from crisis to private sector led growth - the world...
TRANSCRIPT
Report No. 54123-TR
TURKEY
Investment Climate Assessment
From Crisis to Private Sector Led Growth
May 2010
Europe and Central Asia Region
Document of the World Bank
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
CURRENCY EQUIVALENTS
Currency Unit – New Turkish Lira (TL)
EXCHANGE RATE
May 12, 2010 TL 1.53= USD 1.00
WEIGHTS AND MEASURES
Metric System
FISCAL YEAR
January 1 – December 31
Vice President: Philippe H. Le Houerou Country Director: Ulrich Zachau
Sector Director: Fernando Montes-Negret Gerardo Corrochano
Sector Manager Lalit Raina Task Manager: Donato De Rosa
ACRONYMS AND ABBREVIATIONS
ABGS Secretariat General for EU Affairs KOSGEB Small and Medium Enterprises Development Organization
ABİGEM EU Turkish Business Centers KSS Small Industrial Estates
ACTAL Dutch Advisory Board on Administrative Burden MAS Manufacturing Advisory Services
BILGE Computerized Customs Activity MEKSA Training and Small Industry Support Foundation of
Turkey
BRSA Banking Regulation and Supervision Agency MEP Manufacturing Extension Partnership
CBRT Central Bank of Turkey MLT Medium- and long-term
CC Commercial Code MNC Multinational Corporation
CEDPL2 Second Competitiveness and Employment
Development Policy Loan MPM National Productivity Center
CGF Credit Guarantee Fund MTP Medium Term Program
CIF Cost, Insurance & Freight NIS Networks and Innovation Survey
CMB Capital Markets Board NPAA National Plan for Adoption of the Acquis
CPRR Japanese Council for the Promotion of Regulatory Reform
NPL Non-Performing Loans
DA Development Agency NSC National Steering Committee
D&Q Design and Quality OECD Organization for Economic and Cooperation
Development
DB Doing Business OLS Ordinary Least Squares
ECA Europe and Central Asia OSB Organized Industrial Zones
EFCAS Enterprise Financial Crisis Assessment Survey PMR Product Market Regulation
EIF European Investment Fund R&D Research and Development
ES Enterprise Survey RIA Regulatory Impact Analysis
EU European Union SBA Small Business Administration
EURADA European Association of Development Agencies SCM Standard Cost Model
FDI Foreign Direct Investment SIC Standard Industrial Classification
FOB Free on Board SME Small and Medium Enterprise
FX Foreign Exchange SPO State Planning Organization
GDP Gross Domestic Product TASB Turkish Accounting Standards Board
GERD Gross Expenditures on R&D TESK Merchants and Artisans Confederation of Turkey
GoT Government of Turkey TFP Total Factor Productivity
GOV Governance TFRS Turkish Financial Reporting Standards
GVA Gross Value Added TOBB Union of Chambers and Commodity Exchanges of
Turkey
GVC Global Value Chains TOSYOV Small and Medium Industry Owners and Managers Foundation of Turkey
IAC Investment Advisory Council of Turkey TPI Turkish Patent Institute
ICA Investment Climate Assessment TSI Turkish Studies Institute
ICS Investment Climate Survey TUBITAK Scientific and Technological Research Council of
Turkey
ICT Information and Communication Technologies TÜRKAK Turkish Accreditation Agency
IFRS International Financial Reporting Standards TURKSTAT Turkish Statistical Institute
IIPR Intellectual and Industrial Property Rights USPTO United States Patent and Trademark Office
ISE Istanbul Stock Exchange VEDOP Tax Offices Automation Project
ISO International Organization for Standardization WDI World Development Indicators
ISPAT Investment Support and Promotion Agency of
Turkey WFE World Federation of Exchanges
JASME Japan Finance Corporation for Small and Medium
Enterprises WIPO World Intellectual Property Organization
KADIM Struggle against Unregistered Employment Project YOİKK Coordination Council for the Improvement of the Investment Environment
KKB Credit Bureau of Turkey
ACKNOWLEDGMENTS
This report was prepared in close cooperation with the YOİKK Secretariat. Preliminary and intermediate
findings of the report were presented to YOİKK Steering Committee meetings on June 11 and November
13 2009. Intermediate findings were presented to various stakeholders in a series of roundtables held in
November 2009. The World Bank team also performed field visits to Izmir and Adana, where it greatly
benefited from discussions with local stakeholders.
The World Bank team was led by Donato De Rosa and included Dragana Pajovic, Paulo Correa, Murat
Seker, Federica Saliola, Delia Rodrigo, Carlos Piñerúa, Jorge Peña, Alvaro Escribano, Manuel de Orte,
Baris Dincer, Irem Guceri, Selma Karaman, Ayse Seda Aroymak, Zeynep Lalik, Erkan Erdil, Murat Ucer.
The report was undertaken under the guidance of Ulrich Zachau, Fernando Montes-Negret, Gerardo
Corrochano, Lalit Raina and Keiko Sato. Indispensable contribution and insight came from İbrahim H.
Çanakcı, Undersecretariat of Treasury, Cavit Dağdaş, Deputy Undersecretariat of Treasury, Berrin
Bingöl, Murat Alıcı, Mehmet Dündar, Özge Dumlupınar, Serenay Usta, Gamze Özdurgutlu, Gönül Bakır
Kartal, Bahar Konak, Başak Ünal and Can Gürlek (Undersecretariat of Treasury team). In addition to the
Treasury team, recognition is extended to all YOİKK member institutions. Willem van Eeghen and Stefka
Slavova (World Bank) and Rauf Gönenç (OECD) were peer reviewers for the report. In addition, the team
received helpful comments and suggestions from Mark Roland Thomas, Cihan Yalcin, Kamer
Karakurum-Ozdemir, Mediha Agar, Muammer Komurcuoglu, Jesko Hentschel, Cristobal Ridao-Cano,
Raif Can, Steen Byskov, Jean-Louis Racine, Cemile Hacibeyoglu, Andres Federico Martinez, Mahesh
Uttamchandani and Anthony Ody.
TABLE OF CONTENTS EXECUTIVE SUMMARY ............................................................................................................................................i
OVERVIEW AND POLICY OPTIONS ...................................................................................................................... ii
Chapter 1. CHALLENGES FACING THE TURKISH BUSINESS SECTOR ............................................................. 1
1.1 Macroeconomic Setting ....................................................................................................................................... 4
1.2 Effects of the Global Crisis on the Corporate Sector: Preliminary Evidence ...................................................... 9
References .......................................................................................................................................................... 12
Chapter 2. EVOLUTION OF THE TURKISH INVESTMENT CLIMATE ............................................................... 13
2.1 The Investment Climate and Business Sector Performance .............................................................................. 13
2.2 The Regulatory Environment ............................................................................................................................ 21
2.3 Labor Market and Skills .................................................................................................................................... 36
2.4 Innovation ......................................................................................................................................................... 45
2.5 Access to Finance and Corporate Governance .................................................................................................. 52
References .......................................................................................................................................................... 62
Annex 2-A. Econometric Methods ..................................................................................................................... 64
Annex 2-B. Investment Climate Variables Used for the Econometric Analysis ................................................ 76
Chapter 3. PROMOTING SME GROWTH ................................................................................................................ 79
3.1 Patterns of Firm Growth in Turkey ................................................................................................................... 80
3.2 Investment Climate Constraints to SME Growth .............................................................................................. 84
3.3 Enhancing the Ability of SMEs to Access Bank Credit .................................................................................... 87
References .......................................................................................................................................................... 94
Annex 3-A. Descriptive Statistics and Econometric Analysis ........................................................................... 95
Chapter 4. STIMULATING KNOWLEDGE FLOWS .............................................................................................. 102
4.1 Production Networks and Knowledge Flows .................................................................................................. 104
4.2 Increasing the Absorptive Capacity of SMEs.................................................................................................. 110
References ........................................................................................................................................................ 117
Annex 4-A. Econometric Methodology and Results ........................................................................................ 120
Annex 4-B. International and Domestic Trade Flows: A Regional Perspective .............................................. 121
Chapter 5. ENHANCING REGULATORY CAPACITY FOR BETTER REGULATION ...................................... 125
5.1 Regulatory Policies and Institutional Drivers in Turkey ................................................................................. 128
5.2 Administrative capacities to prepare new regulations ..................................................................................... 134
5.3 Administrative capacities to review existing regulations ................................................................................ 136
5.4 Policy Options for Regulatory Reform............................................................................................................ 143
References ........................................................................................................................................................ 147
Annex 5-1. OECD Principles of Regulatory Quality and Performance ........................................................... 148
Annex 5-2. The Sources of Law in Turkey ...................................................................................................... 149
Figures
Figure 1-1: 5-year CDS spreads, Jan 2007-Dec 2009 .................................................................................................... 1
Figure 1-2: Short-term external corporate debt in Turkey, million USD....................................................................... 1
Figure 1-3: GDP by sector in Turkey, 1998-2009, current prices, billion TL ............................................................... 6
Figure 1-4: Value added as share of GDP by sector, 2008 ............................................................................................ 6
Figure 1-5: Turkey‘s exports by industry 2000-2009 .................................................................................................... 7
Figure 1-6: Turkey‘s exports by partner country, 2000-2009 ........................................................................................ 7
Figure 1-7: Exports by technology level, share of GDP, 1996-2008 ............................................................................. 8
Figure 1-8: FDI net inflows/GDP, 2009 ........................................................................................................................ 8
Figure 1-9: Effects of the Crisis ................................................................................................................................... 10
Figure 1-10: Sales in the corporate sector: June 2008 - June 2009 .............................................................................. 10
Figure 1-11: Structure of corporate liabilities .............................................................................................................. 11
Figure 1-12: Firms‘ survival strategies, forms of adjustment ...................................................................................... 11
Figure 2-1: Major obstacle levels for firms in Turkey ................................................................................................. 15
Figure 2-2: The IC and TFP: Cross-country comparison ............................................................................................ 16
Figure 2-3: TFP in Turkey ........................................................................................................................................... 17
Figure 2-4: IC percentage contributions to aggregate and average productivity ......................................................... 19
Figure 2-5: Share of Investment Climate effects on the sample mean of employment and probability of exporting
and receiving FDI ........................................................................................................................................................ 20
Figure 2-6: Firms perceiving tax rates and tax administration as a major or very severe obstacle, by country........... 22
Figure 2-7: Tax rates as an obstacle, by firm size region and industry ........................................................................ 23
Figure 2-8: Corporate tax rates 2005-2009, Turkey and comparator countries, percent ............................................. 23
Figure 2-9: Tax wedge, 2008 ....................................................................................................................................... 24
Figure 2-10: Share of firms facing informal competition ............................................................................................ 25
Figure 2-11: Development of product market regulation since 1998, Turkey and comparator countries ................... 29
Figure 2-12: Days to obtain operating license, country comparison ........................................................................... 30
Figure 2-13: Days to obtain operating license, by region, city, size, ownership and exporting status ........................ 30
Figure 2-14: Days to obtain import license, by country .............................................................................................. 31
Figure 2-15: Days to obtain import license, by size, ownership, exporting, region, city and industry ........................ 31
Figure 2-16: Days to clear customs for imports........................................................................................................... 32
Figure 2-17: Access to land as an obstacle, by city ..................................................................................................... 32
Figure 2-18: Access to land as an obstacle, by firms size, ownership, exporting status and industry ......................... 33
Figure 2-19: Days to obtain a construction-related permit, by country ....................................................................... 33
Figure 2-20: Number of inspections per year, selected countries ................................................................................ 34
Figure 2-21: Average time spent dealing with each inspection (days), by firm size ................................................... 34
Figure 2-22: Labor regulations as an obstacle, country comparison ........................................................................... 38
Figure 2-23: Temporary employment as a share of total employment, country comparison ....................................... 40
Figure 2-24: Labor regulations as an obstacle, by size, ownership and exporting status ............................................ 41
Figure 2-25: Population with at least upper secondary education, percent, 2007 ........................................................ 42
Figure 2-26: Distribution of permanent workers, by type and firm size, 2005 ............................................................ 42
Figure 2-27: Distribution of permanent workers, by type and firm size, 2008 ............................................................ 42
Figure 2-28: Education of workforce as an obstacle, by country ................................................................................ 43
Figure 2-29: Education of workforce as an obstacle.................................................................................................... 43
Figure 2-30: Share of firms offering formal training to employees, by country .......................................................... 44
Figure 2-31: Share of firms offering formal training to employees, by firm size, ownership form, exporting status,
region and industry ...................................................................................................................................................... 44
Figure 2-32: Share of production and non-production workers that received training, country comparison .............. 44
Figure 2-33: R&D expenditure as a share of GDP - total, private and public, 2000-2008 .......................................... 46
Figure 2-34: R&D investment as a share of total sales, 2005 and 2008 ...................................................................... 46
Figure 2-35: Patent applications, per million population, 2000-2008 ......................................................................... 47
Figure 2-36: Share of firms developing new products and R&D spending, by firm size, ownership form, exporting
status and region .......................................................................................................................................................... 48
Figure 2-37: Share of firms using technology licensed by foreign firms, country comparison ................................... 49
Figure 2-38: Share of firms using foreign licensed technology, by size, foreign ownership, exporting status and
region ........................................................................................................................................................................... 49
Figure 2-39: Share of firms with internationally recognized quality certifications, country comparison .................... 50
Figure 2-40: Share of firms with internationally recognized quality certifications, by region .................................... 50
Figure 2-41: Use of websites in communication with clients and suppliers, country and small firm comparison ...... 51
Figure 2-42: Use of ICT in communication with clients and suppliers, by size, ownership form and exporting status
..................................................................................................................................................................................... 51
Figure 2-43: Access to finance as an obstacle , country and small firm comparison .................................................. 53
Figure 2-44: Turkish banks by ownership, 2005 and 2009 .......................................................................................... 54
Figure 2-45: Market capitalization of firms listed on the Istanbul SE/GDP, 2005, 2007, 2008 and comparator
countries ...................................................................................................................................................................... 55
Figure 2-46: Firms listed on the ISE, by industry, 2009 .............................................................................................. 55
Figure 2-47: Claims on the private sector, % of GDP, 2008 ....................................................................................... 56
Figure 2-48: Private sector credit by recipient (biannually, billion TL) ...................................................................... 56
Figure 2-49: Share of firms with loans, country and small firm comparison .............................................................. 56
Figure 2-50: Share of firms with loans, by region ....................................................................................................... 56
Figure 2-51: Sources for finance of fixed assets, country comparison ........................................................................ 57
Figure 2-52: Private credit coverage, percentage of adults .......................................................................................... 58
Figure 2-53: Public credit coverage, percentage of adults ........................................................................................... 58
Figure 2-54: Firms having financial accounts externally audited, country comparison .............................................. 59
Figure 2-55: Share of firms for which collateral was required for their latest loan, country comparison ................... 59
Figure 2-56: Type of collateral, 2005 and 2008 .......................................................................................................... 60
Figure 2-57: Protecting investors index, scale 1-10 ..................................................................................................... 61
Figure 3-1: Employment growth at different sizes and ages (2004-2007) ................................................................... 81
Figure 3-2: Growth rates relative to micro firms (2004-2007) .................................................................................... 82
Figure 3-3: Percentage of firms that are above 16 years old ....................................................................................... 82
Figure 3-4: Growth rates of firms: Cross-country comparison .................................................................................... 83
Figure 3-5: Single most severe investment climate obstacles, by firm size ................................................................. 84
Figure 3-6: Access to Finance for SMEs ..................................................................................................................... 85
Figure 3-7: Loans to SMEs .......................................................................................................................................... 86
Figure 3-8: Guarantees provided by the CGF as a percentage of the total SME loan volume .................................... 91
Figure 3-9: Flow of funds in guarantee agreements under treasury support ................................................................ 92
Figure 4-1: Industrial clusters in Turkey ................................................................................................................... 103
Figure 4-2: Value chain governance modes............................................................................................................... 107
Figure 4-3: Governance modes features, across regions ............................................................................................ 108
Figure 4-4: Governance modes features, across industries ........................................................................................ 109
Figure 4-5: Governance modes features, by firm size ............................................................................................... 109
Figure 4-6: Trade and governance modes .................................................................................................................. 110
Figure 5-1: Phases of the preparation of laws in Turkey ........................................................................................... 130
Figure 5-2: Institutional capacity for managing regulatory reform ........................................................................... 133
Figure 5-3: Explicit program for reducing administrative burdens ........................................................................... 137
Tables
Table 1-1: Turkey: Key Economic Indicators, 2002-2009 ............................................................................................ 5 Table 2-1: Summary of econometric methods ............................................................................................................. 13 Table 2-2: IC variables‘ relative contributions to TFP in 2008 and 2005 (percent) .................................................... 18 Table 2-3: Investment climate variables‘ relative contributions to sample averages of employment, exporting and
FDI in 2008 and 2005 (percent) ................................................................................................................................... 20 Table 2-4: Summary of the effects of the regulatory environment in 2008 ................................................................. 21 Table 2-5: Bankruptcy procedures (duration and recovery rate) comparison of selected countries ............................ 36 Table 2-6: Summary of the effects of labor and skills ................................................................................................. 37 Table 2-7: Key labor market indicators, Turkey and selected OECD countries, 2008 ................................................ 37 Table 2-8: Summary of the effects of innovation ........................................................................................................ 45 Table 2-9: Turkey's patent applications, 1995-2008 .................................................................................................... 47 Table 2-10: Summary of the effects of finance and corporate governance ................................................................. 53
Table 2-11: Structure of the financial system in Turkey, 2002-2009 .......................................................................... 54 Table 2-12: Protecting investors index, scale 1-10 ...................................................................................................... 61 Table 3-1: Growth rates by industry and region (2004-2007) ..................................................................................... 81 Table 3-2: Summary of investment climate effects on firm growth, percent .............................................................. 87 Table 3-3: Comparison of CGFs .................................................................................................................................. 93 Table 4-1: Summary of investment climate effects on knowledge flows .................................................................. 111 Table 5-1: Product Market Regulation - Barriers to Entrepreneurship 2008: Regulatory and administrative opacity
................................................................................................................................................................................... 126
Boxes
Box 1-1: Turkey‘s Income Gap ..................................................................................................................................... 2 Box 1-2: Recommendations in key reform areas from the 2007 Investment Climate Assessment ............................... 3 Box 1-3: The Enterprise Financial Crisis Assessment Survey (EFCAS)..................................................................... 10 Box 2-1: The 2008-2009 Enterprise Survey ............................................................................................................... 14 Box 2-2: Tax reforms .................................................................................................................................................. 23 Box 2-3: The Government‘s struggle against informality ........................................................................................... 26 Box 2-4: Development Agencies (DAs) ...................................................................................................................... 27 Box 2-5: Indicators of product market regulation (PMR) ........................................................................................... 29 Box 2-6: Employing workers in Turkey: rules of hiring, work schedules and redundancy ......................................... 39 Box 2-7: Labor market reforms in Turkey ................................................................................................................... 40 Box 2-8: Technology and innovation reforms ............................................................................................................. 52 Box 2-9: The Turkish banking sector in the wake of the crisis ................................................................................... 55 Box 3-1: Data for the analysis of firm growth ............................................................................................................. 80 Box 3-2: Main obstacles to SME lending: views from banks and SME representatives ............................................. 88 Box 3-3: The Credit Guarantee Fund (CGF) ............................................................................................................... 91 Box 3-4: Credit guarantee schemes: International comparison ................................................................................... 92 Box 4-1: Global value chains .................................................................................................................................... 105 Box 4-2: The ―Networks and Innovation‖ Survey (NIS) ........................................................................................... 106 Box 4-3: SME support programs: The international experience ............................................................................... 114 Box 5-1: Regulatory reform to improve the business environment: International experiences ................................. 129 Box 5-2: International examples of oversight bodies for regulatory reform .............................................................. 131 Box 5-3: Coordination between levels of government .............................................................................................. 134 Box 5-4: Comprehensive administrative simplification efforts in OECD countries .................................................. 137 Box 5-5: Benefits of licensing reform ....................................................................................................................... 138 Box 5-6: Dealing with construction permits .............................................................................................................. 139 Box 5-7: The use of SCM: International experiences ................................................................................................ 141
i
EXECUTIVE SUMMARY
This Investment Climate Assessment draws on the analysis of firm-level survey data collected during
April 2008-January 2009, supplemented by other sources, to provide a comprehensive and up-to-date
description of the investment climate facing Turkish firms of all size classes, including the impact of
government regulations and recent reforms. An important feature of the analysis is extensive use of data
from comparable countries to benchmark Turkey‘s performance. Beyond description, the report seeks to
identify key priority areas where further policy reform and institutional development could help
strengthen Turkish firms‘ performance in such areas as productivity, export competitiveness and
employment creation. A special aspect of the report is its focus on Turkey‘s small and medium scale
enterprise (SME) sector.
Since late-2007, global conditions have taken their toll on the Turkish business sector. Turkey‘s
economy contracted by 4.7 percent, with unemployment reaching 14 percent in 2009. Sustainable growth
in the post-crisis environment will require continuation of reforms aimed at promoting healthy business
sector development.
Continued commitment to business environment reforms will help support a sustainable recovery. Building on recent reforms, actions to improve the regulatory framework need to be sustained to reduce
incentives for firms to remain informal. Business registration has been simplified, but administrative
procedures still impose a high ―time tax‖ on firms. Improved availability of skilled labor will be crucial
for improving productivity. Sustained encouragement of firm-level innovation would also have positive
effects on enterprise performance. The availability of credit to the corporate sector, especially SMEs, has
been negatively affected by the crisis. With a competitive global environment expected in the post-crisis
period, the report identifies three priority areas to help the economy achieve sustainable, broad-based
growth that incorporates SMEs and is more evenly distributed across the country.
A first key priority is to alleviate the, mainly financial, obstacles that constrain SMEs‟ growth, thus
preventing the largest portion of the enterprise sector from reaping scale economies. A healthy SME
sector could improve employment opportunities and promote regional development. SMEs grow more
slowly than micro or large firms in Turkey, unlike SMEs in comparator countries. Existing policies and
regulations may impact SMEs more than either micro or large firms. Access to finance appears to be the
single most important constraint to SME growth. Enhancing banks‘ ability to assess borrowers‘
creditworthiness, plus a more active Credit Guarantee Fund, should help ease lending to SMEs.
A second priority is to increase Turkish SMEs‟ competitiveness by enhancing their ability to adopt
and use knowledge. Diffusion of the sources of growth beyond firms that are already sufficiently
competitive to be direct exporters (and beyond already-successful manufacturing poles) will increase
Turkey‘s resilience to future external shocks in global demand and ensure that the productive base of
Turkish manufacturing is more evenly distributed geographically. The local business and institutional
environment combines with country-wide features to determine firms‘ incentives to adopt innovative
modes of production and organization. Analysis of Turkish production networks indicates that the
absorptive capacity of local suppliers is key for successful participation in global markets. Existing
government programs aimed at increasing the operational capabilities and absorptive capacity of SMEs
could be improved at the local level in line with international best practice.
A third priority is to further reform and strengthen the regulatory capacity of the government. Turkey‘s recent important steps in establishing institutions and mechanisms for regulatory reform could
be made more effective by refining their strategic vision, improving horizontal and vertical coordination,
and enhancing consultation with the private sector. The establishment of Development Agencies (DAs)
offers an opportunity to ease investment climate constraints through actions at the local level.
ii
OVERVIEW AND POLICY OPTIONS
1. Since late-2007, adverse changes in global conditions have taken their toll on Turkey‟s
previously booming economy and business sector. Prior to the global economic crisis, Turkey‘s
economy had been thriving. Following the 2001 banking crisis associated with a sharp recession and a
restructuring of the financial sector, Turkish GDP growth averaged nearly 7 percent per annum between
2002 and 2007. An important engine of growth was private investment, in part driven by large capital
inflows, which contributed to a trebling of private sector Gross Fixed Capital Formation between 2002
and 2008. Since 2008, though, the external economic environment has deteriorated markedly, with falls in
external demand and international capital flows – and associated declines in domestic demand and credit
availability. Turkey‘s economy contracted by 4.7 percent, with unemployment reaching 14 percent in
2009.The corporate sector, in particular, has been hard hit by the slowdown in global demand. A survey
carried out by the World Bank in the summer of 2009 shows that most enterprises experienced a sharp
contraction in sales, with reported declines between 2008 and 2009 in the region of 40 percent. Almost
half of the firms surveyed (46 percent) reported restructuring their liabilities, while one-third delayed
payments to tax authorities and suppliers.
Tough global conditions heighten the need to attack constraints to firms’ productivity and growth.
2. Turkish firms cite a number of external constraints to their own performance. According to
a survey conducted between April 2008 and January 2009, a majority of Turkish firms see themselves as
held back by problems with access to finance (some 26 percent of firms cited this as their single most
important constraint). Tax rates (18 percent) and political instability (18 percent) rank second and third,
while other important factors are
competition from the informal sector
and an inadequately educated
workforce (15 and 9 percent
respectively). Analysis of survey data
confirms a significant association
between the quality of the investment
climate and performance in areas like
productivity, job creation, export
competitiveness and attractiveness for
foreign investment. Productivity
analysis shows that almost one-third
of variation in the performance of the
business sector in Turkey is explained
by investment climate factors.
3. The regulatory environment is the area of the investment climate with the largest relative
contribution to productivity. Other relevant investment climate areas include infrastructure bottlenecks,
access to finance and corporate governance, the availability of skilled labor and innovation. Aggregate
productivity in Turkey appears to have increased since 2005, driven by improved allocation of resources
towards firms with higher productivity. This has widened the gap between low and high productivity
establishments, with larger firms appearing to benefit from the more positive aspects of the investment
climate and smaller and less productive firms bearing the costs of its less positive features.
4. Analysis of survey data indicates that firm-level productivity in Turkey is negatively
associated with a number of features of the regulatory environment. Some of these include formal
bureaucratic requirements, such as the number of inspections to which businesses are subjected, the
number of compulsory certificates required and the time necessary to obtain them, as well as time
Figure 1: Top Five Investment Climate Obstacles
Source: Turkey ES 2008
25.9
18.2 17.514.7
9.1
0
5
10
15
20
25
30
Access to Finance
Tax Rates Political instability
Practices Informal Sector
Inadequately educated workforce
iii
consuming customs procedures for imports. Burdensome regulatory requirements constitute fertile
breeding ground for the negative consequences of corruption, as it is exemplified by the negative
association between productivity and informal payments to obtain power supply or a contract with the
government. Inefficient regulations also provide incentives for firms to remain partially informal. Firms
that are subject to competition from informal establishments are, in turn, associated with lower levels of
productivity.
5. Problems for Turkish firms posed by tax rates and tax administration seem to have
declined in importance since 2005, at least partly reflecting the impact of recent reforms. When
examining firms‘ perceptions, the relative importance of tax rates in a ranking of obstacles has dropped,
from being the largest obstacle in 2005, to where a relatively low 18 percent of firms in 2008 perceived
tax rates as the single most relevant
impediment to business operations.
The share of enterprises identifying
tax rates as a major or very severe
constraint decreased from 81 percent
in 2005 to 50 percent in 2008. Tax
administration, viewed as a major
constraint by 59 percent of
manufacturing firms in 2005, had
dropped to 19 percent by 2008. These
improvements between 2005 and
2008 can, at least in part, be ascribed
to tax reforms introduced since 2006.
Notable among the reforms are the
introduction of a new corporate tax
code, the reduction of corporate income tax from 30 to 20 percent, and lower taxation on interest. On
another front, despite the lower tax rates introduced with recent reforms and the fall in the headline
measure of informality – from 53 percent to 44 percent between 2004 and 2008 – the share of surveyed
firms complaining about informal competition increased from 44 percent in 2005 to 52 percent in 2008.
6. Turkey has made significant
reforms in some areas of the regulatory
climate, including easing business
registration, but red tape still imposes
significant costs on businesses. The
Government has made progress towards
facilitating the process of business
registration. According to Doing Business
2010, recent Turkish reforms have reduced
the time it takes to register a business from
13 required steps in 2004, down to six steps
in 2009. The Government has initiated e-
Judicial Registrations and Online Company
Registration procedures, with a draft act currently being evaluated by the Prime Minister‘s Office and
expected to be adopted in the National Assembly in early 2010. Manufacturing firms interviewed in the
2008 enterprise survey reported that the time it takes to acquire an operating license had decreased from
66 days in 2005 to 62 days in 2008. Even after the improvement, though, this is still higher than in
comparator countries. More broadly, compliance with administrative procedures remains problematic for
business operation. There is no clear framework for streamlining administrative procedures for
businesses, and operating licenses are issued by various Ministries, each responsible for different business
Figure 2: Share of firms facing informal competition
Source: Turkey ES 2008
Figure 3: The „time tax‟
Source: Turkey ES 2008
33%35%
43% 44%49%
52% 52% 54% 55%
0%
10%
20%
30%
40%
50%
60%
Poland 2009
Romania 2009
Czech Rep. 2009
Turkey 2005
(manuf)
Hungary 2009
Turkey 2008
(manuf)
Turkey 2008 (all)
Bulgaria 2009
Brazil 2009
9% 9% 10% 11%13% 13%
19%
27%
0%
5%
10%
15%
20%
25%
30%
iv
areas. Additionally, in overall terms the perceived „time tax‟ – the share of management time spent
dealing with government regulation – appears to have increased sharply since 2005, from 9 to 27 percent.
The time tax is largest for medium and large enterprises (32 percent and 34 percent respectively), whereas
managers in small firms report spending 23 percent of their time dealing with red tape.
7. A specific area of improvement has been in the cost and time invested in construction of
business premises. According to the Doing Business‘ measure of the time involved in building a
warehouse, Turkish firms were able to shorten their average site development time by 44 days between
2005 and 2009. This said, the time to obtain building permits varies significantly between Turkish cities,
with firms in Istanbul seeming to struggle with construction permits more than firms located elsewhere.
Additionally, a variation is notable by firm size, with SMEs spending nearly twice the time as large firms
dealing with construction permits (60 versus 32 days).
8. Business inspections appear to be somewhat less burdensome in Turkey than in comparator
countries. The average time all employees in a company spend with inspections per year is 6.6 days,
which puts Turkey ahead of several comparator economies. The survey results are similar when firms are
asked about the number of inspections taking place in their organization: the average annual frequency of
inspectors‘ visits has diminished from 4 in 2005 to 2 in 2008. When examining the time that each firm
spends on inspections, medium size firms seem to be more affected. Additionally, regional comparisons
show significant variations in inspection times and duration. In general, the large regional variations
observed for licenses, permits and inspections are a consequence of the fact that municipalities, often
lacking adequate capacity, are frequently responsible for implementing regulations that are established at
the central level. This institutional setup tends to create a burden for businesses and individuals, when
having to comply with procedures established locally without prior agreement at the central level.
Better availability of skilled labor would help improve Turkish firms’ productivity.
9. Firms with a higher share of staff with university education tend to show higher
productivity, according to econometric analysis. Survey results show that larger firms are in a better
position to afford skilled staff with university education.
10. Education levels in Turkey lag behind other OECD countries. OECD data show that 26
percent of the Turkish adult population holds secondary education diplomas. This is well below the
OECD average of 69 percent and the EU19 average of 70 percent. Graduates with tertiary education in
Turkey are also scarce, and the entry rate to higher education programs (tertiary education) is low by
international standards. Only 29 percent
enroll in higher education in Turkey,
compared to a 56 percent average in the
OECD countries. Nearly a quarter of
Turkish firms rate the education and
skills levels of the workforce as a
―major‖ or ―very severe‖ constraint on
operations and growth (See Figure).
Although this is an improvement from
the 33 percent in 2005, the high rate still
requires the attention of policymakers
and shows that measures need to be taken
to better coordinate labor supply with the
demands in the business sector.
Figure 4: Education of workforce as an obstacle
Source: Enterprise Surveys 2008
6%
21% 21%25%
29%33%
36%
42% 43%
0%
10%
20%
30%
40%
50%
Hungary 2009
Bulgaria 2009
Turkey 2008
(manuf)
Turkey 2008 (all)
Czech Rep. 2009
Turkey 2005
(manuf)
Poland 2009
Chile 2006
Romania 2009
v
11. The share of Turkish firms offering formal training to their employees has increased
slightly. Some 24 percent of manufacturing businesses reported offering training to their employees in
2005, and this number had risen to 29 percent three years later. The share of large firms offering training
is three times higher than that among small businesses. Enterprises with exporting activities are also more
active in offering training to workers.
12. Turkey‟s rate of labor force participation, at less than 50 percent, is low by international
standards and has decreased somewhat since 2005. This places Turkey around 20 percentage points
below OECD and EU-15 averages. The employment rate for women is particularly low, standing at 26.7
percent in 2007, far below the OECD and EU-15 averages of 61.3 percent and 65.3 percent respectively.
The 2008 Enterprise Survey shows that on average only 16 percent of production employees and four
percent of non-production employees are female.
13. The 2008 survey found firms in Turkey less likely to regard labor regulations as a serious
obstacle than they had been in 2005, probably an effect of the timing of the survey. Turkey‘s score
on this indicator appears to have improved substantially vis-à-vis the average for the Europe and Central
Asia region. The survey results require cautious interpretation, however, as the inflexibility of labor
regulations is still mentioned by enterprises in face to face interviews as the overarching constraint facing
firm operation and growth. While firms‘ perceptions in the 2008 survey might have been influenced by
labor reforms initiated in early-2008, it is also possible that the timing of the survey (from April 2008 to
January 2009) found business more preoccupied with other, more immediate issues – e.g. loss of market
share requiring downsizing or problems with access to finance – thus decreasing the perceived relative
importance of labor regulations. These concerns may resurface as a major constraint for sustainable
recovery.
Productivity and exports would benefit from policies encouraging innovation.
14. Turkish firms that invest in Research and Development (R&D) tend to show higher
productivity levels, according to analysis of the 2008 survey of Turkish enterprises. Firms that had re-
organized production processes to take advantage of outsourcing were also found to be more productive.
Employment, exports and FDI are all positively associated with innovation at the firm level. Econometric
analysis also points to a positive association between employment levels and the use of ICT in
communication with customers and suppliers. Firms with quality certifications are also more likely to
have a larger workforce. Finally, there is a significantly positive correlation between variables reflecting
innovation (such as quality certification and the use of ICT) and the probability of exporting.
15. Turkey‟s total investment in
R&D has nearly doubled over the past
ten years, reaching 0.73 percent of GDP
in 2008. This is also reflected in the
relative high number of Turkish firms (23
percent) that perform R&D expenditures
(See Figure).Nonetheless, Turkey still lags
behind other middle income countries and
the OECD, which presented an average of
2.29 percent in 2007, compared to 0.71
percent for Turkey in the same year.
16. Turkey‟s application of
international quality standards (ISO 9001) has shown remarkable improvement over the past
decade, with more than 13,200 certificates issued by the end of 2008. This performance compares
Figure 5: Share of firms with R&D spending
Sources: Enterprise Surveys
11%
16%19%
22% 23%
28%
0%
5%
10%
15%
20%
25%
30%
Hungary Poland Romania Czech Rep. Turkey Bulgaria
vi
relatively well to that of other economies. Firm surveys in 2008 found 30 percent of Turkish firms
reporting having an internationally recognized quality certification. This puts Turkey ahead of other
middle-income countries, such as Brazil (26 percent), Bulgaria (20 percent) and Poland (17 percent).
Certifications among small firms lag far behind medium and large companies: about 55 percent of large
firms hold a quality certification, which is three times the share of small firms.
17. The Government has taken steps to encourage the use of ICT. With the implementation of the
e-Transformation Turkey Project, the expansion of ICT in public services has received a boost. The
Government has also stepped up initiatives to raise awareness of ICT among citizens and businesses.
Further support is planned for enterprises in their use of ICT, as well as increased competition in the
electronic communication sector.
Credit availability to firms, especially SMEs, has suffered during the crisis.
18. Turkish firms with good access to finance tend to show higher productivity. Analysis from
2008 indicates that higher productivity was related to several variables representing financial soundness
(e.g., firms with a higher share of sales paid for before delivery, and firms with the ability to finance a
higher proportion of fixed assets purchases with internal funds).
19. Turkey‟s financial sector is relatively small by comparative standards. According to a recent
study by the Banks Association of Turkey, the ratio of financial assets to GDP in 2007 was 150 percent in
Turkey, compared to 246 percent for emerging market economies and a global average of 421 percent.
This said, according to the 2008 enterprise survey, 57 percent of Turkish firms had access to a loan,
which compares fairly well to other middle income economies (see Figure). Nonetheless, firms of all size
categories perceive access to finance as their single most severe obstacle, with medium-sized firms
appearing to be particularly affected (34 percent), followed by micro (26 percent), small (24 percent) and
large firms (19 percent). Turkish firms rely more on bank loans for investment financing (38 percent) than
do firms in other countries. This is
especially true for medium-sized firms,
for which bank finance accounts for 47
percent of total investment funding.
Collateral requirements appear
particularly onerous for SMEs,
compared to both micro and large
firms, amounting to 100 percent of
loan value for small firms and 91
percent for medium firms. The share of
loan applications that is rejected is also
substantially higher for SMEs (17
percent) compared to large firms (12
percent).
20. The credit crunch following the global financial crisis has affected lending to the SME
sector. Starting in late 2007, SMEs‘ share in total credit declined by about 5 percentage points to just over
20 percent, while SMEs‘ share in total corporate credit dropped from about 52 percent to some 44
percent. While growth in total banking sector credit remained relatively high until the escalation of the
global crisis in late 2008, SME credit growth started to decelerate as early as the beginning of 2008.
Between December 2006 and November 2009, cumulative growth in SME credit amounted to some 35
percent, which was only about half the rate of growth in other (non-SME) corporate credit. Non-
performing loans SMEs rose from below 4 percent in the middle of 2008 to almost 8 percent.
Figure 6. Share of firms with loans
Source: Enterprise Surveys
40% 43% 46% 47% 50%57% 57%
65%69%
35% 42% 40% 43%48% 50% 49% 46%
65%
0%
20%
40%
60%
80%
Bulgaria 2009
Hungary 2009
Turkey 2005
(manuf)
Czech Rep. 2009
Poland 2009
Turkey 2008 (all)
Turkey 2008
(manuf)
Brazil 2009
Chile 2006
All firms Small firms
vii
Actions in three priority areas will support broader-based growth that incorporates SMEs and is better
distributed geographically.
A first key priority is to alleviate the, mainly financial, obstacles that constrain the growth of Turkish
SMEs.
21. Difficult access to finance prevents the largest element in the Turkish enterprise sector from
reaping scale economies. SMEs account for 79.4 percent of employment, 44.6 percent of total
investments, 67.4 percent of total sales, 25-30 percent of total exports, 57.3 percent of total value added
and 25 percent of bank credit (indeed, given data limitations and the size of the informal sector, the
contribution of SMEs to the economy may well be somewhat underestimated). Given SMEs‘ scale, the
development of a more productive and more outward-oriented SME sector is a crucial development
challenge for Turkey. A healthy SME sector can not only provide increased employment opportunities for
a rapidly increasing workforce and promote regional development, but is also crucial to increasing the
resilience of the economy to future external shocks.
22. Turkish SMEs grow more slowly than other firms, the opposite of international experience.
Analysis of firm dynamics
indicates that small (11-50
employees ) and, especially,
medium firms (51-250
employees) grow more slowly
than all other size categories,
with employment growth 16
percent lower than micro firms
and 5 percent lower than large
firms. This is contrary to what
is observed in comparator
countries, where SMEs grow
faster than large firms.
Comparison with other
countries also shows that
SMEs in Turkey are, on
average, older. This is especially true for medium-sized firms, with 60 percent in Turkey being more than
16 years old, compared to 20 percent in the EU-10. This might indicate that SMEs in Turkey face barriers
to their expansion that force them to remain at a smaller – and suboptimal – scale of operations. By
contrast, the demographics of Turkey‘s large firms are in line with values in other countries. The slower
growth of Turkish SMEs suggests that existing policies and regulations may have more distortionary
effects for SMEs than for either micro or large firms. It seems likely that SMEs have neither the capacity
of large firms nor the flexibility of micro firms to cope with the effects of these policies.
23. Problems with access to finance seem to be the most important constraint to the growth of
SMEs. According to econometric analysis, one percent more usage of external finance for investment is
related with 0.3 percent higher employment growth. The association of a loan or a line of credit with
employment growth is even stronger and is estimated to have an effect on employment growth of 33
percent.
24. Improved ability on the part of banks to assess borrowers‟ creditworthiness, plus targeted
interventions to ease collateral requirements, could help ease financial constraints to SME growth.
Structural measures to enhance the ability of banks to assess the creditworthiness of SME borrowers
appear necessary to help SMEs to tap into bank credit. Such measures would include (i) encouraging the
Figure 7: Growth rates (percent) of Turkish SMEs relative to SMEs in
other countries (2004-2007)
Source: Turkey ES 2008
-8.5
-9.8-8.8
-7.6
-10.8 -10.6-9.8
-11.2
-12.7
-10.7
-7.7
-11.9-12.4
-11.3
-14
-12
-10
-8
-6
-4
-2
0
Russia Poland Ukraine Romania ECA EU-8 EU-10
Small (11-50) Medium (51-250)
viii
expansion of coverage of existing credit bureaus, the Credit Registry of the Central Bank and the Credit
Bureau of Turkey (KKB), and (ii) accelerating the adoption of the new Commercial Code, in order to
enable SMEs to benefit from a simplified set of financial reporting standards.
25. Enhancing the role of the Credit Guarantee Fund (CGF) may also help improve SMEs‟
credit access. The CGF has played an important role in facilitating SMEs‘ access to credit by easing
collateral requirements (especially since its recapitalization in 2007). The new CGF model, with Treasury
involvement for a period of two years, is a positive initiative that expands the capacity of the CGF to
serve the financing needs of SMEs following the credit crunch in the aftermath of the crisis. Considering
ways to make the new CGF scheme more active would, hence, be a priority. Furthermore its scope could
be enhanced to better target the needs of medium-sized firms. In fact, since 1994, the bulk of credit
guarantees provided by the CGF has benefited micro and small enterprises, with only 11 percent of the
guarantee fund being used by medium-sized firms.
A second priority is to enhance SMEs’ ability to adopt and use knowledge.
26. Access to the sources of higher efficiency needs to be extended to a wider share of firms. Turkey needs to open up the sources of growth beyond firms that are already sufficiently competitive to
be direct exporters (and beyond already-successful manufacturing poles). Success in this effort will
increase the resilience of the Turkish economy to future shocks in global demand. It will also ensure that
the productive base of Turkish manufacturing is more evenly distributed across Turkish regions. Since the
1980s, the liberalization of the Turkish economy has offered new opportunities related to the general rise
in trade in intermediate goods and in international capital mobility. Integration into global trade and
investment flows has been accompanied by a significant spatial transformation of the Turkish economy,
characterized by the emergence of a number of new industrial agglomerations far from the earlier
manufacturing regions. These new centers are the so-called ―Anatolian Tigers.‖ Clusters of industries
have formed in various parts of the country, with specialization in both traditional and more
technologically advanced sectors, and have become the core of manufacturing and exporting activities. In
response to these developments, the government has activated a number of instruments to foster the
ability of SMEs to participate in global markets. The rationale of several such interventions has been to
remove obstacles to the competitiveness of SMEs related to the business environment. For instance,
SMEs in the manufacturing sector have been encouraged to locate in appropriately planned ―small
industrial estates" (KSS) and ―organized industrial zones‖ (OSB) that can ease investment climate
constraints by providing a number of advantages in terms of infrastructure services and regulation of
business activity.
27. The local business and institutional environment combines with country-wide features to
determine firms‟ incentives to adopt and use innovative modes of production and organization. The
availability of a research base at the local level can, for example, encourage innovative behavior, if
contacts exist between firms and local research organizations. The availability of a skilled workforce is
also highly dependent on the quality of the local higher education and vocational training. The ease of
access to bank finance is, on its part, conditional on the development of the local banking sector, as well
as on personal contacts that may facilitate relational lending practices. Local conditions also influence the
effect that the regulatory environment has on firm operations, since a large number of operating licenses
are awarded at the local level. As a result, the effects in terms of knowledge transfer of linkages between
globally connected firms and local suppliers may vary widely depending on local conditions. Analysis of
Turkish production networks indicates that the absorptive capacity of local suppliers, especially SMEs –
i.e. their ability to adopt and use knowledge – is key for successful participation in global markets.
Specifically, together with a more efficient regulatory environment and easier access to finance for
investment, the availability of technical skills and capacity ―to handle technology‖ is conducive to more
knowledge-intensive value chain arrangements.
ix
28. In addition to wider investment climate reforms, scope exists to improve local-level
programs aimed at increasing the operational capabilities and absorptive capacity of SMEs.
Several governmental and non-governmental organizations provide support to firms, especially SMEs,
with the objective of increasing their operational capabilities and absorptive capacity. The largest
government program is offered by the Small and Medium Scale Enterprises Development Organization
(KOSGEB), while the Union of Chambers and Commodity Exchanges of Turkey (TOBB) also provides
such services to its associates. Following international best practices, the government could aim at
reforming existing support programs by: (i) advancing the implementation of a flexible and decentralized
management model to better serve the needs of SMEs on a local level; (ii) ensuring that the services on
offer are not already available on market terms to SMEs, in order not to crowd out private providers; (iii)
rationalizing the services on offer to create a single entry-point which could help SMEs better understand
their business needs and opportunities; (iv) expanding the scope of support schemes beyond micro firms
to better cater to the needs of larger SMEs.
A third priority is to further reform and strengthen the regulatory capacity of the government.
29. Turkey has taken important steps to improve the regulatory environment. The Government
has paid particular attention to establishing institutions and mechanisms for regulatory reform; enacting
legal reforms conducive to simplification of the legal framework; and introducing, through pilot projects,
a number of regulatory tools to improve the quality of regulations. In this process, achieving EU
harmonization has been a key driver of reform and Turkey has partially embraced the EU Better
Regulation agenda in a number of areas. As a result, Turkey has established solid pillars of a regulatory
system that have the potential to develop into a ―whole-of-government‖ approach to regulatory
management and reform. The Coordination Council for the Improvement of the Investment Environment
(YOİKK) has become a key structure where the private sector makes contributions to the process of
improving the investment climate. The Council conducts its agenda with the help of 12 Technical
Committees working on specific issues with participation of both public and private institutions.
However, the different responsibilities allocated to institutions are not always linked towards a single
regulatory reform strategy. This creates difficulties when it comes to establishing priorities and taking the
lead for reform, and it often also results in overlapping responsibilities within and across levels of
government that make implementation cumbersome, thus directly affecting business operation.
30. Building on recent progress, regulatory reform could aim at a clearer strategic vision,
improved horizontal and vertical coordination among levels of government, and enhanced
consultation with the private sector. In order for regulatory reform to produce substantial effects on the
regulatory burden experienced by the business sector, a number of steps are necessary. First, there is a
need for support at the highest political level that is translated into a clear, coherent and comprehensive
strategy for regulatory reform (country-wide and including all components of a regulatory system).
Second, coordination among different institutions and of different initiatives with similar objectives could
be improved. Third, there is a need to link regulatory reform to clear and measurable economic targets
and objectives in the medium and long term. Fourth, capacity building across the administration remains a
crucial element for success. Efforts in different directions to train officials in the use of modern regulatory
tools testify to the need to dedicate resources to this goal. Fifth, consultation with private sector
stakeholders should be mandatory and institutionalized, with the existing YOİKK platform offering a
good starting point.
31. Complementing other reforms, the establishment of Development Agencies (DA) could offer
an opportunity to ease investment climate constraints by providing an interface for businesses at
the local level. The Government, through the State Planning Organization, is currently in the process of
making Development Agencies operational, with the goal of having 26 DAs that will cover the entire
x
country. DAs are potentially well placed to be an interface between business and government, provided
that they can preserve a light organizational structure and remain at arm‘s length from the government,
with the objective of minimizing the risk of capture by local interests. Additionally, an essential condition
for DAs to be able to perform their tasks will be the existence of internal Government regulations
recognizing their formal role vis-à-vis central and local government. Looking ahead, DAs could perform a
number of useful functions. First, as already intended by the Government, they could act as “one-
window” shops. Even short of radical reform of responsibilities for the issuance of licenses and permits,
DAs could perform a useful facilitator role between issuing agencies and firms. Second, DAs could act as
information points for businesses, in close cooperation with TOBB as well as with Government agencies,
such as KOSGEB and TÜBITAK. The objective would be to rationalize financial and non-financial
support initiatives – especially for SMEs who normally face high information costs – available at the local
level. Third, the FDI promotion function via Investment Support Offices could be carried out in close
coordination with ISPAT, the national FDI promotion agency.
xi
Summary of Policy Objectives and Options
Objectives Options
Ease constraints on the ability of
SMEs to grow in size and generate
employment
Encourage the expansion of coverage of existing credit bureaus.
Accelerate adoption of the new Commercial Code in order to
enable SMEs to benefit from a simplified set of financial reporting
standards.
Consider ways to make the new CGF scheme more active, expand
its reach, and allow it to better reach the medium-sized firm
segment.
Increase the “absorptive capacity”
of SMEs at the local level
Advance the implementation of a flexible and decentralized
management model for SME support programs.
Ensure that the services on offer are not already available on
market terms to SMEs in order not to crowd out private providers
of such services.
Create a single entry-point for the various SME support programs,
possibly in cooperation with DAs, in order to help SMEs better
understand their business needs and opportunities.
Expand the reach of support programs beyond micro-enterprises,
to better serve the needs of larger SMEs.
Improve the regulatory
environment for businesses
Map current regulatory reform initiatives in a single strategic
document setting priorities and sequencing of reforms.
Strengthen the institutionalization of regulatory reform by creating
a single oversight body for regulatory reform in the Prime
Minister‘s office.
Strengthen YOİKK‟s role to improve the business environment and
advocate for regulatory reform.
Design a comprehensive administrative simplification strategy
with clear objectives, targets and review criteria for lower levels of
regulation to improve the business environment.
Improve coordination mechanisms inside the administration when
preparing laws and regulations.
Strengthen coordination and cooperation among levels of
government.
Make consultation with stakeholders compulsory for the
preparation of new and amended laws and regulations.
Continue implementation of Regulatory Impact Analysis (RIA).
Use existing e-government strategies to support regulatory reform
and simplification efforts.
Improve access to information on
regulatory requirements and
government support programs
Development Agencies (DAs) could serve as
―one-window‖ shops, without radical reform of competences for
the issuance of licenses and permits.
information point for businesses, in close cooperation with
business associations with a local presence, such as TOBB, as well
as with Government agencies, such as KOSGEB and TÜBITAK,
etc.
FDI promotion, with Investment Support Offices acting in close
coordination with ISPAT.
1
Chapter 1. CHALLENGES FACING THE TURKISH BUSINESS SECTOR
1.1 The global crisis is posing new challenges to the Turkish business sector. The external
economic environment for developing countries has deteriorated markedly since 2008. World economic
growth in 2009 was negative (-0.6 percent), with the Euro Area, Turkey‘s main export market, expected
to experience a contraction in GDP of -4.1 percent.1 As a result, the demand for Turkish exports has fallen
dramatically with serious consequences for industrial production. Exports fell by 23 percent in 2009 and
industrial production by 9.6 percent compared with a year earlier. In addition to slowing demand for
exports, capital flows to Turkey have fallen dramatically from USD 50.3 billion in 2007 to USD 14.7
billion in 20082 and risk premia have escalated significantly, as shown by the CDS spread for Turkey
from 167 basis points end 2007 to 200 in late 2009 (Figure 1-1). During the period of high growth from
2002-07 Turkey relied heavily upon net capital inflows. The corporate sector now faces major challenges
in continuing to attract external finance and in rolling-over its short term external debt, as shown by
Figure 1-2.
Figure 1-1: 5-year CDS spreads, Jan 2007-Dec 2009 Figure 1-2: Short-term external corporate debt in
Turkey, million USD
Source: Bloomberg Source: CBRT
1.2 Effects on the growth prospects of the Turkish economy are likely to be durable. Turkey‘s
economy contracted by 4.7 percent in 2009, and unemployment increased to 14 percent from 11 percent
in the previous year. The Government‘s Medium-Term Program (MTP) sets out a realistic
macroeconomic framework that, conservatively, foresees a rather slow recovery scenario.3 This mirrors
projections for the world economy and for Turkey‘s major trading partners, and foresees a return to
potential growth of the order of 5 percent only by 2012. This growth is associated with a decline in
unemployment of only 1.5 percentage points from the peak of 14 percent in 2009. Economic activity is
projected to recover weakly to 3.5 percent in 2010, 4.0 percent in 2011, and 5.0 percent in 2012. The
growth process is expected to be led by the private sector, with an expected pick-up in private gross fixed
capital formation to 8 percent in 2010.
1.3 Continued reform of the investment climate is key to mitigating the effects of the crisis and
closing the income gap with more developed countries. The MTP outlines a post-crisis reform agenda
for shared growth that follows the development axes established by the Ninth Development Plan for
2007-2013. Many of the planned actions aim at making the investment climate more conducive to private
sector led growth, based on the notion that the investment climate – ―the set of location-specific factors
1 IMF World Economic Outlook Database, April 2010 2 Capital inflows including net errors and omissions, excluding change in official reserves and IMF credits (World Bank data). 3 On September 16, 2009, the government announced its new Medium-Term program (MTP). The MTP was followed on
September 18 by the more detailed Medium-Term Fiscal Plan. The MTP gives aggregate fiscal targets for the period 2009-12.
0
200
400
600
800
1000
Jan
-07
Ap
r-0
7
Jul-
07
Oct
-07
Jan
-08
Ap
r-0
8
Jul-
08
Oct
-08
Jan
-09
Ap
r-0
9
Jul-
09
Oct
-09
Bas
isp
oin
ts
0
10,000
20,000
30,000
40,000
50,000
60,000
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
2
shaping the opportunities and incentives for firms to invest productively, create jobs and expand,‖ as
defined in World Bank (2005) – can significantly impact productivity, growth and economic activity.
Important investment climate reforms contemplated in the MTP include further labor-market reform, tax
administration reform, increased effectiveness of credit guarantees, expansion of education and vocational
training and increased credit access for SMEs. Continued commitment to such reforms is crucial to
narrow the income gap between Turkey and more developed OECD economies, which, despite sustained
economic growth since 2002, remains wide (Box 1-1).
4 Labor productivity is measured as real GDP per worker in 2005 constant prices, USD, as share of U.S. value. Labor force
participation rate is measured as the proportion of the population ages 15 and older that is economically active: all people who
supply labor for the production of goods and services during a specified period.
Box 1-1: Turkey‟s Income Gap
Whereas Turkey has been successful since 2001 in securing a reasonable level of macroeconomic stability,
convergence to the per capita income levels of OECD countries has been slow. In 2007, income per capita
was 18 percent of the US level, up from 17 percent in 1960, while the gap relative to the EU15, as European
countries were converging to US levels, has actually widened, with income per capita relative to the average
of the EU15 economies decreasing from 26.5 percent in 1960 to 21.6 percent in 2007.
Figure A
Source: Penn World Table Version 6.3
Source: WDI
The evolution of income per capita is driven by labor productivity and labor force participation (Figure A).4
Whereas low labor force participation appears to be the main driver of slow convergence in per capita
incomes (Panel 2), labor productivity convergence has also been slow, especially when compared to
countries, such as Korea, that had similar levels of labor productivity and income per capita until the 1970s
(Panel 1). Capital intensity and total factor productivity (TFP) are the drivers of labor productivity and are
affected by the incentives to invest and innovate associated with the policy and institutional framework (the
investment climate) in which the business sector operates.
13.3
26.7
0
20
40
60
80
100
19
50
19
52
19
54
19
56
19
58
19
60
19
62
19
64
19
66
19
68
19
70
19
72
19
74
19
76
19
78
19
80
19
82
19
84
19
86
19
88
19
90
19
92
19
94
19
96
19
98
20
00
20
02
20
04
20
06
%
1. Labor productivity as share of U.S. Value
EU15 Korea Turkey
58.3
50.4
40
45
50
55
60
65
70
75
80
19
80
19
81
19
82
19
83
19
84
19
85
19
86
19
87
19
88
19
89
19
90
19
91
19
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
%
2. Labor force participation rate
United States EU15 Korea Turkey
3
1.4 The objective of this report is to explore empirically the role of the investment climate in
determining the business sector‟s economic performance. The report uses the information contained in
a 2008-2009 enterprise survey of Turkish enterprises to estimate the effects of the business environment
on firm-level productivity and various measures of enterprise performance, such as the ability to generate
employment or the probability to export and attract foreign capital. The main advantage of this type of
surveys is that information is gathered directly from firms‘ managers on the quality of the physical and
institutional infrastructure, as well as on basic economic performance measures at the firm level.
Improved business sector performance, in turn, translates into aggregate economic improvements, with
increased firm-level productivity and capital intensity translating into higher aggregate productivity, thus
contributing to the reduction of income per capita differences.
