the world bank’s poverty reduction strategy paper approach...

G-24 Discussion Paper Series

UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT

UNITED NATIONS

CENTER FORINTERNATIONALDEVELOPMENTHARVARD UNIVERSITY

The World Bank’s Poverty Reduction StrategyPaper Approach: Good Marketing

or Good Policy?

Jim Levinsohn

No. 21, April 2003

G-24 Discussion Paper Series

Research papers for the Intergovernmental Group of Twenty-Fouron International Monetary Affairs

UNITED NATIONSNew York and Geneva, April 2003

CENTER FOR INTERNATIONAL DEVELOPMENTHARVARD UNIVERSITY

UNITED NATIONS CONFERENCE ONTRADE AND DEVELOPMENT

Note

Symbols of United Nations documents are composed of capitalletters combined with figures. Mention of such a symbol indicates areference to a United Nations document.

*

* *

The views expressed in this Series are those of the authors anddo not necessarily reflect the views of the UNCTAD secretariat. Thedesignations employed and the presentation of the material do notimply the expression of any opinion whatsoever on the part of theSecretariat of the United Nations concerning the legal status of anycountry, territory, city or area, or of its authorities, or concerning thedelimitation of its frontiers or boundaries.

*

* *

Material in this publication may be freely quoted; acknowl-edgement, however, is requested (including reference to the documentnumber). It would be appreciated if a copy of the publicationcontaining the quotation were sent to the Publications Assistant,Macroeconomic and Development Policies Branch, Division onGlobalization and Development Strategies, UNCTAD, Palais desNations, CH-1211 Geneva 10.

UNCTAD/GDS/MDPB/G24/2003/2

UNITED NATIONS PUBLICATION

Copyright © United Nations, 2003All rights reserved

iiiThe World Bank’s Poverty Reduction Strategy Paper Approach: Good Marketing or Good Policy?

PREFACE

The G-24 Discussion Paper Series is a collection of research papers preparedunder the UNCTAD Project of Technical Support to the Intergovernmental Group ofTwenty-Four on International Monetary Affairs (G-24). The G-24 was established in1971 with a view to increasing the analytical capacity and the negotiating strength ofthe developing countries in discussions and negotiations in the international financialinstitutions. The G-24 is the only formal developing-country grouping within the IMFand the World Bank. Its meetings are open to all developing countries.

The G-24 Project, which is administered by UNCTAD’s Macroeconomic andDevelopment Policies Branch, aims at enhancing the understanding of policy makers indeveloping countries of the complex issues in the international monetary and financialsystem, and at raising awareness outside developing countries of the need to introducea development dimension into the discussion of international financial and institutionalreform.

The research carried out under the project is coordinated by Professor Dani Rodrik,John F. Kennedy School of Government, Harvard University. The research papers arediscussed among experts and policy makers at the meetings of the G-24 TechnicalGroup, and provide inputs to the meetings of the G-24 Ministers and Deputies in theirpreparations for negotiations and discussions in the framework of the IMF’s InternationalMonetary and Financial Committee (formerly Interim Committee) and the Joint IMF/IBRD Development Committee, as well as in other forums. Previously, the researchpapers for the G-24 were published by UNCTAD in the collection International Monetaryand Financial Issues for the 1990s. Between 1992 and 1999 more than 80 papers werepublished in 11 volumes of this collection, covering a wide range of monetary andfinancial issues of major interest to developing countries. Since the beginning of 2000the studies are published jointly by UNCTAD and the Center for InternationalDevelopment at Harvard University in the G-24 Discussion Paper Series.

The Project of Technical Support to the G-24 receives generous financial supportfrom the International Development Research Centre of Canada and the Government ofDenmark, as well as contributions from the countries participating in the meetings ofthe G-24.

THE WORLD BANK’S POVERTY REDUCTIONSTRATEGY PAPER APPROACH: GOOD

MARKETING OR GOOD POLICY?

Jim Levinsohn

Gerald R. Ford School of Public Policy,Department of Economics,

andWilliam Davidson Institute,

University of Michigan

G-24 Discussion Paper No. 21

April 2003

viiThe World Bank’s Poverty Reduction Strategy Paper Approach: Good Marketing or Good Policy?

Abstract

In 1999, the World Bank and the International Monetary Fund adopted anew set of processes to guide lending to some of the world’s poorest countries.This set of processes is known as the Poverty Reduction Strategy Paper (PRSP)approach.

This study reviews the PRSP approach. The study begins with a primer ofjust what the PRSP approach is. In what ways does it represent a change inpractices and in what ways is it a codification of business-as-usual? The paperthen reviews the recent “mid-term” evaluations of the PRSP approach conductedboth internally by the Bank and Fund as well as by external organizations. It isargued that neither the internal nor external reviews are asking the really hardquestions. To really evaluate the PRSP approach, it is necessary to compareoutcomes to what would have happened but for the PRSP’s implementation. Thatmeans evaluating the marginal impact of the approach. Knowing whether thePRSP process is really addressing the concerns of the poor means being able toidentify the poor, measure changes in their well-being, and then analyse whetherthese changes are in fact due to changes in policy resulting from the PRSPapproach.

ixThe World Bank’s Poverty Reduction Strategy Paper Approach: Good Marketing or Good Policy?

Table of contents

Preface ............................................................................................................................................ iii

Abstract ........................................................................................................................................... vii

I. Introduction .............................................................................................................................. 1

II. The PRSP process .................................................................................................................... 2

A. How the process is supposed to work .................................................................................. 2

B. What should be in a PRSP? ................................................................................................. 4

III. Evaluation from inside ............................................................................................................ 6

IV. Evaluation from outside .......................................................................................................... 8

V. Yet another review ................................................................................................................... 9

Notes ........................................................................................................................................... 13

References ........................................................................................................................................... 14

1The World Bank’s Poverty Reduction Strategy Paper Approach: Good Marketing or Good Policy?

I. Introduction

In 1999, the World Bank and the InternationalMonetary Fund (IMF) adopted a new set of proc-esses to guide lending to some of the world’spoorest countries. Amid the blizzard of acronymsexplaining the new process,1 the Bank and theFund laid out a process that very poor countrieswould need to follow if they wished to make useof various concessionary lending facilities. Abouttwo and a half years later, in the spring of 2002,the Bank and the Fund concluded a review of thisnew process. Contributors to this review includeddozens of non-governmental organizations (NGOs)as well as the Bank and the Fund themselves. TheBank and the Fund, while acknowledging thatthe process could be improved, concluded that itworked pretty well based on the preliminary evi-dence so far available. The NGOs were, on thewhole, less enthusiastic.

Neither the Bank nor the outside commenta-tors are asking the hard questions. The rightquestion to ask is the following: Relative to what

would have happened absent the adoption of thePoverty Reduction Strategy Paper (PRSP) proc-ess, has the implementation of the PRSP processyielded benefits that exceed its considerable ad-ministrative costs?

Instead of tackling this difficult question, theBank and the Fund have basically evaluated thePRSP process noting that some potentially repli-cable good practices have emerged and that theinclusivity that the PRSP process promotes is help-ful. In particular, the process is better than onethat ignores the poor, never solicits outside opin-ion, imposes solutions with no reference to theparticulars of the recipient country, and is derivedwith no consultation with the recipient country.Compared to this picture of international lendinghell, the PRSP process fares quite well indeed.Many of the NGOs, on the other hand seem to beevaluating not the PRSP process, but rather seemto be asking the grand question of whether theBank has achieved its stated mission of “A worldwithout poverty”. By this metric, the process fallsrather short.

