Partial reformis, therefore, an outcome of a government’s necessity to keep the key patron-clientrelations intact. Political risks are thus inherent in the reform programs, andgovernments are selective in implementation. Neo-patrimonies often view donorassistance as largesse for the clients, and covenants are only formalistic governancefeatures that some bureaucrats are keen to implement.Midgel (1998) has also pointed out the triangles of accommodation amongbureaucrats, politicians and strongmen. Partial reform syndrome is often a “way oftailoring donors” proposals and projects to fit unique national circumstances.”13Globalization, modernity and geo-politicsBalochistan possesses deep-water ports on the main Arabian Sea/Indian Ocean thatnot only provide close access to the Persian Gulf, but also better transhipment hubs.It also borders volatile Afghanistan and geo-strategically important Iran. It represents



Page 214an area of great importance in future political and economic rivalry between existingand emerging global players. The independence of Central Asian States (CAS), theemergence of China as a global power, and the international consensus on nurturingstability in Pakistan (as a pre-condition to a peaceful Afghanistan) have all contributedto the implementation of a rather rushed but fundamentally profound developmentagenda in Balochistan. The province is also one of the feasible routes of energy tradefrom CAS, particularly Turkmenistan, which ranks fourth worldwide – after Russia, theUnited States and Iran – in natural gas reserves.14Modernization theory of development argues that economic growth leads totransformation of small, socially cohesive, traditional communities, and that theresistance to change thins away as economic interests generate new social dynamics.The story of Balochistan is a little different, in that its geographical spread, degree ofbackwardness and marginalization, low population density, complex political landscape,and consistent neglect pose key challenges to the advent of modernity. The ongoingstruggles against externally induced development in the province, therefore, areneither new nor exceptional.Several mega-projects – including development of the Gwadar deep-sea port, variousinter-provincial highways, and dams – have been initiated in Balochistan. However,these projects have given rise to political resistance by Baloch tribal leaders, who haveall along feared being further marginalized. Another reason is the limited control thatprovincial governance institutions are likely to have over these mega-projects. A thirdconcern is the uncertainty regarding the future ethnic demography and politicalbalance within the province. The Federal Government’s International donors –technical, financial support – policy transfers

Figure 12.1 Balochistan: global imperatives and national and local systems ofgovernance.



Page 215efforts to establish new military cantonments in the province for protecting key portand energy installations have also been unpopular with the Baloch nationalists. Issuessuch as resettlement difficulties (irregularities in land records, difficulty in gettingownership titles recognized, and compensation for displaced people) and influx ofskilled workers from outside have been cited as causes of “uneasiness and fear ofdisenfranchisement” (ADB 2004b). Figure 12.1 summarizes the governance andpolitical context, particularly the interplay of formal and informal, as well as theinteraction between the local and the global.Externally assisted reforms and institutional change programsWe now examine the progress of provincial reforms supported by the ADB. Inparticular, we focus on the BRMP and DSSP as instruments of policy change. Thissection notes the major successes and challenges, and documents some early signalsof institutional change, notwithstanding the difficult and complex political environmentnoted in the preceding section.Public resource management reformsThe BRMP is financed by ADB as a package for “law and public sector and economicmanagement” sector assistance, and highlights the overall good governance andinstitutional reform agenda underwritten in its policy matrix (ADB 2004b).The rationale for the program is the essential requirement for intervention to improvethe state of economic and social development in Balochistan, which remains far belowits potential, in spite of numerous initiatives, both indigenous and externally financed.The overarching goal is to: create sustainable fiscal space to finance high-priority,poor-targeted social and human development programs, implemented by localgovernments; support increased participation of key stakeholders for transparent,accountable, and sustainable public service provision;

Figure 12.2 PRMP policy goals.



Page 216and generate an enabling environment for the above (that is also supportive ofprivate sector development), with particular emphasis on sustainable watermanagement.The BRMP comprises a set of loans, amounting to US $133 million for public resourcemanagement reforms; these include a technical assistance (TA) project loan forsupporting public resource management reforms – especially to strengthen capacityfor program implementation. Figure 12.2 summarizes the policy goals under BRMP.Improved and efficient social service deliveryThe complementary budget support program – Balochistan DSSP – was approved in2005 with a cost of US $205 million. BDSSP was designed to assist the province inaccelerating progress in meeting the Millennium Development Goals (MDG) particularlyrelated to education, health, water supply and sanitation. The objectives of BDSSP areto: (i) improve access to enhanced-quality education, health, water supply andsanitation in the public and private sectors; (ii) decrease inequities in social servicesbased on income, gender, ethnicity, and geography; and (iii) strengthen the ability oflocal governments, the private sector, and communities to improve social servicedelivery.The BDSSP has four principal policy outcomes15: (i) provincial efficiency incentives tostrengthen and clarify relationships between provincial and local governments; (ii)managerial power to strengthen the relationship between the local governments andthe providers of social services; (iii) citizens” voice and empowerment; and (iv) privatesector engagement in service delivery. Figure 12.3 illustrates these outcomes and thetechnical features of policy reform.We now examine the progress and trends of these policy reforms.Enhanced fiscal space via fiscal reformsWe have noted above that the underlying rationale for reforms has been the rigid andnarrow budgetary space available to the provincial government due to highestablishment costs, unsustainable subsidies, debt servicing bills and lack of rule-based transfers to service delivery agents, i.e. the local governments, since the 2001devolution reforms.Province revenues: There are signs of tax rationalization. Since the commencement ofreforms, the revenue receipts from the tax and non-tax heads registered animpressive increase of 58.2 percent in FY2006 compared to FY2005.16 However, akey feature of planned reform (i.e. reducing exemptions to levy of property tax) hascontinued to face not only a weak political will – as local elites and strongmen resistthis move – but also a lack of persuasiveness on the part of officials, who have beenunable to plead their case based on an indigenous intent and purpose. A FederalCabinet Committee was formed to review the technically sound proposals to increasethe scope of the tax base, but has not been able to meet – like the cabinet itself,which meets only very occasionally. Another reform proposal, the reinstitution of theagricultural income tax (AIT), which has been suspended for