1.5 Both in 2004-05 and 2008-09 total factor productivity in Turkey is highly correlated with
the investment climate. Analysis of firm-level data collected in 2008 and 2009 and presented in this
report finds a strong association between firm-level productivity and the investment climate, confirming
the results of the 2007 Investment Climate Assessment based on 2004 and 2005 data. Both assessments
pinpoint reforms aimed at improving the business environment by analyzing a number of key economic
variables – productivity, employment, wages, exports and FDI – in relation to a number of investment
climate variables. The 2007 study also identified labor productivity as the most critical challenge for the
convergence of Turkey‘s income per capita, showing how it accounts for 80 percent of the per capita
income gap between Turkey and the EU-15. Labor productivity improvements, in turn, are achieved by
Box 1-2: Recommendations in key reform areas from the 2007 Investment Climate Assessment
Regulatory environment
Streamlining of business entry and exit
Reform of the firm registration, licensing and inspections regimes
Reform of taxation
Reform of access to land for business
Streamlining of public institutions involved in customs procedures
Labor market and skills
Reform of the fiscal and institutional framework governing labor
Increasing the flexibility of labor market regulations
Reform of the education and training system, in order to match workers‘ skills with the needs of
enterprises
Innovation, technology adoption and ICT
Legal and institutional reform of the National Innovation System
Reform of the IPR legislation and alignment with the EU
Reform of the telecommunications sector
Adoption of quality standards and certification
Improving the legal and institutional framework governing firm access to and enforcement of
standards
Access to finance and corporate governance
Strengthening financial reporting and credit information on firms
Increasing the use of collateral for lending transactions
Enhancing the legal and institutional framework for corporate governance
Access to infrastructure
Institutional reforms and capital investments in the electricity and transport sectors
Institutional reforms affecting the availability of trade-related services
4
increasing: (a) capital per laborer, that is, investments in physical assets such as machinery, infrastructure,
and buildings, and (b) TFP, the residual contribution from factors such as technology adoption, labor
skills, educational achievement, and competition among – and management of – firms.
1.6 Investment climate reforms implemented since 2007 acquire renewed significance in the
wake of the recent crisis. Consistent with the Ninth Development Plan (2007-2013) and with the
recommendations of the previous investment climate report (Box 1-2) the Government has undertaken a
number of structural reforms in several areas of the business environment. Notable measures have
included (i) tax simplification accompanied with a reduction of the corporate income tax rate from 30
percent to 20 percent; (ii) streamlining of procedures for firm start-up; (iii) adoption of a modern FDI
promotion strategy coupled with the restoration of legal certainty for land ownership by foreigners; (iv)
simplification of customs procedures and e-transformation of customs offices with the introduction of a
Computerized Customs Activity System (BILGE); (iv) planned enactment of a new Commercial Code
improving corporate governance, as well as the protection of investors‘ and minority shareholders‘ rights;
(v) reform of R&D legislation aimed at increasing the private sector share of R&D; (vi) first phase of
labor market reform geared to lower non-wage labor costs, accompanied by actions to strengthen the
development of a competency-based skill-building system and continued reforms of curricula in
secondary school. These reforms acquire renewed significance for the business sector in the current
macroeconomic environment.
1.1 Macroeconomic Setting
1.7 In the aftermath of the 2001 crisis sound macroeconomic management and abundant global
liquidity underpinned steady growth. After a banking crisis in 2001 that led to a sharp recession and a
restructuring of the financial sector, GDP growth averaged nearly 7 percent per annum between 2002 and
2007 (Table 1-1). An important engine of growth was private investment, in part driven by large capital
inflows, which contributed to trebling Gross Fixed Capital Formation by the private sector between 2002
and 2008 (from TL 43 billion, or less than USD 30 billion, to TL 152 billion, or USD 117 billion).
1.8 Sustained GDP growth failed to make a visible impact on the unemployment rate inherited
from the 2001 crisis. Capital and exports intensive economic growth, combined with a rapidly growing
and young labor force (an estimated 700,000 workers join the labor force each year), led to a decline in
labor force participation, from 49.6 in 2002 to 46.9 percent in 2008, compared to an OECD average of
70.8 percent.5 Low labor force participation in Turkey reflects particularly low participation among the
female population. Only 26.7 percent of women participated in the labor force in Turkey in 2008,
compared to an OECD average of 61.3 percent and 65.3 percent of women in EU15.
1.9 Since 2008, the global turmoil has placed strains on Turkey‟s economy. Turkish
manufacturing has been hard hit by the drop in global demand, with the effects felt in terms of
unemployment, economic hardships at the household level and poverty. The Turkish economy had
already begun to slow down from 2007. Annual growth in 2007 fell to 4.7 percent from 6.9 percent in
2006. This slowdown was reflected in, among other things, a build-up of inventories by Turkish firms in
the order of 2 percent of GDP between the first quarter of 2007 and the third quarter of 2008.
Unemployment has also risen sharply. After remaining stable at levels below 10 percent for several years,
the unemployment rate in 2009 averaged 14 percent.
5 OECD Employment Outlook (2009) and World Bank calculations.
5
Table 1-1: Turkey: Key Economic Indicators, 2002-2009
2002 2003 2004 2005 2006 2007 2008 2009
Real GDP (bln USD, current
prices) 230.5 304.9 390.4 481.5 526.4 648.8 742.1 617.6
Real GDP per capita (USD) 3,325.9 4,341.4 5,486.7 6.681.4 7,213.9 8,781.8 10,039.9 8,255.1
Private gross fixed capital
formation (bln TL) 43.4 62.0 97.4 115.1 143.3 151.9 152.4 125.8
(Growth rate in percent)
Real GDP Growth 6.2 5.3 9.4 8.4 6.9 4.7 0.7 -4.7
CPI Inflation (%) 45.2 25.3 8.6 8.2 9.6 8.8 10.4 6.3
Long-term interest rate 63.5 44.1 24.9 16.2 18.0 18.3 19.2 11.9
Short-term interest rate 59.5 38.5 23.8 15.6 17.9 18.3 18.9 10.9
Exchange rate 1.5 1.5 1.4 1.3 1.4 1.3 1.3 1.6
(In percent of GDP unless otherwise indicated)
Labor force participation 49.6 48.3 46.3 46.4 46.3 46.2 46.9 47.6
Unemployment 10.3 10.5 10.8 10.6 10.2 10.3 11.0 14.0
Saving-investment balance
Domestic savings 18.6 15.5 16.0 15.9 16.5 15.5 16.9 14.2
Investment (Contributions to
growth)
Public 0.3 -0.6 -0.2 0.7 0.1 0.2 0.4 -0.1
Private 2.0 3.1 5.6 3.1 3.1 0.6 -2.0 -4-4
Fiscal sector
Primary balance (IMF
defined) 3.3 4.8 5.5 5.0 4.6 3.1 1.7 -2.1
Public external debt 7.1 7.5 8.2 8.0 8.1 6.7 6.8 7.8
Monetary indicators
Broad money 39.9 35.2 34.6 40.5 42.4 43.9 48.7 50.01
Claims on private sector 14.5 14.5 17.3 22.2 25.9 29.5 32.6 33.01
External sector
Current account balance (bln
USD) -0.6 -7.5 -14.4 -22.1 -32.3 -38.3 -41.9 -14.0
Trade balance (bln USD) -6.4 -13.5 -22.7 33.1 41.1 -46.8 -53.0 -24.9
Exports (fob, bln USD) 40.7 52.4 68.5 78.4 93.6 115.4 140.8 109.7
Imports (cif, bln USD) 51.6 69.3 97.5 116.8 139.6 170.1 202.0 140.9
FDI 0.5 0.6 0.7 2.1 3.8 3.4 2.5 1.0
CBRT Reserves (bln USD) 28.3 35.3 37.6 50.2 60.7 74.7 72.9 74.8
Source: CBRT, IMF, OECD, SPO, Turkstat, World Bank. 1May 2009
6
1.10 The current account deficit and inflation persistence have now diminished in immediate
importance while the fiscal challenge has increased. Turkey‘s current account deficits are not expected
to return to pre-crisis levels in the medium term, although higher oil prices could still pose a challenge.
Inflation persistence appears to have receded, with inflation in 2009 still below the inflation target. The
monetary framework has not been altered and will continue to pursue unchanged inflation targets (6.5
percent in 2010 and 5.5 percent in 2011), in line with the framework in place since 2006. On the other
hand, public debt and the budget deficit have increased as a result of the global crisis – the budget deficit
stands at -2.1 percent of GDP in 2009 compared to 3.1 percent in 2007 and public sector debt at 7.8
percent. This has sharpened the challenge to fiscal management and underlined the central importance of
the MTP and the associated budgets to reduce economic uncertainty.
1.11 Concerns over external financing were prominent before and as the global crisis hit. Since
the 2001 crisis, the Turkish Treasury has pursued a conservative strategy of financing itself largely using
domestic currency debt instruments, removing much of the foreign exchange risk from the public debt
portfolio. After declining in 2001-05, the external debt-to-GDP ratio increased by more than 4 percentage
points in 2006, driven by corporate sector external borrowing. External debt-to-GDP stood at 43.9 percent
at end-2009. However, the subsequent sharp contraction in economic activity has reduced the overall need
for foreign financing in 2009, while at the same time most large corporations have either retained market
access or been able to draw on their own FX holdings. Turkey‘s foreign financing needs have fallen but
will remain high in 2010-12. The current account deficit is contracted from USD 41.9 billion in 2008 to
USD 14 billion in 2009. A sharp contraction in domestic investment has cut the demand for medium- and
long-term (MLT) borrowing by the private sector, which has been reflected in lower rollover ratios.
Repayments are also lower beyond 2010. The overall financing gap was USD 6 billion in 2009, and thus
should easily be financed through reserves, which were USD 74.8 billion in 2009. Turkey should be in a
position to attract further FDI given privatization efforts and the potential for mergers and acquisitions.
Similarly, net portfolio flows are assumed to be positive, consistent with continued Eurobond issuance
and the potential of the domestic capital market to raise financing. Projections on capital inflows are of
course conservative if compared to the pre-crisis period of record global liquidity.
Figure 1-3: GDP by sector in Turkey, 1998-2009,
current prices, billion TL Figure 1-4: Value added as share of GDP by sector, 2008
Source: WDI
1.12 The services sector has not only been the fastest growing in the past decade, but is also hit
by the economic crisis to a lesser extent than the manufacturing sector. Since 1998, the services
sector has grown rapidly, with an average 30 percent annual growth and increased relative contribution to
GDP from 51 percent in 1998 to 62 percent in 2008 (Figures 1-3 and 1-4). Meanwhile, the industry sector
in Turkey remains small, both in relation to the domestic services sector as well as when comparing to
0
50
100
150
Agriculture Industry Services
0%
20%
40%
60%
80%
100%
Industry Services Agriculture
7
other emerging economies. Merely 28 percent of Turkey‘s GDP came from the manufacturing sector,
with a real decrease from USD 58 billion in 2008 to USD 53 billion one year on.
1.13 Following high export growth since 2002, changed global demand conditions appear to have
affected the destination of Turkish exports. Turkish exports averaged 20 percent annual growth in the
period 2000-2008 (Table 1-1). Exports continue to be dominated by manufactured goods as well as
machinery and transport equipment, each taking up 28 percent of total exports, with a total value of USD
4.8 billion in 2009. The machinery and transport equipment sector in particular has increased its relative
share of exports, from 21 percent in 2000. For a considerable time, the majority of Turkish exports have
been targeting the EU-market, with the relative share of total exports going to the EU27 region averaging
58 percent in the years 2000-2007. This flow has however experienced a significant drop to 46.3 percent
in 2009. A similar negative development has been noticed in exports to the United States, with a decline
from 11.3 percent in 2000 to 3.1 percent in 2009. Instead, the shares of exports to the Middle East and
Africa have increased from 8 to 16.5 percent and from 4.9 to 9.9 percent respectively in the same period.
Figure 1-5: Turkey‟s exports by industry 2000-2009
Figure 1-6: Turkey‟s exports by partner country, 2000-2009
Source: OECD
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Commodities and transactions, n.e.c
Miscellaneous manufactured articles
Machinery and transport equipment
Manufactured goods
Chemicals and related products, n.e.s
Mineral fuels, lubricants and related materialsAnimal and vegetable oils, fats and waxes
Crude materials, inedible, except fuels
Food and beverages
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Other
Ukraine
United States
Russian Federation
Africa
Asia
Middle East
EU27
8
1.14 The technology content of Turkish exports remains dominated by low-technology
manufactures with recent increases in
medium- and high-technology products. 6
Turkey‘s medium- and high-technology
exports have increased significantly, with
real values reaching 26.5 billion in 2008, a
360 percent increase since the turn of the
century. Nonetheless, exports with higher
technology content are a relatively low
contributor to Turkey‘s GDP, averaging 3.4
percent for the past five years. Medium- and
high-technology imports on the other hand
contributed with 8 percent of GDP in 2008,
creating a substantial trade deficit for higher
technology goods and suggesting the need
for further actions in reforming export flows
from Turkey.
1.15 Inward FDI remains weak in an international comparison, with a recent shift away from
telecommunication and finance. Net inflows of FDI into Turkey reached USD 6 billion in 2009, an
inflow well below the USD 15.4 billion
recorded in 2008 as well as the pre-crisis
level in 2007 (USD 19.1 billion). The sharp
decline is to a large extent a combination of
the negative impact that the financial crisis
has had on overall international investment
flows and the stagnation of the large-scale
privatizations that took place in Turkey in
the years 2005 and 2006.7 Most FDI
inflows during this period were in the
financial services and telecommunications
sectors. Three years on, the relative
significance of FDI in these industries has
diminished and instead, electricity, gas and
water supply as well as manufacturing hold the largest shares of investments from non-residents (33
percent and 28 percent respectively). Within manufacturing, the most significant increase in FDI in the
first ten months of 2009 compared to the same period one year prior was in the chemicals and motor
vehicles sectors (from USD 89 million to USD 306 million and USD 64 million to USD 208 million
respectively). Meanwhile, the manufacturing industry has experienced a decline of FDI inflows into the
sectors ‗other manufacturing‘ (USD 1,689 million to USD 382 million) and ‗food and beverages‘ (from
6 Technology level of exports is here defined according to UNIDO‘s methodology classifying manufactured goods into four sub-
categories: resource-based, low-, medium- and high-tech exports, based on the Standards International Trade Classification
(SITC) Revision 3. Examples of resource-based goods include: beverages, cut gems and glass, petroleum/rubber products,
prepared meats/fruits, vegetable oils and wood products. Examples of low-tech manufactures include: clothing, footwear,
furniture, headgear, leather manufactures, plastic products, pottery, simple metal parts/structures, textile fabrics, toys, and travel
goods. Examples of medium-tech manufactures include: chemicals and paints, engines, fertilizers, industrial machinery, iron,
motors, pipes/tubes, plastics, ships, switchgear, synthetic fibers, vehicles and watches. Examples of high-tech manufactures
include: aerospace, cameras, office/data processing/telecom equipment, optical/measuring instruments, pharmaceuticals, power
generating equipment, transistors, turbines and TVs. For detailed technological classifications of exports, see UNIDO (2009). 7 Privatization implementations in Turkey in 2005 and 2006 totaled USD 16.3 billion in domestic and international sales
(Undersecretariat of Treasury)
Figure 1-7: Exports by technology level, share of GDP,
1996-2008
Source: UN COMTRADE, staff calculation
Figure 1-8: FDI net inflows/GDP, 2009
* 2009 est. Source: IMF International Financial Statistics
0%
2%
4%
6%
8%
1996 2000 2004 2008
Low-tech Medium- & High-tech Resource-based
4.2
3.4
2.8
2.7
2.7
1.6
1.4
1.1
1.0
0.5
0.0 1.0 2.0 3.0 4.0 5.0
Romania
Turkey 2007
Russia*
ECA average*
Poland*
Brazil*
Czech Rep.
Hungary*
Turkey
Slovakia*
%
9
USD 939 million to USD 120 million).8 The recent adoptions of a modern FDI promotion strategy by the
Turkish government, along with the restoration of legal certainty for land ownership by foreigners, are
steps in the right direction to improve investment levels in the country.
1.2 Effects of the Global Crisis on the Corporate Sector: Preliminary Evidence
1.16 Four factors appear to have combined to increase the impact of global events on Turkey.9
Although none of these factors is entirely specific to Turkey, their magnitudes and combination go a long
way towards explaining the impact of global events on the real economy in Turkey.
1.17 First, the speed of the economic expansion in the preceding five year period had led to a marked
inventory and capital build-up in Turkish manufacturing, with inventories accounting for fully one
quarter of aggregate GDP growth in 2007 and the first three quarters of 2008, and capital formation
growing at an average of 15.5 percent per year in 2002-2007.
1.18 Second, the high level of uncertainty experienced by all emerging markets during the crisis was
compounded in Turkey by perceived vulnerability to the scarcity of external financing. Turkey‘s most
notable macroeconomic challenge in the pre-crisis period, the current account deficit (averaging more
than 5 percent of GDP over the five years from 2004-08), had been financed by increasing debt-creating
flows to the private sector. By 2009, Turkish corporations needed to roll over total debt amortizations of
the order of USD 100 billion. About a third of this total was not true foreign exposure, since it captured
offshore lending by Turkish banks to their Turkish clients, and a further (hard to quantify) portion was
secured by assets held overseas by Turkish nationals. Nonetheless, uncertainty over the magnitude of this
exposure and its effect on the economy was high at the outset of the crisis.
1.19 Third, Turkey‟s banking system, having weathered several earlier crises, was quick to cut
lending to all but the most creditworthy borrowers. There was no visible slowdown in credit
intermediation to the private sector prior to September 2008. Domestic credit then retrenched significantly
in late 2008. Turkey‘s domestic financial system, although it is well-capitalized and prudential regulations
meet modern standards, is shallow for an economy of Turkey‘s size. After surviving crisis and
restructuring in 2001-02, the Turkish banking sector was understandably conservative faced with the high
uncertainty of late 2008 and early 2009.
1.20 Fourth, the composition of Turkey‟s exports exacerbated the demand shock. Turkey‘s exports
– concentrated in hard-hit sectors such as automotive vehicles, consumer durables, and capital goods and
machinery – made it vulnerable to a dip in export demand. Export volumes in the first half of 2009 were
down 11 percent and this combined with price effects to create a loss in export earnings of more than a
third (y/y). Exports had continued to perform strongly right up to the crisis: FOB export growth in 2008
(y/y) was 23 percent while imports (CIF) rose by 18 percent. Following the crisis, imports have
contracted even faster than exports: as a result, net exports have contributed positively to GDP growth
since the onset of the crisis.
1.21 A crisis impact survey fielded by the World Bank confirms that the global economic and
financial crisis has had a significant impact on the Turkish enterprise sector. The Enterprise
Financial Crisis Assessment Survey (EFCAS) was conducted in the summer of 2009 in six countries
including Turkey (Box 1-3). Its objective was to identify the channels through which the global economic
crisis is affecting the corporate sector and how firms have reacted to the shock.
8 See Undersecretariat of Treasury, International Direct Investment Information Bulletin (December 2009) 9 Restoring Equitable Growth and Employment Programmatic Development Policy Loan (REGE-DPL), draft December 14, 2009
10
Box 1-3: The Enterprise Financial Crisis Assessment Survey (EFCAS)
The Enterprise Financial Crisis Assessment Survey (EFCAS) was carried out in June and July 2009 and
covers 1,686 enterprises from Bulgaria, Hungary, Latvia, Lithuania, Romania, and Turkey. In all countries,
the EFCAS covers a subsample of the 2008-2009 Business Environment Enterprise Survey (ES), carried out
by the World Bank and the European Bank for Reconstruction and Development in 30 economies of Europe
and Central Asia. Turkey‘s EFCAS sample is composed of 514 enterprises.
The ES and EFCAS samples are representative of the universe of non-agricultural private sector formal firms
(with at least five employees) in the economy for groups D, F, G, H, I and subgroup 72 of the United Nations
Statistics Division ISIC Rev. 3.1. Results are estimated through the application of sampling weights – that
denote the inverse of the probability that the observation is included due to the sampling design – to the
original data. Therefore, results are representative of the non-agricultural private economy in Turkey.
1.22 The main channel through which Turkish firms have been affected is a drop in demand. While a large majority of
firms across countries,
including Turkey, declared
contraction in demand, the
most important effect of the
crisis on their business, the
share of enterprises citing a
combination of supply-side
factors in Turkey – notably
input costs, debt levels and
access to credit (about 19
percent) – was one of the
highest among the surveyed
countries.
1.23 Reported declines in volumes of
sales between 2008 and 2009 have been
substantial. Turkish firms reported a
significant drop in sales amounting to an
average 38.7 percent in June 2009 relative
to June 2008, as shown in Figure 1-9. The
decline presents wide sectoral variations;
with fabricated metal products (32.3
percent) and garments (31.3 percent)
being hardest hit compared to the food
and textiles sectors that have been the
least affected with a drop of 13.2 percent
and 16.6 percent respectively. This
development is supported by the industrial
turnover index for the Turkish
manufacturing sector, when measured in the same time period. The sub-sectors of manufacturing showing
most significant reduction in industrial turnover coincide with sectors in the survey reporting largest drop
in sales (basic metals, fabricated metal products and non-metallic mineral products). The overall decline
notwithstanding, a significant proportion of firms (about 15 percent) reported increases in sales (25.4
percent on average) in the same period, which may be a first indication that the crisis may be
accompanied by structural adjustments, with a redistribution of market shares across firms.
Figure 1-9: Effects of the Crisis
Source: EFCAS
Figure 1-10: Sales in the corporate sector: June 2008 - June
2009
Source: EFCAS
70.32
70.77
71.3
75.43
78.12
78.47
0% 20% 40% 60% 80% 100%
Hungary
Lithuania
Turkey
Latvia
Bulgaria
Romania increase the level of debt
increase input cost
reduce access to credit
drop in demand for its products or services
other
-48.4 -48.0
-38.7-35.9 -35.5
-25.4
-60
-50
-40
-30
-20
-10
0
Latvia Lithuania Turkey Romania Bulgaria Hungary
%
11
1.24 Foreign currency
exposure and short term maturity
highlight potential risks for the
corporate sector. Confirming the
accumulation of foreign debt and
the prudent attitude of the banking
system towards lending in the wake
of the crisis, survey findings in the
summer of 2009 indicate that the
share of foreign currency debt in the
corporate sector is considerable (22
percent), while debt maturity is
concentrated in the short-term (66
percent).
1.25 Debt restructuring has been the predominant form of adjustment on the part of firms. As a
response to the current liquidity constraint, about 33.7 percent of Turkish firms delayed payments to tax
authorities and suppliers. At the same time, 45.9 percent of surveyed companies attempted to restructure
their debt while only 0.2% of firms have filed for insolvency or bankruptcy. A large proportion of firms
(25.4 percent) benefited from some form of state-aid.
Figure 1-12: Firms‟ survival strategies, forms of adjustment
Source: EFCAS
4.10.2
8.7
23.4
45.9 47.9
63.365.8
1.4
25.421.0
3.6
33.3
27.4
33.7
50.7 50.5
28.3
0
10
20
30
40
50
60
70
Romania Bulgaria Turkey Lithuania Latvia Hungary
%
% of firms overdue on obligation in last year which have filed for INSOLVENCY/BANKRUPTCY during the last 12 months% of firms overdue on obligation in last year which have RESTRUCTURED any of their outstanding liabilities in the last 12 months% of firms overdue on obligation in last year which have applied for STATE AID in the last 12 months
% of firms which delayed payments to tax authorities or suppliers for more than one week
Figure 1-11: Structure of corporate liabilities
Source: EFCAS
49.2 49.356.9
66.4 69.4
80.1
39.4
21.4
31.221.8
35.3
20.7
0
20
40
60
80
Latvia Bulgaria Romania Turkey Hungary Lithuania
%
average % of total liabilities with a term of maturity less than one year
average % of total liabilities denominated in foreign currency
12
References
Organization for Economic Cooperation and Development (OECD). 2009. OECD Employment Outlook 2009 –
Tackling the Jobs Crisis. Paris
Republic of Turkey State Planning Organization. 2006. Ninth Development Plan 2007-2013. Decision No: 877
Republic of Turkey State Planning Organization. 2009. Medium Term Programme 2010-2012. Official Gazette No.
27351, 16.09.2009
Republic of Turkey, Undersecretariat of Treasury, International Direct Investment Information Bulletin (December
2009)
United Nations Industrial Development Organization (UNIDO). 2009. Industrial Development Report 2009 –
Breaking In and Moving Up: New Industrial Challenges for the Bottom Billion and the Middle-Income Countries.
Vienna
World Bank. 2005. World Development Report 2005: A Better Investment Climate for Everyone. Washington, DC
World Bank, Poverty Reduction and Economic Management, Turkey Country Management Unit, Restoring
Equitable Growth and Employment Programmatic Development Policy Loan (REGE-DPL). Draft December 14,
2009
13
Chapter 2. EVOLUTION OF THE TURKISH INVESTMENT CLIMATE
2.1. Continued reforms of the investment climate are required to ensure that the Turkish
enterprise sector remains competitive in a more challenging global environment. Since 2001 the
Turkish Government has complemented sound macroeconomic management with more or less
comprehensive reforms of various aspects of the investment climate – including taxation, business
registration, customs, FDI promotion, R&D and labor legislation. Since the second half of 2007, changed
global conditions characterized by declining domestic and global demand, reduced international capital
flows and tighter credit conditions have already taken their toll on the Turkish business sector, as outlined
in the preliminary evidence reported in the previous Chapter. Whereas the depth and duration of the
aftereffects of the global turmoil are still uncertain, a number of investment climate areas emerge as
crucial to make the Turkish business sector more resilient to future shocks and more competitive both
domestically and internationally. As argued in this report, these are related to facilitating the growth of
SMEs, intensifying knowledge flows to Turkish firms, and equipping policymakers with a system capable
of ensuring consistent regulatory quality.
2.1 The Investment Climate and Business Sector Performance
2.2. Business sector performance is associated with the quality of the investment climate.
Whereas Turkey has been successful since 2001 in securing a reasonable level of macroeconomic
stability, convergence to the per capita income levels of OECD countries has been slow. In order to help
close the gap in income per capita, it is essential to tackle the underlying policy and institutional
framework that constitutes the investment climate conditions in which the business sector operates.
Econometric analysis of firm level data collected in 2008 and 2009 (Box 2-1) confirms that productivity,
job creation, the ability of exporters to be competitive on international markets and the attractiveness of
the economy for foreign investment are significantly associated with the quality of the investment
climate. Similar results were obtained by analyzing data of a previous survey, conducted by the World
Bank in 2004 and 2005 (World Bank 2007). In addition, the analysis shows that firms with large market
shares tend to cope better with the bottlenecks imposed by the investment climate and are even able to
benefit from its more positive aspects. The analysis is based on the robust methodology summarized in
Table 2-1.
Table 2-1: Summary of econometric methods
Data
base
(survey)
i. Econometric model used to identify IC
effects
ii. Evaluation of IC contributions Economic
performance:
Left hand side or
dependent variables
Right hand side or
explanatory variables
Man
ufa
ctu
rin
g [
cro
ss-
sect
ion
20
08
fo
r IC
var
iab
les
and
rec
all
dat
a fo
r T
FP
]
Productivity (TFP)
Employment
Probability of
exporting
Probability of
receiving FDI
135 investment
climate variables +
other controls and
more than 20
industry/region/size
controls +
simultaneous effects
IC contributions to aggregate productivity through
average productivity and allocative efficiency.
IC contributions to the sample means of
employment, probability of exporting and
probability of receiving FDI.
I. identification and evaluation of IC effects on economic performance in 2008
II. Comparison of IC effects on economic performance from 2005 and 2008
14
Box 2-1: The 2008-2009 Enterprise Survey
This report draws its data from a core enterprise survey that provides a standardized way of measuring and
comparing investment climate conditions. Private contractors conduct the Enterprise Surveys on behalf of
the World Bank in Turkey and in other countries globally, utilizing a standard instrument that enables
comparisons of investment climate conditions across different regions within a given country, as well as of
a country with its peers globally.
Data were collected in Turkey in between April 2008 and January 2009 as part of the fourth round of the
Business Environment and Enterprise Performance Survey (ES), a joint initiative of the European Bank for
Reconstruction and Development and the World Bank. The objective of the survey is to obtain feedback
from enterprises on the state of the private sector as well as to help build a panel of enterprise data that will
make it possible to track changes in the business environment over time, thus allowing, for example, impact
assessments of reforms. Through interviews with Turkish firms in the manufacturing and services sectors,
the survey assesses the constraints to private sector growth and allows estimation of statistically significant
business environment indicators that are comparable across countries, across 5 Turkish regions, across
industries and firms of different sizes.
The sample for the Turkey survey was selected using stratified random sampling. The whole population, or
universe of the study, is the non-agricultural economy. Three levels of stratification were used in Turkey:
type of industry, firm size, and geographic region. The universe was stratified into 11 manufacturing
industries, 1 retail industry and 6 residual industries. Size stratification was defined following the
standardized definition for the rollout: small (5 to 19 employees), medium (20 to 99 employees), and large
(more than 99 employees). For stratification purposes, the number of employees was defined on the basis
of reported permanent full-time workers. This seems to be an appropriate definition of the labor force since
seasonal/casual/part-time employment is not a common practice, except in the sectors of construction and
agriculture. Regional stratification was defined in 5 regions. These regions are Marmara, Aegean, South,
Central Anatolia and Black Sea-Eastern. The Turkey sample contains panel data. The wave 1 panel
consisted of 1325 establishments interviewed in 2005. Note that there are additional variables for location
(city), industry, and size that reflect more accurately the reality of each establishment.
Table 2-2: The ES 2008-09 Sample Industry Region
Marmara Aegean Central Anatolia South
Black Sea - Eastern Total
Food 40 22 41 34 21 158 Textiles 71 47 11 45 5 179 Garments 64 23 12 23 6 128 Chemicals 49 20 19 14 5 107 Plastic and rubber 21 6 2 9 5 43 Non-metallic mineral products 29 37 18 17 9 110 Basic metals 4 1 5 5 4 19 Fabricated metal products 11 3 11 8 5 38 Machinery and equipment 4 1 11 12 6 34 Electronics 3 4 1 3 2 13 Other manufacturing 20 13 21 6 7 67 Retail 51 7 17 18 14 107 Construction 3 3 1 4 2 13 Wholesale 37 10 6 31 6 90 Hotel and restaurants 0 0 0 1 0 1 Transport 3 0 2 3 0 8 IT 0 0 0 1 1 2 Other services 12 2 5 8 8 35
Total 422 199 183 242 106 1,152
The nature of the sample allows exploration of the effects of the investment climate over time and across
regions. Econometric analysis uses the manufacturing subset of the sample (903 establishments). Of these,
425 panel establishments were present in the 2005 survey. This allows assessing the impact for firm
15
performance of the numerous investment climate reforms that have occurred between 2005 and 2008.
Furthermore, the regional stratification of the sample will allow drawing statistically significant inferences
of the effects of the investment climate across five broadly defined regions (Table 2-2). The regions are
sufficiently diverse in terms of institutional settings, economic structure, income per capita and product
specialization to warrant a meaningful assessment of the investment climate at the regional level.
The source of the sample frame was twofold. Universe estimates were taken from the TOBB database
which contains a full list of establishments in manufacturing sectors. TOBB refers to the Union of
Chambers and Commodity Exchanges of Turkey. Universe estimates for service sectors were taken from
the Statistical Institute of Statistics (SIS) with additional information based on SIC code from the Turkish
Studies Institute (TSI). Comparisons were made between estimates in TOBB and SIS to establish that the
two sources are comparable and hence can be used side by side.
Three additional modules have been added to the ES survey with interviews of over 800 manufacturing
firms in July-August 2009. These modules address (i) the effects of the current global financial crisis
referred to in Chapter 1; (ii) firm-level innovation referred to in Chapter 2 and (iii) production networks
used in Chapter 4.
2.3. Access to finance is perceived as the most serious obstacle by Turkish firms, a negative
development since 2005 and a likely consequence of the credit squeeze in conjunction with the 2008-
2009 global crisis. Figure 2-1 depicts average firm responses when asked which elements of the business
environment represent the biggest obstacle faced by the establishment. Despite the objective improvement
in access to credit documented below, it is noticeable that a majority of firms are obstructed by access to
finance in their business (26 percent), a likely effect of the more restrictive credit conditions in the
aftermath of the 2008-2009 global economic and financial crisis. Tax rates (18 percent) and political
instability (18 percent) rank second and third, while other important factors are informal competition and
an inadequately educated workforce (15 and 9 percent respectively). Moreover, firms‘ opinion has shifted
since 2005, where tax rates were identified as the main obstacle to operations and access to finance was
depicted as the second most relevant.10
Scrutinizing the data by firm size, access to finance as an obstacle
becomes even more evident. 29 percent of SMEs consider finance to be the chief obstacle to business. As
will be discussed later in this Chapter, the econometric analysis has shown that many positive effects on
productivity, as well as other key performance measures, derive from a sound financial system and wide
access to finance for firms. It is therefore essential to recognize the negative development that the survey
results present when addressing access to finance in relation to other key variables.
Figure 2-1: Major obstacle levels for firms in Turkey
Source: Turkey ES 2008
10 Due to methodology differences in 2008 and 2005, a comparison over time is only possible for the two largest obstacles, as
perceived by Turkish firms.
26%
18% 18%15%
9%
3% 3% 2% 2% 2%
29%
18%17%
14%
8%
4% 2% 2% 2% 1%
0%
10%
20%
30%
All firms SMEs
16
Productivity
2.4. In 2008 the investment climate continues to be strongly associated with productivity. Total
factor productivity (TFP) is an important driver of aggregate GDP growth in Turkey (World Bank, 2007).
Several factors may influence the evolution of TFP, in other words the level of efficiency and
technological content of production. Amongst these, the investment climate – related to the regulatory
environment, the availability of skilled labor, the system of incentives to develop new products or
experiment with novel production techniques, the ease of access to external finance – plays a critical role.
This is confirmed by analysis of the 2008 enterprise survey of Turkish manufacturing enterprises. Based
on the concept of demeaned TFP, defined as the share of aggregate log-TFP associated exclusively with
the investment climate, and normalizing aggregate log-TFP to be 100, an estimated 31.4 percent of TFP is
associated with the various aspects of the investment climate and the remaining 68.6 percent with other
factors.11
2.5. The productivity of Turkish manufacturing is, overall, negatively influenced by the
investment climate. Figure 2-2 compares the Olley and Pakes (1996) demeaned decomposition of
Turkey‘s TFP with those of other countries, including Turkey in 2005, for which similar estimates are
available. Both in 2008 and in 2005, the aggregate log-TFP of Turkish manufacturing industry is, overall,
negatively influenced by the investment climate. This does not imply that Turkey is less productive than
other countries but that investment climate conditions, on balance, are less conducive to the efficient use
of resources. In other words, negative investment climate factors tend to dominate over positive ones,
indicating that the investment climate available for doing business is preventing the economy from using
resources as efficiently as it could.
Figure 2-2: The IC and TFP: Cross-country comparison
Source: Staff calculations based on Escribano et al. 2009. See Annex 2-A for details
2.6. The impact of the investment climate is asymmetrical with the productivity of larger firms
positively associated with current investment climate conditions. The Olley and Pakes method allows
decomposing aggregate TFP into an average component and an allocative efficiency component. The
former reflects the productivity of the average firm, while the latter provides a measure of the efficiency
with which resources are distributed among producers. Both in 2008 and 2005, the Turkish investment
climate affects aggregate productivity mostly by negatively affecting the average firm. However, in 2008
the effect of the investment climate on the allocative efficiency component is positive, partially
compensating the increased negative effect on the average firm. In other words, the positive effects tend
to be concentrated in – conceivably larger – high market share firms and the negative effects in low
market share ones. As discussed in Chapter 3, the need to increase average efficiency emphasizes the
importance of policy and institutional mechanisms that stimulate the expansion of small- and medium-
sized enterprises.
11 See Annex 2-A.
-0.6
9
-0.5
8
-0.5
6
-0.3
7
0.3
7
0.5
0
0.6
3
0.7
7
1.0
0 1.5
2
-0.7
0
-0.7
0
-0.5
5
-0.4
0
0.1
0 0.5
0
0.6
0
0.7
0
0.5
0
1.5
0
0.0
1
0.1
2
-0.0
1
0.0
3
0.2
7
0.0
0
0.0
3
0.0
7 0.5
0
0.0
2
-1.2-0.8-0.4
00.40.81.21.6
2
De
me
an lo
g-TF
P
(Dem.) Aggregate log-TFP (Dem.) Average log-TFP (Dem.) Allocative efficiency - logs
17
2.7. Improved allocation of resources appears as the main driver of increased aggregate
productivity since 2005. Aggregate log-TFP of the manufacturing sector grew from 2.22 to 2.42 between
2005 and 2008 (Figure 2-3).12
The increase is driven proportionately more by improved allocative
efficiency – from 0.29 to 0.44 – while average productivity has barely changed. This may be a sign of
improved allocation of resources in Turkish manufacturing. The Black Sea-Eastern region is the most
productive only partly because the average firm of this region is slightly more efficient – 9 percentage
points more than the average firm of South and Central Anatolia regions, 13 percent more than in
Marmara and 25 percent over the level of the average firm of Aegean. Most of the greater productivity
derives from a substantially larger allocative efficiency effect, implying that resources tend to be
concentrated in firms with high productivity. This efficiency effect is twice as large as in Central
Anatolia, and more than three times the value in the other regions. Large variations also appear when
examining sectors, with non-metallic mineral products, other manufacturing and garments presenting
higher aggregate log-TFP. In all three cases, it is the allocative efficiency component that is driving
higher aggregate productivity. The basic metals sector, on its part, stands out because its performance is
driven by high average firm productivity.
Figure 2-3: TFP in Turkey
Source: Staff calculations based on Escribano et al. 2009. See Annex 2-A for details
2.8. Investment climate reforms can positively affect both more efficient allocation of resources
and increased average productivity. The importance of the allocative efficiency effect in Turkey
indicates that the productivity improvements of the past few years have been driven by better allocation of
resources towards the most productive firms (Figure 2-3). This emphasizes the persistence of a large gap
between low and high productivity establishments within and across sectors and regions. Given the large
negative association between current investment climate conditions and productivity (Figure 2-2),
continued momentum in investment climate reforms will facilitate the reallocation of resources towards
more productive firms (increasing allocative efficiency), while, at the same time, closing the gap between
more and less productive firms by increasing the productivity of the average firm.
2.9. The regulatory environment is the largest relative contributor to productivity. Table 2-3
presents the relative contributions of various investment climate areas to demeaned aggregate productivity
(the effect that can be associated with the investment climate). In order to avoid cancellations from
positive and negative effects, the individual contributions are considered in absolute value. As in 2005,
12 Due to a number of methodological differences described in Annex 2-A, it is difficult to establish a direct equivalency between
the demeaned aggregate log-TFP in 2005 and 2008. However, the computed results provide a reasonable approximation.
2.2
2
2.4
2
3.4
2
2.6
5
2.4
5
2.2
6
2.0
7
2.8
3
2.7
3
2.6
7
2.5
0
2.4
7
2.3
9
2.1
3
2.1
1
1.8
1
1.9
3
1.9
8 2.2
2
2.0
4
2.0
4
1.9
7
1.7
8
1.9
8
2.0
3
1.7
9 2.0
7
2.0
9
2.2
6
1.8
9
1.8
7
1.9
3
0.2
9
0.4
4
1.1
9
0.6
0
0.4
1
0.2
9
0.3
0
0.8
5
0.7
1
0.8
8
0.4
3
0.3
8
0.1
3
0.2
4
0.2
4
-0.1
2-0.4
0.0
0.4
0.8
1.2
1.6
2.0
2.4
2.8
3.2
3.6
log-
TFP
Aggregate log-productivity Average log-productivity Allocative efficiency - logs
18
the regulatory environment remains the area of the investment climate with the largest relative
contribution to productivity (Table 2-3).13
This motivates the closer examination, in Chapter 5, of the
mechanisms underlying the formation and enforcement of regulations affecting business activity. Other
relevant investment climate areas include infrastructure bottlenecks; access to finance and corporate
governance; availability of labor and skills of the workforce; and quality and innovation capacity.
Table 2-2: IC variables‟ relative contributions to TFP in 2008 and 2005 (percent)
Variable groups Contributions to TFP
(percent)
2008 2005
Regulatory environment 46.82 72.97
Labor and skills 5.06 6.95
Quality and innovation 3.21 4.74
Finance and corporate governance 9.72 10.60
Infrastructure 11.28 4.08
Other control variables 23.91 0.68
Source: Staff calculations based on Escribano et al. 2009. See Annex 2-A for details
2.10. Analysis of individual effects shows that most statistically significant variables have a
negative effect on TFP. Figure 2-4 reports the percentage contribution to aggregate and average log-TFP
of individual variables. Among regulatory variables, days to clear customs to import has a large negative
contribution to average log-TFP (-73.0 percent). Since the contribution to aggregate log-TFP is almost
identical, the negative effect on productivity is evenly distributed among firms. The number of
compulsory certificates has a contribution to average log-productivity of -52.6 percent. The negative
effect is, in this case, amplified by the allocation effect as the contribution of the variable to aggregate
log-TFP is -60.4 percent, indicating that the negative effect is concentrated in high market share firms.
Firms with larger market shares, on average, are required to obtain more compulsory certificates, and
therefore the negative effect of this variable affects firms using a larger proportion of resources of the
economy. Number of tax inspections accounts for -30.4 percent of aggregate log-TFP. The main
contributor is the average effect with -28.1 percent, whereas the allocation effect contributes only -2.3
percent. Therefore, the problem is slightly concentrated in high market share firms, to some extent
amplifying the negative average effect. Informal competition has a contribution to aggregate log-TFP of -
13.8 percent. In this case, the main contribution derives from average log-TFP (-13.1 percent), whereas
the allocation effect accounts for 1.3 percent. That is, the negative effect is somewhat mitigated because
the problem tends to affect low share of sales firms. The percentage of female workers in staff – an
indication of labor intensive activities such as garments or textiles – is negatively associated with
aggregate log-TFP (-14.2 percent). The positive allocation effect is less important as it accounts for only
0.5 percent of the total contribution to aggregate log-TFP. Exporting is positively related with TFP. Out
of the total 13.6 percent contribution to aggregate log-TFP, 11.5 percent is due to the average effect and
only 2.1 percent to allocation, indicating that the positive effect tends to be concentrated on high market
share firms. Age of the firm, which is probably capturing effects such as the vintage of the firms or
learning by doing externalities on TFP, has a clearly positive association with aggregate log-TFP, 78.4
percent. Furthermore, the contribution to average log-TFP and to the allocative efficiency are respectively
68.7 and 9.7 percent, implying that firm age affects more firms with high market shares.
13 Due to changes in the sampling methodology, analysis based on comparison of 2005 and 2008 results, can only provide an
approximation of the effects of changes in investment climate conditions.
19
Figure 2-4: IC percentage contributions to aggregate and average productivity
Notes: Contributions computed according to equation (9) in Annex 2-A. The IC contributions to the allocative
efficiency term are not included in the table. Nonetheless, they can be obtained implicitly as the IC contribution to
aggregate log-TFP minus the IC contribution to average log-TFP. The productivity measure used is the restricted
Solow residual. 2% of upper and lower bounds of productivity are excluded of the calculations.
Source: Staff calculations based on Escribano et al. 2009. See Annex 2-A for details.
Other Performance Measures: Employment, Exports, FDI
2.11. The investment climate is also associated with employment, exporting and foreign direct
investment. As shown in Figure 2-5, investment climate variables account for a sizeable share of the
sample mean of employment, as well as of the probability of exporting and receiving FDI. The
importance of the investment climate is greatest for employment (65.9 percent), it accounts for almost
half of the probability of exporting, while is has a more limited contribution to the probability of receiving
FDI. TFP, on its part, accounts for a large share of the probability of exporting (37.1 percent), while it
plays a more limited role for employment and FDI.
-38
.0
-5.2 -2.1
-0.4
-5.4 -0
.4
-30
.4
-5.1
-60
.4
-13
.8
-72
.1
7.4
7.1
-1.0
15
.3
8.2
5.6 7.3
-14
.2
6.2 1
3.6
-2.3
1.7
78
.4
-40
.7
-4.7
-2.1
-0.7
-7.9 -0
.6
-28
.1
-4.2
-52
.6
-15
.1
-73
.0
6.3
5.5
-2.1
16
.5
5.0
4.0 6.2
-14
.7
5.0 1
1.5
-6.7
4.5
68
.7
-100
-80
-60
-40
-20
0
20
40
60
80
100
1.1 1.2 1.3 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 3.1 3.2 3.3 3.4 3.5 4.1 4.2 5.1 5.2 6.1 6.2 6.3 6.4
% C
on
trib
uti
on
% Contr. To (Dem.) Aggregate log-TFP
% Contr. To (Dem.) Average log-TFP
1. Infrastructure1.1 Duration of power outages1.2 Shipment losses, exports (interaction with dummy for exporter)1.3 Shipment losses, domestic
2. Regulatory environment2.1 Payments for power supply (dummy)2.2 Payments for government contract (dummy)2.3 Security expenses2.4 Tax inspections (number)
2.5 Compulsory certificates (days)2.6 Compulsory certificates (number)2.7 Informal competition (dummy)2.8 Clearing customs for import (dummy)
3. Finance and corporate governance3.1 Sales paid before delivery3.2 New fixed assets finance - internal funds3.3 New fixed assets finance - state-owned banks3.4 Largest shareholder3.5 Subsidies (dummy)
4. Quality and innovation4.1 R&D (dummy)4.2 Outsourcing (dummy)
5. Labor and skills5.1 Staff - female workers5.2 Staff - university education
6. Other control variables6.1 Share of exports6.2 Decreased sales (dummy)6.3 Decreased prices (dummy)6.4 Age
Infrastructure Regulatory environment Finance and corporate governance
Quality and innovation
Other control variablesLabor and skills
20
Figure 2-5: Share of Investment Climate effects on the sample mean of employment and probability of
exporting and receiving FDI
A. Employment B. Exports C. FDI
Source: Staff calculations based on Escribano et al. 2009. See Annex 2-A for details
2.12. The relative importance of regulatory variables for employment, exports and FDI appears
to have decreased since 2005, while the relevance of innovation and skills for exporting has
increased. The total relative weights of the investment climate block of Figure 2-5 can be decomposed
into the contributions of each group of investment climate variables. The absolute percentage
contributions by blocks of variables (summing up to 100%) of Table 2-4 illustrate the variation between
2005 and 2008 of the estimated effects on average log-employment and on the probability of exporting
and of receiving FDI inflows. In the case of employment, regulatory variables lose relative weight (from
46 to 4.2 percent), with, especially, labor and skills, real wages, and TFP gaining in relative importance
for the sample mean of employment. A similar apparent decline in the relative importance of the
regulatory environment can be observed in the case of exporting and FDI. Such a large fall in the bearing
of regulatory variables may be in part due to the methodological differences between the 2005 and 2008
surveys. It is also worth noticing the large increase of the relative contribution of TFP to the probability of
exporting and of receiving FDI, indicating that greater productive efficiency has become more crucial for
the international integration of the Turkish manufacturing sector, as shown by the ability to compete in
international markets and to attract foreign capital. The increase in the relative weight for exporting of
variables related to innovation and quality (from 5 to 21 percent) is also worth mentioning. As argued in
Chapter 4, in order to achieve broad-based business sector development, it is crucial that the investment
climate is propitious to knowledge transfer from more productive and innovative internationally oriented
firms towards less productive and innovative local suppliers.
2.13. In order to trace the evolution of the business environment over the past few years, the remainder
of this Chapter takes stock of various aspects of the investment climate and their effects on business
sector operation and performance. The areas examined include the regulatory environment; the
availability of labor and the level of skills; innovation capacity and technology; as well as access to
finance and corporate governance.
IC : 65.9%
Other variables: 14%
TFP: 3.5%
Wages: 16.6% TFP:
37.1%
IC: 47.4%
Other variabl
es: 15.6%
TFP: 7.2%
IC: 10.1%
Other variables: 82.7%
Table 2-3: Investment climate variables‟ relative contributions to sample averages of employment, exporting
and FDI in 2008 and 2005 (percent)
Variable groups Employment Probability of
exporting
Probability of
receiving FDI
2008 2005 2008 2005 2008 2005
Regulatory environment 4.19 36.27 8.45 41.04 4.83 26.74
Labor and skills 19.15 10.44 0.00 6.10 7.95 38.85
Quality and innovation 7.75 7.68 20.83 5.17 2.94 13.74
Finance and corporate governance 21.74 21.57 2.78 2.12 4.52 0.00
Infrastructure 2.00 0.00 8.76 10.35 5.43 0.00
Real wages 19.82 13.30 0.00 0.00 0.00 0.00
TFP 4.14 2.17 43.91 13.38 43.27 17.83
Other control variables 21.20 8.57 15.27 21.85 31.05 2.83 Source: Staff calculations based on Escribano et al. 2009. See Annex 2-A for details
21
2.2 The Regulatory Environment
2.14. By improving the efficiency and transparency of its regulatory environment the Turkish
economy can diminish the overall risk of informality and boost productivity and growth. Evidence
from econometric analysis of enterprise survey data supports the notion that business sector performance
benefits from well-designed and enforced rules and regulations. While a detailed treatment of the
mechanisms underlying the formation and enforcement of regulations is provided in Chapter 5, this
section compares the development of the regulatory environment in Turkey between 2005 and 2008.
Following examination of the econometric evidence of the effects of the regulatory environment on firm
performance, the section offers a description of recent progress in the areas of taxation, informality, and
administrative procedures.14
Table 2-4: Summary of the effects of the regulatory environment in 2008
Explanatory variable
Dependent variable
Productivity Employment
Probability
of
exporting
Probability
of receiving
FDI
Tax inspections (number) -
Tax inspections (days) -
Business inspections (number) -
Compulsory certificates (days) - -
Compulsory certificates (number) - +
Compulsory certificates (payments) -
Customs clearance for imports (days) -
Payments for power supply (dummy) -
Payments for government contracts - -
Security expenses - +
Informal competition (dummy) - Source: Staff calculations based on Escribano et al. 2009. See Tables in Annex 2-A for detailed results.
2.15. The productivity of Turkish industry is negatively associated with formal and informal
aspects of the regulatory environment. In addition to hampering the day to day operation of the
business sector, cumbersome and inefficiently enforced rules and regulations de facto create an uneven
playing field among producers, for example by implicitly or explicitly favoring larger or more well-
connected firms with greater capacity to handle bureaucratic requirements. This dampens the benefits of
market competition by reducing the incentives for both incumbents and other firms to adopt best practice
production techniques, ultimately resulting in lower overall productivity.15
Such a mechanism may be at
work in Turkey, where firm-level productivity is negatively associated with a number of features of the
regulatory environment (Table 2-5). Some of these include formal bureaucratic requirements, such as the
number of inspections to which businesses are subjected, the number of compulsory certificates required
and the time necessary to obtain them, as well as time consuming customs procedures for imports.16
Burdensome regulatory requirements constitute fertile breeding ground for the negative consequences of
corruption, as it is exemplified by the negative association between productivity and informal payments to
14 To enable comparison between 2005 and 2008, only the manufacturing sector is measured in the econometric analysis, whereas
descriptive data for 2008 also embrace firms in the services industries. Due to differences in the variables used in 2005 and 2008,
some homogeneity is lost when comparing 2005 and 2008 results. Nonetheless, the comparison presents an opportunity to assess
key differences in the way investment climate variables affect firms‘ productivity, employment, propensity to export and to
receive FDI in the two time periods. For a description of econometric methods and for detailed regression results, see Annex 2-A,
as well as the background paper for this report by Escribano et al. (2009). 15 See Conway, De Rosa, Nicoletti and Steiner (2006). 16 The average number of days needed to clear customs to import is significant both in 2005 and 2008. Moreover, the coefficients
are virtually identical, -.175 in 2008 and -.171 in 2005. However the contribution to average log-TFP became more negative,
from -51.6 to -73, indicating that now the average firm suffers more this problem. See Table 2-A-2.
22
obtain power supply or a contract with the government. Inefficient regulations also provide incentives for
firms to remain partially informal. Firms that are subject to competition from informal establishments are,
in turn, associated with lower levels of productivity.
2.16. The regulatory environment seems to have a differentiated effect depending on firm size. For instance, larger firms must obtain a greater number of compulsory certificates but they spend less
time doing so compared to firms with fewer employees.
2.17. The ability to export and the attractiveness for foreign investors of Turkish firms are
negatively associated with formal and informal regulatory requirements. For instance, export
propensity is negatively associated with the number of business inspections and with the informal
payments required to obtain compulsory certificates or a government contract. At the same time, the
probability of receiving FDI is negative associated with the duration of tax inspections.
Taxation 2.18. Despite the fact that tax inspections are still negatively correlated with the productivity of
Turkish firms, tax rates and tax administration are perceived as less burdensome than in 2005.
Econometric analysis of enterprise survey data shows that the high number of tax inspections experienced
by firms has a negative impact on productivity (Table 2-5). However, when examining firms‘ perceptions,
from being the largest obstacle in 2005, the relative importance of tax rates has dropped, with 18 percent
of firms in 2008 perceiving tax rates as
the biggest obstacle to business
operations (Figure 2-1 above). At the
same time, the share of manufacturing
companies identifying tax rates as a
major or very severe constraint to their
business has decreased from 81 percent in
2005 to 50 percent in 2008 (Figure 2-6).17
Tax administration, viewed in 2005 by 59
percent of manufacturing firms as a major
constraint has now dropped to 19 percent.
Worth mentioning is also that only a
fraction of Turkish firms (0.3 percent)
indicate that tax administration is the
biggest obstacle to their business.
2.19. As depicted in Figure 2-7, the share of firms perceiving tax rates to be a major or very severe
obstacle to operations varies across firm size, with small and medium enterprises being more bothered by
tax rates (55 and 51 percent respectively) than large companies (41 percent). Tax rates are also much less
of an obstacle in Black Sea – Eastern and Central Anatolia (42 and 45 percent) than in for instance the
Aegean region (61 percent). Variations across industries can also be noticed, with particularly firms in the
construction sector being bothered tax rates (86 percent).
17 Note that in Figure 2-1 above, firms are asked to specify the one element in the business environment that represents the
biggest obstacle faced by the establishment, whereas Figures 2-6 and 2-7 present the answers of firms when asked to rate how
much of an obstacle each individual business element is to operations.
Figure 2-6: Firms perceiving tax rates and tax administration
as a major or very severe obstacle, by country
Source: Enterprise Surveys
16%25%
41%50% 52%
59% 59%
81% 84%
8%15%
22% 19% 22% 27%
42%
59%
75%
0%
20%
40%
60%
80%
100%
Chile 2006
Bulgaria 2009
Czech Rep. 2009
Turkey 2008
(manuf)
Turkey 2008 (all)
Poland 2009
Hungary 2009
Turkey 2005
(manuf)
Brazil 2009
Tax rates
Tax administration
23
Figure 2-7: Tax rates as an obstacle, by firm size region and industry
Source: Turkey ES
2.20. Since 2006 Turkey has embarked on a comprehensive overhaul of the taxation system. The
improvement in the perceptions of enterprises between 2005 and 2008 can, at least in part, be ascribed to
the tax reforms that have
been introduced since
2006 (Box 2-2). Notable
among these are the
introduction of a new
corporate tax code,
reduction of corporate
income tax from 30 to 20
percent, as well as lower
taxation on interest.
These achievements
were recognized in the
World Bank 2007 Doing
Business Report, where
Turkey resulted as one of
the top reformers in
2006-2007 thanks to improvements in the ―Paying Taxes‖ indicator. The reduction in corporate tax rates
has been significant and has improved Turkey‘s standing relative to a number of European neighbors
(Figure 2-8).
55%51%
41% 42%45% 48%
56%61%
34%39% 42%
45% 46% 46% 48% 49% 51% 53%59% 59%
86%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
Figure 2-8: Corporate tax rates 2005-2009, Turkey and comparator countries,
percent
Source: Doing Business 2010
Box 2-2: Tax reforms
The Turkish government has recognized the importance of lower tax rates for a better business environment
through the introduction of a new corporate tax code in 2006. The corporate tax rate was reduced by ten
points to 20 percent, broadening the tax base and weakening the incentives for firms to work in the
informal sector, and thus boosting overall growth and productivity. Moreover, the tax on interest was
lowered from 18 to 15 percent.
As a part of the Tax Offices Automation Project (VEDOP), an online tax filing system was also introduced,
reducing the time needed for complying with tax regulations by 31 hours as well as allowing companies to
submit tax reforms and payments through the Internet Tax Office. This, along with other simplifications of
the tax mechanism, initiated through Turkey‘s ―Ninth Development Plan 2007-2013‖ will reduce firms‘
compliance costs as well as provide a more efficient tax collection system, connecting a country-wide
30
35
40
45
50
55
60
2005 2006 2007 2008 2009
Hungary
Czech Rep.
Romania
Turkey
Poland
Bulgaria
24
2.21. The tax wedge on labor – the gap between total labor costs and the employee‟s take home
pay – has been reduced since 2005. An important disincentive to increase formal employment is the tax
wedge, representing the non-wage cost of labor. This has also declined in the past three years from 42
percent in 2005 to 38 percent in 2008 for an average income earner, spouse and two children. When
making an international comparison based on an average income earner with a non-working spouse and
two children, Turkey has seen improvements, from previously having the highest taxation among the
OECD countries (OECD 2005). Nevertheless, the tax wedge is still well above the OECD average of 27
percent. Comparing the tax wedge for a single worker with 67 percent of average earnings, the gap to the
OECD average of 34 percent becomes considerably smaller.