THE WORLD BANK’S POVERTY REDUCTIONSTRATEGY PAPER APPROACH: GOOD

MARKETING OR GOOD POLICY?*

Jim Levinsohn

* This paper was prepared with financial support of the International Development Research Centre (IDRC) of Canada.

2 G-24 Discussion Paper Series, No. 21

Before trying to answer whether the PRSPprocess has been a good thing, this paper first re-views just what this process is supposed to be. Itgoes on to examine some of the existing reviewsof the PRSP approach, tries to explain why thesetend to fall short of their goal, and finally tries toanswer the question formulated above. The con-clusion is that the PRSP approach is full of goodideas. At the end of the day, though, knowingwhether the PRSP process is addressing the con-cerns of the poor means being able to identify thepoor and measure changes in their well-being. Thisis hard enough, but to really answer the questionat hand, one also needs to examine whether thechanges in the welfare of the poor are in fact dueto changes in policy resulting from the PRSPapproach. This is a very tall order to fill. In a sur-prisingly large number of countries, though, it is,at least in principle, doable. (In others, the datasimply do not and for the foreseeable future willnot exist.) The first round of reviews of the PRSPapproach have been dominated by precious littlecareful quantitative analysis and, instead, a lot ofstories, but they are not the way to guide policy.Future reviews of the PRSP process, from all sides,should honestly confront the difficult task of deter-mining the marginal impact of the PRSP approachand refrain from making conclusions based on pre-existing notions of whether the process is or isnot working.

II. The PRSP process2

Poverty reduction strategy papers are not atrivial bureaucratic hoop through which countrieshave to jump. Rather, they comprise major effort.How did it come to pass that countries from Alba-nia to Zambia have, in the last couple years, writtenpapers averaging around a hundred pages or soon how these countries plan to reduce poverty?

A. How the process is supposed to work

At their September 1999 annual meetings, theBank and Fund lined up behind a proposal that“country-owned” poverty reduction strategiesshould form the basis for all Bank and Fundconcessional lending. These strategies would take

the form of papers called Poverty Reduction Strat-egy Papers. Hence was born the PRSP process. Theprocess was essentially a way to implement a setof principles the Bank had earlier adopted. Theseprinciples were called the Comprehensive Devel-opment Framework. The relationship between theComprehensive Development Framework and thePRSP process is confusing,3 but it is probably ap-propriate to think of the Framework as thedestination and the PRSP as the route selected.

In time, the plan is that every country receiv-ing what is called HIPC (highly indebted poorcountry) relief and all countries making use of theIMF’s Poverty Reduction and Growth Facility willneed to author a PRSP that must then be approvedby the boards of the Bank and Fund.4 The expec-tation is that eventually about 70 low-incomecountries will be expected to prepare PRSPs.Clearly, the process is going to be pervasive andwill not be restricted to only the most troubled orvery poorest economies.

The Bank and Fund have gone to some painsto emphasize that there is not a single templatefor a PRSP. Rather, each nation’s PRSP is expectedto follow the following five principles:5

In the language of the PRSP Sourcebook ofApril 2001, PRSPs should be:

• country-driven and owned, based on broadbased participatory processes for formula-tion, implementation and outcome-basedmonitoring;

• results-oriented, focusing on outcomes thatwould benefit the poor;

• comprehensive in scope, recognizing the mul-tidimensional nature of the causes of povertyand measures to attack it;

• partnership-oriented, providing a basis forthe active, coordinated participation of devel-opment partners (bilateral, multilateral, non-governmental) in supporting country strate-gies;

• based on a medium and long term perspectivefor poverty reduction, recognizing that sus-tained poverty reduction cannot be achievedovernight.


In order to focus attention on the marginalchanges brought about by the PRSP process, it isworth considering which of the above represent achange from the Bank’s and Fund’s usual way ofdoing business. This is an analysis of intent, notnecessarily an analysis of outcomes.

Point 1: The notion that a country’s plan ofaction should come first and foremost from therecipient country and not from the Bank and Fundis a fine idea and is, if not new, at least more promi-nently displayed than it has been in the past. Ittakes a bit more reading before one notes that theseplans have to be approved reasonably enough bythe boards of the Bank and Fund. In the end, theplan is going to be a collaborative effort betweenthe poor country and the Bank and Fund. Underthe PRSP process, it is now clear that the first handplayed is that of the country and the lending agen-cies can then respond. One of the real changes thatis highlighted in this first principle is the inclu-sion of all the groups that comprise “broad basedparticipatory processes”. The phrase “countryownership” has come to mean more than just thatthe Finance Ministry has to sign off on the plan.Rather, it is expected that country ownership willcome about through:6

• The participation of many ministries, parlia-ment, and provincial or state governments.

• The inclusion of other “stakeholder groups”.This list includes “civil society groups, wom-en’s groups, ethnic minorities, policy researchinstitutes and academics, private sector, tradeunions, and representatives from different re-gions of the country”.

• The participation of other external aid pro-viders.

• The inclusion of “mechanisms used to con-sult the poor and their representatives”.

Some of these really are changes from theusual way of doing business. In particular, theexplicit inclusion of multiple branches and layersof government and the inclusion of civil societygroups are new ways of trying to do business.Readers from developed countries that tend tohave less geographic and ethnic divisions, bettercommunication infrastructure, and more estab-lished governmental institutions can judge for

themselves how easy it would be to get the abovelist of stakeholder groups to agree on a strategy toalleviate poverty in their own country. Still, inprinciple, it sounds like a good idea to at least askfor everyone’s involvement. The efficacy of thisstrategy in practice is discussed below. The ex-plicit inclusion of the poor in the consultativeprocess is a new thing. How this is actually doneis again tricky, and whether it is effective is againa separate matter that is discussed below. Finally,the idea the first draft of the document that willguide economic policy (which is now the PRSP)be initially drafted by the recipient country andnot the Bank or Fund is also a change from howthese matters often unfolded in the past.

Point 2: The focus on a “results-oriented”process is hopefully not new. It is clear what thealternative is, but few seem to have advocated aprocess-oriented strategy without reference to re-sults prior to 1999. On the other hand, the focuson how the results impact the poor is a marginal(in the positive sense) change in emphasis if notcontent. This is a very good idea in principle. Inpractice it is often hard to analyse in a rigorousway the distributional impact of government ex-penditures and other policies, but the intent is righton target. There is a danger that the parties in-volved will shy away from the difficult questionof just how a policy impacts the poor and insteadtake the much easier route of suggesting that anypolicy that promotes economic growth is good forthe poor since in general growth is good for thepoor. This point, too, is discussed below in theevaluation of the PRSP process.

Point 3: The PRSP process should acknowl-edge the many dimensions of poverty. One way inwhich Bank thinking about poverty today reallydoes seem different than that of ten or twenty yearsago is an increased emphasis on dimensions ofpoverty beyond income-based measures. Point 3,then, does seem like a change in Bank/Fund policy.As will be argued below that this change is notalways for the better and that there are some prettygood reasons for sticking with income (or con-sumption) based measures of poverty. Still, it is achange.

Point 4: The PRSP process should includeparticipation by other aid providers, or, to use thelanguage of the PRSP Sourcebook, “developmentpartners”. The idea here is that the PRSP process


should be such that all the aid providers come to-gether to support the poverty reduction plan.Because the plan is, in principle, the work of thecountry and not the Bank, this means aid provid-ers should be supporting the country’s own plan.This seems ingenious. Because at the end of theday the Bank and Fund have to approve the PRSP,this point could appear to be a strategy for gettingother aid providers to buy into a Bank/Fund de-velopment strategy. But if the PRSP strategy is agood one and if this results in other aid givers alsofocusing on the poor, this is a laudable goal.Whether this point represents a marginal changeis unclear. The fact that PRSPs need Bank/Fundapproval still gives the Bank and Fund the finalsay on the policy environment required to obtainconcessional lending. This has been and contin-ues to be the case. Whether the Bank and Fundhave in the past tried to get other bilateral, multi-lateral, and non-governmental parties to supportthe Bank and Fund’s policy prescriptions variesby instance.