Page 217

Figure 12.3 Conceptual framework of DSSP policy reforms.PRA = policy reform action.Source: Asian Development Bank.over six years – officially on account of drought conditions but more due to politicalexpediency – has also not met with success for a number of reasons.Subsidies – the rents: At the start of the BRMP-led reforms it was clear that theundoing of subsidies, inasmuch as these reflected the rent-seeking nature of provincialgovernance, would be most difficult. Although intense conflict over this thorny issuehas continued between reform agents and entrenched elites, a middle ground – thepartial reform syndrome – was reached with the “capping” of the total amount ofprovincial subsidy in the 2006–07 budget.17 Subsidies were also co-produced by theFederal Government, and the latter also capped its share of the tube-well subsidy (atPRs 2 billion), with the agreement of no further change. Reform to reduce thewastage of water resources, the raison d’etre of the subsidy, has also receivedpriority. Issues relating to conservation of water are being acutely focused on as partof a US $1.92 million grant for the study of three major river basins and also thecomplete tube-well subsidy mechanism.Another drain on the constrained provincial exchequer – the untargeted wheatsubsidy – also underwent a partial re-structuring. With Balochistan being a highwheat-deficit area, the Provincial Government of Balochistan (PGB) had facilitatedprovision of wheat, the staple food item, below market prices to the flour mills withthe intention that the subsidy would trickle down to the consumer through cheaperatta (whole wheat flour). This practice continued in spite of the fact that there was noevidence that atta prices reflected the subsidy transfers. Such a system



Page 218of patronage was mere rent-transfer to local businesses and to bureaucrats whoregulated the subsidy. With technical support from the BRMP, the middle path was tointroduce a poverty-targeted atta subsidy. Although the PGB was persuaded to makean allocation of PRs 641 million in the FY2007 budget for a poverty-targeted attasubsidy, the rent seekers, led by the top food bureaucrat, have lobbied for using amajor part of the allocation for maintaining “strategic reserves”. Thus, instead ofseeing improved institutional arrangements for wheat distribution that involve electedlocal representatives, the re-structured subsidy has not been able to bypass localpatrimonies.The total development program of the province increased by 15 percent in 2005–06compared to the previous year, when the reform started. More importantly, fundswere directed towards the ongoing schemes, thus changing the entrenched practice ofdeveloping new schemes to appease political clients (while ignoring ongoing schemesand allowing partially completed projects to rot away). Thus the PSDP “throwforward18 syndrome” has been addressed and brought down from an estimated tenyears to just over five years over the past two years. Efforts have been made toimprove the development planning process by virtually doing away with “blockgrants”.19 There has been greater delegation to district governments to planschemes, up to a project cost of PRs 10 million. However, sustaining this move will bea challenge, given the multiple claims on the small development budget and theconstraints of a coalition government.Transparent, rule-based system of local government fundingDecentralisation reforms required rule-based systems for local transfers via themechanism of Provincial Finance Commissions (PFCs). It took some time to set up andstrengthen the PFC. By second award, for FY 2003–04, the provincial allocable – aswell as federal general sales tax – were transferred to district and sub-districts, butnot to the lowest tier, i.e. the union council administrations (due to capacity issues,including absence of bank branches/accounts in their areas). However, for FY 2004–05, the village cluster unions started receiving funds. The 2005–06 – the fourth PFCAward – was more sophisticated and included performance-based grants andmaintenance funds for provincial assets that had been earlier transferred to local(district) governments. Over the last two years,Table 12.1 Net fiscal transfers to local governmentsPFC Award Year (PRs, Million)1st 2002-03 8840.0002nd 2003-04 8513.3773rd 2004-05 10593.4004th 2005-06 13071.0005th 2006-07 15707.343Source: Government of Balochistan (2006).



Page 219the local governments” share has gradually increased as a percentage of the totalreceipts of the Government of Balochistan (Government of Balochistan 2006).The fifth PFC award is a regular three-year award, thus indicating a medium-termview and providing the local governments with a predictable fiscal environment forlocal planning. The transfers to the district governments are one-line grants, therebyensuring flexibility and meeting another target of fiscal decentralisation under thedevolution law. Table 12.1 shows the increasing trend across the PFC awards.Performance grant rules – provincial efficiency incentives for service deliveryIn 2006, with external assistance through DSSP, a formula-based fiscal transfermechanism in the form of conditional grants – consistent with the provisions ofBalochistan Local Government Ordinance 2001 – was put into operation. Differenttypes of grants were provided for under these rules, all relating to improving servicedelivery or to achieving provincial policy targets articulated in the PRSP and/orcommitment to MDGs. The PGB disbursed performance grants to all but four of the 29program districts by the end of fiscal 2006 (June 30th).20 Performance grants includefour categories of grants, namely, “Policy Support Grant”, “Capacity Building Grant”,“Municipal Infrastructure Grant” and “Performance Incentive Grant”.21 However, theallocations are to be utilized for health, education, and water supply and sanitationsectors.Furthermore, all the allocations, including the overall budget figures, were placed forthe first time on the official website. Even though this openness was a departure fromestablished practice, the facts of budget documents being technical and complex,limited internet access, and poor English language skills continue to mar publicoversight.Institutional change to convert resources into results22Rationalizing managerial power: PGB has initiated reforms to re-orient serviceproviders to introduce improved service standards, and is working on draft proposalsto improve the incentive system for managers and service delivery staff.Citizens’ voice: PGB under DSSP has committed to improve the relationship betweencitizens and the provincial and district policy makers, and to ensure that citizenfeedback systems are in place. Complaint cells have been created by the education,health, and water and sanitation departments in more than half of the 29 districtgovernments. The Balochistan Education Department formed and trained parent-teacher committees in nine districts. Similarly, guidelines for hospital managementboards for district hospitals have been prepared.A public grievance redress mechanism has been notified for the whole of theGovernment of Balochistan (even though BRMP conditions required this for only asmall number of departments). Although the number of citizen complaints is low asyet (attributable to both an absence of public dissemination and a cultural reluctanceto lodge complaints in what is still a highly tribal culture), a mechanism is now inplace and is likely to yield results. Another notable shift is the adoption