Figure 2-9: Tax wedge, 2008
a) average income earner, spouse and two children,
2008
b) single with no children earning 67 percent of average
wage, 2008
Source: OECD
Informality
2.22. Unfair competition from informal firms is highly obstructive of sound business
development. Strongly connected to the topic of tax payments, and one of the critical issues examined in
the 2007 Investment Climate Assessment, is the negative impact that the informal business sector has on
productivity. Despite lower tax rates introduced with recent reforms, which could contribute to reducing
the competitive advantage of informal firms, analysis of the Enterprise Survey shows that productivity is
still negatively associated with competition from informal businesses (Table 2-5).
2.23. In Turkey the underreporting of revenues and wages and the non-registration of workers
with the social security system are the most important forms of informality. Informality is usually
43.9%42.7%
42.0%38.5%
33.7%31.8%
27.3%27.2%
25.4%20.6%
18.1%15.1%
0% 10% 20% 30% 40% 50%
HungaryGreece
Turkey (2005)TurkeyPoland
SpainOECD
PortugalSlovak RepublicCzech Republic
KoreaMexico
46.7%42.0%
40.0%38.7%
37.6%37.6%
36.1%33.8%33.5%32.9%
17.4%10.9%
0% 10% 20% 30% 40% 50%
HungaryTurkey (2005)
Czech RepublicPolandTurkeyGreece
Slovak RepublicSpainOECD
PortugalKorea
Mexico
network of 599 tax, inspector and regional finance offices. The nationwide communications network has
been implemented by the Ministry of Finance in order to streamline administrative procedures and provide
citizens with a secured and efficient online system to file taxes.
Additional tax reforms simplifying the activities of Turkish firms include amendments to the R&D law and
allow for deduction of all R&D expenditures from the taxable income, with additional tax benefits after
capitalization of profits. Moreover, firms with more than 500 R&D personnel and a 50 percent increase in
R&D expenditure, compared to the year prior can deduct a further 50 percent of their revenues for tax
purposes.
Amendments to the Personal and Corporate Income Tax Codes have also been made in order to cover anti-
abuse legislation, restricting remittances to tax havens. Furthermore, business telecommunication became
more cost effective in March 2009, as the Special Communication Tax was reduced to 5 percent from a
previous 15 percent for land lines and 25 percent for mobile networks.
25
defined as legal economic activity taking place below the radar of government, covering categories that
include unregistered firms, registered firms engaging in unrecorded economic activity, and the self-
employed. Among the many categories, unrecorded sales (to evade taxes) and underreported/unrecorded
employment (to avoid labor regulations and contributions) are the most common in Turkey. Although the
headline measure of informality has been falling, this is almost entirely explained by migration out of
agriculture into more formal sectors. The headline measure of informality fell from 53 percent to 44
percent between 2004 and 2008. 18
However, this aggregate decline hides important patterns. Most of the
decline between 2001 and 2006 is explained by migration of the workforce out of agricultural
employment (where nearly all workers are informal) to manufacturing and services, mainly in urban areas
(where informality rates are below 20 percent for wage earners). Moreover, during this period of rapid
expansion of the economy, urban and non-agricultural informality increased (from 29 percent to 34
percent for non-agricultural employment).
Figure 2-10: Share of firms facing informal competition
By country By size
By region and industry
Source: Turkey ES 2008
18 TURKSTAT measures informality as the proportion of workers unregistered for social security. Issues related to informality in
Turkey are addressed in a forthcoming World Bank Economic Memorandum, Turkey Informality: Causes, Consequences,
Polices.
33%35%
43% 44%
49%52% 52% 54% 55%
0%
10%
20%
30%
40%
50%
60%
Poland 2009
Romania 2009
Czech Rep. 2009
Turkey 2005
(manuf)
Hungary 2009
Turkey 2008
(manuf)
Turkey 2008 (all)
Bulgaria 2009
Brazil 2009
52%52%
55%
49%
50%
51%
52%
53%
54%
55%
56%
Small Medium Large
44%49%
58% 59%62%
41% 43% 45% 46% 50% 51% 51%54%
58% 58%64% 64%
0%
10%
20%
30%
40%
50%
60%
70%
26
2.24. The share of surveyed firms facing informal competition has increased substantially. 52
percent of all firms interviewed in the 2008 survey indicated that they compete with informal companies,
a negative development since 2005, where the corresponding number was 10 points lower (Figure 2-10).
Moreover, 32 percent of all firms claim practices of competitors in the informal sector to be a major or
very severe obstacle. 15 percent of all firms interviewed agree that it is the most serious problem they are
faced with in their business, positioning informal sector practice as the fourth largest obstacle. Figure 2-
10 also shows that larger firms experience more competition from the informal sector than SMEs (55
percent and 52 percent respectively), whereas companies in the South and in the Marmara region face
considerably less informal competition than Black Sea – Eastern, Central Anatolia and Aegean
Additionally, the survey data shows that informal competition is highest in the food and chemicals
sectors.
2.25. When considering that tax evasion is one of the main drivers of the informal sector, the
Turkish government should continue its efforts in tax regulation reforms. Other recently
implemented as well as planned actions are listed in Box 2-3.
Administrative procedures
2.26. Recent successful measures to streamline compliance with administrative procedures need
to be expanded. There is today no clear framework for streamlining administrative procedures for
businesses. Operating licenses are issued by various Ministries, each responsible for different business
areas. For instance, the Ministry of Public Works is responsible for construction permits, in terms of their
design and their applicability. Another example is the Ministry of Environment who is responsible for
environmental licenses. However, there is at this point no single overall strategy for stepping up and
facilitating licensing procedures. Both the Ministry of Agriculture and the Ministry of Health, for
instance, are involved in the processing of licenses in the food industry, with the result of much
overlapping and confusion.
2.27. Notwithstanding successful earlier actions towards regulatory reform, many concerns still
remain to be managed. The last major governmental initiative to limit license and permit requirements
was implemented in 2005, through the regulation on ―Opening a Business Place and Work Licenses‖
(CEDPL2), reducing the number of documents needed to obtain a license, as can be noticed in the survey
2005-2008 comparison. Nonetheless, the weaknesses in the Turkish regulatory system for licensing and
permits are many and much remains to be done. As discussed in greater detail in Chapter 5, some major
Box 2-3: The Government‟s struggle against informality
In 2006 the Government organized and initiated the two-year ―Struggle against Unregistered Employment (KADIM)
Project,‖ with the main objective to reduce informal employment. The target was to prevent 100,000 illegal foreign
workers and replace these jobs with a formal workforce, thus increasing domestic employment. Means to be applied
include information and awareness activities, bureaucracy reduction, efficient inspection, audit and employee
registration controls, as well as labor cost reductions. Between October 2006 and December 2007, more than
730,000 employees and nearly 200,000 workplaces had been audited. Moreover, the Turkish Social Security
Institution increased the number of employees covered by 63 percent. The final results of the KADIM project have
not yet been made public.
The latest initiative against informality was introduced in June 2008 through the Revenue Administration in
cooperation with several key institutions. The development of the ―2008-2010 Action Plan of Strategy for Fight
against the Informal Economy‖ covers 105 actions over a three-year period, with three key objectives: promoting
formal activities; strengthening audit capacities; and procuring and strengthening organizational and societal
consensus.
27
difficulties include: insufficient coordination among agencies responsible for application processing;
inadequately qualified municipalities in dealing with and applying secondary legislation; major gaps in
sector-based strategies; and poorly constructed digitalized systems for application approvals from citizens
and corporations. In addition, involved Ministries do not have access to any standardized procedure
guidelines for implementing the regulatory framework.
2.28. The perceived „time tax,‟ management time spent dealing with government regulation,
appears to have increased since 2005. Management personnel in Turkish businesses declare to spend on
average 27 percent of their working time dealing with bureaucracy, a steep increase from the 9 percent
average in 2005. The time tax is largest for medium and large enterprises (32 percent and 34 percent
respectively), whereas managers in small firms declare to spend 23 percent of their time with regulations.
2.29. The establishment of Development Agencies (DA) could offer an opportunity to facilitate
licensing procedures by providing an interface for businesses at the local level, provided that risks
of capture are minimized. The Government, through the State Planning Organization, has established
Development Agencies, with the objective of 26 Agencies covering the entire country (Box 2-4). Two
Development Agencies are fully operational today – in the Çukurova region (Adana and Mersin) and
Izmir, along with their accompanying Investment Support Offices. The objective is for agencies to be
rolled out across the country building on the experience of Çukurova and Izmir, to serve as ―one-stop-
shops‖ for companies in financial and technical matters and ―execute programs that will support
development, competitiveness and local initiatives‖ (SPO 2009). In connection with the Development
Agencies, and building on the experience of Izmir and Adana, Investment Support Offices are planned to
gradually be established across the country with main focus on aiding investors and assisting them in
licensing and registration procedures. DAs are well positioned to be an interface between business and
government provided that they preserve a light organizational structure and remain at arm‘s length from
the government, with the objective of minimizing the risk of capture on the part of local interests.
Additionally, an essential condition for DAs to be able to perform their role is the existence of internal
Government regulations recognizing their formal role vis-à-vis central and local government.
Box 2-4: Development Agencies (DAs)
A Regional Development Agency – or Development Agency (DA) – can be defined as a regionally based, publicly
financed institution outside the mainstream of central and local government administration designed to promote
indigenous economic development through an integrated use of policy instruments. DAs have been operating in
Western European countries since 1950s and 1960s in order to decrease the negative effects of the Second World
War and keeping up with rapid technological developments in the world. According to European Association of
Development Agencies (EURADA), ‗a DA is an operational structure that identifies sectoral or overall
development problems, chooses a range of opportunities or methodologies for their solution and promotes projects
which can maximize the solutions to the problems‘.
In Turkey, regional development policies have been developed in the quest to eliminate regional disparities, to
accelerate local and regional economic development and to enable sustainable development. However, those policies
developed towards particularly less developed regions have not been very successful, mainly due to the lack of
institutional capacity at the local/regional level, i.e. the lack of effective institutional structures at the local level as
well as that of sufficient financial resources. There had been a number of initiatives of regional development
projects prior to Turkey‘s comprehensive adjustment process towards EU candidacy such as East Marmara Regional
Planning Project (1963), Zonguldak Regional Planning Project (1963-64), Antalya Project (1960-65), Aegean
Regional Development Project (1963-69), Çukurova Region Planning Project (1962), Keban Project (1964) all
aiming at developing policies, plans and proposals for the problems of different regions.
Since Turkey gained European Union (EU) candidate status with the decision of Helsinki Summit in 1999, the
momentum of legal and institutional reforms has increased unprecedentedly. Accession negotiations between the EU
28
2.30. Looking ahead, DAs could more fully perform a number of useful functions. First, as already
intended by the Government, they could act as “one-window” shops. Without radical reform of
competences for the issuance of licenses and permits they could perform a useful facilitator role between
issuing agencies and firms. Second, DAs could act as an information point for businesses, in close
cooperation with business associations with a local presence, such as TOBB, as well as with Government
agencies, such as KOSGEB and TÜBITAK. The objective would be to rationalize support initiatives –
especially for SMEs who normally face high information costs – available at the local level. Third, the
FDI promotion function via Investment Support Office should be carried out in close coordination with
ISPAT, the national FDI promotion agency.
STARTING A BUSINESS
2.31. The Government has made progress towards facilitating the process of business
registration. According to Doing Business 2010, recent Turkish reforms have reduced the time it takes to
register a business from 13 required steps in 2004, down to six steps and six days required in 2009. This
development is supported by the OECD Product Market Regulation (PMR) Indicators, which show a
reduction in number of mandatory registration procedures, from 29 procedures in 2003 to 11 procedures
in 2008, as well as a cut in days needed to complete registration, from 6 days to 2 days. Although Turkey
has experienced an improvement since 2006 as well, the positive six-year development can mainly be
attributed to the 2003 legal reforms, with a modern investment law and new legislation covering
registration procedures (World Bank 2007).
and Turkey that started in 2005 and the issue of Turkey‘s compliance with the ―acquis communautaire‖ became
even more significant for the legal and institutional reform processes. In its National Plan for the Adoption of the
Acquis (NPAA), Turkey has committed itself to legal and institutional changes that will contribute to its adjustment
to EU regional policy. Accordingly, in order to facilitate the development of structures of local/regional governance,
‗The Law on the Establishment, Coordination and Duties of Development Agencies‘, was ratified on January 25,
2006 (Law No. 5449, 2006) to facilitate and regulate the establishment of Development Agencies-DAs.
The Law establishes that the State Planning Organization (SPO) will be responsible for the coordination of DAs at
national level. Agencies in all the 26 NUTS II regions were established. Their main purpose is to accelerate regional
development, promote cooperation between the public and private sectors and contribute to the reduction of inter-
regional disparities. The DAs will be funded in part from transfers from the national budget and in part by the
special provincial administrations (local authorities) and municipalities. The DAs will also be expected to generate
operating revenues, although this is not realistic in the poorer regions.
Two pilot DAs in Izmir and the Çukurova region started operating in 2008 and other agencies are now established in
all provisional NUTS II-type regions. A total budget of nearly €125 million has been earmarked for the development
agencies in the 2009 national budget. Relevant local and regional stakeholders are involved in establishing the
budgets of individual DAs, but not in selecting the provinces to host the DAs.
At present DAs which are aiming to increase regions‘ competitive power and decrease regional discrepancies,
focusing especially on high-technology, innovation and communication are still in their infancy and face criticisms
and legal challenge on the grounds that they will weaken central government.
The organizational structure of the DAs includes a Developmental Board, a Management Board, a General
Secretariat and Investment Support Offices for business support. The Developmental Board members are
representatives of various public and private organizations, NGOs and universities within the region. This board
functions as an advisory board. The Management Board is formed by governors, mayors of the metropolitan
municipalities, the chairmen of the Chambers of Commerce and Industry and three representatives from NGOs or
from the private sector. General Secretariat is the executive body in DAs. Moreover, Investment Support Offices-
ISOs are located in each province of the NUTS II regions.
29
2.32. The 2007 Investment Climate Assessment presented criticism of the lack of transparency in
registration procedures as well as costs for Turkish firms, making among others recommendations
for better use of information technologies. The Government has in fact initiated e-Judicial Registrations
and Online Company Registration procedures, with a draft act currently being evaluated by the Prime
Ministry and expected to be adopted in the National Assembly in early 2010 (Prime Ministry e-
Legislation Seminar, September 2009).
Box 2-5: Indicators of product market regulation (PMR)
The OECD PMR is a comprehensive and internationally-comparable set of indicators that measure the
degree to which policies promote or inhibit competition in areas of the product market where competition is
viable. The indicators measure the economy-wide regulatory and market environments in OECD countries
in 1998, 2003 and 2008 and depict the strictness of the regulatory environment on a scale 0 to 6, with
higher numbers indicating more restrictive policies towards competition. The PMR system has been revised
and updated as of 2008 to better integrate sectoral information and cover additional key regulatory issues,
such as governance and treatment of foreign entities.
The measured variables in the areas of start-up and licensing systems are based on questions on how many
days, procedures and the cost it takes to register a business as well as the presence or absence of ―one stop
shops‖ for licensing information.
As can be seen in Figure a. depicting Turkey‘s ranking in one of the measured variables, administrative
burdens on start-ups, gradual improvements have been made, moving from 3.1 index points in 1998 to 2.6
index points in 2008. Turkey has however still ways to go to the OECD average of 1.53 points. The
conditions for Turkish enterprises are thus more severe when analyzing indicators measuring license and
permit systems, where the results have consistently been meager. For three consecutive measured periods,
Turkey has performed poorly and remains to have a 6.0 index point ranking in the PMR indicators.
Figure 2-11: Development of product market regulation since 1998, Turkey and comparator countries
Index scale of 0-6 from least to most restrictive
a. Administrative burdens on start-ups b. License and permit systems
Source: OECD, Product Market Regulation Database
Methodology derived from Wölfl et al. 2009
2.2
0
2.3
5
2.3
3 3.0
9
3.0
6 3.9
0
3.5
9
2.3
1
1.6
9 2.5
1
2.8
3
2.7
4 3.9
0
3.2
2
1.5
7
1.7
3
2.0
9
2.6
1
2.8
5
3.2
3 3.8
4
0
2
4
6
1998 2003 2008
0
6
4
2
4
6 6
0
2
0 0
4
6 6
0 0 0 0
2
4
6
0
2
4
6
1998 2003 2008
30
LICENSING
2.33. The time needed to obtain an
operating license19
remains high in Turkey
but tends to be shorter among firms in the
services sectors. Manufacturing firms
interviewed in the 2008 enterprise survey
report that the time it takes to acquire an
operating license has decreased from 66 days
in 2005 to 62 days in 2008. Albeit the
improvement, this is still relatively higher
than in comparison countries. However, as
seen in Figure 2-12, when considering the
total sample that includes the services sector,
the number of days to obtain the license
drops to 36 days, which internationally
compared, is more favorable.
2.34. There are signs of variations among regions, as well as by size and ownership form of firms.
Figure 2-13 shows that obtaining an operating license is nearly three times as time consuming in the
South than in Central Anatolia. When comparing operating license procedures by city, the time in
Istanbul has improved but is among the longest in the country. Furthermore, medium firms need more
time than small and large ones to obtain necessary licenses. The same goes for domestic firms that have to
wait 36 days for their operating license, versus 21 days for foreign-owned firms. By sector, enterprises in
chemicals, textiles, and garments have to wait the longest to obtain their licenses, while the metals and
machinery, as well as the service sector have a waiting time far below the average.
19 An operating license is defined as a statement order by a government authority or other authorized legal entity that authorizes
the business activity to be carried out in the country. An operating license is usually required as part of the business startup
process and it verifies that the establishment possesses the operating business licensing requirement. The business activities
which can be covered by an operating license may differ according to the country specific regulations.
Figure 2-12: Days to obtain operating license, country
comparison
Source: Enterprise Surveys 2008
Figure 2-13: Days to obtain operating license, by region, city, size, ownership and exporting status
Source: Turkey ES 2008
20.8 23.7
35.6 36.0
57.4 62.0 66.0 67.7
0
10
20
30
40
50
60
70
Bulgaria 2009
Romania 2009
Hungary 2009
Turkey 2008 (all)
Russia 2009
Turkey 2008
(manuf)
Turkey 2005
(manuf)
Chile 2006
22.427.4
32.138.7
65.6
15.3 17.9 18.8
36.3 37.842.1
78.4
33.1
46.1
33.6
20.7
36.3
50.4
32.3
0
10
20
30
40
50
60
70
80
90
31
2.35. Firms claim to need longer today to obtain import licenses20
than in 2005, but Turkey
compares well internationally. The time has increased from 12 days to 18 days for the manufacturing
sector and the number is even higher
when services industries are included
(Figure 2-14). Medium-sized firms have
to wait on average 32 days, in comparison
with only 8 days for large enterprises.
Figure 2-15 also shows that there are large
variations among cities, with companies
in Denizli having to wait considerably
longer to obtain an import license than for
instance firms in Izmir (42 and 11 days
respectively. Furthermore, food and
chemicals are some of the sectors that
have to wait longer to obtain an import
license, while garments and non-metallic
material experience shorter waiting
periods.
Figure 2-15: Days to obtain import license, by size, ownership, exporting, region, city and industry
Source: Turkey ES 2008
2.36. Clearance of customs procedures has become less cumbersome than in 2005. According to
survey data, firms claim that the time to clear customs for imports has increased from 7 days in 2005 to
10 days in 2008 for the Turkish manufacturing industry (Figure 2-16). However, when analyzing only
panel firms (the firms interviewed in both 2005 and 2008) the survey data shows an improvement in
customs clearance for imports, from 13 days to 8 days.21
20 An import license is defined as a required document issued by a government authority or other authorized legal entity that
authorizes the importation of certain goods into its territory. 21 Considering the smaller sample of panel firms, some caution should be used when interpreting this data. Nonetheless, the
comparison provides useful information when used in used together with other data sources, such as Doing Business.
21.8
32.0
8.0
17.522.2
10.7
15.917.9
23.9 24.2
42.0
6.9
10.7
20.424.0 24.8
28.9
34.1
0
10
20
30
40
50
Figure 2-14: Days to obtain import license, by country
Source: Enterprise Surveys 2008
8.111.2 12.3
16.6 17.721.2
29.632.7
43.1
0
10
20
30
40
50
Bulgaria 2009
Poland 2009
Turkey 2005
(manuf)
Chile 2006
Turkey 2008
(manuf)
Turkey 2008 (all)
Russia 2009
Czech Rep. 2009
Brazil 2009
32
2.37. Doing Business data also present a more positive image, with the number of days to import
decreasing from 25 in 2005 to 15 in 2009. In line with the findings of the EU Progress Report for 2010,
Doing Business indicators also show that Turkey compares more favorably relative to other emerging
economies (the average for the ECA region in 2009 is 18 days). It is worth mentioning that Turkey has
initiated numerous recent actions to simplify firms‘ trading across borders. A Customs General
Communiqué (SPO 2009) was published in April 2008 in the Official Gazette, covering the simplified
practices when dealing with import and export documents. Since October 2008, the e-transformation of
customs offices has been implemented through the Computerized Customs Activity (BILGE) System,
connecting 120 customs directorates and allowing over 95 percent of all trade transactions to be carried
out electronically today (European Commission 2009). Additionally, build-operate-transfer agreements
for renovation of border gates have been initiated with the private sector as well as better coordination of
agencies and tasks involved in the customs clearing process (World Bank 2008).
ACQUIRING LAND
2.38. Land access is in general not perceived as a major problem by Turkish firms. Only 5 percent
of enterprises consider access to land to be a major or severe obstacle to operations. This is not only a
significant improvement from the 20 percent response in 2005, but it also places Turkey in a positive
international position.
2.39. Firms‟ perception of land access
varies by size and sector. The differences
by region depict a mean of 3 percent in Black
Sea - Eastern versus 7 percent in the South,
with a similar pattern depicted by city. As
Figure 2-17 depicts, in Kocaeli and Adana
the ability to acquire land is not seen as a
major issue by many firms (3-4 percent),
while 8 percent of corporations in Malatya do
struggle with land access. A small variation
by firm size can also be noticed, where small
and large firms (6 percent) deem land access
to be a problem to a larger extent than
medium enterprises (Figure 2-18). An
explanation for the higher share for firms
with large number of employees may be that these naturally acquire larger areas for production. Figure 10
Figure 2-16: Days to clear customs for imports
Source: Enterprise Surveys 2008
Figure 2-17: Access to land as an obstacle, by city
Source: Turkey ES 2008
35
78
10 1010 10
1315
17
0
3
6
9
12
15
18
Hungary 2009
Bulgaria 2009
Turkey 2005
(manuf)
Turkey 2008 (panel
firms)
Poland 2009
Czech Rep. 2009
Turkey 2008 (all)
Turkey 2008
(manuf)
Turkey 2005 (panel
firms)
Brazil 2009 Chile 2006
3%
4%5%
5%6% 6%
6% 6% 7%
8%
0%
3%
6%
9%
33
shows that the differences among industries are more noticeable. Access to land is perceived as more of a
hurdle in the machinery and equipment as well as basic metals sectors, whereas garments companies are
not disconcerted by this matter.
2.40. Turkey has also made improvements in the cost and time invested in constructions of
business premises. According to the Doing Business‘ measure of the time involved in building a
warehouse Turkish firms have been able to shorten their site development time by 44 days between 2005
and 2009. However, even though Turkey is below the Europe and Central Asia average, the country‘s
results are mediocre in an OECD comparison. Also, it still takes Turkish firms six times longer to
construct a building than for Finland, the EU best practice economy. The fees involved are also 23 times
higher than in Hungary, for example. Overall, Turkey is ranked as 133 for dealing with construction
permits among 183 measured economies.
2.41. The ES data also has a similar measure, where interviewed firms are asked how many days it took
to obtain a permit that was requested over the last two years. The survey results present a more positive
picture, with an improvement since 2005 from 58 days to 47 days in 2008, which is a leading ranking
relative to other middle income economies, such as Brazil, the Czech Republic, Poland and Romania.
Interesting to see is that the time to obtain building permits varies significantly between Turkish cities,
with Kayseri and Bursa showing the least average time. Meanwhile, firms in Istanbul seem to struggle
Figure 2-18: Access to land as an obstacle, by firms size, ownership, exporting status and industry
Source: Turkey ES 2008
Figure 2-19: Days to obtain a construction-related permit, by country
Source: Enterprise Surveys
6%3%
6%4% 5% 6% 5%
2%3% 4% 4% 5% 6% 7%
11% 11%
17%
22%
0%
5%
10%
15%
20%
25%
41.9 43.0 47.1 47.558.4 58.5
87.0 87.9104.2 104.9
118.1 122.5139.1 142.8
0
20
40
60
80
100
120
140
160
Turkey 2008 (all)
Hungary 2009
Turkey 2008
(manuf)
Bulgaria 2009
Turkey 2005
(manuf)
Czech Rep. 2009
Istanbul 2008 (all)
Istanbul 2005
(manuf)
Russia 2009
Istanbul 2008
(manuf)
Romania 2009
Poland 2009
Brazil 2009
Chile 2006
34
with construction permits and have to wait on average 87 days to obtain a permit (Figure 2-19).22
Additionally, a variation is notable by firm size, with SMEs spending nearly twice the time as large firms
dealing with construction permits (60 versus 32 days).
2.42. Some measures in improving land access conditions for businesses are underway. YOİKK‘s
Location of Investment Technical Committee has set up goals for technical and legal actions in
streamlining land title procedures as well as re-estimate land title fees, according to best international
practice. Plans are also underway to evaluate the procedures involved in acquiring building permits and
the costs for the construction process. The Government is preparing large plots of land for the
development of industrial zones, aimed to attract high-value added or high-tech investments.23
BUSINESS INSPECTIONS
2.43. Business inspections appear to be less burdensome in Turkey than in comparator countries. The average time all employees in a company spend with inspections per year is 6.6 days, which puts
Turkey ahead of such comparator economies as Bulgaria, the Czech Republic and Hungary. The survey
results are similar when firms are asked about the number of inspections taking place in their
organization. The average frequency of inspector visits has also diminished from 4 inspections in 2005 to
2 in 2008 (Figure 2-20). Additionally, regional comparisons show significant variations in inspection
times and duration, with the average number of inspections per year being highest in Central Anatolia and
lowest in the Aegean region. The differences can be explained by the fact that it is often the
municipalities themselves enforcing regulations, without necessarily having higher approval.
2.44. Time spent with inspections varies by industry and firm size. Inspections are least frequent in
the services sectors whereas food and chemicals sectors are inspected more often. This is to be expected
as the time corresponds to the health, environment, or safety risk levels. For the same reasons, large
businesses are inspected more frequently than small and medium firms. However, when instead
examining the time that each firm spends on each inspection, medium firms seem to be more affected.
The average for medium and large businesses is 3.7 days, comparing with 2.2 for small firms (Figure 2-
22 The Enterprise Survey results for construction-related permits correspond better with the Doing Business indicator when taking
into consideration the results for Istanbul, seeing as Doing Business methodology makes the assumption that the firm being
analyzed operates in the country‘s most populous city, in this case, Istanbul (doingbusiness.org). This might suggest that Doing
Business data is not necessarily representative of the regulations in other cities and regions. 23 2009 Technical Committee Action Plans of the Turkish Coordination Council for the Improvement of the Investment
Environment (YOİKK)
Figure 2-20: Number of inspections per year, selected
countries
Figure 2-21: Average time spent dealing with
each inspection (days), by firm size
Sources: Enterprise Surveys and Turkey ES 2008
2.02.4 2.4
2.83.5 3.7 4.0
6.1
0
1
2
3
4
5
6
7
Turkey 2008
Hungary 2009
Turkey 2008, manuf
Brazil 2009
Bulgaria 2009
Czech Rep. 2009
Turkey 2005, manuf
Chile 2006
2.2
3.7 3.7
2.8
0
1
2
3
4
Small Medium Large Total
35
21). The variations by ownership form and exporting status are on the other hand not significant for any
of the variables.
2.45. Further harmonization among institutions and municipalities would further reduce the
burden on businesses. Although the survey results from 2008 show signs of improvements in
enforcement of and compliance with government regulation in Turkey, these remain quite weak and
disorganized areas. As mentioned above, the municipalities are often the ones implementing regulation,
while the inspection system per se is centralized. This institutional setup tends to create a burden for
businesses and individuals, when having to comply with procedures established locally without prior
agreement at the central level. At this stage, there are no clear governmental goals set for improvement of
inspection procedures.
CLOSING A BUSINESS
2.46. Productivity and growth in an economy are to a large extent affected by the efficiency of
firm entry and exit, through improved resource allocation and higher aggregate efficiency. A crucial
feature of a friendly investment climate is the existence of efficient insolvency procedures. Just as
markets with more net entry of firms over time become exposed to greater competitive pressure, so do
economies allowing for efficient exit of less productive firms induce higher market competition, thus
creating higher allocative efficiency and better long-term growth.24
2.47. By having well-designed insolvency and bankruptcy procedures, creditors are insured that
debts from the insolvent firms can be collected efficiently and with the highest recovery rate
possible. In this manner, an economy can improve the degree of competition in product markets and, as a
result, productivity. Nevertheless, it is often the case, particularly in developing and emerging economies,
that the insolvency procedures are too elaborate, with high administrative costs and long delays. The
cause, as presented by Djankov, et al. (2008a), is the practice of developing countries copying
industrialized ones in their often very formal bankruptcy procedures, without having the skills or the
capacity to follow through with saving or swiftly closing the insolvent business.
2.48. The most recent reforms in Turkey in regards to insolvency measures took place in 2003
and 2004, with the main amendment in the Execution and Bankruptcy Law25
focusing on so called
―Suspension of Bankruptcy.‖ The procedure, applied from the Swiss bankruptcy code, presents the option
of suspension of all insolvency proceedings when the possibility of restructuring the company can be
presented to the court. The objective is to give the debtor a chance to restructure and avoid bankruptcy
before creditors force the debtor into liquidation.
2.49. Turkey has to close the gap with OECD standards in insolvency procedures. When taking a
first look at the results for Turkey in the Doing Business 2010 results, firms dealing with bankruptcy
procedures seem to have experienced significant improvements since 2005. The time to close an insolvent
business has in more than halved, from 5.9 years in 2005 to 3.3 years in 2009. This puts Turkey on par
with the average of Europe and Central Asia. Furthermore, the recovery rate has risen from seven to 20
percent in the past four years. Although a strong improvement, this is still well below the average OECD
level and international best practice (Japan). Out of 183 measured economies, this puts Turkey at rank
120 (Table 2-5).
24 See for instance Nicodeme and Sauner-Leroy, (2007) 25 Execution and Bankruptcy Act No. 2004
36
Table 2-5: Bankruptcy procedures (duration and recovery rate) comparison of selected countries
Time (years) Recovery rate
(cents on the dollar)
Country 2005 2009 2005 2009
Turkey 5.9 3.3 7.0 20.2
International Best Practice 0.4¹ 0.4¹ 93² 92.5²
Region (Europe & Central Asia) 3.9 2.9 29.8 31.6
OECD 1.5 1.7 74.0 68.6
Brazil 10.0 4.0 0.5 17.1
Bulgaria 3.3 3.3 33.0 32.1
Chile 5.6 4.5 23.0 21.3
China 2.4 1.7 31.0 35.3
Czech Republic 9.2 6.5 18.0 20.9
Hungary 2.0 2.0 36.0 38.4
Malaysia 2.2 2.3 39.0 38.6
Poland 3.0 3.0 26.5 29.8
Romania 4.6 3.3 17.0 28.5
Russia 3.8 3.9 27.0 28.2 ¹Ireland; ²Japan
Source: Doing Business 2010
2.50. The time connected with bankruptcy procedure has been reduced, possibly because courts
have applied the law in a strict way, demanding from the indebted company to present a serious change
in being able to pay the debt in full before allowing extensions of the reorganization period. Nonetheless,
practitioners seem to agree that the suspension amendment to the Bankruptcy Law has not been applied in
an efficient manner and that many debtors have avoided serious efforts to reorganize, and instead stayed
in business at the cost of their creditors. The possibility of long extensions of bankruptcy suspension is
one of the main problems in the insolvency procedures, allowing for strong depreciation of the recovery
rate over time. Other issues to be addressed include a lack of training among judges and often insufficient
background in international standards in the case of insolvency procedures.
2.3 Labor Market and Skills
2.51. The skill level of the workforce is positively associated with productivity, while smaller
firms are less likely to employ more educated workers and to provide training. Econometric analysis
shows that firms with a higher share of staff with university education can expect a positive impact on
productivity (Table 2-6).26
At the same time, a higher share of female workers is negatively associated
with manufacturing productivity, reflecting the generally lower education levels of female workers,
especially in labor intensive industries, such as textiles and garments. Survey results also show that larger
firms have higher shares of non-production workers (e.g. managers, administration, sales), are in a better
position to afford skilled staff with university education and also have the means to provide training to
workers.
26 Results of the 2005 survey showed that more productive firms tended to have lower levels of part-time and unskilled
employees. The length of the training provided to skilled workers – a variable not available in 2008 – also had a positive
association with productivity levels. See Table 2-A-2.
37
Table 2-6: Summary of the effects of labor and skills
Explanatory variable
Dependent variable
Productivity Employment Probability
of exporting
Probability of
receiving FDI
Non-production workers + +
Staff - female workers -
Staff - university education + +
Training (dummy) + Source: Staff calculations based on Escribano et al. 2009. See Tables in Annex 2-A for detailed results.
The labor market context
2.52. Strong economic growth, averaging 6.5 percent since the 2001 crisis, has not been
accompanied by significant improvements in the labor market. A comparison with the OECD
indicators four years ago as well as with OECD, EU-15 and selected economies shows that the overall
participation rate of 46.9 percent is modest and has actually decreased since 2005 (Table 2-7). Turkey‘s
labor participation is 20-25 percentage points below OECD and EU-15 averages. There has not been any
improvement in unemployment in Turkey since the sharp increase during the 2001 crisis, but
unemployment figures reached instead a record level of 11 percent in 2008.
2.53. The employment gap in an international context becomes even more noticeable when
looking at the employment rate for women, which in 2007 was 26.7 percent, far below the OECD and
EU-15 averages of 61.3 percent and 65.3 percent respectively. The 2008 Enterprise Survey also shows
that on average only 16 percent of production employees and four percent of non-production employees
are female. Earlier World Bank analyses have presented two main reasons for the low participation rates
for women: the transition away from the agricultural sector, where women had higher participation rates,
and selective participation in the urban labor market.
Table 2-7: Key labor market indicators, Turkey and selected OECD countries, 2008
Persons aged 15-64, percent
Labor force participation rate Employment/population ratio Unemployment rate
Total Males Females Total Males Females Total Males Females
Turkey 46.9 74.8 26.7 41.7 66.6 23.5 11.0 11.0 11.9
Turkey (2005) 51.3 76.2 26.5 45.9 68.2 23.7 10.5 10.5 10.6
Hungary 61.5 68.3 55.0 56.7 63.0 50.6 7.9 7.7 8.1
Mexico 62.2 83.5 43.3 59.9 80.7 41.4 3.7 3.4 4.3
Poland 63.8 70.9 57.0 59.2 66.3 52.4 7.2 6.5 8.0
Portugal 74.2 79.5 68.9 68.2 74.0 62.5 8.1 6.9 9.4
Slovakia 68.9 76.4 61.4 62.3 70.0 54.6 9.6 8.4 11.4
Spain¹ 73.7 83.0 64.1 65.3 74.6 55.7 11.4 10.1 13.1
EU-15 72.5 79.7 65.3 67.4 74.4 60.4 7.1 6.6 7.6
OECD 70.8 80.5 61.3 63.9 75.7 57.5 6.0 6.0 6.2 Source: OECD, 2009
1. Indicators are based on values 16-64 years
38
2.54. Higher education levels are positively associated with employment and labor participation. In Turkey‘s case, this is particularly evident for female workers. Labor force participation for women
with upper secondary education was merely 34 percent in 2008, whereas the corresponding rate for
women with tertiary education was 71 percent. As Turkey‘s population expands by 800,000 every year,
the young working population will increase rapidly, making focus on education on all levels and school-
to-work programs even more crucial. Seeing as training opportunities after formal training is limited
today (World Bank, 2008), Turkey needs to put more focus on relevant vocational in-class training as
well as on work training programs.
2.55. Inclusion of workers reallocating from the agricultural sector and informality are
important underlying causes of poor labor market performance. One of the main reasons for
Turkey‘s low job creation in the past seven years can be widely ascribed to the significant transfer that the
country has made away from the agricultural sector. Other factors include the high informal employment
in Turkey, the share of unregistered workers averaging 50 percent in manufacturing and services (OECD
2007), as well as the high income tax rate and tax wedge. Although, as mentioned in the previous section,
the tax wedge in Turkey has decreased since 2005, it still remains among the highest when compared to
other OECD countries. OECD data also shows that restrictiveness of employment protection in Turkey is
highest when measured against the same group of countries, with a 3.5 ranking on a scale of 0-6, with 6
being the most restrictive (Venn, 2009). Important mentioning is also that no change in employment
protection has happened since 2003. The higher strictness does not allow for flexibility for firms in hiring
new employees.
Employment trends and constraints
2.56. Firms appear less bothered by regulatory labor issues, although the result may be affected
by the crisis environment and the issue is likely to become a concern for sustainable recovery. The
2008 survey shows that firms in Turkey to
a much lesser extent than in 2005 regard
labor regulations as a serious obstacle to
business operations. 8 percent perceived
labor regulations to be a major or very
severe obstacle, compared to 46 percent in
2005 (Figure 2-22). Also by international
standards, this has been a great advance,
improving Turkey‘s position vis-à-vis the
average of the Europe and Central Asia
region. Whereas a plausible explanation for
the improvement in perceptions can be the
labor reforms that were initiated in the
beginning of 2008 (Box 2-6), the
inflexibility of labor regulations is still
mentioned by enterprises in face to face interviews as the overarching constraint facing firm operation
and growth. It is possible that the timing of the survey – from April 2008 to January 2009 – found
business more preoccupied with other, more immediate constraints – e.g. loss of market share requiring
downsizing or access to finance – thus decreasing the relative importance of labor regulations. Bearing
the current financial crisis in mind, the 2005 survey results might, in fact, better reflect today‘s labor
regulation environment. It is therefore a concern that should be given more attention as the economy
improves.
Figure 2-22: Labor regulations as an obstacle, country
comparison
Source: Enterprise Surveys
8% 8% 9% 10% 13%
27% 30%
46%
58%
0%
10%
20%
30%
40%
50%
60%
70%
Turkey 2008
(manuf)
Turkey 2008 (all)
Hungary 2009
ECA Region
Bulgaria 2009
Poland 2009
Chile 2006
Turkey 2005
(manuf)
Brazil 2009
39
2.57. The obstacle level does not vary by firm size, whereas ownership form appears to affect to
what extent firms are burdened by labor regulations. While 8 percent of domestically owned firms
find regulations to be an obstacle, close to none of the foreign companies cite them as a constraint. The
2005 Survey addressed the hypothetical issue whether firms would change the number of permanent full-
time employees if labor restrictions did not exist (World Bank, 2007). The response was remarkable,
showing that 62 percent of the companies would in fact increase their employment levels. Unfortunately
this is not comparable with 2008 data, as the hypothetical question was not addressed. Although thought
needs to be given to what the actual effect on employment growth would be if this were the case, the
result still shows that firms were hindered in their operations by the strict labor regulations three years
ago.
2.58. Another sign of the restrictive labor policies is the earlier discussed large informal sector.
With 52 percent of the firms interviewed stating that they compete with unregistered and informal
businesses, this shows that many establishments choose away business restrictions completely. At the
same time, protection of workers through social security networks and unemployment insurance policies
are very weak in Turkey.
27 Box 2-6 presents the methodology of World Bank Doing Business and can be found at www.doingbusiness.org, The specific
labor regulations of the Turkish Labor Code are clarified in Article 11 and 17 of Labor Law No. 4857.
Box 2-6: Employing workers in Turkey: rules of hiring, work schedules and redundancy
One of the indicators in the Doing Business assesses the regulation of employment, with the methodology
specifically measuring the effects on hiring, the rigidity of working hours, difficulty of redundancy and
redundancy cost.
In the case of Turkey, Doing Business depicts a limited liability company, operating in the manufacturing
sector in Istanbul, fully domestically owned, with 60 employees. The business is also subject to collective
bargaining agreements and is assumed to abide by every law and regulation, but does not grant workers
more benefits than mandated by law. The worker is assumed to be a 42-year-old, nonexecutive, full-time,
male employee, has worked at the same company for 20 years, and earns a salary plus benefits equal to the
economy‘s average wage during the entire period of his employment. Additionally, the worker is not a
member of a labor union, unless membership is mandatory.
Fixed-term contracts are not allowed for permanent tasks in Turkey, there is no limit for fixed-term
contracts, and the mandated minimum wage relative to the average value added per worker is 41 percent.
When measuring rigidity of hours, there are restrictions on night work but not on weekly holiday work, 6-
day workweeks as well as 50 hour-workweeks (including overtime) for 2 months are allowed, and there is a
requirement for paid vacation of 26 workdays per annum for an employee with 20 years of service.
Termination of workers due to redundancy is allowed in Turkey and the employer is not required to notify
a third party nor does he need the approval of a third party before terminating one or a group of redundant
workers. The law does, however, mandate retraining or alternative placement before termination and there
are also priority rules applying to re-employment. When it comes to the cost of redundancy, these are more
onerous in Turkey than in many comparator countries. An employer is required to give 2 months notice
before a redundancy termination, and the severance pay for a worker with 20 years of service equals 20
months of wages. However, no penalty is levied for redundancy dismissal. The redundancy cost equals in
total 23 months of salary, a rate quite high in particular when compared to the ECA and OECD averages of
6 to 7 months.
Turkey has an overall rating of 44 index points (on a 1-100 scale, with higher values representing more
rigid regulation). Relative to other economies, Turkey ranks 145 out of 183 measured countries.
Source: Doing Business 201027
40
2.59. The survey results further indicate that the use of temporary employment among Turkish
enterprises has decreased remarkably since 2005. In 2005, 31 percent of the workforce was
temporary.28
This number has diminished to 4
percent, when sampled in 2008. Moreover,
the average length of temporary employments
has also decreased, from 7 months in 2005 to
4 months three years later. This reduction in
use of temporary workers complements the
overall opinion in the survey that the
perception of labor regulations as an obstacle
has dropped from 46 to 8 percent. The change
in temporary employment is depicted in
Figure 2-23. Companies‘ relying less on
temporary workers today could de facto
confirm that they are less burdened with the
obligations usually associated with hiring
permanent employers, such as benefits and payroll taxes. Nevertheless, despite these positive implications
and seeing as the Government is only in the first stages of a comprehensive reform package, more time
needs to be dedicated in order to fully analyze the impact of the labor reforms in Turkey. The main state
employment initiatives taking place are summarized in Box 2-7.
28 Full-time temporary or seasonal employees are defined as all paid short-term (i.e. for less than a fiscal year) employees with no
guarantee of renewal of employment contract and work 40 hours or more per week for the term of their contract (World Bank
Enterprise Surveys, 2008)
Figure 2-23: Temporary employment as a share of total
employment, country comparison
Source: Turkey ES 2008
Box 2-7: Labor market reforms in Turkey
Earlier important legal amendments in Turkey include the introduction of the Labor Law No. 4857
from 2003, which harmonized legal and institutional regulation with the International Labor
Organization (ILO) and the European Union, as well as the restructuring of the Turkish Employment
Agency (ISKUR), which was given approval to the opening of private employment offices.
The Ninth Development Plan 2007-2013 sets out a comprehensive labor market reform, focusing on
three major areas of reform
- Reducing the non-wage cost of hiring. This encourages companies to hire more unskilled
employees and creates incentives for workers to participate in the labor force.
- Increasing employment flexibility while improving worker protection regulations and supporting
disadvantaged workforce groups. Severance payment obligations in Turkey in 2006 were among
the highest in comparison with other OECD countries, creating further disincentives for formal
employment. On the other hand, unemployment insurance policies only reached 4.4 percent of
unemployed workers in 2007.
- Improving coordination between employment and training programs, by adjusting existing
vocational and secondary education programs to the needs of the labor market and providing
assistance in finding jobs to minority groups, such as youth, women and disabled workers.
The Government has initiated the first phase of the labor market reform through the introduction of the
Employment Package, which covers amendments to existing laws, with the objective of increasing
employment in the country. The main actions in the Employment Package include
- Five percent reduction of employer‘s social security contributions
- Further reductions in social security contribution for employment of the young, women and
workers with disabilities
- Legislative changes in employer obligations in regards to inter alia hiring quotas
2%4% 4% 6% 7%
10% 12%
31%
0%
5%
10%
15%
20%
25%
30%
35%
Bulgaria 2009
Turkey 2008 (all)
Turkey 2008
(manuf)
Brazil 2009
Romania 2009
Czech Republic
2009
Poland 2009
Turkey 2005
(manuf)
41
2.60. The change in opinion regarding labor regulations and employment patterns with Turkish
firms might, to some extent, reflect recent reforms. With clear plans on decreasing informality,
applying flexible employment models and focusing on education-to-labor-demand compatibility, Turkey
is taking earnest steps in improving the
unemployment situation in the country. The
shift in firms‘ opinion regarding labor
regulations as a strain on operations could be
an indication of the effects of the labor
reform package that the Government has
initiated, as presented it the Ninth
Development Plan 2007-2013. The result
was particularly positive among foreign firms
where almost two thirds of the companies
responded that regulations were no obstacle
to business (Figure 2-24). This indicates that
Turkey‘s efforts towards supporting
investors, among others through incentives of
insurance premium cuts may be paying off
and should be continued and expanded.29
2.61. Reductions in labor costs will have positive effects on employment levels. Although the
cutback in non-wage labor costs, i.e. tax wedge reduction, is not at this stage reflected in the change in the
employment numbers from 2005 to 2008, it can be expected that this action will result in higher
employment levels. Turkey‘s Action Plan for the 60th Government (2007) in fact analyzes the impact of
non-wage labor cost cuts, with the conclusion that employment in Turkey is positively receptive to
changes in labor costs.
29 Cabinet Decision on State Aids for Investments, July 16 2009
- Transfer of funds from the Unemployment Insurance Fund (UIF) to ISKUR as a support for
training programs and enable other unemployed, registered to ISKUR – alongside the ones who
receive unemployment insurance – to benefit from fund financed education programs.
- Broadening and sustainability of active labor market programs, with the aim of increasing the
number of beneficiaries from 32, 200 in 2008 to 214,000 by the end of 2010.
- Coooperation between the local Provincial Employment and Provincial Vocational Education and
Training Councils has been initiated, with the aim of meeting the needs of the labor market more
effectively
Further flexibility measures have been put forward by the Government in the Medium Term Program
2010-2012, including severance pay reforms and additional modifications to the Labor Law no. 4857.
YOİKK‘s Employment Technical Committee also specifies labor market reforms in its 2009 Action
Plan. These include facilitation of use of the short-term work allowance, elimination of the burden for
employers of filing recruitment and discharging document to multiple institutions and streamlining the
process of work permit applications for foreign workers. The initiatives are planned to be completed in
the fall of 2009.
Figure 2-24: Labor regulations as an obstacle, by size,
ownership and exporting status
Source: Turkey ES 2008
8%9%
8%
0%
8%
11%
7%
0%
2%
4%
6%
8%
10%
12%
42
Education, skills and training
2.62. Education levels are lower than in
comparators. OECD data, illustrated in
Figure 2-25, show that 26 percent of the
Turkish adult population holds secondary
education diplomas. This is not only lower
than in 2005, but well below the OECD
average of 69 percent and the EU19 average
of 70 percent. Graduates with tertiary
education in Turkey are also scarce and the
entry rate to higher education programs
(tertiary education) is well below international
standards. Only 29 percent enroll in higher
education in Turkey, compared to 56 percent
average in the OECD countries.
2.63. The proportions of skilled and unskilled workers have tipped in favor of skilled employees. As seen in Figure 2-26, the distribution of permanent workers between non-production, skilled production
and unskilled production workers was 16 percent, 42 percent, and 42 percent respectively in 2005. The
corresponding shares in 2008 were 25 percent, 46 percent and 29 percent (Figure 2-27). When comparing
worker types by firm size, the variations have remained marginal, with large firms having a higher share
of unskilled production workers than small and medium firms.
Figure 2-25: Population with at least upper secondary
education, percent, 2007
Source: OECD
Figure 2-26: Distribution of permanent workers, by
type and firm size, 2005
Figure 2-27: Distribution of permanent workers, by
type and firm size, 2008
Source: Enterprise Surveys
69
26 2735
49
7986 86 87 89 90
0
20
40
60
80
100
19%16% 15% 16%
48%
41% 39%42%
33%
43%46%
42%
0%
10%
20%
30%
40%
50%
60%
Small (5-19) Medium (20-99) Large (100+) Total
Non-production Skilled Production Unskilled production
27%25%
20%25%
49%44% 43%
46%
24%
31%
38%
29%
0%
10%
20%
30%
40%
50%
60%
Small (5-19) Medium (20-99) Large (100+) Total
Non-production Skilled Production Unskilled production
43
2.64. Nearly a quarter of Turkish firms rate the education and skills levels of the workforce as a
“major” or “very severe” constraint on operations and growth. Although this is an improvement from
the 33 percent in 2005 (Figure 2-28), the
high rate confirms Turkey‘s serious
problem with lagging education and skill
levels. This shows that much remains to
be done in coordinating labor supply
with the demands in the business sector.
Rated by ownership form, foreign firms
perceive the education level to be a
bigger constraint than domestic firms do
(19 percent and 15 percent respectively).
Variation can also be seen among regions
and industries in Turkey. Firms in the
Aegean and Black Sea – Eastern regions
are especially concerned about
inadequately educated workforce (30 and
31 percent respectively), while more than a third of enterprises in the chemicals, electronics, metals and
construction industries perceive skills and education of workforce to be a major or very sever obstacle to
operations (Figure 2-29).
2.65. The share of the workforce with a university degree also remains low, with somewhat more
positive results for the services industry. The survey results show that 10 percent of the total workforce
in the manufacturing industry held a university degree in 2008, which is an actual deterioration from the
11 percent share in 2005. When including the services sector, the share of employees with a university
rises to 18, putting Turkey in a relatively positive position compared to for instance Hungary (9 percent),
Brazil (10 percent) and the Czech Republic (16 percent). The share of highly educated employees is
particularly high among non-exporting firms (19 percent versus 14 percent for exporters). A higher share
can also be found in the Central Anatolia (23 percent) compared to only 11 percent in the South. When
scrutinizing education levels by industry, construction (39 percent), wholesale and retail (27 percent) and
chemicals (22 percent) are the sectors that stick out, compared to relatively low shares in the textiles,
garments and basic metals industries (7 percent, 8 percent and 9 percent respectively). There is no
significant variation by firm size and ownership form. Taking into consideration the econometric results
presented in Table 2-6, a highly educated workforce is a key issue for enterprises, as a high share leaves a
significant impact on the firm‘s productivity and growth.
Figure 2-28: Education of workforce as an obstacle, by country
Source: Enterprise Surveys 2008
Figure 2-29: Education of workforce as an obstacle
Source: Turkey ES 2008
6%
21% 21%25%
29%33%
36%
42% 43%
0%
10%
20%
30%
40%
50%
Hungary 2009
Bulgaria 2009
Turkey 2008
(manuf)
Turkey 2008 (all)
Czech Rep. 2009
Turkey 2005
(manuf)
Poland 2009
Chile 2006
Romania 2009
24% 27%23%19%
15%20%
26%21% 23% 23%
30% 31%
13% 15%20%
23% 24% 25%26% 26% 28% 29%33%
37%42% 45%
0%
10%
20%
30%
40%
50%
44
2.66. The number of Turkish firms offering formal training to their employees has increased
slightly. Figure 2-30 demonstrates that 24 percent of manufacturing businesses claimed to offer training
to their employees in 2005. This number rose
to 29 percent three years later. Training is, as
can be expected, offered more often the
larger the firm (see also econometric analysis
in Table 2-6), where the share of large firms
offering training is three times higher than
with small businesses. Enterprises with
exporting activities are also more active in
offering training to workers. Measured by
industry, training is provided to a lesser
extent in the machinery, garments, food and
textiles sectors, while 83 percent of the firms
in the electronics industry provide training
for their workers (Figure 2-31).
2.67. Survey data also shows that
production workers are more likely to
receive training than non-production
workers. 67 percent of production workers
received formal training while only 50
percent of non-production workers were
provided the same. Put in an international
perspective, both production and non-
production workers in Turkey are provided
training to a larger extent than many other
comparative emerging economies, such as
Romania and Hungary (Figure 2-32).
Figure 2-30: Share of firms offering formal training to
employees, by country
Source: Turkey ES 2008
Figure 2-31: Share of firms offering formal training to employees, by firm size, ownership form, exporting
status, region and industry
Source: Turkey ES 2008
Figure 2-32: Share of production and non-production
workers that received training, country comparison
Source: Turkey ES 2008
15%
24% 25%29% 29% 31%
53%61%
0%
10%
20%
30%
40%
50%
60%
70%
Hungary 2009
Turkey 2005
(manuf)
Romania 2009
Turkey 2008 (all)
Turkey 2008
(manuf)
Bulgaria 2009
Brazil 2009
Poland 2009
17%
34%
54%
18%29%
40%
23% 22% 23% 27%
39% 40%
19% 20% 22% 27% 30% 32%41% 42%
53%
83%
0%
30%
60%
90%
24% 27%
50% 51%
63%67%
71% 71%
9%
38%32% 35%
45%50%
38%
50%
0%
20%
40%
60%
80%
Romania 2009
Hungary 2009
Czech Rep. 2009
Chile 2006
Poland 2009
Turkey 2008
Bulgaria 2009
Brazil 2009
Production
Non-production
45
2.4 Innovation
2.68. Innovation is positively correlated with the productivity of Turkish firms. Innovation –
encompassing product and process innovation, technology adoption, quality upgrading and the use of
information communication technologies (ICT) 30
– has become crucial among world‘s economies and
much focus is put by governments towards promoting and stimulating firm-based technological
development. Analysis of the 2008 survey of Turkish enterprises confirms the importance of innovative
behavior for firm performance. Table 2-8 shows a positive association between the occurrence of
investments in R&D and productivity levels. At the same time, firms that re-organized production
processes by resorting to outsourcing where also more productive.31
Table 2-8: Summary of the effects of innovation
Explanatory IC variable
Dependent variable
Productivity Employment Probability
of exporting
Probability
of
receiving
FDI
Website (dummy) + +
R&D expenditure + +
New product (dummy) +
Foreign technology (dummy) +
Outsourcing (dummy) +
Quality certification (dummy) + + Source: Staff calculations based on Escribano et al. 2009. See Tables in Annex 2-A for detailed results.
2.69. Employment, exports and FDI are also positively associated with innovation at the firm
level. Both the 2005 and the survey present a positive association of employment with the use of ICT
(website n 2008, e-mail in 2005) in communication with customers and suppliers. Also, both periods
measured showed that firms with quality certifications have more employees than firms without.32
Also
for exports there is a significantly positive correlation between variables reflecting innovation and the
probability of exporting. In particular, firms with quality certifications are more likely to engage in
exporting activities, with an increasing importance since 2005. The use of ICT is also positively related
with exports in both years. Additionally, firms that introduced new products in 2008 exported a higher
share of their sales. The econometric analysis also shows that firms that use technology licensed by
foreign firms have higher levels of foreign ownership, a correlation that is consistent with the 2005
Survey. FDI is also more present with companies that spend on R&D activities.
30 The World Bank ICT Sector Strategy define ICT as consisting of hardware, software, networks and media for collection,
storage, processing and presentation of information, including computing, undersea cables, radio waves, spreadsheets and
websites as well as post offices and newspapers. 31 Analysis of the 2005 survey also showed the relevance of ICT use (in the form of conducting business via e-mail) for
productivity. The significance of ICT variables for productivity levels is lost in 2008. This may be explained by the fact that use
of the internet has become more widespread since 2005, thus muting productivity differences due to ICT. The purchase of new
technology also had a positive association with productivity in 2005 and turns out not to be significant in 2008. At the same time,
R&D expenditures –-not significant in 2005—have a positive association with productivity in 2008. See Table 2-A-2 for details. 32 A difference between the two periods is that firms in 2005 that owned technology licensed by foreign companies had higher
employment levels. This is no longer true in 2008, perhaps indicating that the use of foreign technology has become more
common across firm sizes.
46
Performance in innovation
2.70. R&D spending has increased but is still lower than in comparator countries. Gross
expenditures on R&D (GERD) in Turkey have seen a positive development (Figure 2-33). Overall
investment in R&D has nearly doubled over the past ten years, reaching 0.73 percent of GDP in 2008.