Point 5: The last point is a concession to re-ality. Because poverty reduction is a long termprocess, the PRSPs should adopt a medium- andlong-term perspective. This does not seem to rep-resent much of a change from the past.

These are the goals of the PRSP process.Devising the plan to implement these goals hasproven to be a time consuming task. Because thePRSPs are “country-owned”, the burden falls onthe country to develop the plans. This is a cost tothe country as it can easily take about two yearsto develop a PRSP and in many of the very poor-est countries, there is not a surplus of qualifiedtalent to draft a good PRSP. Still, the alternativeis for the Bank and Fund to develop the PRSP it-self. A middle ground, and one which has inpractice been adopted in several countries, is forthe Bank and Fund to provide assistance whenneeded but to leave the main responsibility forwriting the plan in the hands of the country.

The above discussion was focused on howthe process is supposed to work. In the followingwill be discussed the sort of content that a PRSPis supposed to include. The next section will evalu-ate how well the PRSP process has worked.

B. What should be in a PRSP?

A good PRSP should lay out a plan to reducepoverty and increase sustainable economic growth.Of course, if this was easy, the Bank with its thou-sands of PhD.-trained professionals and billionsof dollars of resources would probably have solvedthe issue some time ago. Still, it is important tospecify the goals of the PRSP. The PRSP Source-book focuses on four key topics that a good PRSPshould address. These are:7

• Macro and structural policies to support sus-tainable growth in which the poor countriesparticipate;

• How to improve governance – including pub-lic sector financial management;

• Appropriate sectoral policies and programmes;and

• Realistic costing and appropriate levels offunding for the major programmes.

The guidelines outlining what countriesshould include in their PRSPs also highlight theimportance of prioritizing. Finally, there is men-tion of the need to discuss “appropriate sequencing”of policy actions. This last point is an especiallyimportant and too frequently ignored missive. Inthe critical analysis in the last section, the impor-tance of the sequence of policy actions (which maybe very different that the priority of policy actions)is discussed. Each of the above four areas of con-tent certainly belong in a PRSP. Indeed, given thislist, it is not clear what area of development-related government policy does not belong. Eacharea is discussed in turn.

The first area speaks to the importance ofmacroeconomic policy as well as regulatorypolicy. The macroeconomic focus results from thebelief (which is empirically well-founded) thatmacroeconomic growth is good for the poor. Thispart of the PRSP is intended to address some ofthe concerns that have traditionally been in theIMF’s domain – for example, inflation and ex-change rate policy. Alas, while the evidencestrongly supports the idea that growth is good forthe poor, the record on just what promotes eco-nomic growth is much less clear. The PRSP


Sourcebook is apparently less ambivalent as towhat makes an economy grow or not. For exam-ple, explicit mention is made of constraints ongrowth such as exchange rate controls (tell that toMalaysia), and trade barriers (someone informChina).8 Specific mention is also made of labourmarket policy as well as other regulatory and mar-ket controls. The broad prescription seems to beone in which macroeconomic reform should beplanned out according to the mostly-standard Fundprescriptions, but with concern to how the poli-cies will impact the poor. There is specific mentionof the possible need “to strengthen social safetynet programs prior to embarking on the reformprogram”. This is right, but many and probablymost of the countries writing PRSPs don’t havemuch in the way of social safety nets to start with.To ask a country like Mauritania or Burkina Fasoto be sure to implement policies (that is, expendfunds) to protect the poor from the adverse im-pacts of contractionary monetary policy whilekeeping an eye on the government budget deficitis a mighty tall order to fill.

The discussion of labour market policy is alsoambitious. The PRSP should address issues suchas “minimum wages; payroll taxes; rules govern-ing hiring/firing of workers; labour standards” andother labour market regulations and how theseregulations impact the poor. Doing this for a coun-try like the United States for which there is anastounding wealth of high quality labour marketpanel data would be a real contribution to the stateof the art of economic policy analysis. Doing this ina convincing way for Chad (which acknowledgesin their interim PRSP that they don’t have a lot inthe way of data) is inspirational in its ambition.

Overall, some parts of the discussion of themacroeconomic content of PRSPs strike me asextraordinarily hopeful. It might seem like thereis little downside to asking for countries to at leasttry to address these important macroeconomicconcerns and the linkage of these concerns to pov-erty. That might be right, but an alternative viewis that when one asks for a policy plan for whichthere is no hope of careful analytic support, onejust gets platitudes and, if the country is gamingthe situation, a discussion of what the countrythinks the Bank and Fund want to hear.

The second area of content for PRSPs is aplan to improve governance. The inclusion of this

area in the PRSP process speaks to the growingawareness that the quality of institutions matterfor economic progress. There is little doubt whatthe Bank and Fund are looking for here. The coun-tries are expected to address issues such as theaccountability and transparency of governmentalexpenditures as well as issues regarding the non-functionality (in some cases) of civil service. TheBank and Fund documentation regarding the linkbetween this aspect of the PRSP and poverty isthe somewhat nuanced, but the basic idea isstraightforward and seemingly correct. If the gov-ernment is corrupt, the poor are going to have lessinfluence on government policy and more diffi-cult access to government services for the simplereason that they are less able to afford the accessand services.

The third area that the PRSP should addressis the selection of “appropriate sectoral policiesand programs”. The PRSP Sourcebook includeschapters on several specific policy areas. Theseinclude education, health, environment, mining,and “social protection”, among others. Althoughsuggested policy prescriptions will differ by coun-try and by sector of policy, the general approachsuggested for PRSPs is that countries ascertainwhat makes poor households poor and use poli-cies to mitigate the resulting poverty. There seemto be two broad approaches. First, countries shouldexamine the roles that lack of access to health care,education, credit, sanitation, and the like play indetermining poverty. Countries should then directpolicy appropriately while minding their budgetconstraints. The PRSP Sourcebook makes clearthat “policy and program priorities will not beimplemented unless countries ensure that they canafford the public expenditures they plan”. Second,countries should analyse the distributional impactof the expenditures they plan to make. These dis-tributional impacts can vary by geographic region,income group, gender, and religious or tribalgroup. The PRSP guidelines make clear that coun-tries should analyse these distributional impactsof proposed policies. Furthermore, the priori-tization and sequencing of the different policyoptions should take into account the impact on thepoor. While these are also tall orders to fill, econo-mists know how to undertake these tasks if thedata permit. Conversely, the data requirements arenot trivial and requiring information on the distri-butional impact of policies when the data to fulfilthis request does not now exist and will not exist


in the foreseeable future encourages hollow andunsubstantiated claims that probably sound reallygood on paper.

The last area that the PRSPs should addressis just how much all the grand plans are going tocost and to essentially propose a budget. This re-quirement of the PRSP ties in closely with the firstrequirement, macroeconomic stability. The budg-etary implications for the PRSPs need to takeaccount of the possibility of shifting existing ex-penditures, raising new government revenues, andexternal assistance. Because the Bank and Fundare the ones approving (or not) the PRSP, they arepresumably best able to judge the reasonablenessof the likelihood of external assistance. The re-quest for realistic budget information is bothnecessary and appropriate.