Page 220of the citizen report card as an instrument of citizen feedback on services. The firstcitizen report card was developed for the provincial capital, Quetta, and this reportcard has also been placed on web.Institutional and bureaucratic qualityPublic sector reform requires improved bureaucratic quality as key to the process ofinstitutional change (Keefman 2004). BRMP design encompassed reforming theBalochistan Public Service Commission (BPSC)23 by making it more independent,autonomous, and broad-based, with wider powers to recruit public servants on merit.While the PGB made the BPSC relatively more autonomous (by detaching it from theprovincial secretariat and placing it under the Governor of Balochistan and improvingits membership), it has, to date, not widened its ambit to recruit all public servantsincluding the contractual staff. Lower-level jobs and contractual positions are centralto the politics of patronage. It is all but certain that such changes will not be effectedin the foreseeable future, given the political factors identified earlier.Another crucial reform area is the establishment of a “district cadre” comprising agroup of employees who are recruited at the local level and continue to serve within aparticular district (which is most likely to be their own). However, this is an issue withnational ramifications, as it is linked with the same reform in the other three provinceswhich have also adopted the common Local Government Ordinance. Across Pakistan,this reform has been locked in centre-province tussles over devolution. We noted thedifficulties that the entire reform process has suffered due to the regaining ofprovincial ground by provincial patrimonial elites. Establishment of a district cadrewould translate into provincial elite giving away their hold over civil servicerecruitment, promotions and transfers – all sources of patronage and rent. Futureprogress in this area remains uncertain at the national level. It requires a fundamentalchange in the federal to province, and province to local, governance relationships androles regarding the transfers and postings of civil servants.Financial management and procurement reforms are also under way under BRMP. Theprovincial cabinet has approved a draft bill for the creation of a procurementregulatory authority as well as its ancillary rules. Some effort has been made toenhance local resources for improved asset management. Furthermore, theinfrastructure services departments – the Communication and Works and theIrrigation and Power – have developed proposals to revise yardsticks for operationand maintenance expenditures. In Balochistan, the federal audit and accountsoutreach is extremely limited. The capacity of existing account officers in the districtshas been enhanced, but the task is an onerous one and requires major commitmentsand a strategy by the federal government.Fundamental to any improvement in the quality of institutional systems and processes,including those pertaining to human resources (HR), is the establishment of qualitydatabases. Perhaps the most important among these is a “civil servants” census”. Asis widely known, most developing country governments are beset



Page 221with a common mechanism, used by rent-seeking sections, to drain the alreadyenfeebled system: fictitious employees and pensioners. It is very clear that agovernment that is unaware of the total number of its current and still-living formermembers can hardly implement any improvement in HR or fiduciary instruments. TheBRMP has taken an initiative to enumerate all civil servants in Balochistan, along withrequired personal details.Another area where an updated database was essential to proper management was alinkage of figures of total outlays of PSDP schemes with their chronological phasing, soas to gain a grip on the “throw-forward” phenomenon. Development of such adatabase is now under way. Indeed, even before such a database is in place, thethrow forward has been reduced by way of an across-the-board cessation of newschemes, along with a notification eliminating the co-occupation of schemes indevolved subjects by the provincial and local governments.Working towards the improvement of public service institutions and systems is atransaction-intensive process, as recent evidence shows (Pritchett and Woolcock2004). Whereas a plethora of regulations and rules have been notified and deliberatedupon under the current reform process, the future course of translating suchinstruments into service delivery outcomes is neither certain nor formulaic. Our effortin this section has been to highlight the key reform processes as opposed to “results”and impacts. In any case, search for results would be premature at this stage. Themixed progress on fiscal and institutional reforms is no ordinary achievement giventhe difficulties of Balochistan’s context, which have been compounded by law andorder problems.24 Institutional change is a long-term process and is nurtured bydomestic political imperatives and action.Limits of policy-based lendingPolicy-based lending (PBL) is often employed as an incentive for reform. Theelaborate policy matrices prepared jointly by ADB and the federal and provincialgovernments reflect the best technical solutions to the issues of (mis)governance,public service delivery and overall institutional reforms. However, ADB’s evaluationshave also shown that program loans have made limited contribution to developingcountries’ policy reform to date (ADB 2002a). Their effectiveness is marred due to thenature of the program loan instrument, and because these attempts deal with complexand sensitive political issues. Another study shows that program design andimplementation need to address political economy factors (Abonyi 2002). Sustainingpolicy reform has always been and will continue to be a challenge.It has also been noted that a particular feature of reform programs advanced throughpolicy conditions is the “executive shortcut”: donors designing “reforms” in closeconcert with the executive agencies without formal parliamentary scrutiny andinformal political support. This is done in full knowledge of the possibleimplementation hurdles – and even partial successes – that are likely, but with a viewto avoid a catch-22 situation in which the preparatory negotiations for the reformprograms are likely to be painfully drawn out (and may even lead to



Page 222many “pre-natal project mortalities”). Knowing full well, as development bankers (andtheir implementation partners in client governments) do, a technocratic bias(Girishankar 2001) is recognized to exist when sufficient incentives are not created forthe implementing governments to undertake difficult reforms (Bolt and Fujimura2002). It is also appreciated that pre-program negotiations need to somehow includethe political elite (and even rent-seeking groups). Unfortunately, such incentives canin no way compensate the elite for what they otherwise see as benefits of theircapture.Experience with several PBL programs also showed the reluctance of policyimplementers to sustaining reform: firstly, the entrenched political elite see to it thatreform agents are either removed from positions of influence or that their influence iscurtailed through other road blocks. Secondly, rules for some degree of permanenceof tenure for civil servants are not in place; an undue frequency in transfers/postings(which remains a prerogative of governments), and/or absence of required merit andskill sets in the selected incumbents are enough to undermine even the best designedPBL – hence the call to focus on issues of power, politics and democracy, or workingaround the political economy issues of reform contexts (Ahmad and Porter 2006).Which way now?We noted how the informal governance systems result in “partial reform syndrome”.We have also recounted policy reversals while reviewing the reforms supported byADB-assisted programs. Halfway moves to eliminate rents via subsidies, restructuretaxes and undertake second generation, transaction-intensive reforms for improvedpublic services illustrate the inherent tensions between well-designed, technicallysound policy frameworks and the “messy” business of politics. There are reformchampions within the provincial and local governments who aid the process ofinstitutional change, but they face the mutually reinforcing syndromes of informalgovernance systems and the “larger” – extra-provincial – agenda of acceleratedmodernization, centre-local struggles and embedded insecurity and violence.Balochistan province, with its complexities, is not a static subject of inquiry. Politicaldevelopments in the recent years illustrate the influence of growing global energymarkets on the one hand, and the tussle between the central state’s securityimperatives and local pockets of resistance in Baloch tribes on the other. These localstruggles involve the populist rhetoric of ethnic identity, indigenous cultures (includingthe tribal system) and an “intrusive” central state apparatus. The hitherto unknownBalochistan Liberation Army (BLA) has also made news even though its antecedentsare not clear.The purpose of this paper is not to get into the quagmire of a political polemic, but toillustrate the local context within which the reforms are taking place. Thus, thepolitical economy setting is fluid and transitional. What is clear is that with increasedfocus on the region as a source of natural resources and the energy corridor fromCentral Asia to Gawadar port, the pace of economic transformation