Although a clear improvement, Turkey still lags behind in comparison to other middle income countries
and in particular to the OECD, where gross R&D expenditure amounted to 2.29 percent in 2007,
compared to 0.71 percent for Turkey in the same year. Moreover, the share of R&D investments
accredited to the private sector remains on a low level relative to international standards. ES data show an
improvement in R&D investment as share of total sales in the manufacturing sector, from 2.8 percent in
2005 to 3.4 percent in 2008 (Figure 2-34). Figure 2-33 also illustrates that 0.34 percent of GDP in Turkey
was contributed by industry R&D investments in 2008. The increasing private expenditure ahead of
public R&D spending can to a large part be contributed to the 2007 R&D Law amendment on tax
exemptions for R&D investment.
2.71. Earlier World Bank research has showed that Turkey‟s dynamics in public-private and
university-private collaboration covering innovation and technology needs to be boosted.33
This is
confirmed by the WEF indicator, where Turkey received a mediocre rating in comparison to 132 other
economies.34
Other areas of concern have been the lack of efficient intermediaries for transfer of publicly
funded research to the private sector through for instance spin-off companies, joint research initiatives
and technology transfer offices and relatively low number of patent applications by Turkish firms, both at
home and internationally.35
33 Turkey CEM (2006) 34 The World Economic Forum Global Competitiveness Report 2009-2010 measures private cooperation with universities and
research institutes. Turkey was given a score of 3.4 on a 1-7 scale (1=no collaboration; 7=extensive collaboration). 35 Issues related to innovation for Turkey are addressed in a forthcoming World Bank report, Turkey Innovation Policy for
Competitiveness and Employment Generation.
Figure 2-33: R&D expenditure as a share of GDP -
total, private and public, 2000-2008
Figure 2-34: R&D investment as a share of total
sales, 2005 and 2008
Source: Turkstat Source: Enterprise Surveys
0.0
0.2
0.4
0.6
0.8
2000 2001 2002 2003 2004 2005 2006 2007 2008
%
Total Private Public
2.84%
3.41% 3.41%
0%
1%
2%
3%
4%
2005, manuf 2008, all 2008, manuf
47
2.72. Patent activity picked up after the 2001 economic crisis and has steadily been rising but it
remains lower than in comparator economies. The patent activity in an economy is a valuable measure
of a country‘s technological development and
commercial applicability. While patent
applications by Turkish firms increased in the
past six years, the country still shows lesser
activity than many other emerging
economies. The number of applications by
Turkish inventors per million inhabitants
amounted to 36 in 2008, an increase by over
50 percent in two years. It is also an
international improvement, where Turkey has
surpassed some comparator countries, for
example Brazil. This number is however still
low when compared to for instance 119
applications in the Czech Republic and 76 in
Poland, indicating that Turkey still needs to
catch up internationally (Figure 2-35).
2.73. The overall number of patent applications by Turkish firms locally and abroad reached
nearly 2,700 in 2008. Table 2-9 shows that 85 of these applications were filed to the U.S. Patent Office.
It is however evident that Turkish businesses file considerably more patent applications with the
European Patent Office, where 246 applications were made in 2007. Nonetheless, the application level to
both offices is a remarkable improvement compared to a decade prior (10 patents in the U.S. and 20 in the
European Patent Office). Additionally, the overall number of patents filed in Turkey‘s national patent
office has picked up significantly since the 2001 crisis, from 1,152 in 2003 to 7,137 filed applications in
2008. Another good indicator of technological improvements is industrial design applications, which is
another field where Turkey has advanced, from 2,326 applications in 2000 to 6,870 eight years later.
Table 2-9: Turkey's patent applications, 1995-2008
Patent applications with
Turkey's patent office Patent applications, total US patents European patents Industrial designs
Year (TPI) (WIPO) (USPTO) (Eurostat) (WIPO)
1995 1,690 185 6 5
1996 902 209 7 9
1997 1,531 226 10 20 1,995
1998 2,483 239 16 31 1,917
1999 3,020 332 29 22 1,779
2000 3,433 366 28 44 2,326
2001 3,214 421 28 45 2,708
2002 1,874 503 36 61 3,747
2003 1,152 603 28 85 4,208
2004 2,317 868 48 125 4,834
2005 3,518 1,120 46 163 5,380
2006 5,252 1,426 71 197 6,167
2007 6,247 2,184 66 246 6,552
2008 7,137 2,678 85 6,870
2.74. Turkish firms are to a larger extent introducing new products and are doing better in an
international comparison. The level of innovation in the Enterprise Survey does not focus on patent
Figure 2-35: Patent applications, per million population,
2000-2008
Source: WIPO and WDI
0
20
40
60
80
100
120
140
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Brazil Bulgaria Czech Rep. Poland Turkey
48
applications but instead measures the development of new products in firms as well as the level of R&D
spending. When asked if the company in the last three years has introduced a new product, 45 percent of
Turkish manufacturing firms active in the manufacturing sector responded positively, showing an
improvement since the 2005 Survey as well as in relation to comparative economies.
2.75. There are some variations in the spread of R&D investments and introductions of new
products by size, ownership and exporting status. The survey data, as depicted in Figure 2-36, shows
that R&D spending is more widespread among medium and large firms, where on average 32 percent and
34 percent of the companies respectively invest in R&D, while only 16 percent of small firms do the
same. The variations by size when considering the share of firms with new products or services were
again more prominent the larger the firm. Businesses with exporting activities also tend to invest in R&D
(52 percent) more than non-exporters (39 percent). Additionally, foreign firms tend to introduce new
products to a much larger extent than domestic.
2.76. Investments in innovation activities are more widespread in industries with relatively
advanced technologies and show some variation by region. Two thirds of the firms in the metals and
machinery industries and as much as 90 percent of companies in the electronics sector have introduced
new products in the period 2004-2007. Enterprises in the electronics industry have also shown a
noteworthy spread of R&D spending. Regionally, the largest share of firms introducing new products
could be found in the Aegean region and in Central Anatolia, while the South had the lowest share of
innovation activities among firms.
Technology adoption
2.77. The use of foreign-licensed technologies by firms in Turkey is on par with comparable
emerging economies. Using licensed technology can bring significant know-how benefits for a firm and
is a prime source of technology adoption. In Figure 2-37 we can see that the share of Turkish companies
using technology licensed by foreign-owned firms has increased, from 12 percent in 2005 to 16 percent
three years later.
Figure 2-36: Share of firms developing new products and R&D spending, by firm size, ownership form,
exporting status and region
Source: Turkey ES 2008
41%45%
55% 58%
43%52%
41%
28%
43% 44%50% 50%
16%
32% 34%
24% 23%
38%
19%24% 23% 22% 21% 21%
0%
20%
40%
60%
New product R&D spending
49
Figure 2-37: Share of firms using technology licensed by
foreign firms, country comparison
Source: Enterprise Surveys
2.78. There are significant variations by size, ownership form and region in the level of use of
technology licensed from a foreign-owned company. More than 30 percent of large firms in Turkey
apply foreign-licensed technology to their business, compared to only 8 percent of small enterprises.
More foreign firms (26 percent) acquire
technology by foreign-licensing than
domestic (15 percent) as well as
exporting businesses (19 percent)
compared to 14 of non-exporters,
suggesting that local firms lag behind
foreign competitors in the use of
technologically advanced equipment.
Also here, the Central Anatolia and
Aegean regions stick out, with 20 and 22
percent of the establishments in these
regions using foreign-licensed
technology. The Marmara region lags
behind, with a share of 13 percent
(Figure 2-38).
Quality standards
2.79. Firms in possession of quality standards tend to see a positive impact on productivity and
competitiveness. Through the application of standards, a company can facilitate the adoption of
technology and innovation as well as increase productivity and competitiveness by the embedded product
and process information that usually can be found in standards. Additionally, the positive effects are even
stronger when applying internationally recognized quality certification.36
It is therefore likely that Turkish
firms with ISO certifications are more technologically advanced and thus more competitive globally.
36 Blind and Jungmittag (2005).
6%
12% 12% 13% 13%16% 16% 16%
25%
0%
5%
10%
15%
20%
25%
Poland 2009
Czech Rep. 2009
Turkey 2005
(manuf)
Brazil 2009
Hungary 2009
Turkey 2008 (all)
Turkey 2008
(manuf)
Romania 2009
Russia 2009
Figure 2-38: Share of firms using foreign licensed technology,
by size, foreign ownership, exporting status and region
Source: Turkey ES 2008
8%
19%
31%26%
15%19%
14% 13% 14%17%
20%22%
0%
10%
20%
30%
50
2.80. The application of ISO
9001 in Turkey has shown
remarkable increases since the
2001 crisis, with more than 13,200
certificates issued by the end of 2008
and comparing relatively well to
other economies. The survey data
depicts additionally improving
figures, with 30 percent of all Turkish
firms in 2008 claiming to have an
internationally recognized quality
certification (Figure 2-39). This puts
Turkey ahead of other middle-income
countries, such as Brazil (26 percent),
Bulgaria (20 percent) and Poland (17
percent).
2.81. Certifications among small firms
lag far behind medium and large
companies. 55 percent of large firms hold a
quality certification, which is three-fold of
the share of small firms. Nevertheless, Figure
2-39 shows that small firms in Turkey seem
to be doing much better than small firms in
for instance Brazil, Bulgaria and Poland.
Both foreign-owned companies (55 percent)
and firms that export (52 percent) tend to
have such certificates to a significantly larger
extent than domestic (29 percent) and non-
exporting firms (24 percent). Regionally, a
larger share of quality certified enterprises
can be found in Central Anatolia (40 percent). The share of firms with certificates in the Aegean region is
on the other hand only 22 percent (Figure 2-40).
2.82. The industry variations in Turkey confirm that sectors with higher technologies tend to
hold quality certifications more often than traditional lines of business. For instance, 90 percent of
firms active in the electronics industry hold internationally recognized quality certifications, while the
corresponding numbers in for example garments and textiles are 18 and 28 percent. When interpreting the
survey data for Turkey it is however important to keep in mind that the firms‘ answers reflect not only
formally recognized quality certifications, but can also include expired certificates as well as certification
bodies not accredited by TÜRKAK, the national accreditation body. The market for certification services
is in general relatively broad in Turkey but the quality of the certifications is not always strictly examined.
Many smaller certification bodies are subsidiaries of European bodies accredited in their home market
and seem to be approving certificates more quickly than the standard time for review and audit, thus
creating an anticompetitive environment.
Figure 2-39: Share of firms with internationally recognized quality
certifications, country comparison
Source: Enterprise Surveys
Figure 2-40: Share of firms with internationally
recognized quality certifications, by region
Source: Turkey ES 2008
17%20%
25% 26% 26%30%
39% 39%43%
8%12%
24%
8%
21% 20%
27% 27%30%
0%
10%
20%
30%
40%
50%
Poland 2009
Bulgaria 2009
Turkey 2005
(manuf)
Brazil 2009
Romania 2009
Turkey 2008 (all)
Turkey 2008
(manuf)
Hungary 2009
Czech Rep. 2009
All firms Small firms
22%27%
30% 31%
40%
47%
35%33%
46% 45%
0%
10%
20%
30%
40%
50%
Aegean Marmara South Black Sea -Eastern
Central Anatolia
Total Manufacturing industry
51
Information and communication technologies
2.83. Turkish firms continue to increase the use of ICT in their operations. The enterprise survey
measures ICT among firms by the level of e-mail and website usage in communication with clients and
suppliers. The results suggest that both ICT factors have increased since 2005, with 75 percent of Turkish
firms having their own websites and as many as 89 percent communicating per e-mail.
2.84. The level of internet use is
influenced by firm size as well as type. As expected, large and medium firms use
ICT more extensively than small firms,
although there has been a ―catch up‖ by
SMEs to the level of larger enterprises.
This catch up is seen more clearly when
comparing small enterprises in Turkey to
small firms in other emerging economies.
As was the case with quality
certifications, small firms in Turkey seem
to use ICT to a larger extent than in for
example Brazil and Poland (Figure 2-41).
By ownership form and exporting status,
it is evident that foreign firms and
exporters are more advanced in their
application of internet. Especially in the
share of firms having their own website
do non-exporters seem to be lagging
behind (Figure 2-42).
2.85. There are regional and sectoral
differences in e-mail and website usage,
with firms the Aegean region using e-mail
in business communication the most.
Interestingly enough, companies in the
same region seem to have the least spread
of website utilization. Finally, when
looking at industry variations, it appears
that the electronics and metals sectors in
Turkey are very advanced in their use of ICT, while less technology-intensive industries, such as food,
garments and textiles are well below the overall country average in both e-mail and website usage.
2.86. With the implementation of the e-Transformation Turkey Project, the expansion of ICT in
public services has received a boost. The Government has also stepped up initiatives to raise awareness
among citizens and businesses. Further support is planned for enterprises in their use of ICT as well as
increased competition in the electronic communication sector. The overall reform initiatives by the
Turkish Government are summarized in Box 2-7. One can see that the Ninth Development Plan 2007-
2013 among others includes ambitious objectives in the rise of internet usage in Turkey.
Figure 2-41: Use of websites in communication with clients
and suppliers, country and small firm comparison
Figure 2-42: Use of ICT in communication with clients and
suppliers, by size, ownership form and exporting status
Sources: Enterprise Surveys and Turkey ES 2008
35%
48%
61% 65%75% 75% 75% 76%
29%
41%47%
56%
68% 68%
56%
66%76%
0%
20%
40%
60%
80%
Romania 2009
Bulgaria 2009
Turkey 2005
(manuf)
Poland 2009
Hungary 2009
Turkey 2008 (all)
Brazil Turkey 2008
(manuf)
Czech Rep. 2009
All firms
Small firms
85% 93% 96% 93% 89% 95% 87%
68%82%
91% 86%75%
92%
70%
0%
20%
40%
60%
80%
100%
E-mail Website
52
Box 2-8: Technology and innovation reforms
The new R&D law in Turkey, implemented in March 2008, presents an extensive incentive and fiscal policy
program, with the objective of reducing the R&D gap and encouraging economies of scale in the sector. Main
features in the R&D law include (i) tax exemptions for R&D personnel; (ii) R&D and innovation expenditure
discounts; (iii) social security payment deductions for R&D personnel; (iii) stamp tax exemption; and (iv) techno-
entrepreneurship capital support for eligible recent university graduates.
The Government‘s comprehensive research, technology and innovation objectives are described in the Ninth
Development Plan 2007-2013 (exact goal as share of GDP), the Medium Term Program 2010-2012, the National
Science and Technology Strategy 2005-2010, TÜBITAK‘s National Innovation Strategy 2008-2010 as well as
YOİKK‘s Action Plan for 2009.
Concrete policy goals in the area of R&D, stretching until 2013, are presented in the Ninth Development Plan and
cover: (i) increase in the share of R&D expenditure from 0.7 to 2 percent of GDP; (ii) raise in the number of full
time researchers from 24,000 to 80,000; (iii) upsurge in the share of R&D expenditures financed by the private
sector from less that 29 to 60 percent; (iv) increase in the broadband subscriber penetration rate from 3.5 to 20
percent; as well as (v) improvement in the internet user penetration rate from 20 to 60 percent.
The YOİKK Technical Committees for Intellectual and Industrial Property Rights (IIPR) as well as for R&D have
presented short-term goals for IIPR, innovation and technology in the 2009 Action Plan. Amendments to the Patent
Law, Patent Authority Law and Trademark law will be drafted for harmonization with the EU Acquis and
international regulations. Improvements to the university curricula on the subject of IIPRs and advancement of the
qualifications and skills of IIPR courts and law enforcers are also proposed. The R&D Committee is also preparing a
legal amendment, which will simplify the process of employing foreign research staff in Turkish firms and R&D
institutions.
Strong emphasis has also been put on better university-industry cooperation. The Government is planning to
establish Technology Development Regions, bringing together academic and private sector establishments as well as
Technology Transfer Centers, conveying new technologies established at universities to the industry. Cooperation
between the public and private sector in regards to IIPR and copyrights will also expand through Coordination
Councils.
2.5 Access to Finance and Corporate Governance
2.87. The econometric analysis from the survey confirms the positive association of access to
finance and corporate governance with productivity. There is much empirical evidence supporting the
positive correlation between finance, productivity and growth, corroborating the idea that an efficient
financial system helps firms to access credit and thus increase their investment and productivity.37
A
sound financial system promotes economic growth by facilitating firms‘ access to external financing,
allowing them to become more productive and expand.38
The analysis, as summarized in Table 2-10,
indicates that higher productivity in 2008 was related to several variables representing financial
soundness. Higher levels of productivity were observed in firms with higher share of sales paid for before
delivery and with the ability to finance a higher proportion of fixed assets purchases with internal funds. It
is also notable that firms are less productive when financing new investments through state-owned banks.
Also, apparently, firms in receipt of state financial support were more productive. The level of ownership
concentration also has a positive correlation with productivity levels.39
37 See Levine (2005) for a comprehensive survey of the theoretical and empirical literature. 38 See for instance Demirgüç-Kunt and Levine, 2008 39 In the analysis of 2005 data, the only significant variable in the finance and corporate governance block was the use of external
auditory, with a positive effect on TFP. This variable was not significant in 2008 regressions.
53
Table 2-10: Summary of the effects of finance and corporate governance
Explanatory IC variable
Dependent variable
Productivity Employment Probability
of exporting
Probability
of
receiving
FDI
Sales paid before delivery +
Purchases paid after delivery +
New fixed assets finance - internal funds +
New fixed assets finance - equity +
New fixed assets finance - state banks - -
Loan (dummy) +
Land/buildings as collateral (dummy) +
Subsidies (dummy) +
Largest shareholder +
External audit (dummy) + +
Source: Staff calculations based on Escribano et al. 2009. See Tables in Annex 2-A for detailed results.
2.88. Financing structure presents a strong association with employment and exporting, while a
more limited relationship emerges with foreign ownership. Larger firms are more likely to pay sales
after delivery – with effects on the availability of working capital. Confirming the findings of the 2005
survey, employment levels are also positively associated with access to a bank loan and with externally
audited accounts, which, in turn, favors the likelihood to access bank finance. By virtue of greater
availability of internal and external financial resources, larger firms also less likely to finance their
purchases of fixed assets with recourse to finance from a state bank. The probability of exporting is
positively associated with the use of immobile assets, i.e. land and buildings, as collateral and with the
ability to finance the purchase of fixed assets with new equity. Finally, there is a positive correlation
between the level of FDI and firms‘ having their financial statements verified by an external auditor.
Overview of the financial sector
2.89. Despite the recent financial
turmoil, Turkish firms‟ perception of
access to finance improved
significantly in the period 2005-2008. The continued recovery of the economy
after the 2001 crisis, extensive
privatization activities and more sound
fiscal policies had a positive impact on
the enterprise sector. Turkey is also
experiencing growing domestic credit to
private enterprises and a transition from a
state-dominated financial system to more
private sector involvement. Improved
access to finance for firms is today one of
the Government‘s main objectives set up
in the Ninth Development Plan 2007-
2013, under the Increasing Competitiveness axis. The positive change in opinion can be noticed in the
survey firm responses, where the share of firms considering access to finance to be a major obstacle to
Figure 2-43: Access to finance as an obstacle , country and
small firm comparison
Source: Enterprise Surveys
12% 13% 14%17%
22% 24%
37%
47%
56%
13% 16%16%
18%22%
21%
40%
49%50%
0%
20%
40%
60%
Hungary 2009
Turkey 2008
(manuf)
Turkey 2008 (all)
Bulgaria 2009
Poland 2009
Czech Rep. 2009
Romania 2009
Turkey 2005
(manuf)
Brazil 2009
All firms
Small firms
54
operations had gone down from 47 to 13 percent in the measured periods (Figure 2-43). The financial
system is also much less of an obstacle in Turkey than in comparative countries, such as the Czech
Republic, Romania and Brazil. The data also indicate that smaller enterprises tend to be obstructed by
financial factors to a larger extent than large and medium firms.
2.90. Nonetheless, the financial sector in Turkey is yet to reach international development
standards. According to a recent study by the Banks Association of Turkey, the ratio of financial assets
to GDP in 2007 was 150 percent in Turkey,
while the responding share for emerging
economies was 246 percent and the global
average was 421 percent. Furthermore, the
Turkish financial sector remains dominated
by banks, which amount to a total of 45, of
which 32 are depository banks. The change
from 2005 has mainly consisted in a shift
from domestic to a majority of foreign-
owned banks (Figure 2-44). Banks
contributed with nearly 78 percent of total
assets in the beginning of 2009, as seen in
Table 2-11. The Central Bank‘s assets
amounted to 12.8 percent, while 3 percent
and 1.8 percent came from securities funds
and financial leasing companies respectively.
Table 2-11: Structure of the financial system in Turkey, 2002-2009
(TL Billion) 2002 2003 2004 2005 2006 2007 2008
2009
/03 % Distribution
Mar 09
CBRT 74.1 76.5 74.7 90.1 104.4 106.6 113.4 122.7 12.8
Banks 216.7 255.0 313.8 406.9 499.5 581.6 732.8 754.2 78.7
Financial Leasing Companies 3.8 5.0 6.7 6.1 10.0 13.7 17.2 16.9 1.8
Factoring Companies 2.1 2.9 4.1 5.3 6.3 7.4 7.8 7.2 0.8
Insurance Companies 5.4 7.6 9.8 14.4 17.4 22.1 26.5 14.8 1.5
Pension Companies 0.0 3.3 4.2 5.7 7.2 9.5 12.2 12.9 1.3
Securities Investment Funds 9.3 19.9 24.4 29.4 22.0 26.4 24.0 28.7 3.0
Other1 2.7 3.5 4.2 7.8 9.1 12.3 13.8 13.9 1.4
Total 314.1 370.4 437.7 560 668.6 770.1 935.5 958.4 100
Source: BRSA Financial Markets Report, Mar 2009 1Other include Consumer Finance Companies, Securities Intermediary Institutions, Securities Investment Partnerships and Real
Estate Investment Partnerships
2.91. Market capitalization on the Istanbul Stock Exchange (SE) has been hit by the financial
crisis limiting the possibilities of non-bank finance for the Turkish business sector. Market
capitalization amounted to 15 percent of GDP at the end of 2008 ranking Turkey very low in comparison
to other emerging economies (Figure 2-45). Nevertheless, up until one year prior, the Istanbul SE has
experienced a steady growth, with a 10 percentage point cumulative increase in 2005-2007. Figure 2-46
illustrates that a majority of the 295 firms listed on the stock exchange today are manufacturing
companies, followed by the financial sector with 33 percent, the two sectors best served by the exchange.
The development since the 2001 economic crisis indicates an increasing interest in the Turkish stock
exchange and motivates improved corporate governance, stronger regulations and market-related
information. These types of financial intermediation are becoming increasingly important in today‘s
volatile status of the global economy.
Figure 2-44: Turkish banks by ownership, 2005 and 2009
Source: Banks Association of Turkey
6.4%
36.2%
27.7% 29.8%
6.7%
24.4%
37.8%31.1%
0%
20%
40%
60%
State-owned banks
Private, domestic banks
Foreign banks Development and investment banks and SDIF
controlled banks2005 2009
55
Box 2-9: The Turkish banking sector in the wake of the crisis
The Turkish banking sector has proven resilient to the effects of the global credit crisis. The system remains
highly-capitalized and profitable, despite a nontrivial deterioration in asset quality. Unlike in other emerging
market economies, the Turkish banking system relies on a strong and stable domestic deposit base for funding, as
reflected in relatively low loan/deposit ratio of 126 percent, compared to a regional ECA average of 171 percent.
Wholesale funding, including syndications and securitizations, is relatively low at about 15 percent of non‐equity
liabilities. The performance of the banking sector has been aided quite significantly by prompt and effective
policy action on the part of the government and the monetary authorities to maintain stability in financial markets.
Specifically, profitability has been supported in 2009 by a rapid widening of net interest margins as the policy rate
has been cut, enabling banks to reduce their deposit costs and compensate for higher provisioning charges. Capital
adequacy was relatively high at 19 percent as of end‐June 2009.
Nevertheless, as a result of the global credit crisis and the ensuing uncertainty, lending volumes have stalled
and exposure to government securities increased. Since end‐2008, commercial bank loan volumes have been
flat, whereas securities holdings increased by 14 percent (31.5 percent of total assets as of end-December 2009
compared to 26.5 percent end-2008). Lending to SMEs, a sector that had been growing significantly since 2005,
has actually dropped in relative terms, now accounting for less than 22 percent of total lending, compared to
almost 24 percent at end-2008. In this connection, while lending rates, after peaking in October 2008, have been
on downward trend, there is now evidence of increased segmentation in the market, with rates charged to SMEs
clients exceeding by far those offered to larger corporate clients. So far, the performance of corporate and
commercial loans and mortgages has held up relatively well, whereas NPLs on SMEs, credit cards and unsecured
consumer loans has grown considerably. The NPL ratio for the system as a whole was 4.9 percent at end-June
2009 (compared with rates around 3 percent before the crisis).
2.92. Bank credits to the private sector are increasing, but lag behind other emerging economies. After the record low of 12 percent of GDP in the aftermath of the 2001 crisis, claims on the private sector
reached 32 percent in 2008 (Figure 2-47). Compared to other economies however, such as Russia, Brazil
and Chile, Turkish credit to the private sector remains relatively low.
Figure 2-45: Market capitalization of firms listed on the
Istanbul SE/GDP, 2005, 2007, 2008 and comparator countries
Figure 2-46: Firms listed on the ISE, by
industry, 2009
Source: WFE and WDI Source: Istanbul Stock Exchanges
12%15%
17% 18%22%
33%37%
44%
0%
10%
20%
30%
40%
50%
Hungary Turkey 2008
Poland Ireland Mexico Turkey 2005
Brazil Turkey 2007
Manufacturing, 51.9%
Financial institutions, 32.9%
Wholesale, retail,
hotel and restaurant
s, 5.8%
Other services,
5.4%Technolog
y, 4.1%
56
Figure 2-47: Claims on the private sector, % of GDP, 2008
Source: IMF International Financial Statistics
2.93. This growth in credit to the private sector is however increasingly targeted towards
individual consumers rather than enterprises, especially SMEs. Figure 2-48 shows that consumer
credits have surged by more than 700 percent
since 2004, compared to firm loans that grew
by 400 percent, the latter in fact showing a
decline since the beginning of 2009. This
could be a reflection of the improvement of
lending technologies to consumers but at the
same time, the lack of adequate institutions
for credit issuance to SMEs have slowed the
expansion of firm loans. Furthermore,
entrepreneurs with small startup that have
trouble receiving firm loans can receive
credit for their business by instead using
private immobile assets as collateral. This
would also as a result create a bias towards
consumer loans.
2.94. The survey data show that the share of Turkish firms that do have loans is ahead of
comparator countries, even when consider smaller firm sizes. Although the share of small enterprises
17.3%25.0%
31.6%41.9%
50.4%
83.9%
0%
20%
40%
60%
80%
100%
Mexico Turkey 2005 Turkey Russia Brazil Chile
Figure 2-48: Private sector credit by recipient (biannually,
billion TL)
Source: Central Bank of Turkey
Figure 2-49: Share of firms with loans, country and small firm
comparison
Figure 2-50: Share of firms with loans,
by region
Source: Enterprise Surveys Source: Turkey ES 2008
0
20
40
60
80
100
Dec
-04
Ap
r-0
5
Au
g-0
5
Dec
-05
Ap
r-0
6
Au
g-0
6
Dec
-06
Ap
r-0
7
Au
g-0
7
Dec
-07
Ap
r-0
8
Au
g-0
8
Dec
-08
Ap
r-0
9
Consumer Loans
Firm loans
40% 43% 46% 47%50%
57% 57%65%
69%
35% 42% 40% 43%48% 50% 49%
46%
65%
0%
10%
20%
30%
40%
50%
60%
70%
80%
Bulgaria 2009
Hungary 2009
Turkey 2005
(manuf)
Czech Rep. 2009
Poland 2009
Turkey 2008 (all)
Turkey 2008
(manuf)
Brazil 2009
Chile 2006
All firms
Small firms
38%
55%60% 61% 62%
0%
10%
20%
30%
40%
50%
60%
70%
Black Sea - Eastern
Central Anatolia
Marmara South Aegean
57
with loans (50 percent) is smaller than that of medium and large firms, amounting to 68 and 62 percent
respectively, this is still better than in other comparative emerging economies, as for instance the Czech
Republic, Poland and Brazil (Figure 2-49). There are significant differences in firms with loans measured
by region, as depicted in Figure 2-50. 38 percent of enterprises in the Black Sea – Eastern region have
loans while the corresponding number for firms in the Aegean region is 62 percent.
2.95. Turkish firms rely more on bank loans for investment financing than do firms in other
countries but internal funds remain the main source of finance for investments. Internal funds and
bank loans are by far the most important financial sources for purchase of investments, amounting
together to 93 percent of financing. As shown in figure 2-51, more than half of the funds are internally
generated, while issuance of new equity shares and advances from customers play a smaller role for firms
funding.
Figure 2-51: Sources for finance of fixed assets, country comparison
Source: Enterprise Surveys
2.96. In spite of the severe effects of the crisis on Turkey, its private sector firms continued to
have access to international borrowing. Corporate rollover of foreign debt amortizations fell below 100
percent in December 2008 for the first time since January 2007. However, firms continued to roll over the
bulk of their debt service, with drawings tracking repayments quite closely month on month. Over the
first nine months of 2009 the rollover ratio of the non-banking corporate sector‘s external debt was just
over 70 percent.
2.97. Recent actions in facilitating credit access, in particular to SMEs, have been initiated,
among others through increased funding to the public guarantee fund. As discussed in greater detail
in Chapter 3, the Credit Guarantee Fund of Turkey (CGF) supports SMEs‘ access to credit, and through
planned legislative action funding of CGF will be increased by an additional TL 1 billion in the next two
years for leveraging additional resources for SMEs. The Government is mitigating risk-averse behavior in
the banking system by limiting theses guarantees to 65 percent of loan amounts and sharing the burden
with the Treasury and through participation of commercial banks for the guarantees. Funding to the Small
and Medium Industry Development Organization (KOSGEB) has also been increased by 48 percent for
support of existing credit subsidies and technical support programs. Moreover, the credit line with the
CBRT of Turkey‘s Eximbank, which provides export credit guarantees for Turkish exporters was
increased to USD 1 billion in 2008.
2.98. Planned cooperation between the Capital Markets Board and the Istanbul Stock Exchange
could reap benefits for SMEs in future capital markets inclusion. Studies are underway to establish
0%
20%
40%
60%
80%
100%
Brazil 2009
Chile 2006 Turkey 2008
Poland 2009
Hungary 2009
Bulgaria 2009
Czech Rep. 2009
Russia 2009
Other
Advances from customers
Bank loans
New equity shares issued
Internal funds
58
the ISE Emerging Companies Market (ECM),40
as a projected collaboration between ISE and CMB.
Companies that have been registered with the CMB but have not fully satisfied the listing requirements of
ISE will be able to trade on the exchange. The initiative aims to create a transparent and organized floor
where SMEs with growth potential can secure funding from the capital markets to lower registration and
transaction costs and with more lenient auditory requirements. A market advisory mechanism will be
created in order to provide firms with information on capital markets and exchange regulation, both prior
to and during the application process, as well as regulatory support after admission. The technical and
regulatory preparations of the ISE ECM are currently ongoing.
Credit Information
2.99. Turkey‟s two large credit bureaus are the Credit Bureau of Turkey (KKB) and the Credit
Registry of the Central Bank. KKB is owned by nine banks and shares information among 38 financial
institutions. The bureau has in a short time increased its coverage, from 27 percent in 2005 to 43 percent
in 2009 (Figure 2-52). The Credit Registry of the Central Bank covers both individuals and firms, with
information collected from both bank and non-bank financial institutions. However, the Credit Registry‘s
coverage, as seen in Figure 2-53 is albeit improvements lower than KKB. Moreover, the Banks
Association of Turkey is expected to outsource the technical operation to KKB in order to expand
coverage of debtors (HEG-DPL PD 2009). The KKB has established the framework to expand beyond
coverage of individual consumer credit information to also cover the corporate sector. Although this
initiative has not yet been implemented, the current monitoring of individual consumers is also very
useful, especially for information on small enterprises. Since entrepreneurs, as earlier mentioned, do not
face easy approval of corporate loans, they often finance their investments through personal loans, and the
credibility of the owner is thus often reflected on the credibility of their operations. As discussed in
Chapter 3, continued progress in enhancing credit information is crucial to facilitate access to bank credit,
especially for SMEs.
Figure 2-52: Private credit coverage, percentage of
adults
Figure 2-53: Public credit coverage, percentage of
adults
Source: Doing Business 2010
Financial Reporting
2.100. Use of external auditors by Turkish firms has increased notably since 2005. 55 percent of the
firms have their financial accounts externally audited, compared to 42 percent in 2005, showing a
favorable international comparison (Figure 2-54). This has been reflected in an improvement in access to
bank finance for firms and aided credit providers in their quest to assess firms‘ abilities in meeting credit
40 The legal background for the ECM initiative has been published on August 18 2009, as the ―Emerging Companies Market
Regulation‖ in the Offical Gazette, no. 27323. More information can be found on the CMB website
http://www.spk.gov.tr/duyurugoster.aspx?aid=200994&subid=0&ct=c&submenuheader=null (Turkish)
6% 10%
27% 30% 34%43%
59%68% 73%
0%
20%
40%
60%
80%
0% 0%5% 5% 6%
16%
24%
33% 35%
0%
10%
20%
30%
40%
59
obligations. As is often the case in comparable economies, small and medium sized firms in Turkey use
external auditors to a lesser extent than larger firms.
Figure 2-54: Firms having financial accounts externally audited, country
comparison
Source: Enterprise Surveys
2.101. Audit reforms introduced in the past years in Turkey, have had a positive initial influence. Such reforms as required use of International Financial Reporting Standards and International Standards
of Audit have had a strong impact on the increased number of externally audited financial accounts. Since
October 2006, the Turkish Accounting Standards Board (TASB) has been publishing the Turkish
Financial Reporting Standards (TFRS), based on the above mentioned IFRS. The TFRS is currently
solely directed by such regulatory agencies as the Capital Markets Board (CMB) and the Banking
Regulation and Supervision Agency (BRSA), but is expected to be applied in a larger scale with the
ratification of the Commercial Code (CC), which has been pending in Parliament since 2005. The draft
CC also stipulates external auditing of all firms, including those not covered by CMB and BRSA. As
argued in Chapter 3, swift adoption of the Commercial Code with its provisions regarding the wide
adoption of financial reporting standards will be beneficial to SME borrowing.
Use of Collateral
2.102. About two thirds of
firms in Turkey provided
collateral for their latest
loan in 2008, using mainly
personal and immobile
assets. It is well known that
firms that are able to commit
collateral obtain larger loans
and on more beneficial terms
than firms without collateral.
Collateral not only provides
the creditor additional
security in the case that the
loan cannot be repaid but
also improves the borrower‘s
access to credit. Collateral
24% 26% 28%37%
42%47%
55% 55%64%
69%
0%
20%
40%
60%
80%
Poland 2009
Brazil 2009
Bulgaria 2009
Czech Rep. 2009
Turkey 2005
(manuf)
Chile 2006
Turkey 2008
(manuf)
Turkey 2008 (all)
Argentina 2006
Hungary 2009
Figure 2-55: Share of firms for which collateral was required for their
latest loan, country comparison
Source: Enterprise Surveys
32%
47% 49%
63% 65%71% 73% 76%
83% 85%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
Brazil 2009
Chile 2006
Turkey 2005
(manuf)
Poland 2009
Turkey 2008 (all)
Turkey 2008
(manuf)
Czech Rep. 2009
Romania 2009
Bulgaria 2009
Hungary 2009
60
was used for 71 percent of loans in the manufacturing sector, an improvement from less than 50 percent
in 2005 (Figure 2-55), placing Turkey on par with other middle income countries.
2.103. Meanwhile, the most common type of asset provided as collateral is still land and buildings,
in fact gaining importance
since 2005 (Figure 2-56).
This could be a reflection in
the numerous actions taken
to improve housing finance
and thus improving access to
finance for firms pledging
their real estate as collateral
for smaller firms, where
entrepreneurs commit their
personal assets.41
These have
become more common as
use for collateral, while
movables, which were the
second largest group of
collateral in 2005 are today
least significant. Considering that machinery and equipment were the most common asset type among
firms in 2005, the shift away from this group may indicate a tightening of credit standards.
Corporate governance
2.104. Economies are increasingly recognizing the importance of corporate governance in
achieving policy goals and improving growth and efficiency. The concept of corporate governance
incorporates the relationship between a firm‘s management and its stakeholders, among others board,
shareholders, employees, labor unions, customers and suppliers. It also determines the structure of a
company‘s objectives and monitors overall business performance.42
Through its concept of accountability,
corporate governance ensures that the interests of all stakeholders are taken into consideration by the
management in its decision making process.
2.105. Strong corporate governance helps firms attract investment, improves private sector
productivity and strengthens a country‟s financial stability. Through high levels of transparency and
accountability in business procedures as well as improved monitoring of financial flows and firm
performance, an economy can increase access to external financing by attracting investors; develop a
more prosperous stock market, improve operational performance; reduce risk of financial crises; and build
a better relationship of all stakeholders in a firm.
2.106. The development of Turkish corporate governance has been positive on the extent of
disclosure, but remains poor in investor protection when considering director liability and the
possibility of bringing shareholder suits. Doing Business measures the strength of minority shareholder
protections against directors‘ misuse of corporate assets for personal gain. The indicators distinguish 3
dimensions of investor protection: transparency of transactions (extent of disclosure index), liability for
41 Turkey has in the past years taken several actions to promote housing finance. With the passing of a new mortgage law in
2007, the Government has taken an important step in the facilitation of the housing market. Banks are now allowed to charge
floating rates on mortgages and to introduce prepayment penalties. A secondary regulation, allowing banks more flexibility in the
design and securitization of mortgages as well as better risk management was issued in August 2007 (World Bank 2008). 42 See OECD Principles of Corporate Governance (2004) for a detailed overview in the subject as well as OECD Corporate
Governance in Turkey (2006).
Figure 2-56: Type of collateral, 2005 and 2008
Source: Turkey ES 2008
43%
31%
8%
30%
61%
15%19% 20%
39%
55%
Machinery and equipment
including movables
Other Accounts receivable and
inventories
Personal assets of owner
Land and buildings
2005 2008
61
self-dealing (extent of director liability index) and shareholders‘ ability to sue officers and directors for
misconduct (ease of shareholder suits index). The indices score each country on a scale of 1-10, where the
highest represents the most favorable level.
2.107. The 2010 Doing Business Report, as illustrated in Figure 2-57, gives Turkey an average score
5.7. Table 2-123 depicts the Turkey has improved its level of disclosure, by requiring an external body to
review business transactions before they are processed. The country also ranks well on immediately
disclosing transactions to the public and/or shareholders through published periodic filings. However,
Turkey does still not hold directors adequately liable, among others in the case of transactions that were
unfair, oppressive or prejudicial to minority shareholders as well as when requiring beneficiaries to return
excess profits realized through unfair transactions. Turkey also scores lower than most comparator
economies in the possibilities for shareholders to pursue lawsuits. The low score comes from not allowing
shareholders with less than 10 percent interest to request an inspector to investigate the transaction or
inspect relevant documents before filing suit.
2.108. Several measures have been initiated recently to improve corporate governance among
Turkish enterprises. The Capital Markets Board, responsible for monitoring and increasing transparency
in Turkish capital markets, has introduced mandatory corporate governance compliance reports and
requires the establishment of audit committees. Further improvements came through the accounting and
auditing reforms discussed in the financial reporting section above. Moreover, rating principles for
guarantee funds were brought up to international standards in 2008. The issue is widely covered by
several recent Governmental policy publications, such as the Ninth Development Plan 2007-2013 and the
YOİKK Action Plans 2009 and 2010. Upcoming reforms include legislative amendments on ―Director‘s
Liability‖ and ―Ease of Shareholder Suits‖ as presented by Doing Business, the establishment of
corporate reporting standards for firms not listed on the stock market, as well as preparation of a new
corporate governance system for companies listed on the ISE.
Figure 2-57: Protecting investors index, scale 1-10 Table 2-12: Protecting investors index, scale 1-10
Country
Extent of
disclosure
Extent of director liability
Ease of shareholder
suits
Hungary 2 4 7
Czech Rep. 2 5 8
Russia 6 2 7
Brazil 6 7 3
Turkey 2005 8.0 4 4
Turkey 9 4 4
OECD avg. 5.9 5.0 6.6
Bulgaria 10 1 7
Chile 7 6 5
Poland 7 2 9
Romania 9 5 4
Source: Doing Business 2010
4.3
5.0 5.05.3 5.3
5.7 5.8 6.0 6.0 6.0 6.0
0
1
2
3
4
5
6
7
62
References
Banks Association of Turkey. 2009. The Banking System in Turkey. June 2009.
Blind, Knut and Andre Jungmittag. 2005. "Trade and the impact of innovations and standards: the case of Germany
and the UK." Applied Economics, Taylor and Francis Journals. vol. 37(12)
Cameron, A.C. and P.K Trivedi, 2005. ―Microeconometrics: Theory and Applications‖. Cambridge University
Press.
Conway, Paul, Donato De Rosa, Giuseppe Nicoletti, and Faye Steiner. 2006. ―Regulation, Competition and
Productivity Convergence,‖ OECD Economics Department Working Paper, No. 509. Paris.
Demirgüç-Kunt, Asli and Levine, Ross. 2008. ―Finance, Financial Sector Policies, and
Long-Run Growth.‖ The World Bank. Policy Research Working Paper No. 4469
Djankov, Simeon, Rafael La Porta, Florencio Lopez-de-Silanes and Andrei Shleifer. 2002. ―The Regulation of
Entry.‖ The Quarterly Journal of Economics, Vol. 116
Djankov, Simeon, Caroline Freund and Cong S. Pham. 2008. ―Trading on Time.‖ Review of Economics and
Statistics. Nov 2008.
Escribano A. and J. Luis Guasch. 2005. ―Assessing the Impact of the Investment Climate on Productivity using Firm
Level Data: Methodology and the Cases of Guatemala, Honduras and Nicaragua‖. Policy Research Working Paper #
3621, The World Bank, June.
Escribano A. and J. L. Guasch. 2008. ―Robust Methodology for Investment Climate Assessment on Productivity:
Application to Investment Climate Surveys from Central America‖. Working Paper # 08-19 (11), Universidad
Carlos III de Madrid.
Escribano, A., J. L. Guasch, M. de Orte and J. Pena. 2008a. ―Investment Climate and Firm‘s Performance:
Econometric and Applications to Turkey‘s Investment Climate Survey.‖ Working Paper # 08-20 (12), Universidad
Carlos III de Madrid.
Escribano, A., J. L. Guasch, M. de Orte and J. Pena. 2008b. ―Investment Climate Assessment Based on Demean
Olley and Pakes Decompositions: Methodology and Applications to Turkey‘s Investment Climate Survey.‖ Working
Paper # 08-20 (12), Universidad Carlos III de Madrid.
Escribano, A., J. L. Guasch and J. Pena. 2008. ―Assessing the Impact of Infrastructure Quality on Firm Productivity
in Africa‖ Mimeo.
Escribano, Alvaro, Manuel de Orte, Jorge Pena. 2009. ―Econometric Analysis of Turkish Investment Climate –
Analysis based on firm level data of the manufacturing sector in 2008 and 2005.‖ Background paper. World Bank:
Washington, DC.
European Commission. 2009. eGovernment Turkey Factsheet.
European Patent Office. 2009.
Griliches Z. and J. Mairesse. 1997. ―Production Functions: The Search for Identification.‖ In S. Strom (ed.) Essays
in Honor of Ragnar Frisch, Econometric Society Monograph Series, Cambridge University Press, Cambridge.
International Organization for Standardization. 2009. The ISO Survey of Certifications 2008. Genève
63
Levine, Ross. 2005. ―Finance and Growth: Theory and Evidence.‖ Handbook of Economic Growth, in: Philippe
Aghion & Steven Durlauf (ed.), Handbook of Economic Growth, edition 1, volume 1, chapter 12, pages 865-934
Elsevier.
Nicodeme, Gaetan and Jacques-Bernard Sauner-Leroy. 2007. ―Product Market Reforms and Productivity: A Review
of the Theoretical and Empirical Literature on the Transmission Channels.‖ European Communities. Brussels
Magnus J. R. and A. L. Vasnev . 2007. ―Local Sensitivity and Diagnostic Tests‖. Econometrics Journal, V. 10, 166-
192.
Marschak J. and W. H. Andrews. 1944. ―Random Simultaneous Equations and the Theory of Production‖.
Econometrica, 12 (3,4), 143-205.
Olley, G. S. and A. Pakes. 1996. ―The Dynamics of Productivity in the Telecommunications Equipment Industry‖.
Econometrica, Vol. 64, 6, 1263-1297.
Organization for Economic Cooperation and Development (OECD). 2004. OECD Principles of Corporate
Governance. Paris
Organization for Economic Cooperation and Development (OECD). 2006. Corporate Governance in Turkey: A
Pilot Study. Paris
Organization for Economic Cooperation and Development (OECD). 2007. ―Enhancing Turkey's Growth Prospects
by Improving Formal Sector Business Conditions.‖ Economics Department Working Papers no. 542
Organization for Economic Cooperation and Development (OECD). 2009. Population and Employment Statistics
Organization for Economic Cooperation and Development (OECD). 2009. Science and Technology Statistics
Organization for Economic Cooperation and Development (OECD). 2009. ―Ten Years of Product Market Reform in
OECD Countries - Insights from a Revised PMR Indicator.‖ Economics department working Papers no. 695
Republic of Turkey State Planning Organization. 2006. Ninth Development Plan 2007-2013. Decision No: 877.
Prescott, Edward C. (1998) "Needed: A Theory of Total Factor Productivity". International Economic Review, 39,
525-552.
Republic of Turkey State Planning Organization. 2008. 2009 Annual Programme. Decree Number: 2008/14200.
Republic of Turkey State Planning Organization. 2009. Medium Term Programme 2010-2012. Official Gazette No.
27351, 16.09.2009
Republic of Turkey. 2009. Technical Committee Action Plans of the Turkish Coordination Council for the
Improvement of the Investment Environment (YOİKK).
Republic of Turkey. 2010. Technical Committee Action Plans of the Turkish Coordination Council for the
Improvement of the Investment Environment (YOİKK).
Schafer, J, 1997. ―Analysis of Incomplete Multivariate Data‖. London: Chapman and Hall.
United States Patent and Trademark Office. 2009.
Venn, Danielle. 2009. ―Legislation, collective bargaining and enforcement: Updating the OECD employment
protection indicators.‖ OECD Social, Employment and Migration Working Papers no. 89
64
World Bank. 2006. Report on the Observance of Standards and Codes (ROSC) - Accounting and Auditing.
Washington, DC
World Bank. 2007. Turkey Investment Climate Assessment. Washington, DC
World Bank. 2008. ―Investing in Turkey‘s Next Generation: The School-to-Work Transition and Turkey‘s
Development.‖ Report no. 44048 – TÜ. Washington, DC
World Bank. 2008. Program Document for a Second Competitiveness and Employment Policy Loan to the Republic
of Turkey. Report No. 46436-TR Washington, DC
World Bank. 2008. Turkey Country Economic Memorandum – Sustaining High Growth: Selected Issues.
Washington, DC
World Bank. 2009. Turkey National Innovation and Technology System Report. Washington, DC
World Bank. 2009. Programmatic Human Capital, Employment and Growth Development Policy Loan for Turkey.
October 1, 2009. Concept Draft
World Bank, 2009, Turkey Innovation Policy for Competitiveness and Employment Generation. June 2009. Decision
Meeting Draft
World Bank Doing Business. 2010
World Economic Forum. 2009. The Global Competitiveness Report 2009-2010. Geneva
World Intellectual Property Organization. 2009
Annex 2-A. Econometric Methods
In the identification of the statistically significant investment climate effects on productivity and other performance
measures, the main concern is to use the information contained in the enterprise survey to perform the estimation
and, at the same time, control for all the contingent problems that may be encountered: measuring productivity,
simultaneity, endogenous regressors, selection of the relevant model, as well as data quality issues such as missing
values, outliers or measurement errors. This Annex briefly describes the methodology used and proposes solutions
for a number of methodological problems. The methodology, together with complete results, is detailed in the
background paper for this report by Escribano, de Orte and Pena (2009).
Summary of the econometric methods used to assess IC effects The econometric methodology consists of two steps:
a. Identification of statistically significant IC effects on productivity and other performance variables. For the
identification of the statistically significant IC effects on economic performance, analysis uses a
simultaneous equations system that relates the interactions between the investment climate (IC) with
productivity, demand for labor, exports and FDI inflows. Estimation always controls for firms‘ size, region
and sector and yields elasticities and semi-elasticities of investment climate variables with respect to
productivity, employment, wages, export propensity and FDI propensity. The IC elasticities and semi-
elasticities provide a measure of the sensitiveness of outcome variables when the IC changes marginally.
b. Evaluation of relative IC contribution to aggregate (weighted average) productivity.
Analysis also evaluates the IC in terms of the Olley and Pakes (O&P) decomposition of aggregate
productivity (or weighted average using as weights the share of sales) into average productivity and
allocative efficiency.
65
Estimation of IC effects on productivity and other performance measures
In the system of equations investment climate information (IC) is used as proxy of idiosyncratic firm level
differences. We assume that cross-sectional TFP is determined by a wide set of firm level attributes such
as location, sector of activity and firm size (all represented by dummy variables included in the matrix D),
by other firm level attributes (ai) normally unobserved by the econometrician, and by an error term (TFP
i ).
In this case we use the information of the survey to approximate ai and therefore can represent firm level
productivity processes as follows:
logi P
TFP TFP
i D i iTFP a D (1a)
TFP TFP TFP
i IC i ia IC v (1b)
Similarly, the demand for labor is assumed to be determined by firm level productivity (logTFPi), by real
wages in logs (logWi), by firm level differences approximated by the IC, sector/location/size information
(D) and by an error term (L
i ):
= logL L
i L i P i w i D i ilogL a TFP logW D (2a)
L L L
i L i ia IC v (2b)
The probability of firms entering the export market is described by the next pair of equations
logExp Exp Exp
i Exp i P i D i iy TFP Da (3a)
Exp Exp Exp
i IC i ia IC v (3b)
Finally, the probability of receiving foreign direct investment (FDI) is given by
logFDI FDI FDI
i FDI i P i D i iy TFP Da (4a)
FDI FDI FDI
i IC i ia IC v (4b)
Note also that Exp
iy and FDI
iy are used as covariates in all the equations as both are included in the matrix
IC, within the group of other control variables. On the other hand, since the variables yrit, with r = Exp or
FDI, are binary random variables taking 0 and 1 values, 1( / ) ( / )r r
it itP y x E y x , the conditional probability is
equal to the conditional expectation which is usually assumed to follow a Probit or a Logit model. In
general, linear probability models (LPM) approximate well Probit and Logit nonlinear models when the
variables are evaluated close to their sample means. Since we are interested in the mean IC contribution
relative to the mean values of the dependent variables of (1a) to (4a), we will concentrate only on linear
probability specifications, like (3) and (4).
The estimation of the productivity (TFP) equation is at the core of this econometric process of estimating the
relationship of the investment climate with the firm´s economic performance ( , , , IC L IC IC ). We follow a
sequential procedure in which we first estimate the IC effect on TFP and then we proceed by estimating the
remaining equations in the system.
66
The Productivity equation derives from a structural system of equations, such as:
log log log log logi L i M i K i i
Y L M K TFP (5a)
logi P
TFP TFP
i D i iTFP a D (5b)
TFP TFP TFP
i IC i ia IC v (5c)
Where, (5a) is the production function, (5b) relates productivity with firm/sector/region/size fixed effects and (5c) is
a formulation for firm-level fixed effects in terms of the investment climate. Y is firm output (sales), L is
employment, M denotes intermediate materials, K is the capital stock, IC and C are time-fixed effect vectors of other
investment climate and control variables and D are the vectors of state, industry and size dummies. Since the current
literature does not provide a univocal theory of TFP (Prescott, 2004), TFP obtained as a residual from the production
function (5a) is a ―black box‖ that may contain any factor that affects the way firms transform inputs into outputs.
Against this backdrop, the approach implicit in system (5) is to use the information contained in the enterprise
survey to empirically test TFP. That is, the initially undefined TFP measure is filled with enterprise level data
reflecting the way firms transform inputs into output, as well as the factors underlying differences observed between
more and less efficient firms. The robust IC elasticities and semi-elasticities with respect to productivity are
obtained from the following regression equation, obtained by combining (5b) and (5c):
logi
TFP
IC i D i P iTFP uIC D (6)
and are reported in Tables Annex 2-A-2 and 2-A-6. Parameter estimates are interpreted as the average individual IC
effects on productivity (TFP), after controlling for the other IC and control variables. Whereas a causal
interpretation is not possible, the robustness of empirical results allows interpretation of the estimated coefficients.
Summary of econometric issues in productivity estimation
Econometric issues in the productivity equation:
No single salient productivity measure Since there is no single salient measure of productivity, any empirical
evaluation of the productivity impact of the IC may critically depend on the particular productivity measure used.
o Solution : Escribano and Guasch (2005, 2008) – following the literature on sensitivity analysis of Magnus and
Vasnew (2006) – suggest looking for empirical results (elasticities and semi-elasticities) that are robust to several
productivity measures. The different productivity measures derive from: i) different functional forms of the
production function: Cobb-Douglas and Trasnlog; ii) different sets of assumptions (technology and market
conditions) yielding two different approaches to estimate the system: the two step estimation by applying the
Solow residual to estimate firm level productivity and the single step estimation by parametric techniques of; iii)
different levels of aggregation in measuring input-output elasticities of equation (6) (at the industry level or at the
aggregate country level). The productivity measures used in this report are summarized below:
Table 2-A-1
Functional forms of
production function
Estimation
procedure
Aggregation level of
coefficients of PF
Result
1. Solow´s Residual Two-step
estimation
1.1 Unrestricted coefficients 2 (P) measures; 2 (IC)
elasticities 2.2 Unrestricted coefficient
2. Cobb-Douglas Single-step
estimation
2.1 Restricted coefficient 2 (P) measures; 2 (IC)
elasticities 2.2 Unrestricted coefficient
3. Translog Single-step
estimation
3.1 Restricted coefficient 2 (P) measures; 2 (IC)
elasticities 3.2 Unrestricted coefficient
Total 6 (P) measures; 6 (IC)
elasticities
67
Endogeneity of the inputs. There is an identification issue separating TFP from the production function (PF);
when a PF input is influenced by unobserved common causes affecting productivity – such as a firm‘s fixed
effects – there is a simultaneous equation problem in equation (6) and hence in the single step estimation
procedure.
o Solution : To address this well-known problem (Marschak and Andrews, 1944, and Griliches and Mairesse, 1995)
analysis follows the approach proposed by Escribano and Guasch (2005, 2008). That is, the usually unobserved
firm-specific fixed effects (which are the main cause of inputs‘ endogeneity) are proxied by a long list of observed
firm-specific fixed effects deriving from the enterprise survey. Controlling for the largest possible set of IC
variables and plant characteristics (C), under standard regularity conditions, estimation yields consistent and
unbiased least squares estimators of the parameters of the PF and the IC elasticities.
Endogeneity of IC variables: For consistency in estimation, the error term must be uncorrelated with any
variable contained in the IC vector. It can be argued that the error term may contain alien, unmeasured effects
correlated with IC, thus rendering the OLS estimator of IC effects inconsistent.
o Solution 1: Correction for observable fixed effects; by using the full set of information contained in the IC
variables we are able to control for more than 130 variables in the estimation, eliminating a large degree of
endogeneity and spurious correlations as expectation of the outcome variables is conditioned on as much
information as possible.
o Solution 2: In spite of the observable fixed effect correction, the exogeneity condition does not hold for all the
variables in the model. In this case industry-region (or industry-region-size) counterparts of firm level IC variables
are used, which is equivalent to applying a general IV estimator.
Selection of the relevant model. The population model is unknown and needs to be approximated based on a
broad set of more than 130 variables, including IC and other controls.
o Solution: The econometric methodology applied for the selection of the variables goes from the general to the
specific. Otherwise, an omitted variables problem would generate biased and inconsistent parameter estimates.
Estimation proceeds by removing from the regressions the less significant variables one by one, until a final set of
variables is obtained, all significant in at least one of the regressions and with parameters varying within a
reasonable range of values.
Heteroskedasticity in the error term.
o Solution: the heteroskedasticity of the error is addressed by using robust (White) standard errors. In response to
the fact that data was collected using a random sampling by clusters, cluster standard errors are also computed,
allowing for correlation within industry and region.
Summary of econometric issues regarding data quality Although enterprise surveys are valuable instruments improving our understanding of the investment climate factors
affecting economic growth, particularly in emerging and transition economies, at the same time they present a
number of issues related with the quality of the information provided; measurement errors, outlier observations and
missing data are frequently found in this datasets. The enterprise survey of Turkey is not an exception.