The remainder of this paper is devoted to try-ing to analyse the effectiveness of the PRSPprocess. That is, how well has it worked and whatlessons are emerging after about two and a halfyears experience? The simple volume of workproduct stemming from the PRSP process is over-whelming. There are tens of thousands of pagesof PRSPs, interim PRSPs, Bank and Fund jointstaff assessments of the PRSPs, progress reports,status reports, policy papers, and notes to guideBank and Fund staff as well as the PRSP Source-book to guide nations as they prepare PRSPs. Evenif one restricted one’s attention to, say, just thePRSPs (final and interim) themselves, there areabout four dozen countries with quite varying ex-periences. Here, we will begin by evaluating theexisting evaluations by taking a critical look atwhat those inside the process (the Bank and Fund)and those outside the process (mostly NGOs) thinkof the PRSP approach.

III. Evaluation from inside

In the summer of 2001, the Bank and Fundundertook an extensive review of the PRSP proc-ess. That review was completed in March 2002and the results of the review, in keeping with thespirit of inclusion, are available on the web.9 TheBank and Fund findings are presented in both anexecutive summary and a lengthy detailed report.The summary of this internal review is taken fromthe “Review of the Poverty Reduction Strategy

Paper (PRSP) Approach: Early Experience withInterim PRSPs and Full PRSPs”, dated 26 March2002.10

There are two broad approaches one can takein evaluating the PRSP approach. The first is toexamine the multitude of experiences and thenhighlight best practices. The second is to step backand ask if the process is working and, relative towhat probably would have happened anyway,whether the process is truly beneficial to the poor.It’s possible to do the former and never reallytackle the latter. The Bank and Fund internalreview illustrates the point. Highlighting bestpractices is, for sure, useful. Over time, dozens ofcountries will move from their interim PRSPs topreparing the real thing. Learning from the expe-riences of those who have come before makesperfect sense. But this is not the same as askingwhether the entire approach is working. This criti-cism of the internal review, though, is admittedlytoo harsh. The internal review takes the stance thatit is simply too early to know if the process isimpacting poverty. Instead, since many countriesare in the process of either writing interim PRSPsor moving from the interim to full PRSP, it is moreuseful to simply highlight best practices.11 It isprobably too early to make firm judgments on howthe process is impacting poverty. It is not, though,too early to ask whether we’ll ever be able toanswer this question. The Bank and Fund are op-timistic as they predict that a “rich informationbase should be available from a range of coun-tries during the next three years to allow thesequestions (concerning poverty outcomes) to beposed more precisely”. The more cautious con-clusion is that time-series data on income orexpenditure-based poverty outcomes is currentlyscarce in most countries writing PRSPs. Countriesare being urged to collect the needed data, but itremains to be seen whether this will in fact hap-pen. While some data on things like primary schoolcompletion rates and changes in government ex-penditure patterns are likely to be available, dataon changes in poverty outcomes over time are lessavailable.

The internal review of what’s happened sofar, though, focuses on highlighting best practices.(There are dozens of these and countries that arepreparing interim or full PRSPs would be welladvised to review the entire Bank and Fund re-view.)


Those countries that have completed fullPRSPs have often built upon existing data andanalyses as well as on prior strategies. That is,many of the countries that did full PRSPs alreadyhad poverty reduction programmes in the worksbefore the PRSP process was announced. One ofthe countries often mentioned as doing a prettygood job is Uganda. Uganda, though, put togethertheir “Poverty Eradication Action Plan” in 1997so they were ready to go when the PRSP approachwas announced in 1999. This speaks to the mar-ginal impact of the PRSP approach. In the case ofUganda, much of what showed up in their PRSPwas going to happen anyway. The organization ofthe plan and perhaps the increased emphasis oninclusivity were impacted by the PRSP process,but the overall plan was already well under way.Similarly, in Mozambique, the PRSP built uponthe government’s 1999 poverty reduction plans.In Mauritania and Burkina Faso, the governmentsset up plans to attack poverty in 1998, while Bo-livia did so in 1997 and Honduras and Nicaraguaput together anti-poverty strategies followingHurricane Mitch in 1999. Hence, if one were tolook only at those countries that either have al-ready or have almost completed full PRSPs, manyare articulating plans that pre-dated the PRSP ap-proach. In these cases, the marginal impact of thePRSP approach is diminished.

The Bank and Fund internal review does sug-gest that the PRSP approach has yielded somechanges in process. These include:

(i) PRSPs have led to wider governmental in-volvement. Before the PRSP process, it wascommon for Bank and Fund borrowing issuesto be handled almost entirely by the financeministry. The new process has led to ministe-rial-level involvement by several ministries.It remains the case that the “core” ministries(for example, planning or finance) are moreinvolved than sectoral ministries, but thechange is in the right direction.

(ii) PRSPs have led to greater involvement bycivil society organizations. These organiza-tions frequently did not have an opportunityto offer meaningful input prior to the PRSPapproach. This is a two-edged sword. On thepositive side, it means that more segmentsof society have an opportunity to weigh inwith opinions and concerns. On the negative

side, the additional participation tends to“compound rather than resolve the problemof prioritization”.12

(iii) The PRSP approach has highlighted the needfor decent data if one is going to implementcareful poverty diagnostics. The internal re-view notes that the PRSP process has “beenmarked by useful steps toward better povertydata and diagnostics”. So far, every one ofthe completed full PRSPs has drawn upon“nationally representative household surveysfrom which income/consumption poverty es-timates were derived”. While some of thesesurveys were almost a decade out of date, itis nonetheless impressive that all the coun-tries even had a national household income/consumption survey on which to draw. Thereis a concern, though, that those countries thathave completed their full PRSPs are not rep-resentative of poor countries in terms of data.Rather, those that had the necessary datawere, not surprisingly, the first to get theirPRSPs completed. It might seem overly ob-vious that careful poverty measurement anddiagnostics will require a national-levelhousehold survey, but the point is an impor-tant one. Simply focusing attention on theneed for this sort of data, all by itself, is acontribution of the PRSP process.

(iv) Finally, the PRSP process has highlighted theimportance of better understanding and ana-lysing the distributional consequences ofgovernment expenditure. While issues suchas efficient public administration, honest andtransparent government expenditure systems,and the distributional aspects of revenue col-lection are mentioned in the PRSP approach,none of these are really new. An explicit fo-cus on the distributional aspects of the ex-penditure side of the budget is, if not new, atleast more prominent. This is a great idea,but the data requirements necessary to im-plement this analysis are substantial. Doingthis analysis correctly and informativelymeans analysing who benefits from govern-ment expenditure. Good analysis probablyrequires more than just examining the frac-tion of the government budget that is devotedto, say, health and education (which is whatis presented in the internal review.) By pre-senting “evidence” in their own report that


doesn’t even come close to the level of careand detail that the PRSP Sourcebook sug-gests, the internal review sets a pretty badexample.

IV. Evaluation from outside

As befits an inclusive process, the review ofthe PRSP approach solicited opinions from thoseoutside the Bank and Fund. In an admirable dis-play of non-censorship, these reviews are on thePRSP review website. Most of the reviews sub-mitted by non-governmental organizations (NGOs)do not focus on whether the poor are or are notbenefiting under the PRSP approach. Rather, mostof the NGO reviews focus on the NGOs. Many ofthese reviews, for example, ask whether the NGOsare having the sort of input to the process that theBank and Fund seemed to be promising. (Equat-ing input with perceived influence on output, theiranswer seems to be, “no”.)