Page 223will accelerate. While we refrain from making surmises or sketching futuristicscenarios, suffice it to mention that informal systems of tribal governance and localjustice are severely under pressure. The policy setting in the province will arguablyrespond to external factors, including the national and regional imperatives. Thefuture of devolution reform and the commitment of political elites, especially, at theprovince level, will be key to sustaining current reform trends that, by all accounts,are not too discouraging.ConclusionsThe international development agenda, as adopted and articulated by country PRSPs,has placed governance and institutional reforms at the core of its policy process.Pakistan’s high-level reform drive, albeit in a military regime context, has undertakenlocal governance restructuring. However, the country’s federal context, with historiesof provincial identities and territories of power and clientelism, has interacted with thereform process, particularly where transfer of powers and resources to the localgovernments have been viewed as a zero-sum game. Balochistan province has notbeen an exception to this trend. The context of Balochistan is far more complex,given the historical struggles of centre-province politics, the advance of globalizationand modernity, and the entrenched traditional, informal systems of governance.In assessing the institutional reform processes, we have narrated a mixed story ofprogress. Early as it is to analyze the impacts, positive signals have emerged alongwith policy reversals. The issue of unsustainable subsidies has been brought to thefore at both provincial and federal levels. The provincial development program hasseen an increase, and there is somewhat reduced emphasis on new projects that arepatronage instruments in Pakistan’s governance landscape. Importantly, the provincehas successfully introduced rule-based systems of local government funding andreformed administrative frameworks for effective public service delivery.In addition to the larger issues of regional and global geo-politics defined bycompetition over energy resources, these reforms continue to be impacted by thepolitical milieu and recent resurgence of violence. Further, such reforms are driven byexternally assisted program loans, designs of which are not only technically sound butalso demonstrate provincial executive’s ownership. Unfortunately, what theseprograms overlook are entrenched patrimonies and patron-client systems that areendemic.The debate between the technical and the political dimensions of developmentinterventions is well documented. Increasing evidence points out that for effectivesocial and economic development, politics cannot be bypassed (Pycroft 2006). Whiledevelopment agencies attempt to provide politically neutral assistance even when suchassistance aims to restructure government systems, building capacities withoutaddressing the political economy context can sustain patron-client relations and elitepriorities. For instance, the Drivers of Change (DoC) initiatives introduced by theDepartment for International Development (DFID) at least address politics



Page 224of power and interest. However, these initiatives have not yielded the desired results.The International Finance Institutions being bound by charters and the bi-lateraldonors are, often, smaller players in the larger game of national politics. For example,the relationship between land ownership and poverty is a known reality, and yetexternally assisted programs do not take this into account. The government’s owndata (Government of Pakistan 2002) indicate that almost one half of rural householdsown no land, while 2 percent possess 44 percent of the land, and large farmerscontrol 66 percent of all land. Donors admittedly cannot address agendas that are notarticulated locally, and this is the problem with tailoring development assistance topolitical economy factors (Government of Pakistan 2003b).Even if political economy factors are addressed and investigated by donors, thechallenge for external actors is how such findings are to be mainstreamed into theprograms and projects aiming at society’s transformation. There is, however, a lack ofdevelopment tools and instruments that can promote comprehensive reform ofgovernance institutions with external assistance, without triggering concerns in somequarters over the possibility of undermining sovereignty and government legitimacy.Thus, development agenda in the first place needs to reckon with issues of power,elite interest and informal systems. Converting such analysis into action might meandevelopment of aid instruments that facilitate “transformational change” throughinstitutional reform, and that recognize that development is a political process.The search for politically sensitive aid instruments is a project fraught with uncertaintyto begin with. Even if such tools were identified, the mechanisms cannot replace orbe a substitute for domestic political processes. Political mobilization and action is asine qua non for social and economic change (Hickey and Mohan 2004). Whilstexternal agencies may not be able to intervene in local political issues, there isenough evidence to show that perpetual poverty is a direct result of inequitableaccess to public assets such as water, land, livelihoods and other public goods. Weargue for an approach that addresses the fundamental role of local and national statein securing the poor’s entitlements. Institutional reforms located in this frameworkwould have different implications and design features. More importantly, such anapproach will focus on issues of social regulation at subnational, local levels, andrecognize the links between legislatures, voters and policy imperatives beyond thetechnocratic frontiers of executive agents in national governments and aid agencies.This will be difficult and messy but will generate sufficient spaces for political action inthe medium term. This would be applicable to our area of study – the conflict-ladenprovince of Balochistan – as well.Notes1 The author is employed at the Asian Development Bank (ADB). The viewsexpressed in this chapter are the views of the author and do not necessarily reflectthe views or policies of ADB, or its board of governors, or the governments theyrepresent. ADB does not guarantee the accuracy of the data included in this chapterand accepts no responsibility for any consequence of their use. The countries listed inthis chapter do



Page 225not imply any view on ADB’s part as to sovereignty or independent status ornecessarily conform to ADB’s terminology.2 The seven points were: rebuilding national confidence and morale; strengtheningthe Federation, removing inter-provincial disharmony and restoring national cohesion;reviving the economy and restoring investor confidence; ensuring law and order anddispensing speedy justice; depoliticizing state institutions; devolving power to thegrassroots level; and ensuring swift and across-the-board accountability.3 Separate ordinances were issued for each one of Pakistan’s four provinces.4 The term ‘social gap’ describes the gap between Balochistan’s social developmentindicators relative to other provinces in Pakistan.5 For details, see http://www.decentralization.org.pk.6 Balochistan produces more than 36 percent of Pakistan’s natural gas and almost allof its coal being presently mined.7 Along with oil, gas and coal resources, Balochistan also has significant gold, copper,silver, platinum, aluminum and uranium reserves. For a detailed report onBalochistan’s mineral deposits see http://www.balochistan.sdnpk.org/mineral1.htm8 The Constitution of 1973 was by and large a consensus federal arrangement. Itprovided for considerable provincial autonomy, but the control over natural resourceswas retained by the central government. Provinces were to get royalties in exchangefor natural resource exploitation.9 Of these loans, PRs 10.1 billion carries interest of more than 14.5 percent yearly,PRs 6.3 billion bears interest ranging from 15.1 percent to 17 percent, and PRs 0.76billion bears interest above 17 percent. The servicing cost of all loans in FY2004 wasPRs 2.6 billion (of which PRs 2.3 billion was for CDLs), absorbing 10.7 percent of totalrevenue expenditure.10 The provincial allocable in Balochistan is shared among the local governments bythe PFC according to the governments’ historical expenditure patterns. Such anallocation mechanism is possible because the grants primarily support the salaries oflocal government staff – whose number does not substantially change from one yearto the next – and charged expenditures, such as utility payments.11 The post-colonial state continued with the age-old policy of keeping strategic yetinformal contracts with the key tribes through the tribal chiefs, and repressing anydissent or voices against the hegemonic central state.12 The police-levies divide was mediated through a non-uniform system of justicebased on a procedural exception to the criminal procedure code in the “B”, or non-police, areas. The multiplicity of judicial systems devalues the quality of justice, addsdifficulties in seeking redress, and aggravates problems of access.13 Quoted in Brinkerhoff and Goldsmith 2002.14 Though these are largely unexplored, the resource base range is estimated to beas high as 535 trillion cubic feet. In addition, the country possesses 1.4 billion barrelsof proven oil reserves: http://www.american.edu/projects/mandala/TED/turkmen.htm15 The BDSSP’s four principal policy outcomes are to be achieved through 20 firsttranche conditions and 21 second tranche conditions. All first tranche conditions werecomplied with as required.16 Though the increase in tax revenue is only eight percent, and the bulk of theincrease (107 percent) comes from the non-tax receipts, which also include certainone-off sources, but the trend is positive. Source: Government of Balochistan 2006Budget Documents.17 At PRs 2 billion, which is a substantial amount when seen in the light of the factthat the total revenues of the province from its internal sources are estimated to bejust slightly higher, at PRs 2.4 billion (and a total budget of about PRs 35 billion).18 The amount of money outstanding against unfinished schemes from earlier budgetyears.19 Usually comprising funds promised by the head of government, the Chief Ministerof Balochistan, during his visit to constituencies of his political allies.