Econometric analysis uses the subset of the ES sample (1152 firms in total) covering manufacturing firms. In order
to ensure a sufficient representation of large establishments in the sample, a sampling approach which oversampled
large firms was applied. The result is a sample with 903 manufacturing establishments. Due to the stratified
sampling structure, use of proper weighting to correct for oversampling of large firms when doing descriptive
analysis is advisable. However, the regression analysis uses un-weighted estimation based on the fact that
stratification variables are controlled for in the estimation and that stratification is not based on the dependent
variable of the regression.43
The first Turkey survey was performed in the summer of 2005 surveying 1,323 manufacturing establishments. The
sampling structure was based in the same stratification variables used in 2008. Likewise, the sampling process was
based on the Industry Database of the Union of Chambers and Commodity Exchanges of Turkey (TOBB). However,
there are some differences in the industries included in 2008 and 2005, and in the percentage of establishments of
each category, while the same regions are used in both surveys, and percentages of each region are almost
unchanged in 2008. In 2008 135 IC variables are used, out of which 44 are common to 2008 and 2005, and the
43 See Cameron and Trivedi (2005) for more details.
68
remaining are available in 2008 only. Due to these differences some homogeneity in the methodology applied is lost.
However, comparison of econometric results still allows identification of key differences in the way the investment
climate affected firms in the two years.
Econometric issues:
Missing data. The number of missing values reduces the sample available from 903 to 443 observations in the
complete case (49 percent of the sample). Operating with the complete case is only acceptable if missing data
comprise a small percentage, say 5% or less, of the size of the sample (Schafer, 1996), and preserves the
representativeness of the original sampling frame. In models with a large number of regressors, missing data force
the exclusion from regression analysis of explanatory variables with a high proportion of missing values. As
Cameron and Trivedi (2005) point out, this practice may lead to an omitted variables problem.
o Solution for production function variables: We impute those missing values with the objective of preserving
the sample representativeness, gaining efficiency in the estimation and retrieving for the analysis a large number
of costly interviews. Basically, the imputation mechanism replaces the missing values by the expectation of the
variable conditional on the information we have on sector, size and region (see Escribano and Pena, 2008 for
more details). To check robustness we also estimate the productivity equation under different imputation
procedures and different assumptions on the missing data mechanism (MDM), see Escribano, de Orte, Pena and
Guasch 2008. After the imputation we are able to use 768 observations, 85 percent of the sampling frame.
o Solution for IC variables: Industry/region/size averages are a good solution also for the missing IC values.
Nonetheless, using too many IC variables in average form introduces a high degree of multicollinearity to the
regression. In addition, not all industry-region-size averages are good instruments of plant-level IC variables. As
an alternative solution for the missing values in IC variables, we replace only the missing values by the expected
value conditional on the information on industry, region and size.
Endogenous missing data mechanism (MDM). As Escribano and Pena (2008) signals, the missing data problem
may also have important implications for the consistency of the IC parameters estimates when the pattern of
missing values is determined by the IC variables.
o Solution: We need to control for any IC variable correlated with the MDM to achieve consistency in the
estimation of the system.
Outliers.
o Solution: We exclude from the analysis outlier observations, i.e. observations with ratios of materials and/or labor
cost to sales larger than one.
International comparisons: demean log-productivity Interpretation of TFP is assumed conditional on the understanding of firms‘ operating conditions. Any productivity
measure is subject to measurement errors, unmeasured effects, differences in the deflators used, etc. To make cross-
country comparisons based on IC impacts on productivity it is desirable create an index (demean productivity). After
subtracting the mean (that is, the constant term, time effects, industry effects and country-specific effects) from firm
level log-productivity we can concentrate on the part of log-productivity explained by the IC variables. Thus,
demean log-productivity at the firm level is simply:
ˆlog D TFP
i IC iTFP IC (6)
Expression (7) is comparable across countries because the set of IC variables is very similar in all countries (with
some slight modifications depending on the specific characteristics of each economy) and the same methodology is
applied to select the set of significant IC variables. In addition, the O&P decompositions can be easily computed
based on the demean portion of productivity allowing international comparisons of IC impacts on aggregate
productivity.
Assessment of IC effects on the Olley and Pakes (1996) decomposition The Olley and Pakes (O&P) decomposition of aggregate productivity in logs is
ˆlog log cov( , log )Y
i iTFP TFP s TFPN (7)
69
Where logTFP is aggregate log-productivity (or weighted average productivity, where the weights are given by the
shares of sales), logTFP is un-weighted average log-productivity and the last term is the covariance between share
of sales and firm level productivity, or allocative efficiency term, describing the ability of the markets to reallocate
resources from less to more productive establishments.
The useful additive property of equation (7) in logarithms, allows obtaining an exact closed form solution of the
decomposition of aggregate log productivity. Following Escribano et al. (2008b), we can express aggregate log
productivity as a weighted sum of the average values of two composite terms; a) the IC and dummy D variables, the
intercept and the productivity residuals; and b) the sum of the covariances between the share of sales and IC, D and
the productivity residuals:
ˆ ˆ ˆˆ ˆ ˆ ˆ ˆ ˆ ˆlog cov( , ) cov( , ) cov( , )TFP TFP Y TFP Y Y TFP
IC i D i p it IC i i Ds i j i iu uTFP IC D s IC s D sN N N (8)
The contributions of IC variables to aggregate log-productivity of equation (8) can be computed for the whole
sample, by industry/sector, by region, by firm size, etc. In particular, for international comparisons we compute the
IC contributions relative to demeaned aggregate TFP as follows:
100ˆ ˆ ˆ100 [ cov( , )]TFP Y TFP
IC IC it iD
q
IC s IClogP
N (9)
Note that from (9):
i. We can compare net contributions by isolating the impact of IC variables from the impact of industry dummies,
the intercept, and the residuals;
ii. We can differentiate the portion of aggregate productivity explained by IC, and C variables (demean logP),
from the part attributable to the constant term, industry dummies, etc.;
iii. We can carry out international comparisons of the effects of IC on aggregate productivity;
iv. We can neutralize the different directions (positive or negative) of the various IC effects by taking the
percentage contributions in absolute value;
v. Finally, we can compute the absolute percentage contributions to the average log-productivity by blocks of IC
variables or by regions, sectors or sizes.
IC impact on average employment and on the probability of exporting and of receiving FDI In order to evaluate the impact of the average IC variable on the sample average values of the dependent variables of
the system, we substitute unknown parameters from the system (1)-(4) by their corresponding 2SLS estimated
values. The labor demand equation evaluated at the sample means and expressed in relative terms is
100ˆ
logˆ ˆ ˆ ˆ100 log logL P W IC DIC D
LTFP W
. (10)
Since Exp
ity and
FDI
ityare binary variables, evaluating the impact at the sample mean implies evaluating the
probability (frequency) of exporting and receiving FDI, as follows: 100 ˆ
ˆ( 0)
ˆ ˆ ˆ100 logExp P IC DsIC DP Exp
TFP
(11) 100
ˆ( 0)ˆ ˆ ˆ ˆ100 logFDI P IC DIC D
P FDITFP
. (12)
70
Results of the econometric analysis
Table 2-A-2: Comparison of IC effects on TFP in 2008 and 2005 2008 2005
IC coefficient
on TFP
(IC % cont. on
aggr. log-TFP)
[IC % cont. on
avg. log-TFP]
IC coefficient
on TFP
(IC % cont. on
aggr. log-TFP)
[IC % cont. on
avg. log-TFP]
Regulatory
environment
Payments for power supply (dummy) -0.413 (-0.4) [-0.7] N.A (.) [.]
Payments for government contracts -0.023 (-5.4) [-7.9] N.S (.) [.]
Losses due to criminal activity N.S (.) [.] -0.097 (-150) [-148.9]
Payments for protection N.A (.) [.] -0.254 (-32.4) [-33.7]
Security expenses -0.002 (-0.4) [-0.6] N.S (.) [.]
Tax inspections (dummy) -0.17 (-30.4) [-28.1] N.S (.) [.]
Inspections (number) N.S (.) [.] -0.032 (-11.3) [-12.4]
Compulsory certificates (days) -0.031 (-5.1) [-4.2] N.A (.) [.]
Compulsory certificates obtained (number) -0.363 (-60.4) [-52.6] N.A (.) [.]
Time tax N.S (.) [.] -0.021 (119.9) [123.8]
Informal competition (dummy) -0.158 (-13.8) [-15.2] -0.1 (-6.5) [-5.7]
Sales declared to taxes N.A (.) [.] 0.013 (-13.4) [-15.1]
Absenteeism N.A (.) [.] -0.271 (-16.3) [-10.8]
Lawsuit (dummy) N.A (.) [.] -0.147 (59.5) [54]
Customs clearance for imports (days) -0.175 (-72.1) [-73] -0.171 (-51.7) [-51.6]
Total % contributions regulatory environment (-188) [-182.3] (-102.2) [-100.4]
Labor and
skills
Staff - female workers -0.003 (-14.2) [-14.7] N.S (.) [.]
Staff - unskilled workers N.S (.) [.] -0.182 (-3.4) [-1.9]
Staff - part-time workers N.A (.) [.] -0.005 (21.5) [23]
Staff - university education 0.002 (6.2) [5] N.S (.) [.]
Training of skilled workers (weeks) N.S (.) [.] 0.041 (-0.6) [-0.6]
Total % contributions labor and skills (-8) [-9.7] (17.5) [20.5]
Quality and
innovation
New technology purchased (dummy) N.A (.) [.] 0.187 (-14.1) [-13.3]
R&D (dummy) 0.078 (5.6) [4] N.S (.) [.]
Outsourcing (dummy) 0.137 (7.3) [6.2] N.A (.) [.]
E-mail (dummy) N.S (.) [.] 0.074 (11.9) [10.6]
Total % contributions quality and innovation (12.9) [10.2] (-2.2) [-2.7]
Finance and
corporate
governance
Sales paid before delivery 0.003 (7.4) [6.3] N.A (.) [.]
New fixed assets finance - internal founds 0.001 (7.1) [5.5] N.A (.) [.]
New fixed assets finance - state-owned banks -0.005 (-1) [-2.1] N.A (.) [.]
Largest shareholder 0.002 (15.3) [16.5] N.A (.) [.]
Subsidies (dummy) 0.292 (8.2) [5] N.A (.) [.]
External audit (dummy) N.S (.) [.] 0.769 (14.7) [14.6]
Total % contributions finance and corporate governance (37) [31.2] (14.7) [14.6]
Other control
variables
Share of exports 0.003 (13.6) [11.5] N.S (.) [.]
Decreased sales (dummy) -0.194 (-2.3) [-6.7] N.A (.) [.]
Decreased prices (dummy) 0.274 (1.7) [4.5] N.A (.) [.]
Age 0.143 (78.4) [68.7] 0 (3.3) [0.7]
Privatized firm (dummy) N.S (.) [.] 0.344 (-8.2) [-8.7]
Duration of power outages -0.244 (-38) [-40.7] -0.332 (-12.5) [-14.2]
Shipment losses, exports -0.054 (-5.2) [-4.7] N.A (.) [.]
Shipment losses, domestic -0.012 (-2.1) [-2.1] N.A (.) [.]
Phone connection (days) N.S (.) [.] -0.005 (-10.4) [-8.6]
Total % contributions other control variables (46.1) [30.5] (-27.8) [-30.8]
Grand Total % contribution (-100) [-120.1] (-100) [-98.8]
"." contribution statistically non different from zero; N.S not significant variable; N.A not available variable
71
Table 2-A-3: Comparison of IC effects on employment in 2008 and 2005 2008 2005
IC coefficient
on
Employment
(IC % cont.
on av. log-
employment)
IC coefficient
on
Employment
(IC % cont.
on av. log-
employment)
Productivity -0.152 -9.3 -0.072 -92.1
Real wages -0.171 -44.5 -0.101 -563.6
Regulatory
environment
Court action (dummy) 0.321 3.5 N.S. .
Security expenses (dummy) 0.193 2.6 N.S. .
Compulsory certificates (dummy) (b) 0.139 1.5 N.A. .
Compulsory certificates (days) (b) -0.077 -1.8 N.A. .
Time tax (a) N.S. . 0.018 106.4
Informal competition (dummy) N.S. . -0.077 -24.6
Sales declared to taxes (a) N.A. . -0.015 -510.0
Labor costs declared (a) N.A. . 0.019 876.7
Payments for government contracts (dummy) N.S. . -0.112 -19.5
Transaction fees to obtain a land or a building (a) N.A. . -0.075 -407.9
Total % contributions red tape, informality and
others
5.8 21.0
Labor and
skills
Staff - nonproduction workers (b) 0.016 36.1 N.A. .
Staff - university education (b) 0.008 2.9 -0.013 -93.1
Training (dummy) (b) 0.374 4.0 N.A. .
Management education (dummy) N.A. . 0.451 202.3
Internal training (dummy) N.A. . 0.2 67.9
External training (dummy) N.A. . 0.325 79.0
Total % contributions labor and skills 43.1 256.2
Quality and
innovation
Quality certification (dummy) 0.41 5.4 0.448 118.2
FDI (dummy) 0.41 0.4 N.S. . Website (dummy) 0.463 11.0 N.S. .
New technology purchased (dummy) N.A. . 0.226 66.4
E-mail (dummy) N.S. . 0.267 140.8 Discontinued (dummy) -0.144 -1.0 N.A. .
Total % contributions quality and innovation 15.8 325.4
Finance and
corporate
governance
Purchases paid after delivery (a) 0.019 34.3 N.S. .
Purchase fixed assets (dummy) 0.327 5.1 N.S. .
Loan (dummy) 0.193 3.4 0.156 219.2
Loan - state-owned banks (dummy) -0.32 -1.1 N.A. .
External audit (dummy) 0.275 4.9 0.239 64.1
Credit line (dummy) N.S. . 0.214 71.0
Outstanding loan (dummy) N.A. . 0.333 43.1
Loan in TL (dummy) N.A. . -0.278 -55.4
Collateral (dummy) N.S. . -0.269 -53.3
Rent land (dummy) N.A. . -0.21 -235.6
Total % contributions finance and corporate
governance
46.6 53.1
Other
control
variables
Incorporated company (dummy) 0.743 0.3 0.253 6.7 Limited company (dummy) 0.232 5.2 N.A. .
Exporting experience (years) (b) 0.191 4.3 N.A. .
Capacity utilization (b) 0.006 12.5 N.S. .
Shipment losses, domestic (b) -0.016 -0.5 N.S. .
Power outages (dummy) -0.222 -4.0 N.S. .
Age 0.298 24.9 N.S. .
Privatized firm (dummy) N.S. . 0.722 5.1
Exporter (dummy) N.S. . 0.353 109.0
Previous public ownership (dummy) N.A. . 0.755 6.7
Dummy for young firms N.S. . -0.311 -27.4
Total % contributions other control variables 42.6 100
Grand Total % contribution 100.0 100.0
72
Table 2-A-4: Comparison of IC effects on probability exporting in 2008 and 2005 2008 2005
IC
coefficient
on
exporting
(IC % cont.
on av. log-
exporting)
IC
coefficient on
exporting
(IC % cont.
on av. log-
exporting)
Productivity 0.251 83.7 0.178 50.1
Regulatory
environment Inspections (dummy) -0.047 -6.5 0.046 6.6
Payments for government contracts (dummy) -0.221 -5.2 N.S. .
Payment for compulsory certificates (dummy) -0.027 -4.4 N.A. .
Sales declared to taxes N.A. . -0.003 -21.7
Security expenses N.S. . 0.084 103.8
Customs clearance for imports (dummy) N.S. . -0.075 -18.3
Informal competition (dummy) N.S. . -0.05 -3.2
Total % contributions regulatory environment -16.2 67.2
Labor and
skills
Management education N.A. . 0.128 12.6
Staff - skilled workers N.S. . 0.089 5.6
Training of unskilled workers (weeks) N.A. . 0.015 4.6
Total % contributions labor and skills 0.0 22.8
Quality and
innovation Quality certification (dummy) 0.203 14.4 0.064 3.7
New product (dummy) 0.113 8.2 N.S. .
Website (dummy) 0.134 17.1 N.S. .
E-mail (dummy) N.S. . 0.136 15.7
Total % contributions quality and innovation 39.7 19.3
Finance and
corporate
governance
New fixed assets finance - equity 0.002 1.1 N.S. .
Land/buildings as collateral (dummy) 0.106 4.2 N.A. .
Loan (dummy) N.S. . 0.046 3.6
External audit (dummy) N.S. . 0.073 4.3
Total % contributions finance and corporate governance 5.2 7.9
Other control
variables
Increased prices (dummy) -0.082 -3.4 N.A. .
Age -0.057 -25.7 N.S. .
Inventory (days) 0.032 16.7 N.A. .
Duration of power outages N.S. . -0.123 -38.7
Incorporated company (dummy) N.S. . 0.178 1.0
Competitors (number) N.A. . -0.141 -44.1
Capacity utilization N.S. . 0.003 25.5
Unionized workers N.A. . -0.014 -11.1
Total % contributions other control variables -12.5 -67.4
Grand Total % contribution 100.0 100.0
Table 2-A-5: Comparison of IC effects on the probability of receiving FDI in 2008 and 2005 2008 2005
IC
coefficient
on FDI
(IC % cont.
on av. log-
FDI)
IC
coefficient on
FDI
(IC % cont.
on av. log-
FDI
Productivity 0.067 93.5 0.037 38.3
Regulatory
environment
Tax inspections (number) -0.028 -10.4 N.A. .
Customs clearance for imports (days) N.S. . -0.04 -36.5
Payments for government contract (dummy) N.S. . -0.166 -20.9
Total % contributions regulatory environment -10.4 -57.5
Labor and
skills
Staff - nonproduction workers 0.001 17.2 N.A. .
Management education N.A. . 0.105 37.9
Internal training (dummy) N.A. . 0.148 39.0
Staff - university education N.S. . 0.001 6.6
Total % contributions labor and skills 17.2 83.5
Quality and
innovation
Foreign technology (dummy) 0.031 4.1 0.028 2.2
R&D (dummy) 0.005 2.3 N.A. .
New product (dummy) N.S. . 0.143 27.3
Total % contributions quality and innovation 6.4 29.5
Finance and
corporate
governance
Loan - non-financial institutions (dummy) -0.078 -0.7 N.A. .
External audit (dummy) 0.023 9.0 N.S. .
Total % contributions finance and corporate governance 8.3 0.0
Other
control
variables
Importer firm (dummy) 0.03 11.1 N.S. .
More than 5 competitors (dummy) 0.038 20.9 N.A. .
Old firm (dummy) -0.055 -33.4 N.S. .
Decreased prices (dummy) -0.023 -1.8 N.A. .
Duration of power outages -0.015 -11.7 N.S. .
Incorporated company (dummy) N.S. . 0.065 1.4
Exporter (dummy) N.S. . 0.019 4.7
Total % contributions other control variables -14.9 6.1
Grand Total % contribution 100.0 100.0
73
Table 2-A-6: Robust IC elasticities and semi-elasticities with respect to productivity – OLS Estimation
Restricted estimation Unrestricted by industry estimation
Two step estimation One step estimation Two step estimation One step estimation
Solow's Residual Cobb-Douglas Translog Solow's Residual Cobb-Douglas Translog
Coef. S.E Coef. S.E Coef. S.E Coef. S.E Coef. S.E Coef. S.E
Regulatory
environment
Payments for power supply (dummy) (b) -0.413 [0.149]*** -0.179 [0.218] -0.228 [0.223] -0.441 [0.155]*** -0.107 [0.263] -0.161 [0.294]
Payments for government contracts (b) -0.023 [0.011]** -0.019 [0.011]* -0.021 [0.010]** -0.022 [0.011]** -0.022 [0.011]** -0.025 [0.012]**
Security expenses (b) -0.002 [0.012] -0.016 [0.012] -0.015 [0.011] -0.002 [0.013] -0.018 [0.012] -0.016 [0.011]
Tax inspections (number) (a) -0.17 [0.093]* -0.154 [0.091]* -0.122 [0.091] -0.181 [0.093]* -0.168 [0.087]* -0.134 [0.086]
Compulsory certificates (days) (b) -0.031 [0.031] -0.032 [0.024] -0.051 [0.022]** -0.031 [0.031] -0.036 [0.023] -0.046 [0.024]*
Compulsory certificates (number) (a) -0.363 [0.107]*** -0.383 [0.110]*** -0.294 [0.109]*** -0.373 [0.105]*** -0.417 [0.109]*** -0.345 [0.115]***
Customs clearance for imports (days) (a) -0.175 [0.175] -0.312 [0.169]* -0.252 [0.161] -0.188 [0.177] -0.312 [0.189] -0.17 [0.187]
Informal competition (dummy) (b) -0.158 [0.068]** -0.169 [0.066]** -0.142 [0.063]** -0.156 [0.069]** -0.133 [0.070]* -0.106 [0.073]
Labor and
skills
Staff - female workers (b) -0.003 [0.002]* -0.005 [0.002]*** -0.005 [0.002]** -0.003 [0.002] -0.006 [0.002]*** -0.005 [0.002]***
Staff - university education (b) 0.002 [0.003] 0.003 [0.002] 0.003 [0.002] 0.002 [0.003] 0.004 [0.003] 0.003 [0.002]
Quality and
innovation
R&D (dummy) (b) 0.078 [0.087] 0.114 [0.075] 0.11 [0.073] 0.069 [0.088] 0.106 [0.077] 0.143 [0.081]*
Outsourcing (dummy) (b) 0.137 [0.095] 0.161 [0.084]* 0.159 [0.078]** 0.143 [0.096] 0.15 [0.085]* 0.101 [0.073]
Finance and
corporate
governance.
Sales paid before delivery (b) 0.003 [0.002] 0.003 [0.002] 0.003 [0.002] 0.004 [0.002]* 0.002 [0.002] 0.003 [0.002]
New fixed assets finance - internal founds (b) 0.001 [0.001] 0.001 [0.001] 0.001 [0.001] 0.001 [0.001] 0.001 [0.001] 0.001 [0.001]**
New fixed assets finance - state-owned banks (b) -0.005 [0.004] -0.005 [0.003] -0.005 [0.003] -0.005 [0.004] -0.006 [0.003]* -0.005 [0.003]*
Largest shareholder (b) 0.002 [0.002] 0.002 [0.001] 0.002 [0.001] 0.002 [0.002] 0.002 [0.001] 0.002 [0.001]
Subsidies (dummy) 0.292 [0.120]** 0.205 [0.112]* 0.185 [0.111]* 0.302 [0.122]** 0.198 [0.116]* 0.175 [0.120]
Other control
variables
Share of exports (b) 0.003 [0.001]*** 0.003 [0.001]*** 0.003 [0.001]*** 0.003 [0.001]*** 0.003 [0.001]*** 0.003 [0.001]***
Duration of power outages (a) -0.244 [0.158] -0.282 [0.162]* -0.264 [0.154]* -0.25 [0.159] -0.278 [0.165]* -0.213 [0.175]
Shipment losses, exports (a) - interaction with dummy for
exporter
-0.054 [0.023]** -0.055 [0.019]*** -0.06 [0.019]*** -0.054 [0.023]** -0.042 [0.028] -0.054 [0.024]**
Shipment losses, domestic (b) -0.012 [0.003]*** -0.012 [0.004]*** -0.013 [0.004]*** -0.012 [0.003]*** -0.013 [0.005]*** -0.011 [0.004]***
Decreased sales (dummy) (b) -0.194 [0.095]** -0.26 [0.095]*** -0.228 [0.097]** -0.192 [0.094]** -0.267 [0.094]*** -0.176 [0.097]*
Decreased prices (dummy) (b) 0.274 [0.139]* 0.12 [0.142] 0.114 [0.139] 0.27 [0.136]** 0.108 [0.145] 0.113 [0.144]
Age 0.143 [0.059]** 0.161 [0.056]*** 0.156 [0.057]*** 0.142 [0.059]** 0.161 [0.056]*** 0.13 [0.061]**
Observations 780 780 780 780 780 780
R-squared 0.13 0.76 0.78 0.13 0.77 0.80
NOTES:
Restricted: equal input output for all the establishments in the country; Unrestricted: equal input-output elasticities for all the establishments in the same sector.
Two steps estimation: in the first step estimation of equation (3.5a) by non-parametric techniques to compute productivity (Solow residual), in the second step estimate (3.5b) and (3.5c) by OLS using as
dependent variable the Solow residual from the first step, either restricted or unrestricted. Single step estimation: estimate (3.5a), (3.5b) and (3.5c) in a single step by OLS, where (3.5a) can be a Cobb-Douglas Production function or a Translogarithmic.
*significant at 10%; ** significant at 5%; *** significant at 1% given by robust standard errors corrected for correlation between cluster (industry and region) in brackets.
Each regression includes a set of industry, size and region dummies and a constant term.
(a) Variables instrumented with the industry-region-size average; (b) Variables approximated with a proxy (only missing values replaced by the industry-region-size average).
Source: Authors‘ estimations with Turkey ES 2008 data.
74
Table 2-A-7: IC percentage contributions to aggregate demeaned log-productivity in 2008
Aggregate log-TFP Average log-TFP Allocative efficiency
Regulatory
environment
Payments for power supply (dummy) (b) -0.10 -0.20 0.38
Payments for government contracts (b) -1.31 -2.31 3.61
Security expenses (b) -0.11 -0.18 0.23
Tax inspections (dummy) (a) -7.43 -8.28 -3.29
Compulsory certificates (days) (b) -1.25 -1.25 -1.26
Compulsory certificates (number) (a) -14.77 -15.49 -11.21
Customs clearance for imports (days) (a) -17.63 -21.50 1.33 Informal competition (dummy) (b) -3.37 -4.46 1.94
Labor and
skills
Staff - female workers (b) -3.46 -4.34 0.84
Staff - university education (b) 1.51 1.46 1.72
Quality and
innovation
R&D (dummy) (b) 1.37 1.17 2.32
Outsourcing (dummy) (b) 1.78 1.84 1.52
Finance and
corporate
governance.
Sales paid before delivery (b) 1.81 1.84 1.65
New fixed assets finance - internal funds (b) 1.73 1.61 2.33
New fixed assets finance - state-owned banks (b) -0.25 -0.61 1.49
Largest shareholder (b) 3.73 4.85 -1.77
Subsidies (dummy) 2.01 1.48 4.63
Other control
variables
Share of exports (b) 3.34 3.40 3.03
Duration of power outages (a) -9.29 -11.99 3.92
Shipment losses, exports (a) - interaction with dummy for exporter -1.28 -1.39 -0.75
Shipment losses, domestic (b) -0.51 -0.63 0.09
Decreased sales (dummy) (b) -0.56 -1.98 6.42
Decreased prices (dummy) (b) 0.43 1.31 -3.93
Age 19.16 20.22 13.95
Total contribution of IC (demean log-productivity) -24.45 -35.40 29.19
Other stuff Industry/region/size controls -31.85 -33.60 -23.30
Constant term 141.86 170.80 0.00
Residual 14.45 -1.81 94.11
Total contribution of other stuff 124.45 135.40 70.81
Total 100.00 100.00 100.00
Table 2-A-8: IC elasticities and semi-elasticities with respect to employment – IV Estimation Dependent variable: log-employment (demand for labor) Restricted Solow residual Unrestricted Solow residual
Blocks Explanatory ICA variables Coefficient % Contrib Coefficient % Contrib
Productivity -0.152* -8.8 -0.153* -8.8
Real wages -0.171*** -41.8 -0.171*** -41.9
Regulatory
environment
Court action (dummy) 0.321*** 3.3 0.320*** 3.3
Security (dummy) 0.193** 2.4 0.193** 2.4
Compulsory certificates (dummy) (b) 0.139*** 1.4 0.138*** 1.4
Compulsory certificates (days) (b) -0.077** -1.7 -0.076** -1.7
Labor and
skills
Staff - nonproduction workers (b) 0.016*** 33.9 0.016*** 33.9
Staff - university education (b) 0.008*** 2.8 0.008*** 2.8
Training (dummy) (b) 0.374*** 3.8 0.375*** 3.8
Quality and
innovation
Quality certification (dummy) 0.410*** 5.1 0.409*** 5.1
Discontinued (dummy) -0.144** -0.9 -0.143* -0.9 Website (dummy) 0.463*** 10.3 0.462*** 10.3
Finance and
corporate
governance
Purchases paid after delivery (a) 0.019* 32.2 0.019* 32.3
Purchase fixed assets (dummy) 0.327*** 4.8 0.326*** 4.7
Loan (dummy) 0.193*** 3.2 0.194*** 3.2
Loan - state-owned banks (dummy) -0.320*** -1.0 -0.321*** -1.0
External audit (dummy) 0.275*** 4.6 0.275*** 4.6
Other control
variables
Incorporated company (dummy) 0.743*** 0.3 0.742*** 0.3
Limited company (dummy) 0.232*** 4.9 0.233*** 4.9
FDI (dummy) 0.410*** 0.4 0.408*** 0.4
Exporting experience (b) 0.191*** 4.0 0.191*** 4.0
Power outages (dummy) -0.222*** -3.8 -0.221*** -3.8
Shipment losses, domestic (b) -0.016*** -0.5 -0.016*** -0.5
Capacity utilization (b) 0.006*** 11.8 0.006*** 11.7
Age 0.298*** 23.3 0.298*** 23.4
Instruments
evaluation
First stage R-squared: productivity2 0.324 0.318
Partial R-squared: productivity3 0.27 0.264
Partial R-squared F test (p-value): productivity4 0 0
Hansen test (p-value)5 0.591 0.639
Observations 779 779
75
Table 2-A-9: IC linear probability coefficients with respect to the probability of exporting – IV Estimation Dependent variable: probability of exporting Restricted Solow residual Unrestricted Solow residual
Blocks Explanatory ICA variables Coefficient % Contrib Coefficient % Contrib
Productivity 0.251** 124.6 0.243** 119.9
Regulatory
environment
Inspections (number) (b) -0.047** -9.7 -0.046** -9.6 Payments for government contracts (b) -0.221*** -7.8 -0.222*** -7.9
Compulsory certificates (days) (b) -0.027** -6.6 -0.026** -6.6
Quality and
innovation
Quality certification (dummy) (b) 0.203*** 21.4 0.204*** 21.5
Website (dummy) (b) 0.134** 25.5 0.135** 25.8
New product (b) 0.113*** 12.2 0.113*** 12.3
Finance and
corporate
governance
New fixed assets finance - equity (b) 0.002 1.6 0.002* 1.6
Land/buildings as collateral (dummy) (b) 0.106*** 6.2 0.102*** 5.9
Other
control
variables
Increased prices (dummy) -0.082* -5.1 -0.083* -5.2
Age -0.057** -38.2 -0.055* -37.4
Inventory (days) (b) 0.032** 24.8 0.032** 24.7
Large firm (dummy) 0.211*** 10.7 0.214*** 10.9
Instruments
evaluation
First stage R-squared: productivity2 0.214 0.223
Partial R-squared: productivity3 0.0428 0.0436
Partial R-squared F test (p-value): productivity4 0.0046 0.003
Hansen test (p-value)5 0.431 0.412
Observations 636 636
Table 2-A-10: IC linear probability coefficients with respect to the probability of receiving FDI – IV
Estimation Dependent variable: probability of receiving FDI Restricted Solow residual Unrestricted Solow residual
Blocks Explanatory ICA variables Coefficient % Contrib Coefficient % Contrib
Productivity 0.067* 528.9 0.066* 518.1
Regulatory
environment
Tax inspections (days) (b) -0.028*** -59.0 -0.029*** -59.3
Labor and
skills
Staff - nonproduction workers (b) 0.001* 97.2 0.001* 98.2
Quality and
innovation
Foreign technology (dummy) (b) 0.031* 23.2 0.031 23.3
R&D expenditures (b) 0.005*** 12.8 0.005*** 12.7
Finance and
corporate
governance
Loan - non-financial institutions (b) -0.078*** -4.2 -0.078*** -4.2
External audit (dummy) (b) 0.023* 51.0 0.023 51.3
Other control
variables
Importer firm (dummy) (b) 0.030* 62.7 0.030* 63.2
More than 5 competitors (dummy) (b) 0.038** 118.1 0.038** 116.8
Old firm (dummy) -0.055* -188.8 -0.055* -190.6
Decreased prices (dummy) -0.023* -10.0 -0.023* -9.8
Duration of power outages (b) -0.015** -66.4 -0.015** -65.9
Large firm 0.056** 23.5 0.057** 23.7
Instruments
evaluation
First stage R-squared: productivity2
Partial R-squared: productivity3 0.0233 0.231
Partial R-squared F test (p-value): productivity4 0.0121 0.01
Hansen test (p-value)5 0.987 0.987
Observations 778 778
76
Annex 2-B. Investment Climate Variables Used for the Econometric Analysis
Table 2-B-1: Definition of IC Variables – Regulatory environment
Survey
availability
Name of the variable Description of the variable
2005 ICS &
2008 ES
Time tax Percentage of managers' time spent in dealing with bureaucratic issues.
Informal competition (dummy) Dummy variable that takes value 1 if the firm competes with informal (no registered) firms.
Inspections (number) In the last year, total Inspections (number) (log).
Payments for government contracts (dummy) Dummy variable that takes value 1 if in plant's sector it is common to pay an extra amount of
money in order to obtain a contract with the government.
Payments for government contracts Illegal payment in order to obtain a contract with the government. Related as percentage of
contract value.
Construction related permit (days) Actual delay to obtain a construction related in days (log).
Operating license (days) Actual delay to obtain a main operating license in days (log).
Customs clearance for exports (days) Average number of days to clear customs to export (log).
Customs clearance for imports (days) Average number of days to clear customs to imports (log).
Only in 2008 ES
Import license (days) Current delay to obtain an import license related in days (log).
Permit (days) Number of days it takes to obtain a permit
Payment for import license (dummy) Gifts expected or requested to obtain an import license, conditional on submit an import license.
Share of exports Share of exports over total annual sales.
Time tax Percent of managers' time spent in dealing with bureaucratic issues per week (log)
Inspections (days) Number of working days spent with inspections (log)
Tax inspections (dummy) Dummy variable that takes value 1 if the firm has been visited by tax officials during last year.
Tax inspections (number) Total Inspections (number) of tax officials received by the plant in 2007 (log)
Tax inspections (days) Number of working days spent with tax inspections (log)
Security expenses (dummy) Dummy taking value 1 if the plant has security expenses.
Payment for construction permit (dummy) Gifts expected or requested to obtain a construction permit, conditional on submit a construction permit.
Compulsory certificate (dummy) Dummy variable that takes value 1 if the firm has to have any compulsory certificate to produce or
sell any product
Compulsory certificates (number) Number of compulsory certificates obtained (log)
Compulsory certificates (days) Number of working days spent when obtaining compulsory certificates (log)
Payment for compulsory certificates (dummy) Gifts expected or requested to obtain a compulsory certificate, conditional on submit a compulsory
certificate.
Only in
2005 ICS
Payments for protection Cost due to protection payments e. g. to organized crime to prevent violence (bribery) (log).
Sales declared to taxes Percentage of total sales declared to taxes.
Labor costs declared Percentage of workforce declared to taxes.
Absenteeism Days of production lost due to absenteeism (log).
Lawsuit (dummy) Dummy variable that takes value 1 if the firm has been involved in a lawsuit in the last three years.
Table 2-B-2: Definition of IC Variables – Labor and skills Survey
availability
Name of the variable Description of the variable
2005 ICS & 2008 ES
Staff - skilled workers Percentage of skilled workers in firm's staff.
Staff - unskilled workers Percentage of unskilled workers in firm's staff.
Staff - female workers Percentage of female workers in firm's staff.
Staff - university Percentage of staff with at least one year of university.
Labor regulation Share of firms perceiving labor regulation as a major or very severe obstacle
Only in
2008 ES
Training (dummy) Dummy taking value one if the firm provides formal (beyond on the job) training to its employees.
Manager experience (years) Number of years of experience of Top Manager in the sector
Staff - production workers Percentage of production workers in staff.
Staff - nonproduction workers Percentage of nonproduction workers in staff.
Training - production workers Percentage of production workers receiving formal (beyond on the job) training
Training - non-production workers Percentage of non-production workers receiving formal (beyond on the job) training
Only in 2005 ICS
Staff - part time workers Percentage of part time workers in firm's staff.
Internal training (dummy) Dummy variable that takes value 1 if the plant provides internal training to its employees.
External training (dummy) Dummy variable that takes value 1 if the plant provides external training to its employees.
Weeks of training of skilled workers Number of weeks of training received by the skilled workers during last year.
77
Table 2-B-3: Definition of IC Variables – Quality and innovation Survey
availability
Name of the variable Description of the variable
2005 ICS & 2008 ES
Quality certification (dummy) Dummy variable that takes value 1 if the plant has a quality certification.
New product (dummy) Dummy variable that takes value 1 if the plant has developed a new product or product line.
E-mail (dummy) Dummy variable that takes value 1 if the plant uses email.
Website (dummy) Dummy variable that takes value 1 if the plant has a website.
Product upgraded (dummy) Dummy variable that takes value 1 if the plant upgraded an existing product last year.
Foreign technology (dummy) Dummy variable that takes value 1 if the firm used a licensed technology of a foreign company in
the last year.
Only in 2008 ES
Sales of new products Percentage of total sales corresponding with new products
R&D (dummy) Dummy that takes value 1 if the firm performed R&D activities during last year.
R&D expenditures Total R&D expenditures as percentage of annual sales
Computer Percentage of staff using computer at job.
Outsourcing (dummy) Dummy taking value 1 if the plant subcontracts any part of the activity.
Discontinued (dummy) Dummy taking value 1 if the plant has discontinued at least one product line
Only in
2005 ICS
New technology purchased (dummy) Dummy variable that takes value 1 if the firm purchased any new technology during last year.
Table 2-B-4: Definition of IC Variables – Finance and corporate governance Survey
availability
Name of the variable Description of the variable
2005 ICS &
2008 ES
Loan (dummy) Dummy variable that takes value 1 if the plant reports that it has a bank loan.
Loan – bank (dummy) Dummy variable that takes value 1 if the firm has a loan from a domestic private commercial banks.
Loan – state-owned bank (dummy) Dummy variable that takes value 1 if the firm has a loan from a state owned banks.
Collateral (dummy) Dummy variable that takes value 1 if the loan requires collateral.
Collateral Value of collateral as share of loan value (conditional on having loan with collateral)
External audit (dummy) Dummy variable that takes value 1 if the plant uses an external auditory.
Only in
2008 ES
Purchases paid after delivery Percentage of annual purchases paid for after the delivery.
Sales paid before delivery Percentage of annual sales paid for before the delivery.
Purchase fixed assets (dummy) Dummy variable that takes value 1 if the firm has purchaser fixed assets during last year.
New fixed assets finance - internal funds Percentage of firm's working capital financed with funds from informal sources.
New fixed assets finance - equity Percentage of firm's working capital financed with funds from equity.
New fixed assets finance - state-owned banks Percentage of investments in new fixed assets financed with funds state owned banks.
New fixed assets finance - state-owned banks Percentage of investments in new fixed assets financed with external funds
Loan - non-financial institution (dummy) Dummy variable that takes value 1 if the firm has a loan from a non-financial institutions.
Land/ buildings as collateral (dummy) Dummy that takes value 1 if the firm uses land or buildings as collateral (conditional on having a
loan with collateral).
Largest shareholder Percentage of firm's capital owned by the largest shareholder.
Overdraft facility (dummy) Dummy that takes value 1 if the firm has access to an overdraft facility
Subsidies (dummy) Dummy variable that takes value 1 if the firm receives any subsides from the national, regional and
local government or EU.
Only in
2005 ICS
Outstanding loan (dummy) Dummy variable that takes value 1 if the firm has a loan outstanding from a financial institution.
Rent land (dummy) Dummy variable that takes value 1 if the firm has a loan from a leasing arrangement.
Loan in TL (dummy) Dummy variable that takes value 1 if the loan is denominated in Turkish Lira.
Table 2-B-5: Definition of IC Variables – Other Survey
availability
Name of the variable Description of the variable
2005 ICS &
2008 ES
Duration of power outages Average duration of power outages suffered by the plant in hours (log).
Loss from power outages Value of the losses due to power outages as share of percentage of sales.
Incorporated company (dummy) Dummy variable that takes value 1 if the plant is an incorporated company.
Public ownership (dummy) Dummy variable that takes value 1 if the firm belongs to the government.
78
Age Difference between the year that the plant started operations and current year.
Capacity utilization Average percentage of capacity used during last year.
Inputs imported Share of inputs imported directly
Importer (dummy) Dummy variable that takes value 1 if imports are greater than 10%.
Exporter (dummy) Dummy variable that takes value 1 if exports are greater than 10%.
Phone connection (days) Current delay to obtain a phone connection in days (log).
Only in
2008 ES
Power outages (dummy) Dummy variable that takes value 1 if the plant has suffered any power outages during last year.
Duration of power outages Total duration of power outages suffered by the plant by month, in hours, conditional on the plant
reports having power outages.
Shipment losses, exports Fraction of the value of the plant‘s average cargo consignment that was lost in transit due to breakage, theft or spoilage in the international market.
Shipment losses, domestic Fraction of the value of the plant‘s average cargo consignment that was lost in transit due to
breakage, theft or spoilage in the domestic market.
Limited company (dummy) Dummy variable that takes value 1 if the plant is a limited company.
Exporting experience Number of years of exporting experience.(log)
More than 5 competitors (dummy) Dummy taking value one if the plant has less than 5 competitors in the local market.
Domestic sales (dummy) Firms selling more than 90% of their output in the domestic market
Decreased sales (dummy) Dummy taking value 1 if the plant has decreased its sales
Increased prices (dummy) Dummy taking value 1 if the plant has increased its prices
Decreased prices (dummy) Dummy taking value 1 if the plant has decreased its prices
Only in
2005 ICS
Competitors (number) Number of competitors in the main market (log).
Unionized workers Percentage of workers that belongs to a syndicate.
Privatized firm (dummy) Dummy variable that takes value 1 if the firm previously belonged to the government.
79
Chapter 3. PROMOTING SME GROWTH
3.1. Sustainable growth in the post-crisis environment will crucially depend on easing the
constraints to the expansion of SMEs. SMEs in Turkey account for 79.4 percent of employment, 44.6
percent of total investments, 67.4 percent of total sales, 25-30 percent of total exports, 57.3 percent of
total value added and 25 percent of bank credit. Given the rate of tax evasion and the size of the informal
sector, the contribution of SMEs to the economy may be somewhat underestimated. This implies that the
development of a more productive and more outward-oriented SME sector is a crucial development
challenge for Turkey. A healthy SME sector can not only significantly contribute to capital accumulation,
provide increased employment opportunities for a rapidly increasing workforce, and promote regional
development, but is also crucial to increase the resilience of the economy to external shocks, like the one
represented by the recent global crisis. This Chapter, using data from the most recent Enterprise Survey
for Turkey and other countries,44
contributes to the understanding of the role of the investment climate in
determining the expansion of the Turkish SME sector.
3.2. The 2007 Investment Climate Assessment highlighted the importance of investment climate
factors in constraining the expansion of firms. In addition to the 2005 enterprise survey, the report
based its analysis on 1996-2001 TURKSTAT firm level data,45
finding that the main constraints to
productivity and output growth in Turkey may come not so much from barriers to start up and exit, but
rather from barriers to expansion. The report also pointed out that Turkish firms are heterogeneous,
showing large disparities in size, growth and productivity performance. New entrants in Turkey were
found to grow at a slower pace than firms in comparator countries. Furthermore, in Turkey a large share
of productivity benefits derives from firm churning in low-tech industries, rather than, as it is the case in
most comparator countries, from entry and exit in high-tech sectors, where new technologies are often
better harnessed by new firms. The analysis also indicated that firms face significant barriers to
expansion, suggesting reforms aimed at improving a wide range of investment climate conditions,
including access to credit, adoption of quality standards, technologies and innovation, quality of the labor
force and reduction of administrative procedures.
3.3. Analysis of 2008 data confirms that the current investment climate negatively affects the
productivity performance of small and medium-sized firms. As argued in Chapter 2, analysis of 2008
data shows that, while larger firms benefit from the more positive aspects of the investment climate,
smaller firms bear the burden of its less positive features. The need to increase average efficiency of the
Turkish business sector emphasizes the importance of policy and institutional mechanisms targeted
towards easing the efficient operation and expansion of SMEs.46
44 The Enterprise Survey was conducted in 29 other countries in the East Europe and Central Asia (ECA) region in the same time
period it was conducted in Turkey. This allows comparison of firm evolution in Turkey with other countries in the region. The
analysis compares Turkey with Russia, Ukraine, Poland, Romania, EU-10 (Bulgaria, Czech Republic, Estonia, Hungary, Latvia,
Lithuania, Poland, Romania, Slovak Republic, and Slovenia), EU-8 (EU-10, excluding Bulgaria and Romania), as well as with
the average for the entire ECA region. See www.enterprisesurveys.org for the complete dataset and for a detailed description of
the data and methodology used in sampling. 45 The TURKSTAT database draws from annual surveys of all Turkish manufacturing firms with more than 10 employees
conducted between 1996 and 2001. It is representative of the whole population of manufacturing firms in Turkey and includes
information on micro units. 46 Until a few years ago, the definition of what constitutes an ―SME‖ varied from one agency to another (see OECD (2004)).
Since 2005, however, the Turkish Ministry of Industry and Trade has consolidated the concept by defining SME as a firm with a
less than TL25 million turnover and 250 employees. (In the European Union, the turnover figure is about 4 times the Turkish
figure [€50 million] but the employee size is the same.) This SME definition is further broken down in the Turkish context to 3
80
This Chapter adds to the productivity analysis of Chapter 2 by examining the dynamics of evolution of
Turkish SMEs in terms of employment growth. It finds that (i) medium-sized firms are the slowest
growing group; that (ii) SMEs grow at a slower rate in Turkey than in comparator countries; and
that (iii) improved access to finance significantly increases employment growth rates. It also suggests
a number of measures to ease financial constraints to the expansion of SMEs. It should be noted that the
enterprise survey tracks formal full-time employment, hence investment climate constraints to firm
growth could be more deep-rooted and far-reaching than our analysis suggests, since reduced incentives
to employ formally imply that firms have access to lower skill levels, which, in turn, negatively affects
their productivity potential.
3.1 Patterns of Firm Growth in Turkey
3.4. Smaller and younger firms can be expected to grow faster, but in Turkey there seems to be
a “missing middle” with small and medium firms falling short of prediction. Using the size and age
groups described in Box 3-1, the growth rates of firms are calculated as the annual increase in full-time
employment between 2004 and 2007. Figure 3-1 shows that the growth rates of Turkish firms do not
monotonically decrease as size increases. Micro (1-10 employees) and large firms (more than 250
employees) appear as the fastest growing groups in 2004-2007, while small (11-50 employees) and
medium firms (51-250 employees) display far slower growth rates across different age groups. Young (1-
5 years) medium-sized firms, in particular, display no growth at all in the period under examination.
These findings are in opposition with regularities observed in other countries and predicted in the
theoretical literature on firm growth. For instance, firm growth rates should decrease as size and age
increase.47
This is based on the observation that diminishing returns to scale or bounded efficiency may
limit the rate of expansion for larger firms. Similarly, diminishing returns to learning can explain the
inverse relation between growth and age, with older firms less prepared to implement new modalities of
operation.48
In addition to size and age, numerous other factors can affect firm growth, such as the levels
of technology and human capital, the development level of the country, or the regulatory environment.
Such broad investment climate factors are even more critical for smaller firms, who may lack the capacity
to cope with distortions in access to finance, availability of technology and skills or the regulatory
environment.49
sub-segments, as micro (turnover less than TL1 million), small (turnover between TL1-TL5 million) and medium (turnover
between TL5-TL25 million) enterprises. 47 Dunne, Roberts, Samuelson (1989), Evans (1987a, 1987b) and numerous other studies have provided evidence on these
regularities. Studies like Jovanovic (1982), Cooley and Quadrini (2001), Klette and Kortum (2004), Klepper and Thompson
(2007) construct structural models that can explain these regularities. 48 Jovanovic (1982) argues that firms learn their efficiency levels over time. Least efficient firms exit and more efficient firms
adjust their scale of operations through learning. Hence small and young firms grow faster because they are in the initial process
of uncovering their efficiency levels. 49 Beck, Demirguc-Kunt, Maksimovic (2005) show that financial, legal, and corruption related problems distort firm growth.
They also show that whether these factors affect growth depends on firm size and among all firms, small firms are the most
constrained group. In a more recent study Aterido, Hallward-Driemier, and Pages (2009), analyze how business environment
affects firm growth using a wide set of objective measures for obstacles. They show that poor business environment hurts firm
growth and the amount of distortion varies by firm size.
Box 3-1: Data for the analysis of firm growth
The analysis of firm growth is based on the 2008 Enterprise Survey data used in the rest of this report (Box 2-1). In
analyzing how firm evolution changes with size, we divide firms into four size groups measured in fulltime
employment levels: Micro: ≤10, Small: 11-50, Medium: 51-250, and Large: ≥251. These groups are constructed
according to the employment level of the firms in 2004. The relevant questions in the survey refer to the last
complete fiscal year which is 2007. The survey asks the number of full-time employees in the last fiscal year and
81
3.5. The pattern of firm growth presents sectoral and regional variations, with several sectors
and regions where SMEs grow more slowly than large firms. Table 3-1 shows the growth rates of
firms in different size groups in 2-digit
industries and across five macro-regions. In
the food sector both small and medium size
firms grow at a slower rate in 2004-2007
than large firms – 1 percent and 5 percent,
respectively, compared to a growth of 8
percent for large firms. In the garments
sector, medium firms have not only grown
more slowly than large firms, but their
employment has contracted by 11 percent in
the period. In the plastics and rubber sector,
small firms have shed employment at a faster
pace (-5 percent) than large firms (-1
percent), while in machinery and equipment
employment in small firms has declined by 9
percent, compared to an increase in all other size categories. Turning to the regional breakdown, South
and Marmara have the fastest, while Aegean has the slowest, rate of employment growth in micro firms.
In Marmara and Central Anatolia, small and medium firms have slower growth rates than large firms.
Table 3-1: Growth rates by industry and region (2004-2007)
Micro Small Medium Large
2-Digit
Industry
Food 0.12 0.01 0.05 0.08
Textiles 0.17 0.09 0.01 0
Garments 0.04 0.07 -0.11 0
Chemicals 0.13 0.07 -0.03 -0.12
Plastics & rubber 0.26 -0.05 0.05 -0.01
Non metallic mineral 0.02 0.05 0.1 0
Basic metals 0.09 0.06 0.19 0.17
Fabricated metal products 0.18 0.06 0.05 0.03
Machinery and equipment 0.2 -0.09 0.1 0.16
Electronics 0.27 0.01 0.02 -0.34
Services of motor vehicles 0.13 0.1 0.71 0.1
Wholesale 0.12 -0.03 0.13 0.59
Retail 0.21 0.03 0.11 0.28
Region
Marmara 0.21 0.01 -0.01 0.05
Aegean 0.06 0.03 -0.02 -0.04
Central Anatolia 0.14 -0.01 0.1 0.15
South 0.22 0.14 0.06 0.03
Black Sea - Eastern 0.15 0.03 0.08 0.06
three fiscal years ago, which is 2004. Annual growth rates are calculated using these questions. Firms are also
divided into three age groups as: 1-5, 6-15, and ≥16 years of age. In addition to the cross-sectional data from 2008
survey, 419 firms were also surveyed in 2005, thus giving a panel dimension to the data. In all tabulation and
regressions use is made of probability weights, to ensure that results are representative of the population of firms.
Annex 3-A provides a full description of the data used.
Figure 3-1: Employment growth at different sizes and ages
(2004-2007)
Source: Turkey ES 2008
29
14
0
26
14
2 3
1414
-2
1
-4
18
3 2
6
-5
0
5
10
15
20
25
30
35
Micro Small Medium Large
%
1-5 6-15 >=16 Total
82
3.6. Medium firms (51-250 employees) are the slowest growing group, with employment growth
16 percent slower than micro firms and 5 percent slower than large firms. Based on the size
categories described above, regression analysis allows analyzing how growth rates of firms differ by size
controlling for other factor that may influence employment growth.50
Growth can be affected by firm
characteristics, such as age, trade orientation, and ownership status. For instance, as mentioned earlier, it
is well established that younger firms experience faster growth rates, or that firms that are globally
engaged through exporting or foreign
ownership are presented with more
opportunities to expand than non-trading
and fully domestically owned firms. To
control for these factors, regression analysis
includes dummy variables controlling for
firms that generate more than 10 percent of
their sales from exports and for foreign
ownership defined as 10 percent or more
share in firm ownership. Firms with
government ownership might also grow at
different rates because their objectives
might differ from a fully privately owned
firm. To control for this factor, a dummy
variable is included representing 10 percent
or more government ownership51
. In
addition, analysis controls for external conditions that may have an impact on employment growth, such
as the industry and the region in which the firm operates. Regression results indicate that micro firms are
the fastest growing group, allowing for benchmarking of the growth rates of other groups to that of micro
firms (10 employees or less). Contrary to prediction and as already noted above, the growth rate does not
decrease monotonically as size increases. In particular, medium-sized firms are the least growing group,
growing 16 percent more slowly than micro when the full sample is considered and 23 percent more
slowly when looking only at panel firms (Figure 3-2).
3.7. Comparison with
other countries indicates
that SMEs in Turkey are,
on average, older, i.e. they
remain small for a longer
time. The cross-country
nature of the ES data allows
comparison of the dynamics
of firm growth in Turkey with
other countries in Europe and
Central Asia. In all
comparator countries larger
firms are more likely to have
a higher proportion of firms
that are older than 16 years.
Turkey stands out for the
50 See Annex for detailed econometric results. 51 The firms that are 100 percent government owned are excluded from the survey.
Figure 3-2: Growth rates relative to micro firms (2004-2007)
Source: Turkey ES 2008
Figure 3-3: Percentage of firms that are above 16 years old
Source: Turkey ES 2008
-25
-20
-15
-10
-5
0
Small (11-50) Medium (51-250) Large (≥251)
All Data Panel Firms
26
51 1
11
1
31
9
19
2
913
60
19
25
1520
25
49 48
62
20
32
46
0
10
20
30
40
50
60
70
Turkey Russia Ukraine Romania EU10 All ECA
Micro Small Medium Large
83
longer average life of firms in all size categories, which can, to a large extent, be explained by the fact
that, until the early 1990s, private enterprise was not allowed in most of the countries in the region, whose
economies were governed under central planning arrangements. However, the far larger proportion of old
(above 16 years of age) micro, small and, especially medium-sized firms in Turkey is still remarkable
(Figure 3-3). In Turkey, 26 percent of micro firms and 31 percent of small firms are more than 16 years
old, compared to 11 and 9 percent, respectively, in the EU-10 countries. Even more striking results
emerge for medium-sized firms, with 60 percent in Turkey being more than 16 years old, compared to 20
percent in the EU-10.52
This observation might indicate that micro firms and SMEs in Turkey grow at a
much slower pace than firms in other countries. In other words, they face barriers to their expansion that
force them to remain at a smaller –and suboptimal – scale of operations. The demographics of large firms,
on the other hand, are in line with average values of comparable firms in other countries.
3.8. Small and
medium firms have
much lower growth
rates, while large firms
grow at similar rates as
counterparts in other
countries. In order to
verify whether SMEs in
Turkey face an
environment that is more
hostile to expansion than
for comparable firms in
other countries,
regression analysis
allows benchmarking the
growth rate of Turkish
firms of various size
categories to that of a
number of comparators.
Whereas, in all countries
in the region, the rate of
expansion decreases as
size increases, as
mentioned above,
Turkey stands out for the
slow growth of its SMEs
(Figure 3-4a). Contrary
to what is observed in
Turkey, in all
comparator countries in
the region small- and
medium-sized firms
grow faster than large
firms. Turkish SMEs in 2004-2007 grew 14-16 percent more slowly than micro firms, while growth of
52 Dunne, Roberts, and Samuelson (1989) perform a similar exercise for the US firms with a slightly different size and age
classification. They find that the fraction of firms with 5-19 employees who are between 11-15 years old is 7 percent. In the same
age group, the fraction for firms with 20-49, 50-99, and 100-249 employees vary around 9 percent. These values are consistent
with the average values ECA countries reported here.
Figure 3-4: Growth rates of firms: Cross-country comparison
a: Growth rates relative to micro firms, percent (2004-2007)
b: Growth rates (percent) of Turkish SMEs (2004-2007)
Source: Turkey ES 2008
-14
-16
-11
-18
-16
-14
-12
-10
-8
-6
-4
-2
0
Small Medium Large
%
Russia
Poland
Ukraine
Romania
All_ECA
EU10
Turkey
-8.5
-9.8-8.8
-7.6
-10.8 -10.6-9.8
-11.2
-12.7
-10.7
-7.7
-11.9-12.4
-11.3
-14
-12
-10
-8
-6
-4
-2
0
Russia Poland Ukraine Romania ECA EU-8 EU-10
Small (11-50) Medium (51-250)
84
SMEs in comparator countries was in the range of 5-6 percent slower than that of micro firms. Figure 3-
4b shows that, overall, Turkish SMEs grow substantially more slowly than SMEs in comparator
countries. The difference for both small (11-50 employees) and medium firms (51-250 employees) is of
the order of 10-12 percent, with medium-sized firms confirmed as the slowest growing group.53
These
results can explain why there are more SMEs at old ages in Turkey relative to the comparator countries,
while both the age composition and the growth rates of large Turkish firms are internationally
comparable.