The occasional review, that does focus onpoverty alleviation, concludes that poverty is notbeing alleviated. While this review was necessar-ily limited, it was not selective. There appears tobe no NGO review that provided careful dataanalysis to back up its conclusions. Lack of data,though, didn’t stop some NGOs from concludingthat the PRSP approach was essentially a fraud.One of the most negative of the reviews was jointlysubmitted by “Jubilee South, Focus on the GlobalSouth, AWEPON, and the Centor do EstudiosInternacionales with the support of the WorldCouncil of Churches”. Titled “The World Bankand the PRSP: Flawed Thinking and Failing Ex-periences”, the review is mostly as advertised. Thereview concludes that “Fighting poverty becomesthe newest justification for the aging prescriptionsgeared to increasing the overall opening of the“host country” to external economic actors andfree market rules”. While one can argue aboutwhether open trade is good for poverty allevia-tion, this review makes no reference to any studiesand cites precious little evidence of any sort in itsbroad-brushed conclusions. This review is a studyin how not to do programme evaluation, but it’snot a bad case study for Manifesto Writing 101.

Several of the NGO reviews make the casethat the focus on income-based measures of pov-

erty is too narrow.13 In an especially nicelyexposited review, for example, the Catholic Re-lief Services noted that while income-basedmeasures are necessary, they are not sufficient.“Too often, PRSPs fail to reflect a broader ap-proach to poverty reduction that fully addressesdimensions related to security or empowermentas essential ingredients for poverty reduction”.14

This message is echoed by many groups. None,though, offer suggestions of how one might meas-ure these other “ingredients” so as to be able toascertain whether the PRSPs are actually address-ing these dimensions of poverty.

Another message that permeates many of theNGO reviews is a sense of frustration that theirinput is not taken as seriously as, say, the input ofthe Bank and Fund. Some NGOs express the be-lief that while they are grateful for the opportunityto weigh in, their voice is not having much im-pact. While some NGOs remark that the PRSPapproach has given civil society more room forinput that the old process,15 there seems to be acongruence of complaints that civil society is nothaving much impact when it comes to designingeconomic policy. In particular, NGOs often com-plain that they have little influence when it comesplanning macroeconomic structural adjustmentand other poverty reducing economic policies.Catholic Relief Services wrote, in their review,that “Despite some positive and real openings fordialogue between government and civil society,there has been little noticeable impact on the con-tent of the PRSPs”. Save the Children UK wrotein their review, “The experience to date is thatthere seems to have been little difference betweenpolicies outlined in the I-PRSP, describing currentarrangements under the IMF Poverty Reductionand Growth Facility and the final or draft PRSPs,supposedly written after a particupatory processhas taken place”.16 The European Network on Debtand Development (EURODAD) raised similarconcerns.17

Taking as correct for the moment that civilsociety is not having much impact on the economicpolicy content of the PRSP approach, there are atleast two explanations. First, it is possible thatcivil society is simply being allowed a voice atthe table and, no matter what is said, civil societyis preordained to be marginalized. Second, it mightbe that civil society is just not sufficiently well-trained to take on the tricky task of designing


well-targeted poverty reduction policies under thebudget constraints that are a crucial part of thereality of poor countries’ economic plight.

Some reviews are more nuanced than others,but the depressing message of many of the NGOreviews is that these organizations are woefullyunder-prepared when it comes to a careful evalu-ation of the PRSP approach or a convincinganalysis of poverty dynamics or the distributionalimpact of government expenditures.18 Many of theNGOs that offered well exposited critiques of thePRSP process do an amazing amount of good onthe ground in poor countries. Examples includeOxfam, Save the Children, Catholic Relief Serv-ices, and World Vision. These organizations knowhow to run projects and, on a very personal level,enhance the quality of life of the world’s poor.The knowledge gained in the field is invaluable,and these organizations contribute to the PRSPprocess by commenting on the PRSP approach.At the end of the day, though, they typically donot seem to have the capacity to offer compellingguidance when it comes to making economicpolicy and rigorously evaluating the distributionalimplications of that policy.19

The PRSP process has also been reviewed byother multilateral institutions. One of the mostsensible reviews came from the Inter-AmericanDevelopment Bank (IDB.) The IDB noted that“Making PRSPs a condition for debt relief imposesa timetable that is technically and politically sub-optimal; it undermines the quality of the documentand compromises the social marketing of its con-tent”.20 The IDB pointed out that while asking forinput from many is, in principle, a good idea, ithas led to PRSPs that are essentially a “shoppinglist of problems”.21 The IDB also noted that theprocess itself is simply too time-consuming andtoo expensive. In an understated summary, the IDBsuggested that “Perhaps, less time could be spentin preparing diagnoses and more time in design-ing appropriate responses to the country’s mostpressing problems”.

V. Yet another review

Both the process and the content of the PRSPsare especially sensitive to the plight of the verypoor. This raises the question of why the Bank

and Fund have elected to highlight the link be-tween their lending programmes and poverty inthe late 1990s. Some of the answer is probablywhat a for-profit business would call marketing.A quick read of issues of the World DevelopmentReport of around 20 years ago turns up consistentreferences to poverty reduction, most usually inthe context of rural development. While the WorldBank did not have a website with the lead bannerproclaiming “Our Dream is a World Free of Pov-erty” in 1982, at some level the Bank has beenworking on helping the poor in developing coun-tries for decades. The advent of the inclusive PRSPprocess coincides with a heightened awareness ofthe Bank in civil (and sometimes not-so-civil) so-ciety. Recent anti-globalization demonstrationshave put the Bank in the position of having to re-state and better market its mission. This isprobably a good thing, especially if it can be donerelatively costlessly. (Alas, the PRSP process isnot costless, nor do the Bank and Fund incur allthe costs.) The coincident timing of the adoptionof the PRSP process and the Bank’s need to betterexplain what it does lead some to wonder whetherthe PRSP process is but window dressing. Thatis, is the PRSP process simply good marketing?The answer is clearly “no”.

While the PRSP approach with its emphasison inclusion and focus on poverty alleviation isundoubtedly good marketing, it is not just mar-keting. There is content. Content alone, though,does not mean that the PRSP approach embodiesreal change. The right question to ask is whetherthe content is really different from what wouldhave happened anyway. This concern speaks to theneed to address the question posed at the outset,namely, what are the marginal costs and benefitsof the PRSP programme. Especially because themore cynical observers may believe that the PRSPprocess is simply good marketing in response tothe Bank’s and Fund’s increased visibility, it isimportant to keep the spotlight on the marginalchanges imposed by the PRSP process.

This focus on marginal changes – changesrelative to what would have happened anyway –matters because not all changes are marginalchanges. As noted above, the Bank has focusedon issues of rural development and the poor fordecades. In the month before the Bank and Fundadopted the PRSP process outlined above, thesame economists occupied the same offices22


around 19th Street in Washington, DC and, for themost part, they were thinking about the samethings before and after the PRSP process was in-stitutionalized. In order to address the claim thatthe PRSP process is more than good marketing, itis necessary to speak to the marginal net benefitof the process.

The inclusionary aspect of the PRSP ap-proach, if not new, at least receives more promi-nent billing. The same can be said for the explicitfocus on how outcomes are likely to impact thevery poor. The increased attention given to defi-nitions of poverty that extend beyond those meas-ured by income or expenditure is new. Followingfrom this, the increased use of qualitative or nar-rative approaches to poverty research is new. Thenotion that the lending country should initiallydraft the poverty alleviation plan is, in many cases,new. Claims that the PRSP process is simply a re-labelling of the existing process just do not standup to scrutiny. Not everything, though, is new. Theemphasis on pro-growth macroeconomic policies,fiscal responsibility, and the importance of bothgood policy and good government have beenaround for a while.

Not all that is new is good. Some of the inno-vations embodied in the PRSP approach seemflawed in practice, and some of the flawed partsof previous approaches have survived. But someof the new elements are clearly improvements. Inthe following the weaknesses and strengths of thenew approach will be discussed in greater detail.The strengths clearly outweigh the weaknesses.