Page 22620 See: www.brmp.PGB.pk ; also ADB 2006.21 See Schedule I of these Rules.22 This section draws on the progress report of the B-DSSP. See Balochistan DSSPwebsite for details: http://www.balochistan.gov.pk/New%20Folder/BDSSP/BDSSP-Ist%20Tranche.htm.23 A statutory body tasked with all government recruitments in grades 16 to 22 in a1–22 system.24 The number of explosions in the province during 2005 and the first eight monthsof 2006 have been 619 and 881 respectively. Additionally in 2006, in 95 otherincidents the number of rockets fired were too numerous to be enumerated and arelisted as “hundreds”. In these incidents, at least 198 persons have been killed and 484injured. (Source: various media reports available in tabulated form).ReferencesAbonyi, G. 2002, ‘Toward a Political Economy Approach to Policy-based Lending.’Economics Research Department, ADB, Working Paper Series No.14, Manila: AsianDevelopment Bank.Ahmad R., and Porter D.J. 2006, ‘Justice Sector Reform and Policy Conditionality –Symbiosis or Mutual Denial’ in ‘Searching for Success –Narrative Accounts of LegalReform in Developing and Transitional Countries’; International Law and DevelopmentOrganization, Rome.Alesina, Alberto, Sule Ozler, Nouriel Roubini, and Phillip Swagel 1996, ‘PoliticalInstability and Economic Growth.’ Journal of Economic Growth, 1(2): 189–211.Alesina, Alberto, William Easterly, and Reza Baqir 1999, ‘Public Goods and EthnicDivisions.’ Quarterly Journal of Economics, 114: 1243–84.Asian Development Bank 2002a, ‘Evaluation of Social Action Program: OperationsEvaluation Department.’ Manila: Asian Development Bank.——2002b, ‘Pakistan: Country Strategy and Program.’ Manila: Asian DevelopmentBank.——2004a, ‘Devolution in Pakistan’, Manila: Asian Development Bank. Available at:http://www.adb.org/Documents/Studies/Devolution-in-Pakistan.——2004b, ‘Report and Recommendation of the President to the Board of Directors onProposed Loans and Technical Assistance Grants to the Islamic Republic of Pakistanfor the Balochistan Resource Management Program.’ Manila: Asian Development Bank.——2004c, ‘Decentralization Support Program: Progress Report.’ Islamabad: DSP.——2005a, ‘Sector Assistance Program Evaluation for the Social Sectors in Pakistan.’Manila: Asian Development Bank. Available at:http://adb.org/Documents/Reports/SAPE/PAK/sap-pak-2005-08/default.asp——2005b, “Improving Devolved Social Service Delivery in NWFP and Punjab.’ Manila:Asian Development Bank.Auditor General of Pakistan 1998–2001, Social Action Programme: Third PartyValidation. Government of Pakistan, Islamabad.Banfield Edward C, and Banfield, Laura Fasano 1959, The Moral Basis of a BackwardSociety. New York:Free Press.Bolt, R. and Fujimura, M. 2002, ‘Policy Based Lending and Poverty Reduction: AnOverview of Processes, Assessment and Options.’ Economic Research Division (ERD)Working Paper Series No. 2, Manila: Asian Development Bank. Bourguignon, Francoisand Thierry Verdier 2000, ‘Oligarchy, Democracy, Inequality, and Growth.’ Journal ofDevelopment Economics, 62(2): 285–313.Brinkerhoff D.W. and Crosby B.L. 2002, Managing Policy Reform: Concepts and Toolsfor Decision Makers in Developing and Transitional Countries. Bloomfield CT: KumarianPress.Brinkerhoff, D. and A. Goldsmith 2002, Clientelism, Patrimonialism and DemocraticGovernance: An Overview and Framework for Assessment and Programming.Bethesda