3.2 Investment Climate Constraints to SME Growth
3.9. Reasons for the slow growth of Turkish SMEs can be found in features of the business
environment.54
Comparison of the evolution of Turkish firms with several comparator countries has
shown that growth rates of SMEs relative to micro firms are significantly lower in Turkey. One possible
explanation is that existing policies and regulations can have more distortionary effects for SMEs than for
both micro and large firms, since SMEs have neither the capacity of large firms nor the flexibility of
micro firms to cope with the effects of these policies.55
For instance, incentives to hire new workers and
expand may directly be influenced by labor regulations, which impinge on the cost of hiring and firing, by
broader regulatory issues affecting the cost of doing business or by access to finance, with the availability
of funding determining the pace at which firms can expand production and, in consequence, employment.
Figure 3-5: Single most severe investment climate obstacles, by firm size
Source: Turkey ES 2008
3.10. Access to finance is perceived as the single most severe obstacle by firms of all sizes, and
especially by medium-sized firms. According to the 2008 enterprise survey, firms of all size categories
perceive access to finance as the single most severe obstacle (Figure 3-5), with medium-sized firms
appearing to be particularly affected (34 percent), followed by micro (26 percent), small (24 percent) and
large firms (19 percent). Tax rates also rank quite high as an obstacle, especially for micro and small
firms. Adequate skill levels are perceived as a problem mainly by large firms. The stringency of labor
regulations, a possible direct cause of slow employment growth, rank relatively low among firms‘
concerns, being cited as the most severe obstacle by only 4 percent of medium-sized firms, 3 percent of
53 Table A-3-7 in the Appendix shows that the difference in growth rates between Turkish SMEs and SMEs in comparator
countries is statistically significant. 54 Aterido, Hallward-Driemier, and Pages (2009), using data on more than 56,000 enterprises in 90 countries, confirm the
importance of the business environment for employment growth, with effects varying across firms of different sizes. 55 This ―missing middle‖ hypothesis is discussed in several studies. See, for example, Gauthier and Gersovitz (1997), Sleuwaegen
and Goedhuys (2002), and Van Biesebroeck (2005).
26
20 19
15
5
2 1 0
42 3
24
1918
15
11
3 3 3 21 1
34
14 13 13
8
31
4 5
0
4
19
12
18
15
25
1 13 2
02
0
10
20
30
40
Access to finance
Tax rates Political instability
Informal competition
Unskilled workforce
Licenses and permits
Corruption Customs regulations
Electricity Transport Labor regulations
Micro Small Medium Large
85
micro firms and by a negligible proportion of firms in other size categories. As previously mentioned, this
result should be interpreted with care, since the perceptions of respondents may have be influenced by the
timing of the survey.
3.11. Other indicators in the survey confirm that Turkish SMEs are dependent on bank finance
but their applications for bank credit are faced with onerous collateral requirements and high
rejection rates. The perceived severity of access to finance as an obstacle broadly decreases with firm
size, with 16 percent of micro firms citing it as a major or very severe obstacle, compared to only 10
percent of large firms, and with medium-sized firms appearing marginally more concerned (14 percent)
than small firms (13 percent) (Figure 3-6a). At the same time, Turkish firms – as discussed in Chapter 2 –
are more dependent than peers in other countries on bank finance to fund their investments in fixed assets.
This is especially true for medium-sized firms, for whom bank finance accounts for 47 percent of total
funding (Figure 3-6b). Collateral requirements also appear particularly onerous for SMEs, compared to
both micro and large firms, amounting to 100 percent of loan value for small firms and 91 percent for
medium firms (Figure 3-6c). Notwithstanding the higher collateral requirements, the amount of rejected
loan applications is also substantially higher for SMEs (17 percent) compared to more creditworthy large
firms (12 percent) (Figure 3-6d).
Figure 3-6: Access to Finance for SMEs
Source: Turkey ES 2008
16.4
13.314.1
10.5
0
2
4
6
8
10
12
14
16
18
Micro Small Medium Large
(a) % of firms perceiving access to finance as a "major" or "very severe" obstacle
29 34
4234
5 25 2
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Micro Small Medium Large
(b) Sources of funding for the purchase of fixed assets
Other means
Credit from suppliers/customers
State-owned banks
Private banks
New equity shares issued
Internal funds
76
10091
70
0
20
40
60
80
100
120
Micro Small Medium Large
(c) Value of collateral as % of loan value
1917 17
12
0
5
10
15
20
25
Micro Small Medium Large
(d) % of firms with loan applications rejected
86
3.12. Access to bank credit for SMEs has further deteriorated in the aftermath of the global
financial crisis, but SME lending still constitutes a strong growth area for banks. In terms of loans,
the SME sector has started to lose ground since late 2007, with the slowing of domestic growth and
escalation of the global credit crisis. During this period, SME share in total credit has declined by about 5
percentage points to little over 20 percent, and its share in comparison to total corporate credit dropped by
even more from about 52 percent to some 44 percent. In terms of growth rates, growth in total banking
sector credit remained relatively high and unchanged until the escalation of the global crisis in late 2008,
whereas SME credit growth started to decelerate as early as the beginning of 2008. The differences
become more dramatic when SME and non-SME corporate credits are compared (Figure 3-7). For the
whole period for which data are available (from December 2006 through November 2009) cumulative
growth in SME credit amounted to some 35 percent, but that was about half the rate of growth in other
(non-SME) corporate credit. In terms of problem loans, SMEs also looked to be faring much worse than
the rest of the sector, with the gross NPL ratio of the sector rising to almost 8 percent, from below 4
percent in the middle of 2008. However, despite the fact that SME credit appears more affected than other
types of credit as a consequence of the crisis, SME lending still constitutes a strong growth area for banks,
and banks are probably eager to tap that potential.
Figure 3-7: Loans to SMEs
Source: World Bank (2010b)
3.13. Econometric results confirm that access to finance is the area of the investment climate
most consistently associated with the ability of firms to generate employment and grow in size.56
In
order to test the hypothesis that the investment climate – and finance in particular – have an effect on firm
growth, regression analysis is conducted incorporating variables from the 2008 survey.57
Results indicate
that one percent more usage of external finance for investment – including all sources other than internal
funds or retained earnings, such as issuance of new equity, public and private banks, purchases on credit
from suppliers and advances from customers, money lenders, and non-banking financial institutions – is
related with 0.3 percent higher employment growth. The association of a loan or a line of credit with
employment growth is even stronger and is estimated to have a positive effect on employment growth of
56 Inclusion of the IC variables as explanatory variables is not straightforward. As discussed in the previous Chapter and in this
Annex, the inclusion of the investment climate variables can cause endogeneity problems even when the chosen variables are not
subjective measures that depend on firms‘ perceptions of the business environment. For instance, faster growing firms might be
more likely to have access to external finance or they might be interacting more with government officials. To address this issue,
in this Chapter investment climate variables are used as industry-region averages. 57 See Annex 2-B for a complete list of investment climate variables used for the analysis.
10
15
20
25
30
35
Dec-06 Apr-07 Aug-07 Dec-07 Apr-08 Aug-08 Dec-08 Apr-09 Aug-09
Total
TRY
FX
SME Loans
(as % of Total Loans)
Loans=Cash Loans+Non-Performing Loans
40
42
44
46
48
50
52
54
56
Dec-06 Apr-07 Aug-07 Dec-07 Apr-08 Aug-08 Dec-08 Apr-09 Aug-09
SMEs Loans
(as % of Total Corporate Loans)
87
33 percent (Table 3-2).58
Furthermore, when controlling for access to finance, medium-sized firms are
confirmed as the slowest growing group.59
Estimation based on principal component analysis adds
robustness to the results indicating that a one percent increase in the block of variables connected with
access to finance is associated with 6 percent faster employment growth. 60
At the same time, faster
growing firms are more likely to perceive labor regulations as an obstacle and to spend time clearing
customs to export. Among other investment climate variables the perceived stringency of labor
regulations is strongly and positively associated with firm growth. Unlike other variables in Table 3-2,
this captures the subjective sentiment of firm managers and may indicate that faster growing firms are
more bothered with labor regulations, given their needs to expand employment. The number of days
required to clear customs for export is also positively associated with employment growth. This result
may indicate that exporters, who naturally spend time at the customs to ship their products, are also the
faster growing firms.
Table 3-2: Summary of investment climate effects on firm growth, percent
IC Variable
Effect of IC
variable on
growth rate
relative to
micro firms
(%)
Small Firms:
growth rate
relative to
micro firms
(%)
Medium Firms:
growth rate
relative to micro
firms (%)
Large Firms:
growth rate
relative to micro
firms (%)
Labor regulation +41.4 -13.1 -14.8 -10.7
New fixed asset finance - external 0.3 -13.0 -15.2 -11.3
Collateral (*)
NS NS -15.6 NS
Purchase paid after delivery NS -14.0 -15.7 -11.4
Overdraft facility (dummy) NS -13.7 -15.4 -11.0
Loan (dummy) +33.4 -14.5 -16.3 -12.5
Loss from power outages (*)
NS -21.8 -26.6 -22.0
Informal payments (*)
NS -15.7 -16.7 -9.6
Tax inspections (number) (*)
NS -19.5 -19.9 -15.5
Time tax NS -11.6 -12.3 -8.6
Customs clearance for export (days) (*)
+1.0 NS -24.7 NS
Customs clearance for import (days) (*)
NS -37.9 -48.9 -50.1
Inspections (number) NS -13.0 -15.9 -12.1 Notes: See Table 3-A-8 in the Annex. (*) Less than 600 observations. NS “Not Significant”
Source: Turkey ES 2008
3.3 Enhancing the Ability of SMEs to Access Bank Credit
3.14. Improving the connection between the banking sector and SMEs is a crucial development
challenge. It was mentioned that large Turkish enterprises are able to benefit from the more positive
aspects of the investment climate, including the unprecedented expansion of credit to the private sector
from about 15 percent of GDP in the late 1990s, to over 30 percent today. At the same time, micro firms
are often sufficiently flexible to overcome the more negative aspects of the investment climate and are
able to meet their financial needs with internal resources. Within this environment, small and medium
58 See Tables 3-A-8 and 3-A-9 in the Annex for detailed results. Micro firms are confirmed as the fastest growing group,
followed by small firms, while, in most specifications, medium firms are the slowest growing group. 59 Results in Table 3-A-9 indicate that access to finance does not have a strong differentiating effect on the growth of firms in
different size groups, with some evidence that small and large firms find the absence of a line of credit more constraining than
firms in other size groups. Given the relatively small number of observations available we prefer to use as main reference the data
in Table 3-A-8 where the absence of interaction terms allows preserving more of the information contained in the data. 60 The Principal Component method transforms a number of possibly correlated variables – as the variables in the block of access
to finance – into a smaller number of uncorrelated variables named principal components, which are ranked depending on the
amount of variability in the data each of them contains. See Table 3-A12 in the Annex for details.
88
firms appear to bear a large portion of the negative effects of an unfriendly investment climate. This
affects the ability of SMEs to grow and the factor most clearly associated with firm growth appears to be
inadequate access to finance. These conditions have certainly been exacerbated by the repercussions of
the crisis, which has caused a drop in domestic and external demand and a tightening of credit markets.
As a consequence, lending to SMEs, a sector that had been growing significantly since 2005, has actually
dropped in relative terms,61
now accounting for less than 22 percent of total lending, compared to almost
24 percent at end-2008. At the same time, while lending rates, after peaking in October 2008, have been
on a downward trend, there is anecdotal evidence that there is segmentation in the market, with rates
charged to SME clients exceeding by far those offered to larger corporate clients. Moreover, lending
maturities have been shortening and collateral requirements increasing.
3.15. The preceding analysis revealed that access to finance is the investment climate obstacle
most consistently associated with the ability of SMEs to generate employment and grow in size. In
the context of Turkey, with a reasonably developed and sophisticated banking system, increasing the
ability of SMEs to have access to affordable bank credit at convenient maturities appears as a crucial
development challenge. Banks, however, are reluctant to cater to SMEs due to the difficulties they face in
assessing their creditworthiness and their ability to repay debt (Box 3-2).
61 Akbank, the third largest private bank, reported a cut of almost 50 percent in its lending to SMEs in the nine months ending in
September 2009.
Box 3-2: Main obstacles to SME lending: views from banks and SME representatives
Banks are aware that the credit boom of the past several years is over, and the credit environment will be
tighter going forward. But they seem to have nevertheless positioned themselves for stepped up lending to the
SMEs. The lessons of Turkey‘s devastating 2001 banking/financial crisis appear learnt. Banks are conscious of
risks, while at the same time they seem aware of the need to enhance capacity in the SME sector, and are therefore
willing to act as the financial advisory to potential borrowers. Having been in the field for several years, they feel
they know the clientele reasonably well despite the many shortcomings related to transparency, lack of collateral and
so on. Bank‘s preparations in the run-up to Basel II transition, which was expected to result in a tightening of credit
to the SMEs, also helped to enhance the general level of knowledge on the sector.
From the SME side, banks pointed out three major obstacles to lending. Interestingly, the first obstacle has been
put forward – broadly defined – as the “absorptive capacity” of the sector. Banks are aware that credit is just one
ingredient to a very complex problem of supporting viable and productive firms. The SMEs in Turkey are typically
seen strong on the technical side (especially family businesses that have been going on for generations with
expertise transferred from father-to-son) but weak on the managerial side. Through what they call ―academies‖,
several banks help to develop this expertise and raise awareness in their clientele through intensive training
programs, offered on a regional basis.
Second and third constraints in SME lending are very well-known: lack of transparency of financial statements
and lack of collateral. As both are closely related to the informality issue in Turkey (tax and wealth evasion), a
quick resolution to these problems is not expected too soon. Banks are therefore taking a pragmatic stance: as much
as possible, they stick to the usual procedure of ―credit scoring‖ in their lending practices to SMEs, but relationship-
building and establishing a track record for the client seem to be on their agenda as well. The issue of collateral also
appears to be pervasive. Real property (homes, shops, factories) is the most common form of collateral, but personal
letters, guarantees through third persons are also used. Support from government agencies also help to alleviate
problems like KOSGEB support, which primarily provides interest subsidy, and collateral provided through the
Credit Guarantee Fund (CGF) are helpful but they do not yet make up a significant portion of the SME credit
market. According to KOSGEB data for 2004 for instance, only 0.5 percent of the guarantees are provided by the
CGF. The new leg of the CGF that involves a TL1 billion contribution from the Treasury has recently become
operational but is yet to be fully utilized.
89
3.16. Proposal: Enhance the ability of banks to assess the creditworthiness of SME borrowers by
expanding the coverage of credit bureaus and ensuring the wide adoption of financial reporting
standards. Transparent and accessible credit information on firms and improved accounting and auditing
standards are crucial for financial institutions extending loans, particularly to SMEs. Empirical research
has found that the presence of transparent and accessible credit registries is associated with higher share
of private credit to GDP.62
Lenders are able to access borrowers‘ credit history through credit bureaus and
are thus able to estimate their ability and willingness to pay back loans. This information is especially
important to SMEs, which have smaller loan volumes and often shorter credit backgrounds, making easy
to access information essential to lenders. At the same time, ensuring the wide adoption of financial
reporting standards would increase the ability of banks to assess the creditworthiness of SME borrowers.
In particular, the following developments would be appropriate:
Encourage the expansion of coverage of existing credit bureaus, the Credit Registry of the Central
Bank and the Credit Bureau of Turkey (KKB). Whereas the Credit Registry of the Central Bank
covers both firms and individuals, the private Credit Bureau of Turkey (KKB) has not yet
expanded to monitoring of the corporate sector. Although a planned expansion to also cover firms
has not yet been implemented, the current monitoring of individual consumers is also very useful,
especially for information on small enterprises. Since entrepreneurs, as earlier mentioned, do not
necessarily get their business loans approved, they often finance their investments through
personal loans, and the credibility of the owner is thus often reflected on the credibility of their
operations. Nonetheless, a completion of the planned expansion will significantly improve
Turkish financial institutions‘ possibilities to efficiently direct loans to the most creditworthy
customers, both individuals and corporations.
Accelerate adoption of the new Commercial Code in order to enable SMEs to benefit from a
simplified set of financial reporting standards. The Turkish Accounting Standards Board (TASB)
established in 2002, has been publishing Turkish Financial Reporting Standards (TFRS) based on
International Financial Reporting Standards (IFRS) since October 2006. TASB regularly updates
62 See for instance, Djankov et al. 2008.
SME representatives mention a host of obstacles that complicate the operating environment of a typical SME,
such as very high labor taxes, energy costs, the difficulty of adhering to ILO standards (as part of EU accession)
and the statutory minimum wage that drives many entities to lesser developed countries like Egypt (in textiles), or
Bulgaria in agriculture. But finance appears to be a central problem, too, and complaints there seem two-fold:
insufficiency of funds, and as a corollary to that the steep interest rates the banks charge on SME clients, and
inadequacy (or inaptness) of banks‘ risk metrics in evaluating SMEs‘ prospects.
Why did SME credit suffer more than the rest during the past two years? The understanding from interviews with
banks is that both supply and demand factors were at play. One large bank has explicitly emphasized this by stating
that as a business strategy (i.e. to reduce exposure to the more vulnerable clientele) they have curbed lending to the
SME sector more than to the rest of the corporate sector. At the same time, they noted, demand for certain types of
credit collapsed from SMEs too, because of production cuts (a result of a wait and see attitude) and the collapse in
global trade (affecting demand from those engaged in foreign trade). In this connection, banks mentioned that micro
firms (those with turnovers less than TL1 million) were the most affected segment, and that having crossed a certain
threshold in size appears to have increased the survival rate of firms during the crisis. But bank behavior appears to
have played a role in this outcome as well. For instance, some SME representatives expressed the view that banks
were willing to extend credit, but that their methods of risk analysis clashed with the dynamics of SME behavior,
especially the very small ones, where borrowers were judged on the basis of their turnovers. They also stated that as
the size of the firm declines, banks asked more personal (owners‘) collateral, regardless of the business prospects.
Source: World Bank (2010b)
90
and publishes the standards in line with the revisions in the IFRS. The draft Commercial Code
(CC), which has been pending enactment in Parliament since November 2005, accepts TASB as a
sole and exclusive authority to set Turkish Accounting Standards and requires that the financial
statements of all financial entities regardless of public interest or not be prepared in accordance
with TFRS. The code enables simplified set of standards to be adopted by SMEs if endorsed by
TASB. This authority that will be appointed to TASB with the enactment of the CC requires
TASB to have adequate technical and financial resources. However, currently TASB lacks such
resources that will enable it to provide technical support and guidance to the market for the
adoption of TFRS.
The role of Government: expanding the reach of existing schemes in support of SME lending
3.17. A number of government schemes are aimed at addressing financial bottlenecks to the
expansion of SMEs. Several organizations currently provide financial and non-financial assistance for
the development of SMEs. They include the Small and Medium Enterprises Development Organization
(KOSGEB)63
, Halkbank, Turkey's SME and Tradesman Bank, and the Credit Guarantee Fund (CGF).64
3.18. The Credit Guarantee Fund supports SMEs by providing a guarantee for their financing
thus increasing their ability to benefit from bank credit. The CGF became operational in the early
1990s, but it has grown substantially since 2007. Access to CGF support makes it possible for SMEs with
inadequate collateral to apply for bank credits. The CGF provides guarantees through banks to SMEs with
fewer than 250 employees and turnover below TL 25 million. Application for CGF support is carried out
with partner banks as well as through a network of banks with which CGF has an agreement. The upper
limit for guarantees provided by CGF amounts to TL 1 million. In its mission, CGF gives high priority to
supporting young and female entrepreneurs, promoting innovative investments, supporting export,
increasing the rate of employment, and contributing regional development. There is no restriction
imposed on the type of credit for which enterprises seek a CGF guarantee. In order to ease the impact of
the crisis on SMEs, the Medium Term Program for 2010-2012 calls for a stronger and more influential
CGF backed by the Undersecretariat of Treasury. (Box 3-3).
63 The mission of KOSGEB and its focus on ―SME Development and Support‖ is specified in the Council of Ministries‘ Decision
No. 2008/13524 as ―promoting investment, export and production effectiveness of SMEs.‖ Since 2004, KOSGEB provides
financial support by covering interest payments on credits, while leaving the credit risk to the bank. The upper limit for this type
of credit is TL 300,000. Applications are processed through a network of public, private and branches of foreign owned banks
and the benefiting SME remains responsible for the principal amount according to a predetermined payback schedule with the
limit of 48 months. In 2009 the Turkish government increased its funding to KOSGEB by 48 percent. Moreover, through the use
of a web-based system, application processes and monitoring of loans has become highly efficient. 64 Halkbank provides financial support to SMEs for export credit and capital investment along with credit for investment and
operations to improve productivity and increase employment. Credit is given for 5 years for investment purposes, without
requesting any repayment in the first year, and for 4 years for operations purposes. Credit on similar terms – 5 years of maturity
without repayment in the first year – is offered to manufacturing firms for purchases of machinery and equipment. Firms in
manufacturing sector, agribusiness, tourism, education, energy, IT sectors can benefit from medium- to long-term credit with 7
year maturity and no repayment required for the first 2 years. A detailed treatment of KOSGEB‘s activities is provided in Ch. 4.
91
65 Total loan volumes are calculated including cash and non-cash loans. End-year values were taken into account. 2005 SME loan
volume uses an estimate based on the SME loan volume/Total loan volume in the following years. 66 TOSYOV: Small and Medium Industry Owners and Managers Foundation of Turkey
TESK: Merchants and Artisans Confederation of Turkey
MEKSA: Occupational Training and Small Industry Support Foundation of Turkey 67 Banks accepted guarantees up to 5 times the total capital of CGF (Leverage multiplier is 5). 68 Maximum loan volume was TL 750,000. In this Chapter, amounts legally declared in TL terms have been converted into US
dollars using the interbank rates on dates which apply to associated legal arrangements. 69 Law no. 5909 published on Official Gazette no. 27268 dated 06/24/2009. 70 Law no. 4749. 71 The principles of the reformed CGF is set out in the Council of Ministers Decision no. 2009/15197 published on Official
Gazette no. 27289 dated 07/15/2009. 72 As the multiplier effect generates TL 1.2 billion of guarantees out of TL 240 million, roughly.
Figure 3-8: Guarantees provided by the CGF as a
percentage of the total SME loan volume65
Source: Credit Guarantee Fund, BRSA
Box 3-3: The Credit Guarantee Fund (CGF)
The idea of a credit guarantee scheme in Turkey was
initiated in 1991 with the motivation to ease the credit
constraints faced by SMEs. A partnership was later formed
between TOBB, KOSGEB, occupational groups ―TOSYOV,
TESK, MEKSA‖66
and Halkbank to realize this initiative.
CGF became operational in early 1990s, but it has been
growing substantially since 2007. By 2009, the CGF had
accumulated a total capital of USD 165 million, generating a
guarantee volume of USD 426.5 million, and with a
potential67
of USD 586 million. Figure 3-8 demonstrates the
increased capacity of the CGF after a capital injection from
its shareholders in 2007.
Under the scheme, only loans with a maximum volume of
about USD 835,00068
were collateralized, with the amount
of support provided by the CGF ranging between 50-80
percent of the loan, in return for a commission of 1-2 percent of the outstanding loan amount. In this setup, CGF worked
with 38 financial institutions through individual protocols, and was the responsible body for undertaking the follow-up
procedures of non-performers of its collateral share of the loans.
With the aim of providing liquidity to the real sector, in June 2009, the Parliament approved the law69
amending the Law
on Public Finance and Debt Management70
and called for the provision of guarantees totaling USD 650 million via a
reformed Credit Guarantee Fund71
. 20 banks became shareholders to the CGF, and its structure changed radically. In this
new setup, the CGF is envisaged to undertake two separate functions:
(i) Traditional role: The CGF will carry on extending guarantees under the traditional scheme. Its capacity will
allow for holding a maximum risk of about USD 825 million.72
For the utilization of this fund, banks may
send a demand to CGF for it to support the debtor under its own risk, abiding by the previous protocols.
Decisions over loan guarantees are taken by credit committees, which are composed of 5 members, namely,
the General Manager of the CGF and one representative from each of TOBB, KOSGEB, banks and other
shareholders.
(ii) Fund under Treasury support: After a long period of negotiations at the ministerial level between the
Treasury, CGF (including each of its shareholders) and banks (under the umbrella of the Banks‘
Association), all parties agreed on the establishment of a new credit guarantee system, for which the
Treasury support under Law no.4749 will remain in effect for 2 years following its establishment. A major
reform in the new setup is, in addition to the endorsement of the Treasury for extending guarantees, the
participation of the banks to the shareholding structure of the CGF. Under this arrangement, 20 banks have
now become shareholders in the CGF with equal shares, quadrupling its paid-in capital and lending
37,280 42,735 44,590
217,561
0.068% 0.060%0.051%
0.249%
0.00%
0.05%
0.10%
0.15%
0.20%
0.25%
0.30%
0
50,000
100,000
150,000
200,000
250,000
2005 2006 2007 2008
Th
ousand T
L
Real guarantee volume (2005 prices) (LHS)
Ratio of CGF collaterals / SME loans
92
3.19. Placing Turkey in an international perspective, as presented in Box 3-4, the Turkish CGF seems
to have lower share of GDP in guarantees, with the outstanding stock amounting to 0.09 percent of GDP.
However, combined capacity of the traditional and Treasury-funded loans is 0.19 percent, putting Turkey
closer to the average of higher-income countries. As to funding and management of the Turkish CGF,
responsibilities are shared, with the Government – through KOSGEB – together with TOBB representing
almost exclusive involvement in all four areas (funding, management, risk assessment and recovery).
Additionally, the European Investment Fund (EIF) has committed to counter-guaranteeing up to 50
percent of CGF‘s investment credit guarantees.
capacity. This gave these banks the right of representation in the CGF Board of Directors with 2 members
out of 9. As concerns the remaining members, 2 are appointed by the Treasury, 2 by KOSGEB and 3 by
TOBB, with TOBB also appointing the chairman of the Board.
Figure 3-9: Flow of funds in guarantee agreements
under treasury support
In this new fund structure, all loan applications are
planned to be channeled through the 20 banks which
participate in the system. Remaining banks that have a
protocol with the CGF may obtain guarantees for their
creditors through the traditional scheme. Decisions on
loan guarantees will be made by ―credit approval
committees‖, composed of 5 members, namely, the
General Manager of the CGF and one representative
from each of (i) TOBB, (ii) KOSGEB, (iii) banks and
(iv) Treasury. The representative of the Treasury is
empowered with a veto power over decisions related to
the loans under its guarantee.
For each ―Treasury-supported‖ loan, 65 percent of the
risk will be borne by the Treasury and the remaining 35
percent will be collateralized, leaving no risk to the CGF.
Loans with maturities less than 4 years and a maximum
volume of about USD 685,000 will be supported. Unlike the traditional role of the CGF, follow-up procedures will be
fully undertaken by the associated banks, with a payback mechanism channeled from the Treasury to the banks. The
banks will later transfer the collected collateral amount to the CGF, for it to pass on to the Treasury. The support will be
provided to beneficiaries in exchange for a ―guarantee commission fee‖ of 0.5-1.5 percent on the loan amount.
Source: World Bank, REGE-DPL
Box 3-4: Credit guarantee schemes: International comparison
The form and scope of credit guarantee funds (CGFs) has been proven to vary significantly across countries, both in
coverage and actual implications on firm performance. A World Bank survey on CGFs from 2008 indicated that
outstanding guarantees in relation to GDP averaged 0.35 percent for transition economies, whereas the corresponding
share in high income countries was 0.21 percent. The survey also shows that the median age of interviewed CGFs is 15
years, with an average of 27 years in high-income and 13 years in low-income countries. As presented in Table 3-3,
Asian economies have particularly active CGF presence, with the guarantee stock reaching 3.5 and 5 percent of GDP in
Japan and Korea.
There is a large variation in ownership structure, pricing control and risk management but the World Bank survey does
show an overall important role of government in funding of CGFs. Nevertheless, relative to the private sector,
government seems to have less influence in management, risk appraisal and recovery, indicating that the lending banks
are usually left with the responsibility of credit risk assessment and recovery of defaulting loans.
93
3.20. Proposal: Consider ways to make the new CGF scheme more active, expand its reach, and
allow it to better reach the medium-sized firm segment. Collateral requirements, traditionally onerous
for smaller enterprises, have become even more of a constraint to bank lending in the aftermath of the
2008-2009 economic and financial crisis. For many years, and especially since its recapitalization in
2007, the CGF has played an important role in facilitating access to credit for SMEs. The new CGF
model, with Treasury involvement for a period of two years, is a positive initiative that expands the
capacity of the CGF to serve the financing needs of SMEs following the credit crunch in the aftermath of
the crisis. Furthermore, a collateral rate of 35 percent demanded of beneficiary firms is a wise measure
that will minimize the risk of misuse of funds. The Government could build upon these positive
Table 3-3: Comparison of CGFs
Guarantee
Maximum
loan; max as
share of GDP
per capita
Cost of loan
guarantee Target firms Industries
Total
outstanding
guarantee
value/GDP Risk management
Turkey CGF Up to 80% USD 835,000;
53.5%
1-2% per
annum
Less than 250
employees and less than TL
25m annual
turnover
Agriculture,
manufacturing, services and
mining sectors,
limited to capital
formation
0.09% (2009) Counter guarantee
by Treasury and remaining part by
the guaranteed
firms.
U.S. Small
Business
Administration
Preferred Loan
Program
75%; USD 2.2m; 46.2%
1.3% one-time subsidy
of the guaranteed
loan value
Less than 100 employees
Emphasis on women,
veterans & undeserved
markets
N.A. Lender oversight office supervises
lender activity & more than half of
guarantees
extended to preferred lenders.
UK Enterprise
Financing
Guarantee
Scheme
75% USD 1.6m;
35.3%
2% per
annum
Less than GBP
25m annual turnover
(increased
from GBP 5.6m in 2009)
All but
agriculture, transport,
finance and
insurance sectors
0.07%
(Guarantee stock 2005-
2008)
Risk assessment of
borrower delegated to lender. Time
restricted program
until March 2011.
Japan JASME
Agency Loans
100% USD 1-6m, depending on
end use; 26-
157%
Set by JASME
50-300 employees,
depending on
sector
Manufacturing, wholesale,
retail &
services
5% (2007) Mainly providing guarantees during
periods of tight
credit. No direct risk monitoring
system.
Korea CGF 50-100% USD 1.9m, with
exceptions to
USD 7.5m;
88.2%
0.5-2% per annum based
on risk
Unspecified, but 99% of
guarantees to
SMEs
Emphasis on innovative
enterprises and
new firms in all
sectors, except
luxury goods
3.5% (2008) Lower guarantees over time to induce
lenders' monitoring
and credit
evaluation of
guaranteed firms.
Sources: Honohan (2008), Heron & Co. (2007), Beck et al. (2008), Turkey CGF, Korea CGF, JASME
Firms regarded as eligible for guarantees also differ greatly across countries, with dissimilarities in firm size, annual
turnover and industry. Maximum loan guarantees reach as high as USD 2.5 million in the United States (firm size is
however limited to 100 employees). In regards to risk assessment and credit appraisals of final borrowers, this is in many
best-practice cases not done by the CGFs, but instead performed by the intermediary financial institutions giving out the
actual loan (Honohan, 2008), usually creating higher efficiency and lower operating costs for the CGFs. An incentive for
lenders to carry out credit appraisals is shared risk with the CGF (partial credit guarantee funds). On average, the CGFs
guarantee 70-80 percent of loans, with shares going as high as 100 percent in Japan and Korea. In the case of very high
guarantees, the presence of broad and high-functioning credit information systems is desirable in order to ensure
eligibility of borrowers.
94
developments by accelerating the implementation of the new CGF scheme. Furthermore, the CGF could
more fully perform its role if it was better targeted to serve the needs of medium-sized firms. In fact, since
1994, the bulk of credit guarantees provided by the CGF has benefited micro (40 percent) and small (49
percent) enterprises. Since 1994, 23 percent of the guarantee fund has been used by medium-sized firms
(50-249 employees).
References
Aterido, Reyes, Mary Hallward-Driemeier, Carmen Pages. 2009. ―Big Constraints to Small Firms‘ Growth?
Business Environment and Employment Growth across Firms,‖ World Bank Policy Research Working Paper 5032.
Beck, Thorsten, Asli Demirgüç-Kunt and Vojislav Maksimovic. 2005. ―Financial and Legal Constraints to Growth:
Does Firm Size Matter?‖ Journal of Finance 60(1): 131-177
Beck, Thorsten, Leora F. Klapper and Juan Carlos Mendoza. 2008. ―The Typology of Partial Credit Guarantee
Funds around the World,‖ World Bank Policy Research Working Paper 4771.
Cooley, Thomas F. and Vincenzo Quadrini. 2001. "Financial Markets and Firm Dynamics," American Economic
Review, 91(5):1287-1310.
Dunne, Timothy, Mark J. Roberts, Larry Samuelson. 1989. ‖The Growth and Failure of U.S. Manufacturing Plants,‖
The Quarterly Journal of Economics, Vol. 104, No.4, pp.671-698.
Djankov, Simeon, Oliver Hart, Caralee McLiesh, Andrei Shleifer. 2008. ―Debt Enforcement around the World.‖
Journal of Political Economy, 2008, vol. 116, no. 6.
Evans, David S. 1987a. ―The Relationship between Firm Growth, Size, and Age: Estimates for 100 Manufacturing
Industries,‖ Journal of Industrial Economics, Vol. 35, pp.567-82.
Evans, David S. 1987b. ―Tests of Alternative Theories of Firm Growth,‖ Journal of Political Economy, Vol. 95,
pp.657-74
Gauthier, Bernard and Mark Gersovitz. 1997. ―Revenue Erosion through Exemption and Evasion in Cameroon
1993,‖ Journal of Public Economics, Vol.64, pp.407-24.
Heron & Company, for Canadian Small Business Financing Program. 2007. ―Review of SME Loan Guarantee
Programs: Analysis and Comparison of Administrative Features.‖ Ottawa
Honohan, Patrick. 2008. ―Partial Credit Guarantees: Principles and Practice,‖ Prepared for the World Bank
Conference on Partial Credit Guarantees, March 13-14, 2008
Jovanovic, Boyan. 1982. "Selection and the Evolution of Industry," Econometrica, 50(7): 649-67
Klepper, Steven and Peter Thompson. 2007. "Submarkets and the Evolution of Market Structure," Rand Journal of
Economics, 34(4):862-888.
Klette, Tor J, and Samuel S. Kortum. 2004. "Innovating Firms and Aggregate Innovation," Journal of Political
Economy, 112(5):986-1018.
Sleuwaegen, Leo and Micheline Goedhuys. 2002. ‖Growth of Firms in Developing Countries, Evidence from Cote
d‘Ivoire,‖ Journal of Development Economics, Vol. 68, pp.117-35.
Van Biesebroeck, Johannes. 2005. ―Firm size matters: growth and productivity growth in African manufacturing,‖
Economic Development and Cultural Change, Vol. 53(3), pp.545– 83.
95
World Bank. 2007. Turkey Investment Climate Assessment. Washington, DC
World Bank, Poverty Reduction and Economic Management, Turkey Country Management Unit, Restoring
Equitable Growth and Employment Programmatic Development Policy Loan (REGE-DPL). Draft Dec. 14, 2009
World Bank, 2010a. Innovation Policy for Competitiveness and Employment Generation. Forthcoming
World Bank. 2010b. ―The Credit Environment of Small and Medium Enterprises in Turkey: Some Observations,‖
Mimeo, Prepared by Murat Ucer.
Annex 3-A. Descriptive Statistics and Econometric Analysis
Table 3-A-1 shows the number of firms in the sample in each size and age group for both all data and for panel data
only. As expected, larger firms are more likely to be older.
Table 3-A-1 Size and age distribution
ALL FIRMS PANEL FIRMS ONLY
Employ\Age 1-5 6-15 ≥16 Total 1-5 6-15 ≥16 Total
Micro 0.30 0.42 0.28 206 0.18 0.42 0.40 60
Small 0.18 0.48 0.33 374 0.15 0.48 0.37 158
Medium 0.09 0.38 0.53 223 0.09 0.36 0.55 139
Large 0.05 0.38 0.57 111 0.03 0.29 0.68 62
Total 155 395 364 914 50 169 200 419
The survey includes data from manufacturing, retail, and other service sectors. The two digit classification of the
industries is made according to ISIC revision 3.1. The list of industries covered in the data and the number of firms
at different size groups are given in Table 3-A-2. The largest industry in terms of number of observations is Textiles
which is followed by Food industry. For most industries, size distribution has a long right tail. Small firms make the
largest group of firms in almost all industries. The textile industry has the highest share of large firms. The survey
data is also inclusive of all regions in Turkey.
Table 3-A-2 Proportion of firms in industries included in the survey
2 Digit Industries Micro Small Medium Large Total
Food 0.22 0.42 0.26 0.10 130
Textiles 0.07 0.30 0.35 0.28 149
Garments 0.13 0.38 0.35 0.15 96 Chemicals 0.26 0.44 0.22 0.07 95
Plastics & rubber 0.19 0.37 0.33 0.11 27
Non metallic mineral 0.28 0.49 0.22 0.01 83 Basic metals 0.23 0.38 0.31 0.08 13
Fabricated metal products 0.23 0.57 0.10 0.10 30
Machinery and equipment 0.34 0.28 0.28 0.09 32
Electronics (31 & 32) 0.33 0.11 0.44 0.11 9
Services of motor vehicles 0.39 0.45 0.03 0.13 31
Wholesale 0.32 0.57 0.09 0.02 53 Retail 0.38 0.33 0.13 0.16 93
Others* 0.13 0.53 0.25 0.09 68
Total 208 375 223 114 920
* Other industries include Other manufacturing, IT and Transportation sectors
There are five macro- regions covered which are listed in Table 3. Around 40 percent of firms in all regions are
small firms. 40 percent of firms in Black Sea-Eastern region are micro firms. This region has the lowest share of
large firms.
96
Table 3-A-3 Proportion of firms in regions included in the survey
Regions Micro Small Medium Large Total
Marmara 0.14 0.39 0.28 0.19 334
Aegean 0.18 0.39 0.29 0.14 156
Central Anatolia 0.28 0.39 0.26 0.08 145 South 0.29 0.45 0.19 0.07 196
Black Sea - Eastern 0.40 0.44 0.12 0.03 89
Total 208 375 223 114 920
Table 3-A-4 Growth rates of firms at different size and age groups
ALL FIRMS PANEL FIRMS ONLY
Growth\Age 1-5 6-15 ≥16 Total 1-5 6-15 ≥16 Total
Micro 0.29 0.14 0.14 0.18 0.22 0.13 0.24 0.19
Small 0.14 0.02 -0.02 0.03 -0.02 0.02 0.03 0.02
Medium 0.00 0.03 0.01 0.02 -0.31 -0.08 -0.05 -0.08
Large 0.26 0.14 -0.04 0.06 -0.09 -0.15 0.00 -0.03
Table 3-A-5 Fraction of firms in each size-age cell in comparator countries
Turkey
Russia
Employ\Age 1-5 6-15 >=16 Total
Employ\Age 1-5 6-15 >=16 Total
Micro 0.29 0.45 0.26 16,572
Micro 0.37 0.58 0.05 19,554 Small 0.15 0.54 0.31 17,710
Small 0.28 0.63 0.09 28,318
Medium 0.09 0.31 0.60 5,065
Medium 0.08 0.73 0.19 21,734
Large 0.06 0.45 0.49 1,932
Large 0.02 0.50 0.48 7,387
Total 7,945 19,511 13,822 41,279
Total 17,258 48,659 11,077 76,994
Ukraine
Poland
Employ\Age 1-5 6-15 >=16 Total
Employ\Age 1-5 6-15 >=16 Total
Micro 0.36 0.63 0.01 20,933
<=10 0.09 0.63 0.28 34,911
Small 0.24 0.58 0.19 16,489
11-50 0.05 0.62 0.32 15,928 Medium 0.15 0.60 0.25 4,887
51-249 0.04 0.46 0.50 5,920
Large 0.10 0.28 0.62 1,920
>=250 0.13 0.56 0.31 1,176
Total 12,297 26,218 5,713 44,229
Total 4,336 35,210 18,389 57,935
Romania
All ECA
Employ\Age 1-5 6-15 >=16 Total
Employ\Age 1-5 6-15 >=16 Total
Micro 0.36 0.63 0.01 26,556
Micro 0.25 0.64 0.10 269,735 Small 0.17 0.81 0.02 20,608
Small 0.19 0.68 0.13 225,957
Medium 0.11 0.73 0.15 5,195
Medium 0.11 0.63 0.25 76,951
Large 0.31 0.49 0.20 1,010
Large 0.09 0.45 0.46 20,678
Total 13,944 37,720 1,706 53,370
Total 121,983 384,250 87,087 593,320
EU8
EU10
Employ\Age 1-5 6-15 >=16 Total
Employ\Age 1-5 6-15 >=16 Total
Micro 0.18 0.69 0.13 124,967
Micro 0.21 0.68 0.11 166,805
Small 0.18 0.71 0.11 88,097
Small 0.18 0.73 0.09 121,066 Medium 0.11 0.67 0.21 25,969
Medium 0.12 0.68 0.20 32,964
Large 0.17 0.48 0.35 4,501
Large 0.18 0.50 0.32 5,938
Total 42,038 167,904 33,592 243,534
Total 61,424 227,221 38,128 326,773
Results of econometric analysis The specification used in the regression analysis is given in equation (1)
(1)
Annual growth rate of full time employment of firm i at time t (between 2004 and 2007) is represented by git.
Smallit-3, Mediumit-3, and Largeit-3 stand for the indicators of firm size in 2004, obtained from the retrospective
question on employment asked in the 2008 survey. The omitted group is micro firms. Ageit-3 is the age of the firm in
2004. Exportit, Foreignit, and Govtit are the other firm level control variables which are only available in 2007. In the
97
regressions for the panel data use is made of the values of the control variables from the 2005 survey. The size
groups in the regressions for panel data are constructed according to the values in the 2005 survey. Finally Industryi
and Regioni control for the 2-digit industry and the region of the firm. In all regression analyses performed in this
study, probability weights are used in order to obtain results that are representative for all firms in Turkey. All
standard errors are clustered to allow for possible correlations in growth rates across firms within the same industry
and region. The regression results are given in Table 3-A-6.
Table 3-A-6 Regression of employment growth rates
All Data Panel Firms
Small (10-50) -0.14 -0.149
(0.039)*** (0.083)*
Medium (51-250) -0.155 -0.233
(0.033)*** (0.068)***
Large (≥251) -0.114 -0.191
(0.060)* (0.079)**
Age -0.003 0.002
(0.001)** (0.002)
Foreign -0.032 0.02
(0.072) (0.096)
Govt 0.037 0.23
(0.045) (0.041)***
Exporter 0.018 -0.047
(0.033) (0.052)
Const 0.321 0.184
(0.089)*** (0.125)
Obs 668 416 R2 0.201 0.144
Robust standard errors which are given in parentheses are clustered by region and 2-digit industry for all data and by 2-digit industry for the panel data. Additionally control for 2-digit industry and
region fixed effects. *** p<0.01, ** p<0.05, * p<0.1.
In order to analyze whether the growth rates of firms in Turkey are significantly different from comparator
countries, a dummy variable indicating Turkish firms is included. Interaction of this dummy variable with the size
groups reveals whether the average growth rate in each size group in Turkey differs from the growth of firms in the
same size group in the comparator countries. Extending equation (1) to include TRt, as the dummy variable for
Turkey, we obtain
(2)
Regression results are given in Table 3-A-7. Each column shows the result of the regression that includes only the
data from the comparator country (or group of countries) and Turkey. All_ECA column includes all countries in the
region.
Table 3-A-7 How firms in turkey differ from the firms in comparator countries in their growth rates
Russia Poland Ukraine Romania All_ECA EU8 EU10
Small -0.073 -0.059 -0.077 -0.077 -0.055 -0.056 -0.057
(0.025)*** (0.022)*** (0.023)*** (0.021)*** (0.011)*** (0.019)*** (0.016)***
Medium -0.058 -0.038 -0.066 -0.064 -0.056 -0.051 -0.050
(0.022)*** (0.022)* (0.032)** (0.042) (0.012)*** (0.019)*** (0.017)***
Large -0.063 -0.088 -0.068 -0.182 -0.069 -0.058 -0.071
(0.024)*** (0.040)** (0.037)* (0.049)*** (0.017)*** (0.036) (0.031)**
Small*Turkey -0.085 -0.098 -0.088 -0.076 -0.108 -0.106 -0.098
(0.039)** (0.036)*** (0.036)** (0.035)** (0.031)*** (0.035)*** (0.034)***
Medium*Turkey -0.112 -0.127 -0.107 -0.077 -0.119 -0.124 -0.113
(0.041)*** (0.035)*** (0.042)** (0.049) (0.031)*** (0.034)*** (0.033)***
Large*Turkey -0.081 -0.048 -0.074 0.069 -0.073 -0.083 -0.060
(0.048)* (0.057) (0.050) (0.063) (0.045) (0.055) (0.052)
Turkey 0.037 0.148 0.109 0.073 0.068 0.120 0.109
(0.046) (0.032)*** (0.032)*** (0.034)** (0.037)* (0.033)*** (0.033)***
Constant 0.103 0.111 0.061 0.190 0.105 0.064 0.075
(0.037)*** (0.025)*** (0.045) (0.032)*** (0.033)*** (0.052) (0.030)**
98
Obs 1594 1088 1462 1128 7392 2292 2758
R2 0.188 0.191 0.187 0.188 0.117 0.130 0.121
Robust standard errors clustered by region and 2-digit industry are in parentheses. Additionally control for foreign and government ownership,
age, export status, 2-digit industry, and region fixed effects. Regressions for EU8, EU10 and All_ECA include controls for countries. *** p<0.01, ** p<0.05, * p<0.1.
In order to account for investment climate effects, investment climate variables are included one at the time, with
ICvarjct representing the value of the average investment climate variable at industry j, region c, and time t:
. (3)
Since the main data source is cross-sectional, inclusion of the IC variables can cause endogeneity problem even
when the chosen variables are not subjective measures that depend on firms‘ perceptions of the business
environment. For instance, faster growing firms might be more likely to have access to external finance. To alleviate
this problem, we use industry-region averages for the IC variables, capturing the total effect of the business
environment in which firm operates. Cells with less than 5 observations are excluded from the analysis.
Table 3-A-8 Firm growth and investment climate variables
IC Variable Used Smallt-3 Mediumt-3 Larget-3 Aget-3 IC Vart Const Obs R2
Labor regulations -0.131 -0.148 -0.107 -0.003 0.414 0.170 668 0.214
(0.040)*** (0.034)*** (0.060)* (0.001)** (0.195)** (0.085)*
New fixed asset -0.13 -0.152 -0.113 -0.003 0.003 0.316 668 0.21 finance - external (0.040)*** (0.032)*** (0.060)* (0.001)** (0.001)** (0.090)***
Collateral -0.10 -0.156 -0.153 -0.005 -0.001 0.475 219 0.35
0.102 (0.092)* 0.127 0.004 0.001 (0.129)***
Purchase paid after
delivery -0.14 -0.157 -0.114 -0.003 0.000 0.333 660 0.20
(0.039)*** (0.033)*** (0.060)* (0.001)** 0.002 (0.150)**
Overdraft facility -0.137 -0.154 -0.11 -0.003 -0.144 0.348 656 0.21
(dummy) (0.040)*** (0.033)*** (0.062)* (0.001)** 0.122 (0.076)*** Loan (dummy) -0.145 -0.163 -0.125 -0.003 0.334 0.216 665 0.22
(0.039)*** (0.034)*** (0.064)* (0.001)** (0.079)*** (0.087)**
Loss from power -0.218 -0.266 -0.22 0.000 0.003 0.272 223 0.33 outages (0.094)** (0.073)*** (0.081)*** 0.002 0.006 0.204
Informal payments -0.157 -0.167 -0.096 -0.003 0.029 0.281 551 0.23
(0.044)*** (0.041)*** (0.053)* (0.001)** 0.024 (0.084)***
Tax inspections -0.195 -0.199 -0.155 -0.004 0.000 0.525 324 0.30
(number) (0.084)** (0.057)*** (0.084)* (0.001)*** -0.001 (0.133)***
Time tax -0.116 -0.123 -0.086 -0.003 0.002 0.207 617 0.18
(0.037)*** (0.032)*** 0.063 (0.001)** -0.002 (0.104)*
Customs clearance -0.171 -0.247 -0.229 -0.004 0.010 0.316 336 0.38
for exports (days) 0.13 (0.114)** 0.138 (0.002)** (0.004)** (0.112)*** Customs clearance -0.379 -0.489 -0.501 -0.005 0.002 0.414 208 0.52
for imports (days) (0.142)** (0.125)*** (0.122)*** (0.002)** 0.003 (0.139)***
Inspections -0.13 -0.159 -0.121 -0.003 -0.005 0.253 636 0.19 (number) (0.043)*** (0.037)*** (0.063)* (0.001)** 0.003 (0.065)***
Robust standard errors clustered by region and 2-digit industry are in parentheses. Additionally control for foreign and government ownership, export status, 2-digit industry, and region fixed effects. *** p<0.01, ** p<0.05, * p<0.1.
Each row shows the IC variable included in the regression equation.
In order to analyze whether the IC variables have any differential affect on the growth rates of firms of different
sizes, interaction terms of the IC variables with the size dummies are included in the equation, yielding:
. (4)
99
Table 3-A-9 Firm growth and investment climate variables with interactions (all data) IC Variable Small Medium Large IC IC*Small IC*Medium IC*Large Const Obs R2
Labor
regulations
-0.167 -0.122 -0.098 0.230 0.457 -0.238 -0.076 0.181 668 0.220
(0.070)** (0.052)** (0.089) (0.378) (0.466) (0.431) (0.555) (0.093)*
External finance
-0.031 -0.223 -0.149 0.004 -0.005 0.003 0.002 0.158 668 0.23 0.078 (0.089)** (0.082)* (0.002)* 0.004 0.004 0.004 (0.074)**
Collateral -0.086 -0.244 -0.044 -0.001 0.000 0.001 -0.001 0.086 219 0.35
0.196 0.168 0.208 0.001 0.001 0.001 0.001 0.219 Purchase
paid after
delivery
-0.251 0.015 0.32 0.000 0.002 -0.002 -0.006 0.319 660 0.21
0.196 0.232 0.234 0.003 0.003 0.003 (0.004)* (0.175)* Overdraft
facility
(dummy)
-0.112 -0.426 -0.32 -0.161 -0.038 0.382 0.286 0.352 656 0.21
0.132 0.28 0.265 0.168 0.196 0.387 0.412 (0.103)***
Line of
credit
0.261 0.019 0.378 0.783 -0.629 -0.294 -0.79 -0.105 665 0.24
0.185 0.165 0.236 (0.252)*** (0.320)* 0.262 (0.413)* 0.163
Loss from power out.
-0.263 -0.362 -0.255 -0.001 0.01 0.023 0.007 0.165 223 0.33 (0.119)** (0.109)*** (0.117)** 0.009 0.015 0.019 0.019 0.11
Informal
payments
-0.054 -0.133 0.006 0.14 -0.171 -0.08 -0.196 0.226 551 0.28
0.036 (0.054)** 0.051 (0.051)*** (0.059)*** 0.048 (0.052)*** (0.085)*** Tax
inspections
-0.193 -0.2 -0.148 0.000 0.000 0.000 -0.002 0.524 324 0.31
(0.088)** (0.058)*** (0.087)* 0.001 0.001 0.001 (0.001)** (0.135)***
Time tax -0.094 -0.109 0.338 0.003 -0.001 -0.001 -0.015 0.174 617 0.19 0.095 0.136 0.216 0.003 0.003 0.005 (0.007)** 0.132
Customs cl.
- exports
-0.095 -0.237 -0.162 0.016 -0.013 -0.001 -0.012 0.296 336 0.39
0.171 0.151 0.181 0.012 0.012 0.011 0.013 (0.121)** Customs cl.
- import
-0.606 -0.796 -0.903 -0.025 0.025 0.032 0.045 0.885 208 0.55
(0.218)*** (0.198)*** (0.212)*** (0.013)* (0.014)* (0.015)** (0.017)** (0.194)***
Inspections (number)
-0.111 -0.162 -0.09 0.003 -0.008 0.001 -0.013 0.236 636 0.19 (0.054)** (0.056)*** 0.095 0.016 0.015 0.017 0.024 (0.070)***
Robust standard errors clustered by region and 2-digit industry are in parentheses. Control for foreign and government ownership, export status, 2-
digit industry, and region fixed effects. *** p<0.01, ** p<0.05 p<0.1 Each row shows the IC variable included in the regression equation.
Robustness Although averaging the firm level values of IC variables over industry-region alleviates the endogeneity problem, it
might not remove it completely. We perform the same regression analysis specified in equation 3 restricting the
sample to panel firms. In the panel regressions, we use the values of firm characteristics and IC variables from the
2005 survey. Here we cluster the standard errors at 2-digit industry level. Despite of the small sample size for the
panel data and the inclusion of the controls for IC variables, for most of the regressions, medium size firms have the
lowest growth rate. About the IC variables, time to clear customs for both exporting and importing is negatively
related to firm growth.
Table 3-A-10 Effects of IC Variables on firm growth – panel firms Small Medium Large Age IC Const Obs RSqr
New fixed asset financing -0.149 -0.233 -0.19 0.002 0.000 0.196 416 0.14
- external finance (0.084)* (0.067)*** (0.079)** (0.002) (0.001) (0.100)* Collateral -0.177 -0.235 -0.098 0.002 0.000 0.212 116 0.51
(0.067)** (0.043)*** (0.040)** (0.002) (0.000)** (0.101)*
Purchase paid after delivery -0.148 -0.233 -0.19 0.002 -0.001 0.232 414 0.15
(0.084)* (0.067)*** (0.080)** (0.002) (0.000) (0.142)
Overdraft facility (dummy) -0.143 -0.241 -0.207 0.002 0.048 0.127 416 0.15
(0.081)* (0.068)*** (0.082)** (0.002) (0.039) (0.110)
Loan (dummy) -0.143 -0.241 -0.207 0.002 0.048 0.127 416 0.15
(0.081)* (0.068)*** (0.082)** (0.002) (0.039) (0.110)
Loss from power outages -0.202 -0.343 -0.276 0.003 0.004 0.171 308 0.22
(0.080)** (0.071)*** (0.080)*** (0.003) (0.002) (0.143)
Informal payments -0.15 -0.23 -0.208 0.002 -0.001 0.205 368 0.15
(0.089) (0.073)*** (0.089)** (0.002) (0.003) (0.145)
Tax inspections -0.145 -0.23 -0.2 0.002 0.001 0.175 406 0.14
(0.087) (0.070)*** (0.083)** (0.002) (0.002) (0.079)**
Time tax -0.144 -0.241 -0.202 0.003 0.001 0.198 403 0.14
(0.088) (0.077)*** (0.082)** (0.002) (0.001) (0.121)
Customs clearance for export -0.111 -0.295 -0.188 0.003 -0.008 0.25 201 0.27
(0.049)** (0.139)* (0.111) (0.001)** (0.002)*** (0.085)**
Customs clearance for import -0.065 -0.091 -0.027 0.001 -0.004 1.405 139 0.37
100
(0.095) (0.133) (0.097) (0.002) (0.002)** (0.128)***
Robust standard errors clustered by region and 2-digit industry are in parentheses. Control for foreign and government ownership, export status, 2-digit industry, and region fixed effects. *** p<0.01, ** p<0.05 p<0.1
Each row shows the IC variable included in the regression equation.
As an alternative to four size groups, we constructed three size groups measured as total fulltime employment. In
this grouping, size classes are Small: ≤19, Medium: 20-99, Large: ≥100. This is the classification used in ES surveys
for stratifying the population of firms in size. In Table 3-A-11, for brevity, we only present the results that include
external finance and line of credit as IC variables. These IC variables were the ones that had significant coefficients
in the main regression results and they are the only variables that persisted to be significant in this alternative size
grouping. In these regressions, the group of firms with less than or equal to 19 employees is omitted. The non-
monotonic decrease in growth rates as size increases is observable in this size grouping as well.
Table 3-A-11 Employment Growth with Three Size Groups
(1) (2)
Medium (20-99) -0.097 -0.103
(0.032)*** (0.031)***
Large (≥100) -0.085 -0.088
(0.043)* (0.044)**
Age -0.003 -0.003
(0.001)** (0.001)**
External Fin 0.004
(0.001)***
Loan (dummy)
0.311
(0.073)***
Constant 0.197 0.09
(0.084)** -0.081
Observations 668 665 R-squared 0.188 0.185
Robust standard errors clustered by region and 2-digit industry are in parentheses. Additionally control for foreign and government ownership, export status, 2-digit
industry, and region fixed effects. *** p<0.01, ** p<0.05, * p<0.1.
To provide further support to the significance of access to finance for firm growth, we used principal component
method. First we determined the principals of the IC variables that measure access to finance. The first regression
uses six of the seven IC variables that are related to access to finance. The excluded variable is the use of collateral.
Collateral variable is included in the second estimation. Since very few firms answered the question for collateral
usage, the sample size in the second regression is much smaller. The principal component analysis allows explaining
the overall effects of variables representing access to finance. In the regressions, only the first component is
included, since higher ranked components were not statistically significant. In both regressions, access to finance
contributes positively to growth.