The current PRSP approach does not placesufficient emphasis on the importance of policysequencing and instead highlights the need toprioritize policy decisions. These are not one andthe same and the distinction seems especially im-portant when it comes to government spending andpoverty alleviation. While sequencing is men-tioned a couple times in the voluminous PRSPbackground material,23 it is hardly highlighted. ThePRSP approach forces the country to explicitlystate how poverty is to be addressed. Adding anincreased emphasis on the tricky issue of policysequencing would be a good idea.

The PRSP process discusses the role of tradepolicy in poverty alleviation with too much assur-edness. See, for example, the PRSP Sourcebook

chapter on trade policy.24 While links betweentrade and poverty are nicely highlighted in back-ground documents (cited but not well-representedin the Sourcebook) by Alan Winters, the tradepolicy guidance in the PRSP process doesn’t ad-equately reflect the lack of evidence on the linksbetween trade and poverty. While the focus onknowing what the poor consume and how theyearn their livelihoods is right on target, the sug-gestion that more open trade is good for the pooris based more on faith than evidence. Althoughthe PRSP documents are peppered with referencesto moulding the particulars of a poverty allevia-tion programme to the details of the country, thediscussion of trade policy (as well as some of themacroeconomic prescriptions) seem to come froma one-size-fits-all mentality.

The PRSP process also stresses the need tohear from the poor so as to be able to better ad-dress their concerns. To a degree, this is a greatidea, but when anecdotal and qualitative ap-proaches begin to guide policy making, it’s timeto step back and re-evaluate. This approach is evi-denced by the much-heralded “Voices of the Poor”project in the Bank.25 (This project is not part ofthe PRSP programme but is clearly related.) Theadvantage of a project like this is that it gives ahuman face to an issue that might otherwise beobscured by facts and figures. That project alsobrings to the fore the sense of hopelessness thatcomes from simply not having a voice due to howpoverty interacts with the politics of decisionmaking at the village, provincial, and federal lev-els. That project also highlights gender-relatedaspects of poverty. It makes for engrossing andsad reading. This sort of work is a useful comple-ment to more rigorous quantitative work, but it isnot a substitute. The qualitative work suggestssome alternatives to income- or expenditure-basedmeasures of poverty. The problem with measuressuch as, for example, disenfranchisement or op-portunities-based measures of poverty is that theyare very hard to measure. If one cannot measurethese aspects of poverty, it is hard to know whetherthese aspects of poverty are being alleviated. Ab-sent data, it’s all too easy for advocates or scep-tics of one stance or another to simply assert thealleged effectiveness of their preferred poverty al-leviation policies. While income- and expenditure-based measures of poverty have their drawbacks(they don’t address the issue of voicelessness tothe extent that it’s unrelated to income or expendi-


ture), they also have their advantages. De-empha-sizing these traditional measures seems a bad idea.Supplementing them with capabilities-relatedmeasures of poverty, many of which can be meas-ured (e.g. health, education, nutrition, or evensome self-reported measures of well-being) seemslike a good idea. Adding to the mix metrics thatcannot be measured is more problematic. It shouldbe noted that the PRSP Sourcebook chapter on“Well-being measurement and analysis” does thejob right. The focus there is solely on measurablepoverty indicators.

There are many innovations in the PRSP ap-proach that count as strengths. These include theinclusivity of the approach, the principle that thecountry (and not the Bank/Fund) initiate the firstdraft of the plan, and, most of all, the focus on theneed for data to evaluate the effectiveness of thePRSP. Each are discussed in turn.

On the whole, the inclusivity of the PRSPapproach is a good thing and a welcome change.This inclusivity extends to both civil society (thatis the obvious part) but also to broader branchesof government than just the finance and planningministry (this part is less obvious.) It is the latterthat will probably yield the greatest benefits. Byat least trying to get multiple ministries and, inmany cases, multiple levels of government in-volved in the development of a poverty reductionstrategy, the likelihood of success of the resultingplans is enhanced. This seems to be one aspect ofinclusivity that is working. The inclusivity of thePRSP approach is a double-edged sword. Wheneveryone’s input is requested, the list of issues tobe addressed starts to look like a shopping list,and many of those weighing in on the process areill-trained to do so. This has been discussed abovein the section on the NGO reviews of the PRSPapproach and won’t be repeated here. The alter-native is to proceed without soliciting the inputof civil society, and this seems even more prob-lematic. The uncomfortable truth is that the Bank/Fund and recipient government are still free toignore what they perceive as bad advice. The prob-lem arises when, for political reasons or otherwise,good ideas get equal billing with silly ones.

The PRSP process, in principle anyway, isstructured so that the first draft of the poverty al-leviation plan is written by the country and notthe Bank or Fund. It doesn’t always work this way

since in some countries, there just is not enoughcapacity. Still, when there is capacity, this order-ing of matters is an improvement and, in theabsence of capacity, the process is no worse thanbefore. Here, it is too early to know whether thismove to enfranchise the country is working. Theidea is that by instilling a greater sense of owner-ship, the country will feel more committed toimplementing the plan.

A common complaint about asking the coun-try to initially draft the PRSP is that some countriessimply don’t have the capacity. While this is un-doubtedly true, it is not a fair complaint. Whenthe capacity is not there or is not forthcoming,experience suggests that the Bank/Fund teams willstep in and assist. It is hard to see how this out-come is worse than the alternative in which theBank/Fund teams initially draft all the plans forall the countries. The other alternative, conditionalon not being able to magically create local capac-ity in an instant, is to not make the PRSP approachso demanding. It is true that a careful analysis ofthe distributional impact of myriad policies withan emphasis on quantitative analysis is difficult.But relaxing these requirements is a bad idea.Absent hard quantitative goals and metrics forsuccess, it is too easy for the PRSP approach tobe waylaid by political concerns with proclama-tions of success or failure motivated by dogmarather than evidence.

The focus on poverty as opposed to, say,broadly defined economic growth is so pervasivein the PRSP approach that it could almost gounmentioned. This focus is real, not just market-ing and it permeates almost every aspect of theprocess. It’s an obvious good idea. An implica-tion of the focus on poverty is that it requires verydetailed household-level (if not individual-level)data. Detailed data are required to identify the poorand even more data are required to identify tran-sitions from or into poverty. These data are neededfor several aspects of successful PRSP implemen-tation. In particular:

• Data on household income or expenditure areneeded to determine just who the poor are.These data are not as available as one mightthink. There are many examples of householdsurveys in which expenditure data is scarceand in which income data is either not re-ported, reported only in broad bands (which


makes identifying the very poor hard) or re-ported only incompletely. (For example,income from self-production is sometimespoorly reported as are transfers from non-resident household members.) These data areusefully supplemented with data on self-re-ported measures of well-being and othercapabilities-based measures.

• Data on household expenditure are neededto determine how poor households might beimpacted by various policy options that arepart of PRSPs. For example, the PRSPSourcebook notes that what a household con-sumes will impact how it is effected by tradepolicy. Similarly, consumption patterns willbe important to the analysis of the distribu-tional impact of price changes that mightarise from shifts in food pricing policy orother tax-related policies. Expenditure datais generally more scarce than income data butthey are going to be essential.

• Data on the household use of public servicesare needed to determine the distributional im-pact of government provided programmesand services. That is, to know whether, say,health clinics or clean running water invest-ments will help the very poor, one requiresdata on the poor of these services by the poor.

• Data on the income sources of the very poorare needed to ascertain how the employmenteffects of various macroeconomic policy op-tions are likely to impact the very poor. Thismeans knowing more than just a household’sincome, and instead knowing the source(s)of that income, often by sector of employment.