Page 227MD: ABT Associates. Available at: http://pdf.dec.org/pdf_docs/PNACR426.pdf.Hasan, K. Daily Times, Pakistan 2006, ‘No Civil War in Balochistan, Governor tellsVOA.’ Sunday, January 8, 2006. Available at: http://www.dailytimes.com.pk.David Martin Jones 1988, ‘Democratization, Civil Society, and Illiberal Middle ClassCulture in Pacific Asia.’ Comparative Politics, vol. 30, no. 2: 147–69.Easterly, William 2003, ‘The Political Economy of Growth Without Development: ACase Study of Pakistan,’ in Danni Rodrik, ed. In Search of Prosperity: AnalyticalNarratives on Economic Growth. Princeton N.J: Princeton University Press. Available athttp://www.nyu.edu/fas/institute/dri/Easterly/File/Pakistan.pdfEasterly, William and Ross Levine 1997, ‘Africa’s Growth Tragedy: Policies and EthnicDivisions.’ Quarterly Journal of Economics, 112(4): 1203–50.Girishankar, N. 2001, ‘Evaluating Public Sector Reform: Guidelines for AssessingCountry-level Impact of Structural Reform and Capacity Building in the Public Sector.’Washington DC: World Bank OED.Government of Balochistan 2002, Development Statistics of Balochistan 2001–02,Bureau of Statistics. Quetta, Pakistan.Government of Balochistan 2003, ‘Balochistan Poverty Reduction Strategy Paper,’November, Quetta, Pakistan. Available at:http://hdr.undp.org/docs/reports/national/PAK_Pakistan/Pakistan_2003_en.pdfGovernment of Balochistan 2005, ‘Planning and Development Department presentationto the President of Pakistan’, Quetta, Pakistan.Government of Pakistan 2002, ‘Household Integrated Survey and Pakistan IntegratedHousehold Survey.’——2003a, ‘Accelerating Economic Growth and Reducing Poverty: The Road Ahead.’Pakistan Poverty Reduction Strategy Paper. Islamabad.——2003b, ‘Between Hope and Despair: Pakistan Participatory Poverty Assessment.’Islamabad: Planning Commission.Haq, Mahbub-ul-Haq 1998, Human Development in South Asia. Karachi: OxfordUniversity Press.Hickey, S. and G. Mohan (eds) 2004, Participation: From Tyranny to Transformation?Exploring New Approaches to Participation, London: Zed Books.Husain, Ishrat 1999, Pakistan: The Economy of an Elitist State. Karachi: OxfordUniversity Press.Kahn, S. R. 1999, Reforming Pakistan’s Political Economy. Lahore: Vanguard Books.Keefer, Philip 2002, ‘Clientelism, Credibility, and Democracy.’ mimeo, DevelopmentResearch Group: The World Bank. Washington DC.Keefer, Philip 2004, ‘What Does Political Economy Tell Us About EconomicDevelopment – And Vice Versa?’ World Bank Policy Research Working Paper No.3250, Washington DC: The World Bank.Keefer, Philip and Stuti Khemani 2003, ‘Democracy, Public Expenditures, and thePoor.’ World Bank Policy Research Working Paper No. 3164. Washington DC: TheWorld Bank.Midgel, Joel S. 1988, Strong Societies and Weak States: State-Society Relations andState Capabilities in Third World, Princeton, NJ: Princeton University Press.Pritchett, L. and Woolcock, M. 2004, ‘Solutions When the Solution is the Problem:Arraying the Disarray in Development.’ World Development, 32(2): 191–212.Pycroft, Christopher 2006, ‘Addressing the Political Dimensions of Development.’ inADB Governance Brief. ADB: Manila.Rabushka, Alvin and Kenneth Shepsle 1972, Politics in Plural Societies. Columbus:Merrill.Social Policy Development Centre 2001, Annual Report 2001. Karachi: SPDC.Titus, Paul (ed) 2002, Marginality and Modernity: Ethnicity and Change in Post-Colonial Balochistan. Karachi: Oxford University Press.Titus, Paul and Nina Swidler 2002, ‘Knights, Not Pawns: Ethno-nationalism andRegional



Page 228Dynamics in Post-colonial Balochistan.’ International Journal of Middle East Studies,32: 47–69.Titus, Paul 1998, ‘Honor the Baloch, Buy the Pushtun: Stereotypes, SocialOrganization and History,’ in Western Pakistan in Modern Asian Studies, 32,Cambridge: Cambridge University Press.UNDP 2003. Pakistan National Human Development Report: Poverty, Growth andGovernance by Akmal Hussain et al.; UNDP, Islamabad.Wilder, Andrew 1999, The Pakistani Voter. Karachi: Oxford University Press.Zahid Hasnain 2005, ‘The Politics of Service Delivery in Pakistan: Political Parties andthe Incentives for Patronage, 1988–1999.’ PREM South Asia Working Paper Series,Washington DC: The World Bank.Zaidi, A. 2005a, ‘State, Military and Social Transition: Improbable Future of Democracyin Pakistan.’ Economic and Political Weekly, Special Articles, vol. 40, no. 49.Zaidi, A. 2005, ‘The Political Economy of Decentralisation in Pakistan: TransversalTheme Decentralisation and Social Movements.’ Development Study Group WorkingPaper No. 1, Zurich: Development Study Group. Available at: http://www.nccrnorth-south. unibe.chZiauddin, M. 2000, ‘Musharraf announces partyless local bodies polls,’ Dawn, 16August.



Page 229Index30 Baht Scheme, see Universal Coverage (Thailand)

Abdullah Badawi 19, 23–4, 36absenteeism, India 47, 54accounting practices, corporate 9, 74–8, 83–5, 87, 91–2, 94–5ADB (Asian Development Bank) 14, 67–8, 206–8, 215, 221adult literacy, India 52–3Afghanistan 214agencies:autonomy of 168–70, 172, 174–7;co-ordination of 181;composition of 175–6;control of 168–9, 176–7;district-level 178–9, 181, 183;financing of 177;integration with parent body 177;service delivery by 179;state-level 177–9, 181, 183agencification:definition 13, 168–70, 189;empirical evidence for 168–9, 171–2;expectations of 171;in Korea 189;literature on 169–71, 173;networks in 172;in Rajasthan 14, 173–4, 182–3;rationales for 174;supported by donors 175amoral familism 211Anand Panyarachun 122Anti-Corruption Agency (Malaysia) 32–3, 35anticorruption measures: 120–1, 130;Thailand 10, 121Asian financial crisis (1997) 74, 92, 117, 126, 128, 157authoritarian regime 125authority, distribution of 1autonomy:of agencies 168–70, 172, 174–7, 182;Filipino irrigation authorities 66;Japanese prefectures 108;of medical professionals 164;in Thai healthcare system 155

Balanced Scorecard (BSC) 194Balochistan: 14;background 208–9;colonial history 210;education in 212;ethnic factionalism 212–13;fiscal conditions 209–10;fiscal reforms 216–18;geopolitical situation 213–14;governance system 213;institutional change 219–20;mega-projects 214;political economy 210;political instability 211;quality databases 220–1;relations with federal government 209, 215, 217, 223;space for political action 224;tribal system 211Balochistan Public Service Commission (BPSC) 220

Bangkok 155, 159Bangladesh 79, 84BDSSP (Balochistan Devolved Social Services Program) 209, 215–16, 219Bharat Nirman 42bribes 77BRMP (Balochistan Resource Management Program) 209, 215–16, 218, 220–1Bumiputra (Malaysia) 26–7, 29, 31bureaucracy:representative 28–30, 40;unrepresentative 6, 28–30bureaupathologies 174

capacity 20–1, 23, 28, 38–9capital flows 3capture, state: 120, 127, 129–31, 212;see also corruption Central Statistical Organisation (India) 51centralization: 9, 14, 104, 108;degree of 104change management 9China 44, 154;accounting practices 92;accounting standards in 74Civil Servant Medical Benefit Scheme (Thailand) 156, 158, 160