Table 3-A-12: Access to finance and growth: principal component method (1) (2)
Small -0.163 -0.177
(0.075)** (0.145)
Medium -0.135 -0.249
(0.055)** (0.103)**
Large -0.159 -0.253
(0.097) (0.159)
Age -0.004 -0.004
(0.002)** (0.005)
Principal component 0.06 without collateral (0.020)***
Principal component
0.067 with collateral
(0.029)**
Constant 0.304 0.002
(0.115)** (0.191)
Obs 419 170
R2 0.248 0.435
101
Robust standard errors clustered by region and 2-digit industry are in parentheses. Additionally control
for foreign and government ownership, export status, 2-digit industry, and region fixed effects. ***
p<0.01, ** p<0.05, * p<0.1.
We also used several alternative definitions of growth rates. One is log of employment growth and another is an
extensively used measure of job creation and destruction in labor economics. It is found as dividing the difference in
employment levels over the past three years by the simple average of employment levels over the same period. This
measure is bounded by values of -2 and 2 and provides more robust estimates to outliers. In either case the relation
between size and employment growth was not sensitive to these alternative measures of growth.
102
Chapter 4. STIMULATING KNOWLEDGE FLOWS
4.1. Since the 1980s, the liberalization of the Turkish economy stimulated the creation of
regional production networks centered on internationally competitive manufacturers. The 1980s,
characterized by a transition from protectionist attitudes towards increased reliance on market forces, are
considered a turning point for the international integration of the Turkish economy.73
The completion of
trade liberalization in 1984 – requiring the dismantling of foreign exchange controls, the abolition of
quotas on imports, and a downward revision of tariffs – was accompanied by the enactment of pro-active
export promotion policies and by the depreciation of exchange rates. This presented the Turkish economy
with new opportunities related to a generalized rise in international capital mobility and trade in
intermediate goods. As a result, Turkey‘s share of world manufacturing exports increased six-fold from
0.15 percent in 1980 to 0.90 percent in 2008.74
Integration into global trade and investment flows has
been accompanied by a significant spatial transformation of the Turkish economy characterized by the
emergence of a number of more recent industrial agglomerations far from the earlier manufacturing
regions, the so called ―Anatolian Tigers.‖ Clusters of industries, with specialization in both traditional and
more technologically advanced sectors, have formed in various parts of the country and have become the
core of manufacturing and exporting activities (Figure 4-1).
4.2. In response to these developments, the government has activated a number of instruments
to foster the ability of SMEs to participate in global markets. Beginning in the mid-1990s the
attention of policy has turned to the triple challenge of more growth, greater competitiveness, and more
jobs through the development of SMEs.75
In this spirit, a number of instruments have been activated with
the intent of stimulating the participation of SMEs in regional production networks.76
The rationale of
several such interventions has been to remove obstacles to the competitiveness of SMEs related to the
business environment. For instance, SMEs in the manufacturing sector have been encouraged to locate in
appropriately planned ―small industrial estates" (KSS) and ―organized industrial zones‖ (OSB) that ease
investment climate constraints by providing a number of advantages in terms of infrastructure services
and regulation of business activity. In a similar vein, the government has encouraged the development of
regional innovation strategies aimed at sustaining the competitive advantage of regions. In the future,
primary responsibility for the preparation of regional innovation strategies is expected to lie with the
newly established development agencies discussed in Chapter 2.
4.3. Global conditions in the aftermath of the crisis highlight the need to further improve the
international competitiveness of smaller local firms in order to achieve sustainable and broad-based
growth. Chapter 2 emphasized how larger Turkish firms are able to benefit from the more positive
aspects of the investment climate. Foreign-owned firms and successful exporters appear at the forefront of
this pattern, with TFP and investment climate variables related to innovation and quality appearing as
crucial determinants of the probability of exporting and, to a more limited extent, to receive FDI (Figure
2-5). This is in line with the empirical evidence across developed and developing countries indicating that
larger and more productive firms more easily manage to penetrate export markets, and that foreign owned
firms display higher productivity and a higher knowledge content of production relative to domestic
counterparts. 77
The 2008-2009 economic crisis – with its widespread decline in international trade and
73 Eraydın and Armatlı (2005). 74 Export data are compiled from the World Bank WDI database. 75 Kuruüzüm, 1998. 76 See Doloreux and Parto, 2005; Bathelt et al. (2003); Malecki and Oinas (2000); and Tallman et al. (2004). 77 See Wagner (2007) for an overview of the literature on the link between productivity and exporting and Markusen (2005) for a
discussion of the peculiarities of foreign-owned vis-à-vis local firms.
103
capital flows likely to continue in the near future – reinforces the relevance of productive efficiency and
of the ability to innovate as crucial conditions for the international competitiveness of the Turkish
manufacturing sector. Whereas these challenges are present for successful Turkish exporters it is even
more urgent, as already emphasized in the 2007 ICA, to ensure that smaller firms that are not directly
integrated into global markets are able to increase the knowledge content of their production in order to
participate in global value chains. Diffusion of the sources of growth beyond firms that are currently
sufficiently competitive to be direct exporters and outside more successful manufacturing poles will
increase the resilience of the Turkish economy to future shocks in global demand and guarantee that the
productive base of Turkish manufacturing is more evenly distributed across Turkish regions.
Figure 4-1: Industrial clusters in Turkey
Structural characteristics of selected clusters
Factors/ Conditions Denizli Bursa Ankara
The type of the
manufacturing
cluster
Industrial district Innovative manufacturing
cluster
High-tech industrial cluster
Area of
specialization
Textiles, especially towels
and bathrobes
Textiles for home furnishing Machinery, electronics,
defense industry and software
The main character
of the cluster
Traditional
Small artisanal, and highly
specialized family owned
firms located in close
proximity
Traditional/Modern
Small Artisanal, and highly
specialized firms as well as
large multinational companies
co-operating with these small
enterprises
Modern/High-tech
High-tech firms of different
size
Main observed
benefit
Co-operation in production
and
marketing for international
markets
Collective competition in
specialized fields
Weak collaborative
environment Market relations
with state institutions
Technical
dynamic Social
capital
Complementarities
Collaborative action, trust
and reciprocity Strong
social networks
Specialization increasing
shares of export in engineering
industries, Adaptation and
product development for
international markets
Adaptation of new
technologies for national
market
Access to qualified labor
Source: Öz (2004) and Eraydın and Armatlı (2005)
104
4.4. Analysis of the case of Turkish production networks indicates that the absorptive capacity –
i.e. the ability to adopt and use knowledge – of local suppliers, especially SMEs, is key for successful
participation in global markets. Global buyers (domestic or foreign-owned manufacturers, retailers or
brand-name companies) have come to play an increasingly important role in the organization of global
production and distribution systems. In this light, this Chapter examines knowledge flows in Turkish
manufacturing by focusing on the ability of local manufacturers to participate in global markets via their
participation in production networks. A Global Value Chains (GVC) perspective is used to analyze the
mechanisms of knowledge transfer at firm level, and the intensity with which buyers – usually larger
firms with direct links to domestic or international markets –are engaged in transferring knowledge and
technology to smaller suppliers. Analysis shows that (i) suppliers endowed with some technical skills
and capacity “to handle the technology” are more likely to enter knowledge intensive value chain
relationships. Furthermore, (ii) more capable suppliers are more likely to engage in value chain
relationships characterized by transfer of design, quality standards and technology from the buyer to
the supplier. A more efficient regulatory environment, specifically operating licenses and permits, and
higher shares of investment financed from bank loans are also positively associated with knowledge
intensive value chain arrangements. The latter result confirms the central role of access to external finance
already highlighted in Chapter 3.
4.1 Production Networks and Knowledge Flows
4.5. Production networks can play a crucial role in diffusing technological knowledge and
enhancing learning and innovation, with the intensity of knowledge transfer depending on the
complexity of the task, the extent to which information can be codified, and the capabilities of
potential suppliers. Recent developments in the GVC literature have shown the importance of value
chain governance modes, especially those between suppliers and buyers, in enhancing technology
diffusion, learning and innovation (Box 4-1). 78
Value chain governance modes refer to linkages among
buyers, sellers and service providers that influence the range of activities required to bring a product or
service from inception to its end use. The key parameters shaping the nature of the underlying
arrangements among firms are what, how, when, and how much is to be produced. Alternative types of
governance modes can thus be defined based on the ―intensity‖ of knowledge transfer between buyers and
suppliers participating in a value chain and are determined by: (i) the complexity of information and
knowledge transfer required to sustain a particular transaction, especially with respect to product and
process specifications; (ii) the extent to which this information and knowledge can be codified and,
therefore, transmitted efficiently and without transaction-specific investment between the parties involved
in the transaction79
; and (iii) the capabilities of actual and potential suppliers in relation to the
requirements of the transaction.80
Governance modes are thus considered a key vehicle for knowledge
transfer from global buyers towards local firms, as well as key determinants of entrepreneurial growth for
local suppliers.81
78 Governance of the value chain can be defined as ―authority and power relationships that determine how financial, material, and
human resources are allocated and flow within a chain‖ (Gereffi, 1994). Also see Altenburg, 2006; Gereffi, 1999; Giuliani et al.,
2005; Kaplinsky, 2000; Humphrey & Schmitz, 2002a,b; Pietrobelli and Rabellotti, 2007. 79 Typically, the knowledge base of traditional industries is highly dependent upon local and tacit forms of knowledge, whereas
the knowledge base of firms in high-technology sectors is more codified allowing firms to establish networks to access distant
knowledge sources (Vale and Calderia 2006). 80 Sturgeon, 2001; Gereffi and Kaplinsky, 2001; Saliola and Zanfei, 2009. 81 Kappel and Brach 2009.
105
Box 4-1: Global value chains
Recent developments in the studies on Global Value Chains (GVC) have shown the importance of value chain
governance modes, especially those between suppliers and buyers, in enhancing technological knowledge diffusion,
learning and innovation. Governance modes are considered to be a key vehicle for knowledge transfer from global
buyers towards local firms and to enhance entrepreneurial growth of local suppliers. From this perspective, the GVC
literature has used the tool of the governance to examine how firms active in global markets organize the transfer
and coordination of complex and strategic information along value chains and their implications for knowledge
transfer and development.
There is a long tradition of studies investigating the channels through which knowledge is transferred to local
companies in global markets. In particular, a plethora of contributions have focused on Multinational Companies
(MNCs) generating knowledge (and pecuniary) spillovers to the host economy through linkages with local firms
(Hirschman (1958), Lall (1978), Rodriguez-Clare (1996), Markusen and Venables (1999), Görg and Strobl (2001)
and Jarvocik-Smarzynska (2004).). The theoretical debate is dominated by arguments predicting positive effects,
including the transfer of new technologies and management skills and the increase in the level of competition.
However, there is little empirical evidence suggesting that domestic firms benefit from multinational presence, and
the controversy is far from resolved.
The GVC approach can provide complementary insights to these studies. The GVC perspective is useful for various
reasons. First, because the focus moves from manufacturing only to the other activities involved in the value chains
of goods and services, including distribution and marketing. Second, GVC emphasizes the nature of the
relationships among the various actors involved in the chain. Moving beyond firm-specific analysis and
concentrating on inter-firm linkage allows to better capture dynamic flows of economic and organizational activities
between producers within different sectors even on a global scale. Finally, GVC studies identify distinct types of
relationships characterized by a different level of involvement of suppliers in knowledge intensive activities. This
helps capture the essential features of knowledge dissemination and absorption, which are likely to enhance the
development of local suppliers.
4.6. Turkey offers an interesting case study to analyze knowledge flows within production
networks. First, the increasing presence of firms, both domestic and multinational, linked to
different global value chains has provided learning opportunities for local Turkish enterprises. Such
firms act as focal or lead firms for the local network, and perform the key functions of generating new
knowledge and technologies, spinning out innovative companies, attracting researchers, performing
investments in R&D and related research facilities, enhancing R&D activities in networked firms, and
stimulating demand for new knowledge.82
Second, networks of firms, with the knowledge flows that
characterize them, have proved to be a valid alternative to vertical integration in a number of
sectors. A case in point is that of Turkish machinery producers, which, despite weak vertical linkages in
globally integrated supplier networks, which are considered the primary providers of innovation, have
been able to integrate alternative global value chain arrangements that, short of vertical integration, offer
SMEs an opportunity to expand their ability to innovate. 83
Finally, international retailers play an
important role in the development of Turkish industrial districts, by transferring tacit expertise in related
fields, such as marketing, to firms participating in a local production network.84
4.7. Turkish subcontracting relationships can be portrayed to have three main characteristics.85
First, some of the networked relationships are long-term and duration is connected to product-life
cycles. Each time a new product is designed and manufactured, the large firm makes a call for the best
82 Erdil 2009; Autio et al., 2004; Bergman and Feser, 1999 and Ceglie and Dini, 1999. 83 See Kozan et.al. 2006, Ulusoy 2003, Erdil and Çetin 2008. 84 See Öz 2004, Nadvi 1995. 85 See Ulusoy 2003.
106
offer from suppliers. At that stage, suppliers are put into competition. However, the firm generally
continues subcontracting relationships with the same suppliers from a product to another, in order to
avoid costly and time consuming renegotiations. Such duration of relationships allows deriving some of
the benefits of vertical integration. Second, some of the Turkish networked relationships are
institutionalized and hierarchical. Such hierarchy of subcontractors is defined according to the type of
product bought by the large firm. In this case, the subcontractors are autonomously chosen on the basis of
quality. Product design is often carried out jointly by the supplier and the main firm. In the latter case, the
supplier only executes orders from the firms according to its production definitions, and is highly
dependent on the large firm. Third, the Turkish networked relationships are contractual and
characterized by specific procedures. Usually a supplier is contracted when the new product is in the
development phase (with no specification of quantities to be delivered, nor of prices, etc.) providing
flexibility and adaptation capability to possible changes in the specification of products.
4.8. Analysis of survey data confirms that buyer-supplier relationships in Turkey are quite long-
term with an average duration of 12 years. The NIS survey (Box 4-2) allows mapping the nature of
buyer supplier relationships in Turkey. By looking at the sample distribution by firms‘ age, on average,
young companies (six years or less in business) have worked with their largest buyer for 5 years (73
percent of their business life) and firms with more than 40 years of activity for 22 years (55 percent of
their business life time). This result is in line with the argument that Turkish buyers, once they have
chosen their suppliers, tend to continue working with them.86
Relationships tend, therefore, to be based on
trust between the parties, which avoids costly and time-consuming renegotiations and reduces the time
needed to learn the specifications of the product.
Box 4-2: The “Networks and Innovation” Survey (NIS)
This Chapter is based on the ―Networks and Innovation‖ survey (NIS) implemented by the World Bank in the
summer of 2009 which represents a complementary study of the Enterprise Survey (ES) implemented in 2008 (See
Box 2-1). The NIS survey was administered to 514 establishments in the manufacturing sector and it represents a
sub-sample of firms drawn from the set of firms previously interviewed in the ES Turkey 2008. The ES Turkey
2008 data has been used for comparisons in the analysis. The database contains also retrospective information for up
to three periods before the year of reference for 231 establishments.
Sampled firms were stratified by size, location, and sector to ensure that most major types of firms are covered. 15
percent of firms in the sample are classified as micro (less than 10 employees), 41 percent small (11-50 employees),
30 percent medium (51-249) and 13 percent large (250 employees or more). Firms surveyed are active in the
following industries: Food processing (18 percent), Textiles (21 percent), Garments (10 percent), Chemicals (12
percent), Non-metallic mineral products (11 percent), and a residual category labeled ―other manufacturing‖ and
comprising rubber and plastics, basic metals, fabricated metal products, machinery and equipment, electrical
machinery and apparatus, motor vehicles, trailers and semi-trailers, furniture (28 percent). The regions covered are
five: Marmara, Aegean, Central Anatolia, South and Black-Sea Eastern. Marmara includes Bursa, Istanbul and
Koaceli; Aegean includes Denizli, Izmir and Manisa; Central Anatolia includes Ankara, Eskisehir, Kayseri and
Konya; South region includes Adana, Gaziantep, and Kahramanmaras city; and Black Sea-Eastern includes
Erzurum, Malatya, Samsun, and Trabzon city. Only firms with five employees or more were included in the sample.
4.9. The degree of buyer involvement is shaped by the sector in which firms operate, with low-
tech sectors characterized by a more disintegrated supply chain showing greater buyer involvement
and a higher degree of knowledge transfer. On average, 44 percent of firms‘ sales are made according
to largest buyer‘s unique specification. This varies substantially across firms in different sectors. For
example, firms in textile and garments industries make about 55 and 51 percent of their sales,
86 See Kozan 2006.
107
respectively, according to largest buyer unique specification compared to 25 percent of firms in food and
chemicals industries. In 61 percent of the firms in the sample the largest buyer gets involved in defining
design & quality. Labor-intensive industries – such as food, textiles, and garments – show greater buyer
involvement in design and quality standards. Over time, based on the development of relationships based
on mutual trust, suppliers and buyers are expected to share sensitive strategic data. This sharing process is
associated with a number of benefits, notably encouraging suppliers to invest in buyer‘s future needs,
generating entrepreneurial growth, and creating and expanding competitive advantages and synergy
effects.87
33 percent of firms are characterized by buyer involvement in R&D activities, with firms in
food and textiles more likely to receive inputs on process or product R&D activities from their largest
buyer. On average 17 percent of enterprises in the sample received technology inputs from buyers. The
Marmara region stands out in as being the region with the smallest share of firms receiving buyer support
for technology dissemination. Furthermore, buyer involvement in dissemination and diffusion of new
technologies affects sectors unequally, with firms in garments the most likely to receive technology and
firms in chemicals the least likely. This result is obviously linked to the nature of the technology and of
the value chain arrangements typical of each sector, with a sector like garments having a highly
disintegrated value chain and a definition of technology that may include the acquisition of low-tech
equipment from a lead buyer.
4.10. Buyer influence on sales tends to become more prominent as firms become larger, whereas
inputs on R&D are more common for SMEs. Across firm size, large firms exhibit the largest share of
sales made to meet buyer‘s unique specification (51 percent, versus 54 percent for SMEs. On the other
hand, medium firms are more likely to receive inputs from buyers on design and quality standards (77
percent) compared to small and large firms (61 and 70 percent respectively). Inputs on R&D from the
largest buyer are more common with small and medium businesses (38 and 37 percent respectively),
compared to 20 percent for micro firms and 30 percent for large enterprises. Finally, the percentage of
large firms with buyer involvement in technology dissemination is significantly higher than for medium
and small firms (28, 14 and 20 percent respectively). By and large, the survey findings show that micro
businesses are significantly less involved in a close relationship with the main buyer, relative to the other
size groups.
4.11. While all value
chain relationships imply
some transmission of
information between the
parties, the extent to
which knowledge is
actually created,
transferred and adopted
along value chains varies
substantially. From this
perspective, one can
characterize at least four
different types of value
chain agreements, based
on the intensity of
knowledge transfer (Figure
4-2). The first, which we
shall name GOV0, is
characterized by no
87 Kappel and Brach 2009.
Figure 4-2: Value chain governance modes
GOV0
GOV1
GOV2
GOV3
Design and QualityProcess and Product
R&D Activities
Dissemination new technologies into firm’
facilities
Design and QualityProcess and Product
R&D Activities
Dissemination new technologies into firm’
facilities
Design and QualityProcess and Product
R&D Activities
Dissemination new technologies into firm’
facilities
Design and QualityProcess and Product R&D
Activities
Dissemination new technologies into firm’
facilities
108
knowledge transferred from buyer to supplier. The second, GOV1, is characterized by the transfer of a
product specification from the purchasing firm to the supplier, and implies that the latter will carry out a
number of autonomous tasks which will eventually lead to the provision of the required good (or service),
following the directions given by the buyer. Under this circumstance, knowledge transferred is kept to a
minimum, will be mostly codified, and it will flow in one direction only (from the buyer to the supplier).
A third mode of governance, called GOV2, takes place when the buyer provides product design and
quality standards to be followed in the production process. Knowledge transfer is uni-directional here too,
but it may well be partly tacit and it is more detailed than in the former case. Under this circumstance, the
supplier is given strict and mandatory instructions, and is thus granted very little degrees of freedom when
executing his/her tasks. A fourth typology of value chain relationship, GOV3, can be observed when the
buyer disseminates specialized competencies, and involves the supplier in R&D and knowledge
development. One should notice it is not only a matter of ―quantity‖ of knowledge flowing between firms:
a substantial part of the transfer takes place through the mobility of personnel. The knowledge involved is
thus largely tacit in nature, and flows in both directions, although not necessarily in a balanced way.
These typologies of value chain relationships thus correspond to different modes of organizing knowledge
transfer and diffusion.
4.12. The intensity of knowledge transfer varies substantially across regions. Looking at regional
features depicted in Figure 4-3,
the data reveal that Marmara,
Aegean and Central Anatolia
are mainly associated with less
knowledge intensive buyer-
suppliers relationships (GOV0
and GOV1). This is an
indication of the fact that firms
in these more developed
regions are, on average, more
capable of harnessing
knowledge without the need
for buyer involvement and also
of the prevalence of certain
sectors. Buyer-supplier
relationships featured by
transfer of design and quality standards, and a direct involvement in R&D activities but not in technology
dissemination are more common among firms located in the Black-Sea Eastern and South regions.
Figure 4-3: Governance modes features, across regions
Source: Turkey NIS 2009
38%33%
30%
23%
27%25%
21%
35%
29%
23%24%25%
14%
29% 29%
13%
20% 21% 20% 20%
0%
5%
10%
15%
20%
25%
30%
35%
40%
Marmara Aegean Central Anatolia South Black-Sea Eastern
Gov0 Gov1 Gov2 Gov3
109
4.13. Governance modes are also different depending on the organization of production typical
of each industrial sector. Metallic products and other manufacturing industries are highly associated
with little or no knowledge transfer (GOV0). This is an indication that these sectors are, by nature,
characterized by more standardized products which require little or no knowledge transfer from buyers
and that the firms in these sectors that are competitive internationally rely merely on their productivity –
as testified by the vast literature on the self-selection of more efficient firms into export markets – the
quality of their products achieved via successful R&D expenditures, and pricing advantages. Firms in the
textile industry tend to be associated mainly with transfer of design and quality standards and a direct
involvement in R&D activities by the buyer (GOV1 and GOV2). This is also linked to the specific
characteristics of these products, where Turkish manufacturers are often sub-contractors for main buyers,
domestic or foreign, that possess and maintain within their boundaries the functions associated with
higher value added, such as design, quality control and marketing. However, these main buyers have the
need to transfer some of their R&D and quality control competences backward to their suppliers in order
to ensure a certain standard for the product being subcontracted. In food processing 30 percent of firms in
the sample receive
transfer of design and
quality standards, and
45 percent of firms
receive also buyer‘s
involvement in R&D
and technology
dissemination activities.
Firms in the garments
industry show the
highest share of firms
characterized by buyer
transfer of precise
requirements in terms
of design and quality
standards, and
involvement in
technology dissemination and R&D (GOV3). In both of these industries, the buyer is a large retailer or
distributor closer to the final consumer, that needs to ensure maximum consistency of its products with
expected standards and therefore has an interest in maximum hands-on involvement with production in
supplier firms.
4.14. Medium and large firms stand out
for the relatively higher share of suppliers
involved in knowledge intensive
relationships. This is featured by transfer of
D&Q inputs for both medium and large
enterprises (GOV1), whereas input on R&D
investments (GOV2) is more prominent for
medium firms. Additionally, technology
dissemination from the main buyer into
firms‘ facilities (GOV3) seems to be more
frequent in large enterprises. This is probably
an indication of firms active in labor-
intensive sectors, such as textile, garments or
food, which receive a large amount of
knowledge from their buyers. Again, the
results show that knowledge transfer is
Figure 4-4: Governance modes features, across industries
Source: Turkey NIS 2009
Figure 4-5: Governance modes features, by firm size
Source: Turkey NIS 2009
25%
19%
31%
42%39%
41%
30%
35% 34%
38%
29%
19%
30%
35%
3%
17%
25% 24%
15%11%
32%
2%
7%
17%
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
Manufacture of Food
Textile Garments Chemicals Non metallic mineral
products
Other Manufacturing
Gov0 Gov1 Gov2 Gov3
50
31
2124
2722
37 36
14
27 29
139
20
14
28
0
10
20
30
40
50
60
micro small medium large
Gov0 Gov1 Gov2 Gov3
110
relatively less frequent in micro enterprises, whereas a significant number of small firms (27 percent)
receive buyer inputs in D&Q and R&D investments.
4.15. No major differences are observed across firms in terms of share of direct exports based on
governance features. Figure 4-6 shows that the more knowledge intensive the buyer-supplier
relationship is, the higher the share of
indirect exports (sold domestically to third
party). Firms characterized by GOV3
(transfer of design, quality standards, R&D
and technology dissemination) sell 17
percent to distributors compared to 8 percent
of firms characterized by no transfer of
knowledge (GOV0). By restricting the
analysis to those firms that were interviewed
in 2005 and in 2009 (panel firms), we notice
that firms characterized by GOV1 and GOV3
in 2009 have more than doubled the share of
indirect exports. Also, we observe that the
share of direct exports has increased for all
the categories since 2005 except for firms
featured by GOV1.
4.16. Interaction with MNCs is associated with more intensive knowledge transfer. When we
examine the correlation between buyer‘s characteristics and governance features for domestic sales flows
only, we observe that firms characterized by GOV3 are more likely to deal with MNCs operating in
Turkey, and parent companies or subsidiaries: on average 24 percent of their sales is made for these types
of buyers. Firms characterized by GOV2 are less likely to sell their products to government and state
owned firms. Looking at how domestic sales destination for firms interviewed in 2005 and in 2009 have
changed in the past three years, we observe an overall increase of sales made for MNC operating in
Turkey. Firms characterized by transfer of design and quality standards and R&D inputs (GOV2) sell
significantly less to Government and State owned firms company and more to MNCs operating in Turkey
compared to 2005, while firms characterized by GOV3) have increased their sales to MNCs operating in
Turkey and to parent companies or affiliated subsidiaries.
4.2 Increasing the Absorptive Capacity of SMEs
4.17. The local environment can influence the incentives and ability of firms to adopt and employ
new knowledge. The extent to which exposure to advanced technology translates into broad diffusion
across the economy and technological upgrading depends on the absorptive capacity of an economy‘s
firms. For example, the business climate may be too poor to stimulate investment, or the technological
literacy of the local labor force may be too low to successfully adapt the machinery to local conditions.
As a result, local firms may not realize the potential productivity improvements. The absorptive capacity
of a firm, in turn, is determined by two main groups of factors, the first being internal to the firm, in
particular (i) the technological literacy of the firm‘s labor force and management; and the other being
external, notably (ii) the investment climate or broader environment in which firms operate that affects
their incentives to invest and innovate. Furthermore, the local business and institutional environment
overlaps with country-wide features to determine the incentives that firms have to adopt and use
innovative modes of production and organization. The availability of a research base at the local level
can, for example, encourage innovative behavior on the part of enterprises, if contacts exist between firms
and local research organizations. Access to a skilled workforce is also highly dependent on the quality of
the local higher education and vocational training. The ease of bank finance, in turn, is conditional on the
Figure 4-6: Trade and governance modes
Source: Turkey NIS 2009
75
67 66 64
815 15 1717 19 18 19
0
20
40
60
80
GOV0 GOV1 GOV2 GOV3
%
Domestic sales Indirect exports Direct exports
111
development of the local banking sector, as well as on personal contacts that may facilitate relational
lending practices. Finally, the regulatory environment is also largely local since a large number of
operating licenses are awarded by municipalities. As a result, the effects in terms of knowledge transfer of
linkages between globally connected firms and local suppliers may vary widely depending on local
conditions.
4.18. Econometric results indicate that the ability of firms to adopt knowledge transferred in
production networks is strongly associated with features of the investment climate that affect the
firms‟ absorptive capacity. Combination of the 2009 NIS survey with the main 2008 enterprise survey
allows focusing on the effect that investment climate conditions may have on the intensity of knowledge
transfer within production networks. To gauge the relative importance of various investment climate
features, the analysis uses only objective, as opposed to perception based, measures. Various investment
climate features concur to affect a firm‘s absorptive capacity. Some are directly related to the internal
capability of firms, in terms of managerial experience, skill levels of the workforce, provision of training,
prevalence of ICT, possession of an internationally recognized quality certification, use of foreign
licensed technology and R&D expenditures. Other factors that affect a firm‘s ability to adopt and use
knowledge are related to the external environment, namely to the regulatory framework governing the
issuance of licenses and permits or inspections; the availability of bank finance or receipt of direct
government subsidies. Regression analysis also controls for firm size, region and industry and industry
conditions, implying that the results hold regardless of these characteristics.88
The summary results in
Table 4-1 indicate whether each explanatory variable has a positive, negative or insignificant association
with the likelihood of an outcome implying some degree of knowledge transfer (GOV1, GOV2 or
GOV3), relative to the baseline outcome of no knowledge transfer (GOV0).
Table 4-1: Summary of investment climate effects on knowledge flows
GOV0 GOV1 GOV2 GOV3
INNOVATION AND SKILLS
R&D (dummy) - - + +
Foreign technology (dummy) - + NS +
New product (dummy) NS NS - +
Computer - + NS +
Quality certification (dummy) NS NS NS +
Manager experience (years) NS NS NS +
Staff - skilled workers - NS + +
Training (dummy) NS - NS +
REGULATORY ENVIRONMENT
Time tax - + + -
Operating license (days) NS NS NS -
Permit (days) + - NS -
Business inspection (number) + NS NS NS
FINANCE
Loan (dummy) NS NS + NS
New fixed asset finance - bank NS + - +
Collateral NS - + +
TRADE Domestic sales - NS + NS
Inputs imported + - NS -
OTHER Subsidies (dummies) + NS NS NS Notes: See Table 4-A-1 in the Annex. NS “Not Significant”. Regressions include region, industry, size, exporter and foreign
ownership dummies.
4.19. Suppliers endowed with some technical skills and capacity “to handle the technology” and
able to diversify their product offer are more likely to enter knowledge intensive value chain
relationships, because the payoff they can expect to obtain from access to external sources of
88 See Annex for details on the econometric methodology and results.
112
knowledge will be higher.89
It is apparent from Table 4-1 that more capable suppliers are more likely to
engage in value chain relationships characterized by transfer of design, quality standards and technology
dissemination (GOV2 and GOV3). Namely, there is a positive correlation between R&D investments and
GOV2 and GOV3, and a negative correlation between R&D investments and less knowledge intensive
relationship. In the same vein, the usage of foreign technology has a positive correlation with GOV3 and
a negative correlation with GOV0. The share of workforce using computers is positively correlated to
GOV1 and GOV3 and negatively correlated to GOV0. The share of skilled workers turns out to be
significant too, with a positive correlation to GOV3. These results, along with the ones obtained for
innovation variables, seem to confirm the hypothesis, following Cohen and Levinthal (1989), that
domestic firms will need some competencies not only to handle the knowledge they are already endowed
with, but also as a means to gain access to external sources of knowledge. From this perspective, the
greater the technical competencies of local suppliers the more value chain agreements are likely to be an
effective vehicle of knowledge transfer and adoption. Finally, firms that have more diversified products
are also more likely to enter knowledge intensive value chain agreements (positive correlation between
―new products‖ and GOV3).
4.20. A cumbersome regulatory environment negatively affects the likelihood of knowledge
intensive buyer-supplier relations. The time spent by senior management dealing with regulations is
negatively associated with GOV3. An inefficient licensing environment, specifically operating licenses
and permits, does not encourage value chain relationships characterized by high transfer of knowledge.
The higher the number of days to obtain operating licenses, the less the likelihood of observing types of
governance featured by buyers‘ involvement in technology dissemination and R&D activities (negative
correlation with GOV3). In the same vein, the higher the number of days to obtain permits, the less the
likelihood of observing buyer-supplier relationship featured by transfer of knowledge. Finally, the number
of business inspections is positively correlated to GOV0.
4.21. Easier access to bank finance is connected with more intensive knowledge transfer, while
government subsidies appear directed to firms engaged in less knowledge intensive relationship. Providing further support to the importance of bank finance (Chapter 3), higher shares of investment
financed from bank are positively associated with knowledge intensive value chain arrangements
(GOV3). Evidently, access to bank finance goes hand in hand with high collateral requirements, since the
value of the collateral required is also positively associated with more intensive modes of knowledge
transfer (GOV2 and GOV3). Subsidies from the national, regional or local governments instead show a
positive correlation with GOV0.
4.22. Suppliers directly related to international markets tend to be characterized by limited
transfer of knowledge. We find in fact a positive correlation between GOV0 and the shares of inputs
imported directly, and a negative correlation between GOV0 and the share of domestic sales. Turkish
exporters and inputs importers are likely to characterize their cooperation models by arm‘s-length market
coordination or by participation in vertical relationships in which they are allowed a high level of
autonomy. Domestic sales are instead positively correlated with GOV2 (buyer‘s involvement in D&Q and
R&D). From this perspective, as highlighted above, it should be considered the presence of firms linked
to different global value chains (e.g. Turkish companies with extended networks or MNCs located in
Turkey) which provide learning opportunities to local enterprises that are in economic relation with them.
These results suggest that the local relationships among focal firms and other networked local firms (as in
the supplier-buyer network) facilitate inter-firm learning and knowledge exchanges.
4.23. Governments globally have traditionally played a role in facilitating firms‟ access to
knowledge by focusing public interventions in areas where private investment is either sub-optimal
89 Cohen and Levinthal 1989.
113
or non-existent. Firms can acquire technological capability through a number of learning mechanisms.
The most important are private learning mechanisms. These include: (i) internal efforts by firms, such as
in-house training of workers, R&D activities, and hiring of staff with knowledge of more advanced
technologies; and (ii) external learning mechanisms, including contacts with foreign buyers and suppliers
of equipment, interaction with other firms in the industry, training courses and hiring of consultants.
When firms cannot manage the technology transfer process themselves, they can turn to collective
mechanisms, if available, such as technical support services offered by government, business associations,
and NGOs. These collective learning mechanisms can be broad based (e.g., private sector training courses
and consulting services) or they can be highly-focused (e.g., direct technical support to individual firms).
4.24. In Turkey several governmental and non-governmental organizations provide support to
firms, especially SMEs, with the objective of increasing their operational capabilities and
absorptive capacity. A list of these institutions and of their programs cannot be exhaustive, since several
government agencies and other organizations have established such initiatives. Here, it is worth
mentioning the initiatives undertaken by the Union of Chambers and Commodity Exchanges of Turkey
(TOBB), and by two publicly funded programs, under the Small and Medium Enterprises Development
Organization (KOSGEB), and the National Productivity Center (MPM).
4.25. TOBB plays a role in supporting SMEs, by making information on available financial and
non-financial resources more accessible and through the network of EU business centers
(ABİGEM). TOBB has coordinated the establishment by the Prime Ministry of an information site for
SMEs90
in order to provide a foundation to more efficient promotion of publicly funded support programs
for SMEs and to facilitate access to information. Another initiative led by TOBB is the establishment of
European Union Business centers (ABİGEM) in 15 provinces around Turkey with funds provided by the
EU. These institutions provide professional training, consultancy and information services in the regions
since 2002. Their aim is to help SMEs improve their competitiveness within the national and international
markets and increase SMEs‘ contribution to both local and national economies. The total EU financial
investment in the network is exceeding 50 million Euros, making it one of the major EU projects in the
world.
4.26. KOSGEB is a government institution aimed at enhancing the efficiency and competitive
ability of SMEs, including by supporting their technological capacity and innovation potential. KOSGEB was established in 1990 in affiliation with the Ministry of Industry and Trade. It provides
services and programs to SMEs in areas linked to management and business development, as well as
technology upgrading. In April 2009, it included SMEs in the service sector in their scope of firms to
provide support. Support is mainly offered in the form of loans and grants to purchase goods and services
from external providers, but KOSGEB also organizes general training programs. With an operational
budget in the order of USD 60 million, it plays an important role in SME support. Its operational budget
is larger than most comparable non-lending support, including MPM in Turkey, MEP in the US and MAS
in the UK. However, KOSGEB operates a relatively small program when compared to the Small Business
Administration (SBA) in the US, which mostly provides loans to SMEs. While the SBA provided
financial assistance equivalent to more than 3 percent of US GDP in 2008, KOSGEB provided loans of
only 0.02 percent of Turkish GDP. Approximately 70,000 companies have registered in KOSGEB‘s SME
database, out of a total of about 250,000 in Turkey. Support is provided following the completion of
Strategic Roadmaps by candidate enterprises, and reached over 10,000 SMEs in 2008. KOSGEB has
established 22 different targeted programs delivered through 35 Enterprise Development Centers across
90www.kobi.org.tr. The information is provided under 11 headings: ―Enterprise Establishment, Transfer and Dissolution,‖
―Construction and Opening Permits,‖ ―Regulations about Taxes, Health, Labor and Social Security,‖ ―Information About
Production and Quality,‖ ―Foreign Trade,‖ ―Cooperation Between Enterprises,‖ ―Databases,‖ ―Subsidies for SMEs,‖
―Information About Financing,‖ ―Information About EU for SMEs‖ and ―R&D, Innovation and Technology Transfer.‖
114
Turkey. In cooperation with universities, KOSGEB has also established 20 Technology Development
Centers.
4.27. MPM is a smaller organization with a mission to conduct research, disseminate knowledge
and provide technical assistance on productivity to both public and private sector organizations. MPM was established by law in 1965, and operates as an autonomous organization under the
responsibility of the Ministry of Industry and Trade. It performs a broad scope of functions, involving
theoretical, empirical and practical research, as well as monitoring, consultancy and training work. As a
result, the roughly 100 experts employed at MPM – out of a total staff of around 170 – represent a wide
breadth of fields, including various engineering disciplines, economics, statistics, agriculture,
management, sociology, psychology, education, international relations and communications. Consultancy
and training services are offered through a central Ankara office and four regional offices. Productivity
consultancy services are billed at cost to the client, while group training services are offered free. Training
projects are conducted at the provincial level over specific time periods, and each project is tailored to
specific needs of the region. From 1998 to 2009, training projects have covered 50 different provinces.
Box 4-3: SME support programs: The international experience
SMEs are often unaware of the benefits of available support choices and adopting new modes of production may
appear risky. They are constrained to the information they receive from restricted sources, such as personal contacts
and a limited number of suppliers. Successful programs use two types of approaches to help SMEs understand their
needs. The first more traditional approach consists of outreach activities that convey general knowledge about
various topics to firms. The second comprises diagnostic interventions that provide firms with more specific
knowledge about their own potential areas of improvement. Outreach activities based on diffusion of general
knowledge can take multiple forms. These include free seminars and workshops with visual presentations to explain
the use of different technologies or organizational methods to potential users. Extension programs can also promote
awareness of available technologies through practical demonstrations. The diagnostic approach to outreach is
extensively used by the US MEP and UK MAS programs. The UK MAS program provides a free one-day on-site
diagnostic visit by a MAS manufacturing specialist to review a company's entire manufacturing operation, with
3,700 such visits having taken place annually during 2002-05. In 2007, 12 percent of the US MEP‘s assistance was
provided through initial technology assessments, provided at no cost to firms. The diagnostic approach can also take
the form of performance-benchmarking to compare a firm's use of a specific technology to that of a best-practice
firm.
Tailor Interventions to Individual Firm Needs
A key characteristic of the most successful technology extension programs is that they offer services targeted at
problems that are firm-specific. This is in addition to general group awareness and training activities. In the US, the
MEP offered consulting projects to just over 7,500 clients in 2008. During 2002-05, the UK MAS delivered just
over 1,000 consulting projects for clients on average, and 3,051 overall. The average company was involved in two
consulting projects, with each project on average lasting 10.7 days.
Take a Comprehensive Approach
International experience shows that programs need to target a broad range of technological and business areas to be
effective. In the US MEP program, most assistance in 2007 was provided in the areas of manufacturing systems (41
percent) and business services (23 percent), but help was also provided on quality systems (11.5 percent)
engineering services (12 percent), human resources and organizational development (9.4 percent) and IT (3.1
percent). Programs from the US and UK systems have shown that the extension centers are more effective when
they develop a close relationship with their clients. Offering multiple services at the same center, rather than discrete
services in different locations, has the advantage of creating longer periods of interaction between the SME and the
center, strengthening the ties between two, and building the mutual trust necessary for an effective intervention.
Another reason is that many of the barriers to technology absorption are complex and may require addressing
different technological and non-technological aspects within a firm. Moreover, the overlap between management
processes and technology is blurry.
115
Use Different Forms of Interventions
International experience has shown that many problems faced by SMEs are of a very general nature and can be
solved through a single phone call or visit from the manager. Technology extension centers typically find that a
gradual approach to assistance provision avoids wasting valuable resources. The traditional and most common form
of public technology extension services is the provision of technical assistance. Extension services are increasingly
being combined with a variety of other types of assistance, such as demonstration programs, workforce training,
organizational management and networking services that facilitate a more comprehensive approach to technology
diffusion.
Support a Broad Range of SMEs
Many international programs are either tailored to SMEs or offer their services at a discount to SMEs. Their clients
are predominantly in traditional, medium-technology manufacturing sectors, though this can vary widely in certain
countries. Given that these programs are designed to target SMEs, it is interesting to note the range of size of firms
served. The US MEP targets primarily SMEs but also works with a number of large firms. In 2007, 21 percent of
firms served had between 1-19 employees, 19 percent between 20-49 employees, another 19 percent had between
50-99, 23 percent firms had 100-249 employees and the remaining 18 percent had over 250 employees. The UK
MAS program targets small firms more specifically, with 25 percent of its clients during 2002-05 having 1-9
employees, 45 percent of beneficiaries between 10-49 employees, and another 25 percent 49-250, with only 5
percent of clients having more than 250 employees.
Provide Services with Unique Value Added for SMEs
Two factors heavily influence whether a technology extension programs has a unique value added for SMEs. The
first concerns the quality of the knowledge offered through the program. A reasonable proxy for this is the
knowledge and experience of the personnel providing services to the firms. Evaluations from a range of programs
globally show that the most critical factor for the success of the program is the quality of program experts, both in
terms of their technological competence, as well as the quality of their inter-personal relationship with the client.
Personnel should be rigorously screened based on their understanding of local industry needs, technical knowledge
and their willingness to follow general extension programs guidelines during their intervention and to engage the
firm in building self-knowledge. US MEP staff are generally engineers with more than 15 years of industry
experience, sometimes close to retirement age. UK MAS advisors are industry experts including engineers, process
managers, and specialists in specific areas of demand by firms, including exporting/marketing, sales and part
procurement. The second factor influencing the unique value-added of extension programs is whether similar
services are already available to SMEs in the market.
Leverage External Sources of Knowledge
International experience shows that partnerships with external sources of support in the region play an important
role in the delivery of support to enterprises. Technology extension programs make extensive use of networks of
external consultants and experts. In many cases, technology extension programs are affiliated with or managed by
regional technology institutes or universities and can draw heavily on their expertise. This is the case with the
Japanese Kohsetsushi and German Steinbeis programs.
Minimal Response-time
Both the US MEP and UK MAS programs include direct helplines to regional centers where manufacturers can ask
quick questions to an expert. In Japan‘s Kohsetsushi Centers, the standard assistance procedure is similar to the
MEPs in the United States. Many of the consulting services are performed over the phone and last no longer than 20
minutes. If the problem is more complex, the client sends staff members to the center for consultation sessions. If
the problem is still not resolved, a center agent, a registered private manufacturing consultant or a university
professor will visit the firm for an average of 6 days.
Decentralized, Flexible, Autonomous Organization
Decentralization of public extension programs can result in many benefits, such as enhancing efficiency, service
quality and outreach. Comparisons between the more centralized Japanese Kohsetsushi and the US MEP models
find that the US programs exhibit far more in the way of experimentation and innovation than seen in Japanese
efforts, for example in developing assessment tools, benchmarking measures, and telecommunication techniques. A
decentralized operation which maintains network ties maximizes adaptability to local needs and flexibility.
Furthermore, local service providers can focus on industrial sectors relevant to their regions while offering
116
4.28. Existing SME support programs could be improved following international best practices. The study of knowledge flows within Turkish production networks in this Chapter has shown that the
absorptive capacity of firms is essential for the adoption of more knowledge intensive modes of
organization and production. In addition to addressing broader investment climate constraints related to
regulation, finance or the availability of skilled labor, the absorptive capacity of Turkish SMEs would be
further enabled by upgrading and expanding the scope of existing programs. In this regard, a number of
actions inspired by international best practices (Box 4-3) would contribute to enhance the effectiveness of
the assistance received by firms, especially SMEs.
4.29. Proposal: Further decentralize the management model of existing SME support programs
in order to better serve local needs and increase the reach beyond micro firms to cater to the needs
of larger SMEs. Both KOSGEB and MPM have quite centralized organizational model. KOSGEB, with
its 25 regional offices is more able to reach out to different regions, however the programs that are
implemented are generally centrally designed leaving few margins of adaptation to local needs. On the
other hand most KOSGEB support is demand-driven, since it requires a financial commitment from the
beneficiary by being provided in the form of a loan or a matching grant. Moreover, grants for consultancy
support can be used according to pre-defined rules and at support ratios determined according to regions,
which in turn are defined in KOSGEB‘s Supports Regulation and Application Guidelines. One important
limitation is that the firm is limited to using consultants in KOSGEB‘s network for assignments of above
specialized services to all sectors by using other centers' resources. A common feature of extension programs is that
they emphasize the geographical proximity of the local delivery system to the area it is serving. This ensures an
understanding of local industrial needs, and a close and easy interaction with firms that fosters the creation of tight
relationships. Geographical proximity and autonomy also create more of a buy-in from the local community, which
feels that the services are tailored to its needs. In the US, a central MEP agency coordinates and funds a network of
independent ―MEP centers‖ (MECs) mainly state universities and nonprofit organizations. The central MEP
provides program guidelines, training, marketing and research on best practices. Overall, the MEP program is
decentralized and flexible, with individual centers able to develop strategies and services which are appropriate to
state and local conditions. MECs were established in state universities and nonprofit organizations, based on local
needs and the host organization‘s past performance, and they receive funding based on their initial ability to match
federal funding at the time of the application process. The MECs receive one-third of their budget from federal
sources (NIST), one-third from state and local organizations, and one-third from fees collected from client firms. In
order to favor a cooperative relationship with the private sector, MECs generally have governing or advisory boards
which include local public- and private-sector representatives. The decentralization, flexibility and adaptability of
the MEP has enabled centers to respond to local needs and has given the program strong public and private local
support. In Japan national guidance and coordination is provided by the Ministry of Economy, Trade and Industry
(METI) through its Small and Medium Enterprise Agency (SMEA) and by the Agency of Industrial Science and
Technology, but the centers are administered by the prefectural and municipal governments, which provide most of
the funding.
Integrate Evaluation and Assessment in the Program
International experience shows that mechanisms that monitor the impact of interventions on client firms and
periodic independent program assessments provide opportunities to experiment with new program offerings and
continuously improve performance. Both the US MEP and UK MAS programs monitor their effectiveness through
follow-up surveys of participating companies. Such surveys consist of client valuation such as the impact of the
center's intervention on sales, labor costs, material costs, inventory costs, capital investments and jobs, and client
progress such as changes in total sales, productivity and income. The MEP monitors center performance standards,
such as the number of technical projects initiated. Centers also undertake studies in which the performance of
participating manufacturers is compared to others, and case studies linking services to program outcome. In
addition, there have been many third party evaluation of the MEP. The UK Government mandates an annual
evaluation of MAS against yearly targets that relate to numbers of interventions and generation of additional gross
value added (GVA).
Source: World Bank (2010a)
117
US $10,000, and so there is the risk that these consultants may not always have the specific expertise
requested by the firm. In this case, firms may hire outside-network consultants, however in line with
additional KOSGEB requirements. A number of interventions would enhance the effectiveness of both
KOSGEB and MPM programs:
Advance the implementation of a flexible and decentralized management model. A program to
address the specific needs of thousands of clients is best implemented in a flexible and
decentralized way. With this goal in mind, KOSGEB could better leverage existing resources by
decentralizing further the network of technology extension services to universities, industry
associations and private providers, as is the case in a number of OECD countries. It could also
devise a system of monitoring, evaluation and feed-back mechanism to ensure that organizations
delivering the services at the regional level have the ability to tailor programs according to local
needs, as in the case of similar programs in the US, the UK and Japan.
Ensure that the services on offer are not already available on market terms to SMEs in order not
to crowd out private providers of such services. The risk of crowding out private service
providers is less strong in less developed areas, especially for smaller enterprises, where the
availability of managerial and technical skills may be less abundant on the market and SME
clients may not be able to afford market rates. Granting greater autonomy to local offices would
help tailor public services to local demand and supply conditions.
Create a single entry-point which could help SMEs better understand their business needs and
opportunities. This implies improving the capacity to provide a complete diagnosis of clients‘
business needs and opportunities. In the case of KOSGEB, this would require upgrading the
existing ―Business Existing Situation Determination‖ and ―Strategic Road Map‖ questionnaires
into a deeper diagnosis and benchmarking activity. It would also require making corresponding
investments in human resources. The benefit would be the supply of a service that is more
tailored to the specific needs of each company, which has been the main component of programs
worldwide that have been internationally recognized as successful. The newly established
Development Agencies could be a key local partner to coordinate various SME support programs
and better target end-users at the local level.
Expand the reach of support programs beyond micro-enterprises, to better serve the needs of
small (10-49 employees) and medium (50-249 employees) firms. While SME support programs
have been able to reach out to many Turkish firms, a large number of companies still remain
excluded. In Turkey, there are more than 20,000 small enterprises with between 10 and 49
employees, and more than 10,000 medium-sized firms with 50-249 employees, and they are
potentially under-served, since the largest government program, KOSGEB, is mostly used by
microenterprises (those with 1 to 9 employees).
References
Altenburg, T. 2006. ―Governance patterns in value chains and their development impact,‖ The European Journal of
Development Research (18), pp. 498–521
Asanuma, B. 1989. ―Manufacturer-supplier relationships in Japan and the concept of relation-specific skill,‖ Journal
of the Japanese and International Economies, 3, pp. 1-30.
Autio E., Sapienza H.J. and Parhankangas A. 2004, "Knowledge relatedness and post-spin-off growth", Journal of
Business Venturing, Vol. 19 pp.809 - 829.
118
Bathelt, H. 2003. ―Geographies of production: growth regimes in spatial perspectives on innovation, institutions and
social systems.‖ Progress in Human Geography 27 (6), pp.789–804.
Bergman, E.B and E.J. Feser. 1999. ―Industrial and Regional Clusters: Concepts and Comparative Applications. The
Web Book of Regional Science.‖ Regional Research Institute, West Virginia University
Brach J., Kappel R. 2009. "Global Value Chains, Technology Transfer and Local Firm Upgrading in Non-OECD
Countries," GIGA Working Paper Series 110, GIGA German Institute of Global and Area Studies
Ceiglie G. and Dini M. 1999. ―SME Cluster And Network Development In Developing Countries: The Experience
Of UNIDO‖, PSD Technical Working Papers Series, UNIDO
Doloreux D. and Parto S. 2005. ―Regional innovation systems: Current discourse and unresolved issues‖,
Technology in Society 27 (2005) pp.133–153
Eraydın A. and Armatlı-Köroğlu B. 2005. ―Innovation, networking and the new industrial clusters: the
characteristics of networks and local innovation capabilities in the Turkish industrial clusters‖, Entrepreneurship and
Regional Development, 17 (2005), pp.237–266
Erdil, E. And Çetin D. 2008. Innovation and relationships in an organized industrial district: Ankara Sincan
Industrial District, TEKPOL Working Paper Series, 2008/02
Erdil, E. .2009. ―Review of international production and global value chain studies: the case of Turkish regional
networks,‖ prepared for the World Bank Turkey Investment Climate Assessment
Gereffi, G. 1994. ―The organization of buyer-driven global commodity chains: how US retailers shape overseas
production networks,‖ in: G. Gereffi and M. Korzeniewicz (Eds) Commodity Chains and Global Capitalism
(London: Praeger)
Gereffi, G. 1999. 'International trade and industrial upgrading in the apparel commodity chain', Journal of
International Economics 48, pp. 37–70.
Gereffi, G. and Kaplinsky, R. 2001. ―The value of value chains‖, Special Issue, IDS Bulletin, 32
Giuliani E., Pietrobelli C., Rabellotti R. 2005. Upgrading In Global Value Chains: Lessons From Latin American
Clusters, World Development, Vol.33(4), 2005, pp. 549-73.
Göksidan, H.T. 2006. Inter-Organizational relations in an organized industrial district: Ostim case, Unpublished
Master Thesis, METU Source: http://etd.lib.metu.edu.tr/upload/2/12607044/index.pdf
Görg H. & Strobl E. 2001. "Multinational Companies, Technology Spillovers and Plant Survival: Evidence for Irish
Manufacturing," European Economy Group Working Papers 8, European Economy Group.
Hausman J., McFadden D. 1984. Econometrica, Vol. 52, No. 5, pp. 1219-1240
Hirschman A.O. 1958. The Strategy of Economic Development, Yale University, New Haven.
Humphrey, J. and Schmitz, H. 2002a ―How does insertion in global value chains affect upgrading industrial
clusters?‖ Regional Studies, 36, pp. 1017–1027
Humphrey, J. & Schmitz, H. 2002b. ―Developing Country Firms in the World Economy: Governance and
Upgrading in Global Value Chains,‖ INEF Report No. 61, University of Duisburg, Duisburg
Kaplinsky, R. 2000. ―Globalisation and unequalisation: What can be learned from value chain analysis?‖ Journal of
Development Studies, 37, pp. 117–146
119
Kozan M.K., Wasti S.N, and Kuman A. 2006. Management of buyer–supplier conflict: The case of the Turkish
automotive industry, Journal of Business Research, Volume 59, Issue 6, June 2006, pp. 662-670
Kuruüzüm O. 1998. ―Bilimsel Bilginin Ticarileştirilmesinde Yeni Ufuklar: Teknoloji Geliştirme Bölgeleri,‖
Akdeniz Üniversitesi Yayınları
Lall S. 1978. ―Transnational, Domestic Enterprises and Industrial Structure in Host LDCs. A Survey.‖ Oxford
Economic Papers , vol. 30, n.2, pp.217-248
Malecki, E., Oinas, P. (Eds.) 2000. ―Making Connections: Technological Learning and Regional Economic Change‖
Ashgate, Aldershot
Markusen, J. 1995. The boundaries of multinational enterprises and the theory of international trade. Journal of
Economic Perspectives, 9, 169–189.
Markusen J., Venables A.J. 1999. ―Foreign Direct Investment as a Catalyst for Industrial Development.‖ European
Economic Review 43 (1999), pp.335-356.
Nadvi, K. (1995), Industrial Clusters and Networks: Case Studies of SME Growth and Innovation, UNIDO, Vienna
Oz O. (2004), Clusters and Competitive Advantage: The Turkish Experience, New York: Palgrave Macmillan, 2004.
Özkanlı Ö., Erdil E., and Akdeve E. 2008. Innovation and Relationships in Industrial Districts: The Case of Turkey,
Innovative Marketing, 4, pp.114-119
Pietrobelli, C. and Rabellotti, R. 2007. ―Upgrading to Compete. Global Value Chains, Clusters and SMEs in Latin
America,‖ Cambridge, MA: Harvard University Press.
Rodriguez-Clare, Andre. 1996. ―Multinationals, Linkages, and Economic Development,‖ American Economic
Review 86, pp.852-873.
Saliola,F. and Zanfei, A. 2009. ―Multinational firms, global value chains and the organization of knowledge
transfer.‖ Research Policy 38 (2009) pp. 369–38
Smarzynska-Javorcik B. 2004. "Does Foreign Direct Investment Increase the Productivity of Domestic Firms? in
Search of Spillovers Through Backward Linkages." American Economic Review, American Economic Association,
vol. 94(3), pages 605-627, June
Sturgeon, T. 2001. How Do We Define Value Chains and Production Networks, Industrial Performance Center,
Massachusetts Institute of Technology.
Tallman, S., Jenkins, M., Henry, N. and Pinch, S. 2004. Knowledge, clusters and competitive advantage. Academy
of Management Review, 29, (2), 258-271.
Ulusoy, G. 2003. ―An assessment of supply chain and innovation management practices in the manufacturing
industries in Turkey.‖ Int J Prod Econ, 86, pp. 251–70.
Vale M., and Caldeira, J. 2008. ―Fashion And The Governance Of Knowledge In A Traditional Industry: The Case
Of The Footwear Sectoral Innovation System In The Northern Region Of Portugal,‖ 2008, Volume 17, Issue 1-2,
pp. 61-78.
Wagner, J. 2007. ―Exports and Productivity: A Survey of the Evidence from Firm-level Data,‖ The World Economy,
Blackwell Publishing, vol. 30(1), pages 60-82, 01
World Bank, 2010a. Innovation Policy for Competitiveness and Employment Generation. Forthcoming
120
Annex 4-A. Econometric Methodology and Results
Based on the insights from the GVC literature, we define a measure of value chain governance based on three
variables:91
(i) Buyer‘s provision of information on Design/Quality (i.e. product characteristics) and product quality
standards imposition; (ii) Buyer‘s engagement in the supplier process or product R&D activities; (iii) Buyer‘s
provision of employees to (or organized personnel exchanges with) suppliers as a means to disseminate and diffuse
new technologies into the local firms‘ production facilities.