All of the above are data needed to establishbaseline estimates of poverty and its correlates.To really evaluate the impact of the PRSP poli-cies, one needs either repeated cross-sectional dataon the above items or, ideally, panel data. Inmost developing countries, the former are moreavailable than the latter. Nationally representativepanel surveys are pretty rare. Cross-sectional sur-veys (often annual) are more available but by nomeans common. Absent panel data, it is going tobe impossible to investigate transitions into or outof poverty. Questions about whether the poor area stable population or are comprised of differenthouseholds moving into and out of poverty remainunanswerable without panel data.

The PRSP documentation is admirably up-front about the need for good household data ifone is going to be able to identify the poor andmonitor changes in their (individual or collective)welfare. There are myriad references to the ne-cessity of using and, in many cases, collectinghousehold-level data. This is exactly the right wayto focus the evaluation of the PRSP plans. By fo-cusing on poverty, which occurs at the level ofthe household or individual and not at the level ofthe nation, the PRSP approach redirects attentionto micro-data and away from macroeconomic data.This is a good thing. Micro-data are required formost poverty-related policy analysis and with theBank and Fund arguing for its collection, it is morelikely that such data might be collected.

If there were very little micro-data available,one might wonder whether the PRSP approachcould be evaluated, but the World Bank appearsto have access to a much larger amount of datathan one would expect.26 On a helpful web pagetitled “Household Surveys for Poverty Monitor-ing”,27 the Bank notes that fully 97 per cent of thepopulation of low and middle income countriesare covered by at least one survey with data onconsumption and/or income as of the year 2000.Furthermore, the comparable figure is 86 per centfor those covered by comparable data for at leasttwo points in time. In Africa, where some of thepoorest countries are, 93 per cent of the populationis covered by at least one survey measuring in-come and/or consumption while 51 per cent arecovered by comparable data for at least two pointsin time. These are hopeful figures. They suggestthat it is, in practice as well as in principle, possi-ble to evaluate the efficacy of the PRSP approachover time.

In order to conduct the evaluation of the ap-proach, it will be necessary to continue to collecthousehold survey data and not just rely on theexisting stock of data.

This is acknowledged in the PRSP docu-mentation and assuming it happens, this is a niceside benefit of the approach.28

In the Bank’s and Fund’s common review ofthe PRSP programme, they noted that it was tooearly to examine whether the approach was work-ing. That is why they instead focused on bestpractices. The quantitative evaluation that was pre-


sented was minimal and did not really address theissue of how changes in government spendingwere impacting poverty-related outcomes. TheNGO reviews did not contain even bad quantita-tive analysis. At the end of the day, there seem tobe no evaluations of a particular PRSP programmethat really examine the effectiveness of the pro-gramme at the level of detail that the programmerecommends.

Just as it is useful to highlight best practicesso that other countries can gain from the experi-ences of the first completed PRSPs, it would bevery helpful if the Bank and Fund or an NGO couldprovide one, just one, careful, detailed, and rigor-ous analysis of a PRSP programme. Such an analy-sis would identify the poor and note the useful-ness (or not) of expansive definitions of povertyrelative to more narrow ones. The analysis wouldlook carefully at how the poor spend their incomesand from whence the incomes come. The analysiswould examine the distributional consequences ofgovernment spending and would try to distinguishbetween marginal changes in policy and expendi-tures or programmes that would have happenedanyway. Ideally, this analysis would be based onmultiple waves of a household survey. The resultswould of course be preliminary. Still, providingan example of how it is done would be a helpfuladdition to an ambitious and, on the whole, admi-rable programme. The Bank and Fund have theexpertise and resources to conduct such an analy-sis, have access to the necessary data, and pre-sumably know what constitutes a convincinganalysis of a PRSP programme since, after all, theywrote the Sourcebook.

Notes

1 For example, the overview of the new programme, sum-marized in about one page and available at www.worldbank.org/poverty/strategies/overview.htm, refersto the CAS, CDF, HIPC, I-PRSP, JSA, PRGF, PRSC andPRSP.

2 Much of this section is drawn from material on the WorldBank’s PRSP website. This website, devoted to the PRSPprocess, is a tremendous resource. There, one can findlinks to all the original PRSP reports, the interim reports,the evaluation of these reports, the many contributedpapers evaluating the PRSP, and hundreds of related links.The address is: www.worldbank.org/poverty/strategies/index.htm.

3 The Bank has provided the following incomprehensiblediagram to explain the relationship between the Com-

prehensive Development Framework and the PRSP proc-ess: www.worldbank.org/cdf/cdfprsplink.pdf. Some helpin figuring all this out is found at: www.worldbank.org/cdf/overview.htm.

4 Countries can still obtain concessional lending beforetheir PRSP is complete and one way to do this is to pre-pare what is called an Interim PRSP (I-PRSP). This isbasically like a PRSP but much shorter, less detailed,and easier to prepare. It’s not too wrong to think of it asa draft mini-PRSP.

5 In some documents, there are five key principles, in somethere are six or seven, but in substance all the lists arequite similar. To the Bank’s and Fund’s credit, there is avery clear set of expectations associated with what thePRSPs are supposed to accomplish. The PRSP Source-book draft of April 2001 is available at www.worldbank.org/poverty/strategies/sourctoc.htm.

6 See page 6 of the Overview chapter of the PRSPSourcebook, April 2001 draft for public comments.

7 Taken verbatim from the PRSP Sourcebook, page 17 overthe “Overview” section.

8 The PRSP Sourcebook makes explicit reference to thewell-publicized findings of Dollar and Kraay and theircontention that open trade regimes promote growth whichalleviates poverty. No mention is made of the careful andfairly devastating critique of that work.

9 The review has its own homepage: www.worldbank.org/poverty/strategies/review/.

10 The 103 page review is downloadable at: www.worldbank.org/poverty/strategies/review/ earlyexp.pdf.

11 In 2005, another review of the PRSP approach is sched-uled and in the 2002 review, it is noted that the next re-view will need to “examine changes in poverty out-comes”.

12 See paragraph 42 of the internal Bank and Fund review,cited above.

13 The same would apply to expenditure-based measures.While economists tend to differentiate between the twoand prefer expenditure-based measures since they aremore resilient to transient shocks, the complaints aboutincome-based measures seem to apply to expenditure-based measures too.

14 “Review of the Poverty Reduction Strategy Paper Initia-tive”, December 2001, Catholic Relief Services, page 12.Available at: www.worldbank.org/poverty/strategies/re-view/crs1.pdf.

15 For example, Catholic Relief Services note that in Hon-duras, Bolivia, and to some extent in Zambia, civil soci-ety groups had voices in the PRSP that they had not hadprior to the PRSP approach.

16 This review is available at: www.worldbank.org/poverty/strategies/review/scuk1.pdf.

17 Their report is available at: www.worldbank.org/poverty/strategies/review/eurodad1a.pdf and www.worldbank.org/poverty/strategies/review/eurodad1b.pdf.

18 A synopsis of dozens of the submitted reviews of thePRSP process by organizations and, in some cases, indi-viduals, is found at: //www.worldbank.org/poverty/strat-egies/review/imfieo1.pdf. Some are quite good (e.g.Oxfam) and some are quite not-so-good (e.g. AfricanCitizens Development Foundation). The Lagos-basedgroup concluded that the root cause of all poverty in Af-rica is corruption.).

19 With huge budgets and hundreds if not thousands of de-velopment specialists including some of the best poverty


researchers in the world, the Bank/Fund group bringstremendous expertise to the table. It’s just not realistic toexpect NGOs to be able to employ similar expertise. Thisis not to say that the Bank/Fund is always right and theNGOs never so. Hardly. But it is intended to point outthe reality that there are vastly different levels of exper-tise at the NGOs when it comes to policy analysis.