Page 230civil society:India 49;and institutional arrangements 104;Pakistan 207;Thailand 116, 120–1, 125–30co-production, agenda 4coercive imitation 197–8, 200–2, 204community level of government 1Confucian tradition 139consensual regimes 104Constitutional Court (Thailand) 118, 122contracting, agenda 4corruption 20, 23;and accounting practices 74;administrative 119–20, 127, 129– 31;Balochistan 211–12;India 54, 182;and information asymmetry 76;Korea 11, 137, 145, 148;Malaysia 15, 19, 34–5, 37, 39;Pakistan 211;Phillipines 8;Thailand 10, 11, 116–18, 124, 127, 129–30;types of 118–19

debt, as means of supranational influence 126decategorization 69decentralisation 15;agenda 3;India 7, 43–4, 46–9, 52–4;Indonesia 48;Latin America 48;Phillippines 48, 64–6Democratic Party (Thailand) 158democratization 2–4;of civil service 12;and mass education 212;raising expectations of government 2deregulation, agenda 4developmental state 47DMHS (Directorate of Medical and Health Services, India) 173, 177–8drivers of change 6DSSP, see BDSSP

econometric models 87economic freedom 4Education Guarantee Scheme (India) 52efficiency 19, 20E-Japan strategy 101, 109–10Election Commission (Thailand) 10entrepreneurship 100environmental pressure 21ethnic conflict 15, 22executive dominance 6expectations of public services 171

faith in government 171financial disclosure 78–9, 83fiscal crisis, permanent, 2fiscal deficits 48Fiscal Responsibility and Budget Management Act (India) 45Five Year Plans (India) 43, 54foreign aid 8, 58

free market:principles 3;reforms 4fungibility of aid 8, 59–61, 68, 70–1

games 6, 8, 11, 60good public administration, definition 19governance:definition of 5, 16;corporate 9government failure, see state failure

Hara, Professor 98–103, 105, 108–9, 111, 113HelloBaby system (Japan) 97–9, 101–2, 105, 109–10hierarchical organisation 190–1, 193House of Representatives (Thailand) 122

incentives, perverse 8, 58–9, 61India: 10, 15;access to schools 51–2;bureaucracy 46–7, 54;constitutional changes 8, 50, 53–4;crisis of governance 42;decentralisation of education 50–1;federal system 8, 50, 54–5;fiscal deficit 45;Naxalite insurgency 7, 42–3;public health system 10, 13, 44–5, 54, 154, 169;ratings by international bodies 42Indian local government 7Indonesia 79, 83information 8, 60;asymmetries 9, 75, 77innovation 9;breakthrough and incremental models 105–6, 110;environment of 102;literature on 97, 103, 112;loci of 103;myths of 99–102;and personal networks 112;and risk-taking 113;and team-based management 197Integrity Institute of Malaysia 35International Accounting Standards 9, 74, 85, 89, 91–2, 94–5International Monetary Fund 126Iran 214Italy 43

Jamiat-i-Ulama-i-Islam (Pakistan) 213Japan 43;innovation in 97–8;Ministry of Agriculture 101;Ministry of Health 101;prefectures 108;rating by international bodies 146;social structure 102Japan Bank for International Cooperation (JBIC) 67–8



Page 231judiciary:Korea 140–1, 144–5, 151;reputation of female judges 140;role in good governance 136

Kagawa University Hospital 98, 99, 108Kampung Medan (Malaysia) 29Karnataka 43Kazakhstan 79Keiretsu 111Kerala (India) 43K-Mind (Japan) 99K-Mix system (Japan) 97–9, 109Koizumi administration (Japan) 101, 108Korea 10, 74;accounting standards in 74–5;civil service reforms 14;electoral democracy 12;electoral law 140–1, 144–5, 149;electoral law 140–1, 144–5, 149;financial crisis in 83;indictment of National Assemblypersons 141, 144, 148–9;judiciary 10, 11, 135, 139;pardons 149–51;penalties for corruption 145–6, 148–9;public sector 135;rating by international bodies 146;rule of law 136, 151;as socially homogenous society 43;team-based management 14, 190–8, 200–4Korean Trade Organization 200

leadership 196, 199;as driver for team-based management 200;as driver of team-based management 196, 198–9, 202, 204;in innovation 111;in Japanese government 108Liberal Democratic Party (Japan) 108licence-raj (India) 42local government:Balochistan 208, 210, 215–16, 219, 221–3;and foreign aid 71;India 7, 8, 44, 46–7, 49, 50, 52–3;and institutional arrangements 104;Korea 190–2;Latin America 48;Pakistan 208;Thailand 155Lok Sampark Abhiyaan 51Low Income Card (Thailand) 157–8, 162

Madhya Pradesh (India) 7, 50–3Madison, James 21, 30, 37Mahathir Mohamad 19, 27, 29maintenance of effort (MOE) requirement 70–1majoritarian regime 104, 108Malaysia 83;elections 7, 31–2, 35, 37;ethnic disparities in wealth 25, 30;as “flawed democracy”, 32, 37;government 7;police 28–9, 34;quality of graduates 24;

ratings by international bodies 23, 35;schools 28Malaysian civil service:balanced reform strategy 39;competition with private sector 24;executive control 32;Malay preference 6, 25–7, 38–9;performance 19, 23, 39;policy capacity 25;public control 7, 12, 31–3, 35–8;racial bias of bureaucrats 28–9;role in driving development 6;slack and indulgence 33–5;unbalanced reform strategy 6, 21–2, 38Malaysian Remuneration System 22Mazdoor Kisan Shakti Sangathan (MKSS) (India) 50media:Korea 151;power of 2;Thailand 116, 125–6, 130Memorandums of Understanding 168, 179, 181, 183merit pay 22meritocracy 4Millennium Development Goals (United Nations) 46ministers:and bureaucrats 47, 105;Korea 192–3, 196–200;Malaysia 35Ministry of Agriculture (Japan) 109;of Economy, Trade and Industry (Japan) 109;of Government Administration and Home Affairs (Korea) 191–2, 196–9, 201–3;of Health (Japan) 109–10;of Health and Family Welfare (India) 169, 178, 182;of Labor (Korea) 194;of Planning and Budget (Korea) 193;of Public Health (Thailand) 155–6, 159monopolies 22moral hazard 8, 59–61, 63, 68, 70–1motivation 8, 60multi-valued choices 39Musharraf, Pervez 206–7Muslim League (Pakistan) 213