20 Several NGOs also made this point. A more subtle ques-tion is whether, conditional on the requirement that aPRSP or IPRSP be completed to obtain concessional lend-ing, is it perhaps good public policy to rush the comple-tion of the document in order to get the needed funds tothen move forward with the real business of poverty al-leviation and economic growth.

21 The IDB review is available at: http://www.worldbank.org/poverty/strategies/review/imfieo1.pdf.

22 Well, maybe not the same offices given the Bank’s andFund’s propensity to constantly “re-organize”, but theidea still stands.

23 See, for example, the brief discussion of sequencing inthe “Overview” chapter of the PRSP Sourcebook,page 18.

24 This is available online, like all the chapters. The over-view of the trade chapter is at: www.worldbank.org/pov-erty/strategies/chapters/trade/trade.htm The entire chap-ter is available at: www.worldbank.org/poverty/strategies/chapters/trade/trad0613.pdf.

25 This project has its own web page. The address is:www.worldbank.org/poverty/voices/index.htm.

26 While much of the data is probably of questionable qual-ity and much of it is not publicly accessible, the list ofhousehold surveys that exist, many of which are less than10 years old, is surprisingly extensive.

27 See www.worldbank.org/poverty/data/index.htm andwww.worldbank.org/poverty/data/census/part2.htm#Findings as well as the links contained there. These arevery useful resource pages for those looking for house-hold-level data.

28 The Bank and Fund will be in a position to make newlycollected data available to the NGO and academic com-munities and should exercise its power to persuade. Insome parts of the world, the household data is not madepublicly available, and there is little justification for thisonce household identifiers have been removed.

References

All references in this paper are available online and wereaccessed on the Internet.

Catholic Relief Services, “Review of the Poverty ReductionStrategy Paper Initiative”, December 2001. www.worldbank.org/poverty/strategies.review/crs1.pdf.

EURODAD, “Many Dollars, Any Change?”, October 2001,www.worldbank.org/poverty/strategies.review/eurodad1a.pdf and www.worldbank.org/poverty/strategies.review/eurodad1b.pdf.

International Monetary Fund, IMF Independent EvaluationsOffice, “Synopsis of Contributions to the PRSP Review”,2001. www.worldbank.org/poverty/strategies.review/imfieo1.pdf.

Jubilee South, Focus on the Global South, AWEPON, and theCentor do Estudios Internacionales, and The World Coun-cil of Churches, “The World Bank and the PRSP: FlawedThinking and Failing Experiences”, 2001. www.worldbank.org/poverty/strategies/review/jsouth1.pdf.

Save the Children UK, “Save the Children UK submission tothe IMF/WB review of PRSPs”, December 2001. www.worldbank.org/poverty/strategies.review/scuk1.pdf.

World Bank, “PRSP Sourcebook”. Draft, April 2001. www.worldbank.org/poverty/strategies/courctoc.htm.

World Bank, “Overview of Poverty Reduction Strategies”,www.worldbank.org/poverty/strategies/overview.htm.Last accessed 21 November 2002.

World Bank, “Poverty Reduction Strategies and PRSPs”,www.worldbank.org/poverty/strategies/index.htm. Lastaccessed 21 November 2002.

World Bank, “Comprehensive Development Framework”, www.worldbank.org/cdf/cdfprsplink.pdf. Last accessed 21 No-vember 2002.

World Bank, “Overview and Background of the Comprehen-sive Development Framework”, www.worldbank.org/cdf/overview.htm. Last accessed 21 November 2002.

World Bank, “IMF/World Bank PRSP Comprehensive ReviewHomepage”, www.worldbank.org/poverty/strategies/re-view/. Last accessed 21 November 2002.

World Bank, “Review of the Poverty Reduction Strategy Paper(PRSP) Approach: Early Experience with Interim PRSPsand Full PRSPs”, 26 March 2002. www. worldbank.org/poverty/strategies/review/earlyexp.pdf. Last accessed 21November 2002.

World Bank, “Poverty Reduction Strategy Sourcebook, Over-view of Trade Policy”, www.worldbank.org/poverty/strat-egies/chapters/trade/trade.htm. Last accessed on 21 No-vember 2002.

World Bank, “Voices of the Poor”, www.worldbank.org/pov-erty/voices/index.htm. Last accessed 21 November 2002.

World Bank, “Household Surveys for Poverty Monitoring”,www.worldbank.org/poverty/index.htm. Accessed on20 August, 2002.


G-24 Discussion Paper Series*

Research papers for the Intergovernmental Group of Twenty-Four on International Monetary Affairs

No. 20 February 2003 Devesh KAPUR Do As I Say Not As I Do: A Critique of G-7 Proposalson Reforming the Multilateral Development Banks

No. 19 December 2002 Ravi KANBUR International Financial Institutions and InternationalPublic Goods: Operational Implications for the WorldBank

No. 18 September 2002 Ajit SINGH Competition and Competition Policy in EmergingMarkets: International and Developmental Dimensions

No. 17 April 2002 F. LÓPEZ-DE-SILANES The Politics of Legal Reform

No. 16 January 2002 Gerardo ESQUIVEL and The Impact of G-3 Exchange Rate Volatility on Felipe LARRAÍN B. Developing Countries

No. 15 December 2001 Peter EVANS and Organizational Reform and the Expansion of the South’sMartha FINNEMORE Voice at the Fund

No. 14 September 2001 Charles WYPLOSZ How Risky is Financial Liberalization in theDeveloping Countries?

No. 13 July 2001 José Antonio OCAMPO Recasting the International Financial Agenda

No. 12 July 2001 Yung Chul PARK and Reform of the International Financial System andYunjong WANG Institutions in Light of the Asian Financial Crisis

No. 11 April 2001 Aziz Ali MOHAMMED The Future Role of the International Monetary Fund

No. 10 March 2001 JOMO K.S. Growth After the Asian Crisis: What Remains of the EastAsian Model?

No. 9 February 2001 Gordon H. HANSON Should Countries Promote Foreign Direct Investment?

No. 8 January 2001 Ilan GOLDFAJN and Can Flexible Exchange Rates Still “Work” in FinanciallyGino OLIVARES Open Economies?

No. 7 December 2000 Andrew CORNFORD Commentary on the Financial Stability Forum’s Reportof the Working Group on Capital Flows

No. 6 August 2000 Devesh KAPUR and Governance-related Conditionalities of the InternationalRichard WEBB Financial Institutions

No. 5 June 2000 Andrés VELASCO Exchange-rate Policies for Developing Countries: WhatHave We Learned? What Do We Still Not Know?

No. 4 June 2000 Katharina PISTOR The Standardization of Law and Its Effect on DevelopingEconomies

No. 3 May 2000 Andrew CORNFORD The Basle Committee’s Proposals for Revised CapitalStandards: Rationale, Design and Possible Incidence

No. 2 May 2000 T. Ademola OYEJIDE Interests and Options of Developing and Least-developedCountries in a New Round of Multilateral Trade Nego-tiations

No. 1 March 2000 Arvind PANAGARIYA The Millennium Round and Developing Countries:Negotiating Strategies and Areas of Benefits

* G-24 Discussion Paper Series are available on the website at: www.unctad.org. Copies of G-24 Discussion Paper Series maybe obtained from the Publications Assistant, Macroeconomic and Development Policies Branch, Division on Globalization andDevelopment Strategies, United Nations Conference on Trade and Development (UNCTAD), Palais des Nations, CH-1211 Geneva 10,Switzerland (Fax: +41-22-907.072; E-mail: [email protected]).

the world bank’s poverty reduction strategy paper approach...

Documents