NACO (National AIDS Control Organisation) 178, 182National AIDS Control Programme (India) 173National Committee on Human Rights (Thailand) 122National Counter Corruption Commission (Thailand) 10, 122National Economic Consultative Council (Malaysia) 32National Election Commission (Korea) 141National Election Commission (Thailand) 122



Page 232National Front (Malaysia) 31national governments:and agencification 168, 170;India 43–5, 49;and institutional arrangements 104;in institutional arrangements 104;Japan 108–9;Pakistan 15, 224;Philippines 65, 67–8;Phillippines 63;and reform agendas 103National Health Rural Mission (India) 45National Health Security Office (Thailand) 159, 161, 163–4national level of government 43–5, 68, 103–4National Literacy Mission (India) 53networks, in management 190New Economic Policy (Malaysia) 21, 26, 29, 32New Public Management: 22, 100, 103–4, 106, 111, 154, 168;reforms 13, 14, 22New Remuneration System, see merit pay NGOs (Non-Governmental Organizations) 2,36–7, 179, 182

Office of the Counter Corruption Commission (Thailand) 122Ombudsman, Thailand 10Operation Barga 50

Pakistan: 10, 14, 15;devolution 206–8, 220, 223;ethnic distinctions in 212;federal structure 210;federal system 207–8;international consensus about 214;political instability 211panchayat 7, 50, 52, 176panchayati raj institutions (PRIs) 43, 50partial reform syndrome 15, 213, 222participation 4Patron’s Dilemma 60pay gap between public and private sector, Malaysia 24PBL programs 222people sector 1–4performance-based management 193Philippines:bureaucracy 63;corporate accounting practices in 83Philippines irrigation:alternative modes of provision 67;evolution of 63;foreign aid for 8, 58, 61, 64;incremental improvement phase 65;loans for 62;operation and maintenance 66;relationship with donors 68;underinvestment problem 61–3, 67, 69Planning Commission (India) 43policy-based lending 209, 221policy design 13Polyani, Karl 5post-colonial state structure 47power 1PRIs 43, 50–1, 53–4private sector:Balochistan 216;

China 9;expansion of 3;India 46, 175, 179;innovation in 101–3;interaction with other sectors 1, 2, 4;Korea 192;Malaysia 22, 24–6, 29, 30;as model for public sector 4;quality of governance 9;team-based approach in 189;Thailand 123, 155, 163privatization, agenda 3PTD (Perkhidmatan Tadbir dan Diplomatic) 25public choice theory 100Public Complaints Bureau (Malaysia) 32public goods 48, 59, 60, 211–13public health system:India 10, 13, 44–5, 54, 154, 169;Japan 108–9;Rajasthan 168–9, 173, 183;Thailand 10, 12, 154–6, 159–60, 162–4, 166Public Management Reforms 105public school system, India 44, 54public sector:Asian countries 74, 77, 92, 94;extent of 2, 3;innovation in 103;innovation theories 102;Korea 135;Malaysia 24, 26, 30, 34;opacity 92, 95;risk-averseness 101;Thailand 124, 129–30public services, see social services punitive action 35–6, 47

Quetta 220

Rajasthan (India) 7, 13, 14, 50, 53, 168–70, 173–4Reagan, Ronald 2reform agendas:Balochistan 206, 213;executive shortcut 221;and innovation 112–13;Japan 97, 111;Malaysia 21, 23;and national governments 103;New Public Management 106reinvention, agenda 4Remote Visual Consultation System 99rent-seeking 9, 15, 95, 211, 221resources 1responsiveness 19, 20risk, and innovation 101RMRS (Rajasthan Medicare Relief Societies) 173, 175–6, 179Russia 214



Page 233

Samaritan’s Dilemma 59Sarva Shiksha Abhiyan 45Scheduled Tribes (India) 42;participation in formal education 52Senate (Thailand) 122separation of powers 104, 120Shikoku (Japan) 9, 98, 102, 108, 110Singapore 79, 146social audits 50social capital: 60, 107, 126, 172, 183;and innovation 108, 111–12Social Health Insurance (Thailand) 156–60social networks 9, 107–8, 111–13, 120social services 44, 49;state provision of 44, 46ST-Net (Japan) 99State Audit Commission (Thailand) 122state governments, India:and agencies 176, 182;capture 43;delivery of social services 43;education 51–3;fiscal deficit 45, 54;governance reform 44;health expenditure 169, 173, 176–7;relations with civil society 49;relations with federal government 8, 42, 55;relations with local government 49state failure 46STCS (State Tuberculosis Control Society) (India) 179supranational level of government 1, 3sustainability: 14;of team-based management 202

Takamatsu (Japan) 98–9task-specific path dependency (TSPD) model 171team-based management, Korea, see Korea, team-based managementteam leaders 195–6technocratic approach 11Thai Rak Thai party (TRT) 117, 158Thailand 10;accounting standards in 74;anti-corruption measures 125;anticorruption measures 10, 121, 124, 126–8, 131;constitution 10, 116, 122, 124, 130;as flawed democracy 125, 127;Good Governance regulations 122–4;public health system 10, 12, 154–6, 159–60, 162–4, 166;public ownership of hospitals 155–6, 163;public service 124;rating by international bodies 161;relationship between government and business 128transformation, definition of 5Transparency International 23, 35, 144, 152transparency, public sector 78–9;and accounting practices 75–6, 78, 92, 95;in Asia 79;and democratization 4;incentives for financial disclosure 77, 94;Korea 135, 140;measures of 89;and predictability 76, 91;

and reform process 9;Thailand 79, 117, 120, 122, 124, 126trust in government 2

uncertainty, see predictability and unpredictabilityunions 198, 203United Malays National Organisation (UMNO) 31–2, 35Universal Coverage (Thailand healthcare) 12, 154–6, 158–64Uthai Pimchaichon 122

vicarious learning 201–2, 204vicious cycle 63, 67virtuous cycle 45voice:as BDSSP outcome 216, 219;civil and political rights 137;and civil society 49;and decentralisation 54;at local level 46, 52;for poor 48;and state governments 50–1;as World Bank indicator 42, 136Voluntary Health Card (Thailand) 157, 162

Washington consensus: 2;see also free market principles West Bengal 50whistleblowers: 11;rewards for (Korea) 12, 145will to perform: 20–1;Malaysian civil service 22, 30, 33, 38–9;and unrepresentative bureaucracy 28World Bank 2, 67, 78, 126, 136, 175World Business Environment Survey 78–9

Yes, Minister 105


< previous page page_234

Page 234This page intentionally left blank.

< previous page page_234