GOV is regressed on the different measures of IC. The equation to be estimated will thus be of the following type:
ICvarjc is the average value of the average investment climate variable at industry j and region c. We also control for
size, region and industry fixed effect, – which means that results hold regardless of these characteristics. Additional
controls are included for the share of foreign and domestic ownership, and for the share of output produced for
export. To account for endogeneity, we construct a measure of the IC conditions faced by firm i on dimension k, by
averaging the responses of firms in the same location- sector. To ensure adequate numbers of firms in each location-
sector cell average, we drop those cells with less than 5 observations. Standard errors are also clustered at location-
industry level. We use a multinomial logit model approach to investigate the correlation between value chain
governance modes and the characteristics of firms and IC in Turkey. Since GOV0, GOV1, GOV2 and GOV3
represent qualitatively different modes of organizing knowledge transfer, which may each be influenced by different
combinations of explanatory variables, a multinomial logit model is the most suitable for this analysis. We estimated
the multinomial logit model by maximum likelihood method. As we are interested in the change in absolute
probability of the outcome governance induced by the regressors, we shall calculate the marginal effects. We define
a multi-category variable (y) as dependent variable. Let x be the vector of explanatory variables, the multinomial
logit model response probability takes the following form:
for k=0, 1, …K
where Pnk is the probability that the dependent variable (Yn) takes value k at nth observation, with k ranging from 0
to K. In our case k will take values 0, 1 or 2 to identify four different governance modes: GOV0 (k=0), GOV1 (k =
1), GOV2 (k = 2) and GOV3 (k = 3). It should be mentioned that the point estimates of a multinomial logit tell us,
for each choice k, the change in probability of the outcome k, relative to the baseline outcome (k = 0), induced by a
unit change in the explanatory variables. In a multinomial framework, this does not assure that the absolute
probability of outcome k will increase or decrease, but that k will be more or less likely relative to baseline outcome.
As we are interested in the change in absolute probability of the outcome k induced by the regressors, we shall
calculate the marginal effect:
Estimation of the multinomial logit model is based on the assumption that probabilities of the alternative choices are
independent of each other. This property is called the independence from irrelevant alternatives (IIA). The validity
of this assumption is checked using the test introduced by Hausman and McFadden (1984). The available data allow
91 The survey included questions on the establishment‘s sales made to the largest buyer, the number of years of the relationship
with the largest buyer, and the percentage of sales made according to largest buyer‘s unique specification. In additions, firms
were asked whether the largest buyer was engaged in the provision of information on Design/Quality (i.e. product characteristics)
and imposed product quality standards, whether the largest buyer provided inputs on firm‘s process or product R&D activities,
and whether the buyer provides employees to (or organized personnel exchanges with) the firms as a means to disseminate and
diffuse new technologies into the local firms‘ production facilities. Using this information, we investigate the characteristics of
buyer-supplier inter-firms linkages and the intensity with which buyers are engaged in transferring knowledge and technology to
Turkish suppliers. The analysis is restricted to those firms which claimed to have a largest buyer and to sell at least 5 percent of
their sales to the largest buyer (90 percent of firms in the sample).
)]exp(1/[)exp()(0
k
K
k
knknr xxPkyP
n
nk
kx
P
121
us to carry out only a cross-section analysis which highlights simple correlations and no causal links between
variables. Along with multinomial logit regressions, we run probit regressions for each of the three governance
components separately. The purpose of the additional empirical exercise is to complement the analysis by
investigating possible additional features in the relationships between IC variables and governance components, if
any.
Table 4-A-1: Summary of multinomial logit results
GOV0 GOV1 GOV2 GOV3
Variable Coef. P>z Coef. P>z Coef. P>z Coef. P>z
REGULATORY
ENVIRONMENT
Business inspections (number) 0.025 0.048 -0.008 0.459 -0.009 0.354 -0.007 0.489
Operating licenses (days) 0.029 0.454 0.009 0.108 0.002 0.475 -0.010 0
Permits (days) 0.009 0.002 -0.004 0.037 0.000 0.708 -0.005 0.08
Time tax -0.006 0 0.005 0 0.003 0.081 -0.002 0.055
LABOR &
SKILLS
Skilled workers -0.012 0.005 0.002 0.475 0.008 0 0.002 0.227
Training (dummy) 0.010 0.8 -0.098 0.002 -0.025 0.403 0.113 0
Manager experience -0.024 0.474 -0.010 0.724 0.006 0.716 0.028 0.088
INNOVATION
Quality certifications (dummy) -0.139 0.791 -0.156 0.647 -0.524 0.321 0.819 0.004
R&D (dummy) -0.174 0.000 -0.037 0.126 0.137 0.000 0.074 0.000
Foreign technology (dummy) -1.938 0.000 0.881 0.000 0.178 0.516 0.878 0.000
New product (dummy) -0.109 0.598 -0.125 0.466 -0.419 0.038 0.653 0.000
Computer usage (share of workforce) -0.025 0.001 0.011 0.024 -0.004 0.518 0.019 0
ACCESS TO
FINANCE
Bank financing 0.312 0.675 1.343 0 -3.990 0.001 2.334 0
Loans (dummy) 0.444 0.478 -0.335 0.576 0.472 0.062 -0.581 0.46
Collateral (value as share of loan) -0.001 0.252 -0.002 0.015 0.002 0 0.001 0.12
OTHER
Inputs imported 0.818 0 -0.256 0.047 -0.133 0.272 -0.429 0.005
Crime (dummy) 0.144 0.863 0.503 0.166 -1.142 0.315 0.496 0.039
Losses from outages 0.033 0 -0.017 0.056 0.001 0.889 -0.017 0.105
Subsidies (dummy) 2.310 0.014 -0.703 0.396 -1.316 0.191 -0.291 0.705
Domestic loss -0.088 0 0.037 0 0.038 0 0.012 0
Domestic sales, more that 90% of
total (dummy) -0.499 0 0.055 0.519 0.496 0 -0.051 0.483
Note: Regressions include region, industry, size, exporter and foreign ownership dummies
Annex 4-B. International and Domestic Trade Flows: A Regional Perspective
The Aegean and Marmara regions are the most export oriented, but both experienced a decrease in exports
between 2007 and 2008, with Central Anatolia being the only region showing an increase. In each of the five
regions included in the survey sample, firms were asked to provide the share of sales made for direct exports,
exports through a distributor (indirect exports) and domestic sales in fiscal year 2008. Overall, firms in the sample
sell 71 percent of their sales into the domestic market, 12 percent of sales are made through distributors and 17
percent of sales are exported. In addition to having the highest share of exporters (50 percent of the total number of
firms), the Aegean and the Marmara regions are the most export oriented by value, showing the highest share of
direct exports and exports made through distributors (Figure 4-B-1a). It is interesting to note that Central Anatolia
exhibits the lowest share of exports through distributors and the highest portion of national sales (85 percent). The
Black-Sea Eastern region places right after Central Anatolia with a share of 79 percent of domestic sales and 12
percent indirect exports.
122
Figure 4-B-1: International and domestic trade flows (sales)
a. Fiscal year 2008 (%) b. Fiscal year 2008 vs. 2007 (% change)
Source: Turkey NIS 2009
Figure 4-B-1 provides a regional synthesis of the redirection of exports occurred as a consequence of the decline in
global demand discussed in Chapter 1. Between 2007 and 2008 firms in Marmara experienced a decrease in their
direct exports and an increase in domestic sales. Firms in Central Anatolia are the only ones which experienced an
increase in their direct exports. Interestingly, Aegean, firms experienced an increase in their exports through
distributors, while firms in the South region did not experience any significant change in the composition of their
sales flows.
The South is the region that is most
dependent on foreign inputs. In terms of
number of firms importing inputs, firms in
the South region are the most exposed to
trade (55 percent of firms are importers),
followed by Aegean and Marmara. Firms in
Central Anatolia and Black Sea Eastern
confirm to be the least exposed to trade with
external markets. Paralleling the analysis of
exports, firms were also asked to provide the
value share of inputs imported directly and
purchased in the domestic market. Businesses
in the sample import 16 percent of their
inputs directly and purchase 84 percent in the
domestic market (Figure 4-B-2). Central
Anatolia and the Black-Sea region exhibit the
highest share of national inputs purchases.
The South region seems to be the most
dependent on foreign inputs with 22 percent of inputs imported directly, followed by Marmara (18 percent).
Central Anatolia, Black-Sea Eastern and Marmara are the most closed in terms of inter-regional sales flows,
selling about three-quarters of their output within regional borders. An interesting finding is that Marmara
represents the second most important domestic sales destination market for each of the other regions in the sample.
In addition, Aegean and South regions are the most connected areas to the Marmara market, selling on average 26
and 27 percent of their total domestic sales to the Marmara region, respectively. It is also worth mentioning that
regions with the least diversification in inter-regional sales flows – Central Anatolia and Black Sea Eastern – are
also the least exposed to foreign trade. On the other hand, the Aegean and South regions show significant shares of
inter-regional trade flows (about 50 percent). The Aegean region also represents the most open area with respect to
export flows.
71 6657
8574 79
12 15 17
313 1217 20
26
12 13 9
0
20
40
60
80
100
National Indirect exports Direct exports
6.8
-2.4 -3.4
0.3
19.6
-0.5
7.8
-1.8
0.1
-15.0
-6.3 -5.4
5.3
-0.4
-4.7
-20
-15
-10
-5
0
5
10
15
20
25
National Indirect exports Direct exports
Figure 4-B-2: International and domestic trade flows (input
purchases)
Source: Turkey NIS 2009
84 82 8490
78
91
16 18 1610
22
9
0
20
40
60
80
100
Total sample
Marmara Aegean Central Anatolia
South Black Sea-Eastern
Purchased in the domestic market Imported inputs
123
Figure 4-B-3 Intra and inter-regional trade flows: sales and input purchases
Source: Turkey NIS 2009
Marmara represents the main site for sales and input purchases for a majority of industries. Marmara stands
out for the high portion of domestic sales in the Textile sector directed to the region (57 percent). The South region
represents a very important site along with Marmara for the non-metallic products industry receiving about as much
domestic sales as the Marmara region (31 percent Marmara and 29 percent South). In the Chemicals industry, the
Central Anatolia region represents the most important destination of sales (50 percent Central Anatolia and 24
percent Marmara). Central Anatolia represents also the second most important region, after Marmara, for Food (28
percent), Garments (27 percent) and other manufacturing (20 percent) sales. Regarding input purchases, especially
in the Garments industry, the percentage of inputs purchased in Marmara is substantial (81 percent on average).
Central Anatolia represent the second most important site, after Marmara, in the Food, Chemicals and Other
124
manufacturing industries, while the South region represents the second most important site, after Marmara, in the
Textile and Non-metallic industries.
Marmara is the core location for domestic input purchases while Black–Sea Eastern displays the highest
share of regional diversification in terms of inputs purchases. In terms of material input purchases, in the
Aegean, Central Anatolia and South regions at least 50 percent of inputs are purchased within regional borders. At
the same time Black–Sea Eastern appears as the region that is most connected to other regions, as its enterprises
purchase only 37 percent of inputs internally, acquiring a significant share from Marmara (35 percent) and from
Central Anatolia and South (23 percent in total). It is also interesting to note that Black–Sea Eastern is the region
with the highest share of non imported inputs and with highest regional diversification in terms of inputs purchases.
The region of Marmara, on the other hand, appears to be the most self-sufficient in terms of inter-regional inputs
purchases supporting 87 percent of its 82 percent of non-imported inputs. The Marmara region is the also core
location for domestic input purchases for all the regions included in the sample. On average, in fact, each region
purchases about 30 percent or more of its inputs from Marmara
125
Chapter 5. ENHANCING REGULATORY CAPACITY FOR BETTER
REGULATION
5.1. Productivity, employment and the ability to export as well as to receive FDI in the Turkish
business sector are all negatively affected by long and complex regulatory procedures with large
regional differences. The econometric analysis of the Turkey Enterprise Survey (ES) confirms that the
measured regulatory environment variables have a significantly negative impact on productivity, as well
as other performance measures, such as employment and export propensity. The business sector overall
proves to function on a higher productivity level with well-functioning regulatory mechanisms in place.
As presented in Chapter 2, examples of variables with a negative impact on productivity include more
frequent inspections, a higher number of compulsory certificates, longer time dealing with these, as well
as long customs clearance procedures. Additionally, the econometric results indicate that smaller firms
struggle more obtaining necessary certificates, whereas businesses‘ export propensity is negatively
associated with business inspections and payments for compulsory certificates. There are also variations
across the five Turkish regions (Aegean, Black Sea-Eastern, Central Anatolia, Marmara and the South)
when looking at such concerns as obtaining operating and import licenses, construction permits, clearing
goods through customs, getting access to land and dealing with inspections. This could be seen as a sign
of inconsistency and lack of coordination across Turkish regions, a subject that will be discussed in detail
later in this chapter.
5.2. Turkey still lags behind developed OECD economies in offering efficient license and permit
procedures. A measurement of regulatory systems in OECD countries is available through the Product
Market Regulation (PMR) indicators (see Box 2-5 for description of PMR). When examining regulatory
and administrative opacity, there seem to have been significant improvements in Turkey in regards to
communication and simplification of rules and procedures in the period 1998-2008. Although this has
contributed to an overall improvement in lowering barriers to entrepreneurship for Turkish enterprises,
dealing with licenses and permits remains cumbersome, with a continuous 6.0 index point ranking.
Furthermore, Turkey is still lagging behind higher income OECD economies, such as France, Germany
and the UK, as can be seen in Table 5-1.
5.3. This quantitative background prepares for a discussion on the actual policies in place in
Turkey and the country‟s capacities to organize and implement high quality regulation. In recent
years, Turkey has made progress in improving its business environment, tackling several areas identified
as constraints for doing business, such as taxation procedures and customs regulations.92
Through
Turkey‘s e-Government program, an online tax filing system has been introduced, significantly cutting
the time to comply with tax regulations. Furthermore, an automated customs clearance system has been
implemented, allowing for most customs procedures to be carried out electronically.
5.4. One area that has been highlighted as particularly cumbersome is the amount of time that
businesses spend in complying with paperwork and business regulations. According to responses in
the ES conducted for Turkey in 2008, the share of senior management‘s time spent dealing with
requirements imposed by government regulations averaged 27 percent, a steep increase from the 2005
Survey, where the average was 9 percent. Moreover, Turkey ranks much higher than any other
comparative emerging economy interviewed for the ES, where the average for the ECA region is 15
percent. Accepting a degree of subjectivity in survey responses, this concern has led to investigating the
causes for this perception in terms of the institutional framework underlying the formation and
implementation of regulation affecting businesses.
92 The issues of tax and customs procedures are discussed among others in the World Bank Country Economic Memorandum –
Sustaining High Growth: Selected Issues, 2008
126
Table 5-1: Product Market Regulation - Barriers to Entrepreneurship 2008: Regulatory and administrative
opacity
France Germany Hungary Italy Portugal Spain Turkey UK
Licenses and permits system
Is the ―silence is consent‖ rule (i.e. that licenses are issued automatically if the licensing office has not acted by the end of
the statutory response period) used at all?
no no yes yes yes yes no yes
Are there single contact points (―one-stop shops‖) for getting
information on notifications and licenses?
yes yes yes yes yes yes no yes
Are there single contact points (―one-stop shops‖) for issuing or
accepting notifications and licenses?
yes no yes yes yes yes no no
Country scores (0-6) 2.00 4.00 0.00 0.00 0.00 0.00 6.00 2.00
Communication and simplification of rules and procedures
Communication
Are there systematic procedures for making regulations known
and accessible to affected parties?
yes yes yes yes yes yes yes yes
Is there a general policy requiring "plain language drafting of regulation?
yes yes yes yes yes yes yes yes
Do affected parties have the right to appeal against adverse
enforcement decisions in individual cases?
all
cases
all cases some
cases
all
cases
all cases all
cases
all
cases
all
cases
Regulations published or otherwise communicated to the public in a manner accessible at the international level: There are inquiry
points for information on the operation and enforcement of
regulations
yes yes yes n.a. yes yes yes yes
Does government policy impose specific requirements in relation
to transparency and freedom of information government wide?
yes yes yes yes yes yes yes yes
Simplification of rules and procedures
Does the national government (all ministries and agencies) keep a
complete count of the number of permits and licenses required?
yes no no yes yes yes no no
Is there an explicit program to reduce the administrative burdens
imposed by government on enterprises and/or citizens?
yes yes yes yes yes yes yes yes
Is there a program underway to review and reduce the number of
licenses and permits required by the national government?
yes yes yes yes yes yes yes no
Country scores (0-6) 0.00 0.09 1.11 0.00 0.00 0.00 0.50 0.23
Source: OECD. Note: The indicators depict strictness of regulatory environment on a scale 0 to 6 (0=Yes; 3=In some cases; 6=No),
with higher numbers indicating more restrictive policies towards competition. For a detailed description of methodology, see Wölfl
et al. (2009)
5.5. Regulatory quality is generally recognized not only to be a key component of good
governance, but also clearly correlated to measures of growth and productivity.93
For example,
93 Several authors and institutions have explored these links. According to Kaufman et al (2002), the quality of regulation and
governance is correlated to better economic outcomes. OECD has found that ―reforms that reduce competition-restraining
regulations, cut tariff barriers and ease restrictions on foreign direct investment to ―best practice‖ levels in the OECD area could
lead to gains to GDP per capita of up to 2 to 3 per cent in the European Union, where productivity is already higher than in most
developing countries‖ OECD (2005b). The 2005 Doing Business report found that indicators of cost to business in the areas
covered by the report seem correlated with higher rates of job creation. However, rather than showing that cost-cutting in these
areas can produce macroeconomic effects, the correlation might be due to the fact that countries regulate consistently, using a
national style that is reflected in the Doing Business indicators. This would support the OECD‘s 1999 conclusions about the pro-
competitive style of regulation in the United States. A group of researchers finds a strong correlation between regulation of entry
127
administrative, compliance and economic costs generated by government regulations are large. In
developed countries, calculations of administrative costs to business of complying with regulatory
obligations suggest that a 25 percent reduction in the Netherlands will reduce administrative burdens from
3.7 percent to 2.8 percent of GDP. This is still significantly above the UK (1.5 percent), Denmark (1.9
percent) as well as several other European countries.94
In Australia, regulatory compliance costs on
business are estimated to be 4 per cent of GDP.
5.6. Turkey has made important steps in establishing a system and key institutions for
regulatory reform. Since the 2001 OECD Report on Regulatory Reform,95
the Government has paid
particular attention to establishing institutions and mechanisms for regulatory reform; enacting legal
reforms conducive to simplification of the legal framework; and introducing, through pilot projects, a
number of regulatory tools to improve the quality of regulations. In this process, achieving EU
harmonization has been a key driver of reform and Turkey has partially embraced the EU Better
Regulation agenda in a number of areas. As a result, Turkey has established solid pillars of a regulatory
system that have the potential to develop into a ―whole-of-government‖ approach to regulatory
management and reform.
5.7. After an economic crisis in 2000, Turkey embarked in a deep and comprehensive economic
reform program comprising institutional changes in the way regulations were prepared and
revision of the existing regulatory framework.96
Among the different challenges Turkey still faces to
increase productivity and employment to EU levels97
, regulatory reform and in particular reducing red
tape, appears to be fundamental. A number of formal mechanisms have been established to promote a
better investment climate in Turkey. In order to rationalize bureaucratic procedures and reduce red tape, a
comprehensive reform program was launched in 2001 by a Council of Ministers Leading Decision. The
reform program has been implemented by two platforms which provide a channel to reflect views and
priorities of the international and national private sector representatives:
5.8. The Coordination Council for the Improvement of the Investment Environment (YOİKK)98
has become a key structure where the private sector makes contributions in the process of
improving the investment climate. The Council conducts its agenda with the help of 12 Technical
Committees working on specific issues with participation of both public and private institutions.99
into markets and the regulation of labor. As they explain this: ―countries have regulatory styles that are pervasive across activities
and shaped by the origin of their laws.‖ Juan Botero et al (2004). 94 Kox (2005) 95 OECD (2001) 96 In many countries, crises open up the possibility to be innovative and introduce changes in the system. Korea, for instance,
designed a comprehensive strategy for regulatory management and reform after the Asian crisis in 1997. At the core of the
reforms introduced in Korea, regulatory reform was fundamental to overcome the financial crisis. The 1998 regulatory reform
program included two key initiatives. The first was a massive deregulation initiative in which the president ordered each
government ministry to eliminate 50 percent of its regulations. The second was an enduring institutional reform that established
institutions and mechanisms at the center of government to promote reform and monitor and guarantee the quality of regulations
and the regulatory process. World Bank (2008), p. 1 97 World Bank Group (2007) 98 The following institutions are part of YOİKK: Undersecretary of the Ministry of Finance, Undersecretary of the Ministry of
Industry and Trade, Undersecretary of the State Planning Organization, Undersecretary of the Treasury, Undersecretary of
Foreign Trade, Chairmen of the Technical Committees, Chairman of The Union of Chambers and Commodity Exchanges of
Turkey (TOBB), Chairman of The Turkish Industrialists‘ and Businessmen‘s Association (TÜSİAD), Chairman of the
International Investors Association (YASED), and Chairman of the Turkish Exporters‘ Association (TİM). The secretariat works
of YOİKK Platform are conducted by the Undersecretariat of Treasury. 99 The technical committees are: Technical Committee on Company Establishment, Technical Committee on Employment,
Technical Committee on Sectoral Licenses, Technical Committee on Investment Location, Technical Committee on Taxes and
Incentives, Technical Committee on Foreign Trade and Customs, Technical Committee on Intellectual Property Rights,
Technical Committee on Investment Promotion, Technical Committee on Foreign Direct Investment Legislation, Technical
128
YOİKK aims to rationalize the regulations on investments in Turkey, develop policies by determining the
necessary arrangements that will enhance the competitiveness of the investment environment, generate
solutions to the administrative barriers encountered by the domestic and international investors in all
phases of the investment process including the operating period. The Investment Advisory Council of
Turkey (IAC) is an international platform established to receive the recommendations of executives of
high ranking multinational companies and international institutions regarding the Turkish business
environment. The recommendations stated in annual meetings by the Council members regarding
business environment in Turkey are taken to the agenda of YOİKK Technical Committees and to the
following Investment Advisory Council meeting. YOİKK Technical Committees have conducted studies
to identify solutions to the different constraints that businesses face in Turkey. The YOİKK Technical
Committee Action Plans, initially prepared in 2007, were reviewed on the 14th YOİKK Meeting on
March 2010 to tackle the bottlenecks faced by the investors. As a result, the YOİKK Technical
Committee Action Plans for 2010 have been prepared, comprising the activities planned to be realized in
2010 by YOİKK Technical Committees to streamline the investment environment.
5.9. Another recent development in the interface between the government and the business
sector is the set up of development agencies at provincial level. As discussed in Chapter 2, these
agencies could act as an information point and to rationalize initiatives, especially for SMEs, at the local
level, in close cooperation with TOBB and other government agencies. These agencies could act as ―one-
window‖ shop, facilitating the acquiring of licenses and permits between issuing agencies and firms.
5.10. A fundamental driver for regulatory reform in Turkey is commitment to EU
harmonization. The National Program for the Adoption of the EU Acquis (2007-2013) aims at achieving
harmonization with the acquis communautaire with the perspective of full membership of the EU. The
Program includes all chapters of the acquis envisaged to be adopted following a screening process.
Legislative measures, secondary legislation, and the main strategy and policy papers required in the
relevant chapters and financial requirements and resources needed are included in this Program.
Furthermore, the Program has identified the responsible institution and the timeframe for the adoption of
legislation. Eleven chapters are open for EU accession. Legal reforms continue in accordance with the
National Program. The quality of new or amended regulations has also benefited from the EU accession
process: any law or regulation amended for harmonization purposes is sent to the Secretariat General for
EU Affairs (ABGS) in accordance with the Decree Law No. 2008/14481 and Regulation No 2005/9986
(Article 6 (f)), before being sent to Prime Ministry for review of compliance with the acquis
communautaire. ABGS has also permanent contact points at different political levels in all Ministries and
institutions to ensure that EU legislation is considered when preparing regulations. ABGS also
participates in commissions when draft laws are discussed in the National Assembly. Despite these
efforts, the EU Better Regulation agenda has not been fully adopted in Turkey.
5.1 Regulatory Policies and Institutional Drivers in Turkey
5.11. Pursuing private sector development and reforming regulation are two interconnected
policy areas that should have clear objectives. Some principles and objectives of reform, as well as the
responsibilities of the groups involved in reform should be identified by governments. The most effective
way to do this is to establish an explicit regulatory reform policy, based on internationally accepted
principles of good regulation (Annex 5-1). Despite the achievements mentioned above, a number of
institutional challenges remain to improve the quality of regulation and regulatory processes in Turkey.
In particular, regulatory reform is not yet an integral policy area linked to the improvement of the
business environment, as it is the case of many other OECD countries (Box 5-1). Even if efforts have
Committee on Small and Medium Size Enterprises, Technical Committee on Corporate Governance and Technical Committee on
Research and Development.
129
been undertaken to link both issues, and YOİKK seems to be an appropriate platform for this purpose, the
lack of a clear strategy for regulatory reform to promote a friendlier business environment in the country
is evident.
Box 5-1: Regulatory reform to improve the business environment: International experiences
Many OECD countries have established wide and comprehensive regulatory reform programs with a strong focus on
establishing a more friendly environment for businesses:
In Australia, governments have undertaken major policy reforms over the past two decades, which have contributed
to the country‘s strong economic performance. These reforms have included regulatory changes to expose the
economy to greater competitive pressures and to provide firms with greater flexibility to respond. In the same
period, however, Australia has experienced a dramatic rise in the volume and reach of regulation, in response to a
variety of social, environmental and economic issues. Since 1990 the Australian Parliament has passed more pages
of legislation than in the nine preceding decades. In this context, Australian governments have acknowledged that
the country could not function well without regulation. But there is too much regulation and, in many cases, it
imposes excessive and unnecessary costs on businesses. The Regulation Taskforce established in 2006 identified a
forward agenda comprising some 100 reforms to existing regulation that could provide relief to businesses. It also
considered how the processes and institutions responsible for regulation could be improved. As a result of this
research, the Australian Government, the Council of Australian Governments (COAG) and most Australian states
and territories have integrated approaches recommended by the Taskforce into their best practice regulation
processes.
The Netherlands has focused its regulatory reform program in reducing the regulatory burden for businesses and
society. Between 1994 and 2003, a reform approach was incrementally developed, consisting of the following
characteristics: a quantitative method, quantified and individualized targets, external pressure from a watchdog
(Actal), internal organization coordination (the IPAL-unit) and monitoring using the budget cycle. The program was
reviewed in 2007 and the current Dutch Cabinet has committed to simplify the regulatory burden by another 25
percent in 2011, for example by reviewing contradictory rules, cancelling permits and licenses, reducing
administrative and supervisory burdens and improving services. Some of the innovations included in the current
program are: the program was taken beyond administrative burdens, involving the measurement, quantitative
targeting and reduction of substantive compliance costs and increasing the quality of services delivered to businesses
with regards to regulation affecting them. At institutional level, the coordination was facilitated by merging the
project groups in a new unit dealing with all aspects of regulatory reform for businesses. The tasks and
responsibilities of Actal, the watchdog, were also broadened, including substantive compliance costs and an audit
role of regulatory impact analysis.
Source: Government of Australia (2006) and Nijland (2008)
5.12. In Turkey, there is an important disconnection between the efforts undertaken to improve
the legal aspects of laws and regulations and the purpose of regulatory reform, which is to create a
legal framework that is conducive to sound economic activity and social welfare. The gains made in
improving the legal aspects of regulation have not been embedded in the way regulations impact the
economy. The Turkish legal system is composed of a number of regulatory instruments stemming from
various sources of law (Annex 5-2). In the preparation of laws and regulations, the following phases can
be identified (Figure 5-1).
130
Figure 5-1: Phases of the preparation of laws in Turkey
5.13. Following the OECD Principles for Regulatory Quality and Performance, a number of
important challenges remain in order for regulatory policy to find its place as a key priority for
government action. One initial constraint is to find high political support for regulatory reform as a key
priority area of government action. One of the fundamental elements for success in any regulatory reform
is the strong commitment and political support from the highest political level. This has been particular
important in regulatory reform programs established in emerging countries such as Mexico or Korea,
where the involvement of the President made possible the introduction of a regulatory policy. In Turkey,
regulatory reform has permeated certain senior management levels in different ministries, a number of
technical experts and various stakeholders that are aware of the importance to embark in broad regulatory
reform. However, one of the remaining challenges is to ensure ownership of regulatory reform at high
political level in the country.
5.14. Regulatory policy should serve clearly identified policy goals, and be effective in achieving
those goals. In many OECD countries, such as the UK, regulatory policy is clearly defined as a way ―to
eliminate obsolete and inefficient regulation, create user-friendly new guidelines and tackle
inconsistencies in the regulatory system.‖100
In Turkey regulatory policy remains a fragmented area in
which different institutions establish their own policy goals and agenda.
5.15. Regulatory policy should also include the establishment of frameworks for implementation.
This means that governments should ensure that institutional frameworks and resources are adequate, and
that systems are in place to manage regulatory resources effectively and to discharge enforcement
responsibilities. In Canada, for instance, The Regulatory Affairs Sector, transferred in July 2006 from
Privy Council Office, supports the Treasury Board Committee in its role as the "Queen's privy council for
Canada" by providing advice to the Governor General and by providing management and oversight of the
government's regulatory function. In addition, it provides policy leadership on the federal regulatory
policy contained in the Cabinet Directive on Streamlining Regulation. In Turkey the institutional
framework for regulatory policy remains weak and strong leadership is needed. Turkey has the
opportunity to build over the existing pillars to link the private sector and investment climate initiatives
with a more comprehensive approach to regulatory policy.
100 The Better Regulation Executive in charge of the regulatory reform program in the UK has identified the following actions in
order to implement regulatory policy: i) using targeted measures to simplify and improve existing regulation; ii) communicating
more clearly with businesses, to help them understand what they must do to comply; iii) carefully assessing the impact of any
new regulations and iv) working with the EU to improve European guidelines.
http://www.berr.gov.uk/whatwedo/bre/policy/page44059.html
Preparation by Ministries
Request of opinion from other
institutions and stakeholders
Ratification of line minister
Submission for signature to Cabinet
Submission to National Assembly
Submission for ratification to the
President
Publication in Official Gazette/Notification
131
5.16. Regulatory reform needs appropriate institutions at the centre of government to lead this
process. Many emerging countries are setting up regulatory reform units that establish mechanisms with
procedures that spell out explicit responsibilities for managing and tracking regulatory reform inside the
administration. This is done with the objective to keep reform on track and on schedule, and to ensure that
regulatory quality standards improve. Many developed countries have strong oversight bodies for
regulatory management and reform that are responsible for a ―whole-of-government‖ approach to this
policy area.
5.17. Different regulatory institutions exist in Turkey, having different roles and responsibilities,
with mandates covering different aspects of regulatory reform. There is, however, no single oversight
body for regulatory reform that concentrates all responsibilities to oversee the quality of the regulation, as
it is the case in other OECD countries (Box 5-2).
Box 5-2: International examples of oversight bodies for regulatory reform
Many OECD countries and emerging economies have established oversight bodies for regulatory reform with the
aim of institutionalizing regulatory reform, by allocating clear responsibilities to a single unit in charge of regulatory
policy and its implementation:
In the UK, the Better Regulation Executive (BRE) is the institution responsible for regulatory reform issues across
government. Located in the Department for Business, Enterprise and Regulatory Reform (BERR), it works with
government departments and regulators to scrutinize new policy proposals, to achieve effective new regulations, to
make it easier to change or remove regulation where beneficial, to reduce existing regulatory burdens affecting
business, the third sector and frontline staff in the public sector, to improve transparency and accountability for
regulation, to effectively communicate regulatory changes and to drive forward the better regulation agenda in
Europe.
In the United States, the Office of Information and Regulatory Affairs (OIRA) is the oversight body for
regulatory reform. As a federal office that Congress established in the 1980 Paperwork Reduction Act, it is part of
the Office of Management and Budget, which is an agency within the Executive Office of the President. In addition
to reviewing collection of information, OIRA reviews draft regulations and develops and oversees the
implementation of government-wide policies in the areas of information technology, information policy, privacy,
and statistical policy. OIRA reviews agency draft regulations before publication to ensure agency compliance. The
review includes consideration of alternatives to the rulemaking and analysis of the rule‘s effects on society, both its
benefits and costs. OIRA has more than 50 full-time professionals who work with agency professionals on specific
issues and decisions. The majority of OIRA employees are career public servants.
In Italy, the Simplification Unit (Unità per la Semplificazione) has been moved from the Legal Department of the
Prime Minister‘s Office to the newly created (May 2008) Ministry of Simplification. The Unit is in charge of cutting
red tape, simplification measures, very basic quantitative assessments linked to simplification and the
implementation of the taglia-legge (guillotine law) at federal level. The Better Regulation Unit is composed of 30
experts nominated for three renewable years, with proved expertise on legal, economic and organizational issues.
In Korea a Regulatory Reform Committee (RRC) was set up by law in 1997 with a ―general mandate to develop
and co-ordinate regulatory policy and to review and approve regulations.‖ Its main functions are to give some
strategic perspective in the regulatory reforms, to undertake research, to monitor improvement efforts of each
agency and to make sure there is coherence between their actions. The RRC is composed of 25 members, 18 of
whom are from the private sector and 7 are government officials from various departments. The RRC is jointly
chaired by the Prime Minister and one member from the private sector appointed by the President. It is one of the
cases where more power has been given to this kind of institution, multiplying the „engine of reform‟ effect. The
secretariat function supporting the RRC is undertaken by the Regulatory Reform Office which is located in the
Prime Minister‘s Office. This unit includes around 40 civil servants and 3 professional experts, under the direction
of the Deputy Minister for Regulatory Reform.
132
5.18. Four main institutions are active within the Prime Ministry‟s office in charge of improving
the quality of new and existing laws and regulations. These institutions have made progress in
introducing principles of high quality regulation and establishing clearer criteria to prepare and review
laws and regulations. These are:
General Directorate for Laws and Decrees. This Directorate is mainly responsible for reviewing
the quality of new laws and decrees in Turkey. Their review concentrates on the legality of the
drafts and on the consistency of regulation with the Government‘s main development plans. The
Directorate also ensures that consultation among government agencies has been conducted during
preparation of new laws and decrees.
General Directorate for Legislative Development and Publication. This Directorate is in charge
of reviewing quality of by-laws proposed by ministries and public agencies; developing and
managing the Legal Information System (electronic inventory of stock of all primary and
secondary regulations) and keeping it up-to-date by collecting, consolidating and identifying
existing regulations; introducing simplification measures to existing regulation; and publishing
the Official Gazette.
Regulatory Reform Group. The Group is mainly concerned with the introduction of Regulatory
Impact Assessment (RIA) in Turkey. It is currently in charge of the implementation of a training
program of government officials, financed by the EU. It is a small group of experts with primary
responsibilities in other departments. The duties and responsibilities of the Group have been
defined by a memo which was signed by the Undersecretary, but in the current structure the
Group is not attached to other formal institutions.
Quality Legislation and RIA Group. This most recent regulatory oversight body was established
under the Prime Ministry in March 2010. The Group is responsible for preparing draft legislations
in line with recognized national and international standards as well as evaluating and guiding
ministries in RIA processes. The institution will be chaired by Director General of General
Directorate of Laws and Decrees, with two vice presidents from the Directorate of Laws and
Decrees as well as the Diretory of Legslative Development and Publication.
5.19. Institutional fragmentation remains an important challenge to establish a single,
comprehensive and coordinated regulatory policy. Compared to other OECD countries, the current
structure puts Turkey in a weak position of institutional capacities for managing regulatory reform. The
OECD Principles of Regulatory Quality and Performance suggest that countries should ―create effective
and credible coordination mechanisms, foster coherence across major policy objectives, clarify
responsibilities for assuring regulatory quality, and ensure capacity to respond to a changing, fast-paced
environment‖. It is good practice for governments to support reform at all levels. This difficult task is
increasingly important as regulatory responsibilities are shared among many levels of government,
including supranational, international, national, and sub-national levels. High quality regulation at one
level can be undermined by poor regulatory policies and practices at other levels, while, conversely,
coordination can vastly expand the benefits of reform. The policies and mechanisms for coordination
across, as well as between, levels of administration are thus becoming increasingly important for the
development and maintenance of an effective regulatory framework.
5.20. Coordination mechanisms across the administration are fundamental to have a
comprehensive regulatory reform strategy that applies to the whole administration and intends to
achieve common objectives. In many OECD countries, coordination among ministries is ensured
through clear mandates and allocation of responsibilities at high political level. In countries with more
consensual traditions, coordination is ensured through soft mechanisms in the system. Several institutions
133
deal with regulatory reform in Turkey and they do not always act in a coordinated manner. Some of the
programs aimed at improving the regulatory system overlap, such as some areas of e-government and
revision of procedures for economic licenses. The different responsibilities allocated to institutions are not
always linked towards a single regulatory reform strategy. This creates a challenge when it comes to
establishing priorities and taking the lead for reform, and it often also leads to overlapping responsibilities
that make implementation cumbersome, thus directly affecting business operation.
Figure 5-2: Institutional capacity for managing regulatory reform
Note: The sample includes 31 jurisdictions for 2005 and 2008. Data for 1998 are not available for the European Union,
Luxembourg, Poland and the Slovak Republic. Results have been adjusted for 26 countries following the results of a peer review
process. Results for Spain are pending confirmation. Results for Belgium, Germany, Ireland, Luxembourg and Switzerland
represent the initial answers to the surveys and will be adjusted within the next weeks.
Source: OECD Regulatory Management Systems‟ Indicators Survey 1998, 2005 and 2008. www.oecd.org/regreform/indicators,
in OECD (2009)
5.21. Turkey has gone through a decentralization process in which municipalities play a more
active role in implementing regulation. The Ministry of Interior plays a role in dealing with these
issues, but there is no official forum where different levels of government can discuss implementation
issues. As a result, an important number of conflicts arise, in particular dealing with construction permits,
settlement and land issues.
5.22. Conflict resolution mechanisms are costly and time consuming. Businesses can appeal a
municipal decision to Court, but also file a formal complaint to prosecutors, governors or the Ministry of
Interior. This Ministry is the main body responsible for mediating and solving conflicts that arise from
these relationships. This substantially adds to the regulatory requirements – in terms of both time and cost
– faced by businesses.
0
2
4
6
8
10
12
14
16
18
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
20
05
20
08
LUX NOR TUR FIN NZL GRE HUN POL AUT CZE DNK FRA PRT DEU ESP SVK CAN CHE JPN AUS ICE IRL ITA SWE USA BEL KOR EU MEX NLD UK
5th-95th percentile Index 2005 Index 2008
Is there a dedicated body (or bodies) responsible for promoting the regulatory policy and monitoring and reporting on
regulatory reform and regulatory quality in the national administration from a whole of government perspective?
Is this body consulted as part of the process of developing new regulation?Does this body report on progress made on reform by individual ministries?
Is this body entrusted with the authority of reviewing and monitoring regulatory impacts conducted in individual ministries?
Can this body conduct its own analysis of regulatory impacts?Is this body entrusted with an advocacy function to promote regulatory quality and reform?
Is there an advisory body that receives references from Government to review broad areas of regulation, collecting the
views of private stakeholders? (e.g. past bodies have included the Better Regulation Task Force in the UK, the External Advisory Council on Smart Regulation in Canada and the Regulatory Reform Council in Korea)
If the answer is “yes”: b(i) Does this body have a degree of independence from government (e.g. through a board or
commission structure)? If the answer is “yes”: b(ii) Does this body report its findings publicly?
Is a specific minister accountable for promoting government-wide progress on regulatory reform?If the answer is “yes”: c(ii) Is the Minister required to report to Parliament on progress?
See Question 14 / 2008 OECD Regulatory Indicators Questionnaire
Weights:
if yes, weight=3
if yes, weight=2
if yes, weight=2if yes, weight=2
if yes, weight=1
if yes, weight=1if yes, weight=1
if yes, weight=1
if yes, weight=1
if yes, weight=2
if yes, weight=2
134
5.2 Administrative capacities to prepare new regulations
5.23. Current processes for making legislation and subordinate regulations should support
applications of core principles of good regulation. Good practice shows that systematic capacities to
generate high quality regulation are essential and they should ensure that both processes and decisions are
transparent to the public.
5.24. Clear guidelines are essential to establish systematic procedures to prepare and amend
regulations. A skilled and well-trained civil service recruited on the basis of merit is also a prerequisite
for developing and maintaining high-quality regulations and regulatory policies. Law-making in Turkey
follows many high quality standards. The by-law on the ―Preparation Method and Principles of
Legislation‖ (2005/9986) has established procedures and principles regarding drafting of laws, decree
laws, regulations, by-laws, annexed decisions of Cabinet decrees and other regulatory proceedings to be
prepared by the Prime Ministry, ministries and their affiliated, related and associated public institutions
Box 5-3: Coordination between levels of government
Co-ordination between different levels of government is a political priority for many OECD countries. The principle
of subsidiarity reflects a real concern for clarity and calls for finding more appropriate co-ordination mechanisms
that can help to avoid overlapping and duplication. The devolution process also speeds the need for co-ordination.
Most OECD countries dealing with a multi-level dimension have set up co-operation and co-ordination mechanisms
and permanent institutional bodies to streamline the relationship between levels of government. Those mechanisms
are either formal or informal, depending on the political and legal tradition and tend to have a more permanent
structure, rather than an ad hoc basis. For instance, in Norway, several mechanisms are in place to ensure co-
ordination of regulatory proposals affecting local governments. First, regular formal meetings are held between
representatives from central and local government. At the political level a process of four consultative meetings per
year (since 2000) brings together key ministries of the central government with high level representatives from the
Norwegian Association of Local and Regional Authorities (Kommunenes Sentralforbund, KS). Similar meetings are
held addressing issues pertaining specifically to county and municipality issues. Second – as part of the public
consultation on draft laws and regulations – local government and local government organizations (KS) receive for
comments those government draft regulations considered of special relevance for local governments. Third,
continuous informal dialogue takes place between central and local government representatives at different levels, in
many different forms, and on political as well as technical and professional issues.
In most countries, regulatory co-ordination has been promoted by associations and local authorities. This has
provided a good basis for advice and better understanding of the needs and problems at different levels of
government. But co-ordination has been improved mainly by special bodies and institutional mechanisms that serve
lower levels of government to submit comments, to put forward specific measures and to communicate with the
central level. Co-operation agreements have also improved co-ordination by establishing specific plans with clear
frameworks for implementation and financing. In Denmark, for instance, as a result of a sustained process of
decentralization, particularly since the fusion of local authorities in 1970, much government service delivery is
carried out at lower levels of government. Regulatory policy remains concentrated at the national level, although
there is significant consultation with local government as a result of its major role in implementation. From the
perspective of local government, the key regulatory issue is that of increasing the freedom to act to be able to
achieve efficiency gains needed to allow services to be delivered within tight fiscal restraints. To achieve this goal,
the Government initiated a local government reform and a five-year work reform took place in 2007. Structural
setting and relations between local and central government were redefined. According to the new system, there are
new mechanisms and areas in which national and central governments co-operate and co-ordinate their service
delivery.
Source: Rodrigo, D., L. Allio and P. Andres-Amo (2009)
135
and agencies. The by-law was enacted by the Decision of the Council of Ministers on December 2005.
The main principles that apply in the preparation of drafts are the following:
- Drafts shall not be contrary to high norms of law;
- Drafts shall be prepared in line with purposes of regulation;
- Judicial decisions shall be taken into account in the preparation of drafts;
- All of the legislation on regulated areas shall be reviewed during the preparation of drafts;
- The scope of drafts shall be clearly regulated without leaving any room for misgivings; and
- Articles of draft shall be concise and clear, there shall not be explanatory provisions in
parenthesis.
5.25. Turkey has a good pool of legal drafters that work in line ministries, preparing drafts of
laws and regulations. Good expertise can also be found in institutions dealing with the quality control of
such drafts, such as the Ministry of Justice and the different institutions at the Prime Ministry‘s Office.
This is, however, not the rule and in many cases there is a lack of professionals with other backgrounds,
in particular economics, that could incorporate more evidence-base information to the preparation of laws
and regulations with high potential economic impact for society and businesses. It is therefore not
uncommon that some drafts relay on the work of consultants and are sometimes inspired by laws from
other countries, which reveals the need to strengthen capacities in the country to prepare laws on complex
issues.
5.26. Turkey has devoted particular attention to creating capacities inside the administration to
introduce new tools for regulatory management. This is the case of the use of RIA and Standard Cost
Model, which are essential to improve the quality of regulations affecting business operations by reducing
the time and cost of doing business. The efforts have been limited so far, but this approach has showed
the importance of building capacities and creating awareness at different levels: technical experts,
decision-makers, politicians and the private sector.
5.27. Transparent and consistent processes for making and implementing laws and regulations
are fundamental to ensure public confidence in the rulemaking process. Consultation is a systematic
attempt to discover the opinions of groups affected by regulation and to obtain data useful in regulatory
development and analysis. It may be general (e.g. advertisement for comment) or specifically targeted
(e.g. focus groups, working parties), depending on the type and impacts of a regulation issue or proposal.
5.28. The Turkish Government has created some public-private mechanisms to deal with the
business environment and reduce costs for doing business. These instances participate in consultation
mechanisms about the way regulations affect them. The YOİKK and the IAC have proven to be very
helpful in terms of identifying constraints for doing business in Turkey and increase private sector
participation in the design of solutions to improve the investment climate in the country. They have set a
basis for trust between the government and the business sector. But businesses have reported that not all
of the Committees work at the same pace.
5.29. Public consultation is essential to improve the quality of new regulations. In terms of primary
and secondary laws and regulations, Turkey has made it compulsory to consult draft proposals among
government agencies. Any draft that comes to the Prime Ministry has to be submitted with an explanation
about the consultation among government bodies. But consultation with external stakeholders is not yet
compulsory and the degrees to which ministries consult vary across the administration.
5.30. Regulatory Impact Analysis (RIA) is a fundamental tool to assess the likely impacts of
future laws and regulations. RIA provides decision-makers with valuable empirical data and a
comprehensive framework in which they can assess their options and the consequences their decisions
136
may have. A poor understanding of the problems at hand or of the indirect effects of government action
can undermine regulatory efforts and result in regulatory failures. RIA is used to define problems and to
ensure that government action is justified and appropriate.
5.31. A by-law passed in February 2006 introduced the use of RIA in Turkey for all draft laws
and statutory decrees that could have an impact above TL 10 million. The article dealing with the
regulatory impact assessment of this by-law came into effect as of February 2007.101
Since then, RIA is
required by all ministries before a draft law proposal is sent to the Prime Ministry. Following a Prime
Ministry‘s circular in April 2007, RIA guidelines were prepared, including procedures and processes on
how RIA should be conducted. The Better Regulation Group has been responsible for the RIA work in
Turkey. Its main responsibilities are increasing capacity in the administration, particularly in line
ministries; ensuring quality control of submitted RIAs; raising awareness of RIA issues across the
administration; creating a RIA network in the administration; and maintaining a RIA Turkey website.
5.32. The Better Regulation Group obtained a grant from the EU to conduct a project called
“Introducing Regulatory Impact Analysis into the Turkish Legal Framework,” mainly based on
training and capacity building activities. The project has financed different RIA events (workshops,
training, etc.) and has allowed the training of 370 government officials in the use of RIA. Within the
framework of this project, two pilots have been conducted, on taxation for electricity with the Ministry of
Finance and on rural incentives with the Ministry of Agriculture. The project has also set up a web site for
RIA issues in Turkey102
and a RIA Network to facilitate communication and effective implementation
among ministerial RIA units. The finalization of this project has allowed reviewing the current set up for
RIA. Different recommendations have been made to ensure that RIA is a tool to support policy making.
These recommendations are aligned with international good practice and take into consideration Turkey‘s
current capacities of implementation. They cover a wide range of issues, from design and focus,
improvement of guidelines, communication strategy, institutionalization of the process, data collection
strategies, capacity building and improved coordination.
5.3 Administrative capacities to review existing regulations
5.33. Ensuring regulatory quality refers not only to new, but also to existing regulation. Keeping
regulations up-to-date is important to ensure that the legal framework is relevant and effective for policy
purposes. A number of different techniques and mechanisms exist to conduct periodic evaluations of
whether existing regulations still constitute the best available solutions to the problems they seek to
address.
5.34. Governments worldwide are adopting programs to reduce the administrative burdens
associated with regulatory requirements. Regulatory paperwork and government formalities can be
unnecessarily burdensome on regulated groups if coordination between regulators is lacking, new
technologies are not used to assist in information gathering, and if unnecessary information is sought by
regulators.
5.35. Turkey has made progress in simplifying certain procedures, regulations and laws. This has
however mostly been on an ad-hoc basis and without having a comprehensive approach to administrative
simplification, as it has been the case of many OECD countries (Box 5-4 and Figure 5-3). As part of the
efforts undertaken by the Turkish government, there are a number of issues that can be highlighted:
101 Under Article 24 of the ―Regulations on the Procedures and Principles for Preparation of Legislation‖ published on 17/2/2006,
it is provided that RIA shall be carried out for acts and statutory decrees, and other regulatory acts as deemed appropriate by the
Prime Ministry, to be prepared as of 17/2/2006, with the exception of those on national security matters, as well as budget and
final accounts acts and statutory decrees. 102 www.riaturkey.org
137
- Certain laws and regulations have been amended after some simplification processes;
- Some new laws have been enacted to solve legal ―grey areas‖;
- Several pilot projects have been undertaken in the use of tools for administrative simplification,
such as the Standard Cost Model (SCM);
- A new inventory of public services has been prepared for the entire administration; and
- Establishment of one-stop shops to support businesses to deal with procedures.
Figure 5-3: Explicit program for reducing administrative burdens
Note: The sample includes 31 jurisdictions for 2005 and 2008. Data for 1998 are not available for the European Union,
Luxembourg, Poland and the Slovak Republic. Results have been adjusted for 26 countries following the results of a peer review
process. Results for Spain are pending confirmation. Results for Belgium, Germany, Ireland, Luxembourg and Switzerland
represent the initial answers to the surveys and will be adjusted within the next weeks.
Source: OECD Regulatory Management Systems‟ Indicators Survey 1998, 2005 and 2008. www.oecd.org/regreform/indicators
in OECD (2009)
0
1
2
3
4
5
6
FIN HUN IRL SVK POL CZE LUX CHE NOR TUR ICE JPN PRT DNK NZL AUS ESP USA EU DEU SWE BEL GRE AUT ITA MEX CAN KOR FRA NLD UK
199820052008
Is there an explicit government programme to reduce the administrative burdens imposed by government on enterprises and/or citizens? If the answer is “yes”: Does this programme include quantitative targets?
Which of the following strategies are used? - Information and communication technologies for regulatory administration (e.g. electronic databases, online formats)
- Other streamlining of government administrative procedures- Reallocating powers and responsibilities between government departments and/or between levels of government
See Question 12 / 2008 OECD Regualtory Indicaotrs Questionnaire
Weights:if yes, weight=2
if yes, weight=1
if yes, weight=1
if yes, weight=1if yes, weight=1
Box 5-4: Comprehensive administrative simplification efforts in OECD countries
In many OECD countries, administrative simplification is becoming increasingly embedded within the overall
regulatory quality systems of respective countries. In the past, administrative simplification was often undertaken on
an ad hoc or sectoral basis. In most countries there is now more of a ―whole-of-government‖ approach to reducing
burdens. Simplification is being increasingly embedded in the policy-making process. Simplification strategies focus
on two dimensions: ex ante control of the burden introduced by new regulations (a flow concept) and the reform ex
post of existing burdensome regulation (a stock concept).
Measurement has also become an important part of the burden reduction programs of many countries. The focus of
the measurement exercise (and subsequent burden reduction programs) tends to be on business, often with special
consideration for small and medium sized businesses, but there has also been a trend towards measuring and
reducing the burdens imposed on others, including private citizens and the not-for-profit sector. The sophistication
of the measurement techniques varies between countries, but the trend is clearly towards more sophisticated and
accurate techniques that allow a very detailed examination of the source of administrative burdens.
In the UK, for instance, research and surveys have shown that businesses spend at least £1.4 billion each year on
advice to help them comply with regulation. Businesses will pay for advice if they feel that this is cheaper or easier
138
5.36. In many emerging economies, streamlining some regulatory instruments has been an initial
point for broader regulatory reform efforts. By targeting particular regulations, governments ensure
that the political support is concentrated in few, but sometimes very pervasive regulations that might
cause a great impact in economic terms by imposing excessive costs to businesses. This is the case of
licenses and construction permits. In terms of regulatory instruments, licenses and construction permits
have been identified as two important areas for further improvement by YOİKK. Both regulatory
instruments represent the complexities of the regulatory problems in Turkey and have the potential to
increase a positive direct impact on economic activity if they are properly streamlined.
5.37. Licenses are one of the main areas of complaint by Turkish businesses. Some initial
improvements have been made by the Government, such as having a single Business Opening License,
but there has not been a comprehensive approach to streamline, simplify or eliminate licenses. There is
currently no single institution responsible for licenses, which are granted by Ministries, regulatory
agencies and municipalities. The State Planning Organization, which is responsible for the YOİKK
Technical Committee on Sectoral Licenses, conducts some work on the licenses delivered by Ministries.
In addition, the Ministry of Public Works and Settlement is also involved in the work related to
construction permits, as the Chair of the YOİKK Technical Committee of Investment Locations.
However, no single comprehensive strategy for licensing reform has been undertaken in Turkey and it is
not clear how much licenses cost to businesses and government.
than following regulations on their own. The government has decided to make real reductions in how much
businesses need to spend on regulatory advice by tackling, volume and complexity, low awareness of government
guidance, regulatory change, poor quality government guidance as well as uncertainty, risk and lack of confidence.
As a result, the current strategy applied in the UK relies on the following actions:
- Improving the regulatory process: Plan guidance at an early stage of the policy process and issue guidance
earlier
- Improving communication on regulation: Increase the market penetration of businesslink.gov.uk;
communicate directly with businesses using high-quality, simple guidance and communicate with businesses
through intermediaries
- Improving the quality of government advice on regulation: Improve feedback mechanisms on guidance and
consider joint-badging or outsourcing the design of guidance
- Improving the environment for business advice on regulation: Help businesses become informed consumers
of advice services by increasing understanding of regulatory requirements; take advantage of online forums
for businesses to share information on regulations and provide dedicated guidance for advisors where
appropriate.
Source: www.berr.gov.uk/whatwedo/bre/reviewing-regulation/reducing-cost-business/page44090.html
Box 5-5: Benefits of licensing reform
In many emerging and developing countries, such as Korea, Mexico, Croatia, Ukraine, Moldova, Hungary, etc.,
licensing reform has been a trigger for further reforms because a well designed licensing program can create
momentum and political appetite for other regulatory reforms. Licensing reform is also pertinent as a starting point
for a broader program of regulatory reform, since licenses cut across policy areas and mobilize most government
agencies dealing with economic activities.
Business licensing is a commonly used form of regulation which affects specified businesses and occupations by
regulating entry into markets and conduct within markets. Licenses typically impose on businesses a range of
conditions, obligations and rights — often in the form of a specific license, permit or concession. Licensing can be
distinguished from other types of regulatory requirements by usually obliging the regulated parties to obtain a
certification of compliance with regulatory obligations prior to the commencement of a given business activity.
From an economic and business point of view, licensing is a potentially much more costly and potentially damaging
139
5.38. According to the Doing Business Indicators, Turkey ranks 133 among 183 countries in
obtaining construction permits. As mentioned in Chapter 2, DB shows that the site development time in
2009 amounts to 188 days, a figure that is not equally negative when analyzing the corresponding data in
the Enterprise Survey. Average time spent dealing with construction-related permits was in fact lower
than in most comparator countries. Instead, there seem to be significant variations among regions and
cities, indicating that the system of permit processing has not been sufficiently streamlined across Turkey.
Several areas can be identified as problematic in this field:
- There are major gaps on sector based strategies and on urban development plans from several
Ministries. Coordination mechanisms to prepare strategies and plans are weak.
- Traditional methods to deal with permits are slow. There are not well established digitalized
systems that can ensure faster processing of applications from businesses and citizens.
- Similar authorities conflict on areas of action. Sometimes two agencies deal with the same issue
and there is no planning authority that can help solving the differences.
- The approval of investors‟ plans falls into the responsibility of different agencies, which might
have different solutions for different plans.
- Inspections related to construction permits are in the hands of the Ministry of Interior, which
does not have a specialized group of experts on planning and risk related issues. The Ministry of
Interior inspects only from the administrative side.
- Municipalities are responsible for delivering some permits and licenses but they do not have
adequate capacity to understand and apply secondary legislation.
- There is no standardization of procedures and steps in Ministries involved in construction
permits. There are no clear guidelines and handbooks that can help to implement the regulatory
framework.
regulatory intervention when compared to other types of regulation (such as broadly applied competition law and
accounting rules, or other ‗lighter‘ forms of regulation etc). This is because licensing requirements not only impose
regulatory compliance burdens (as do most types of regulation), but also can restrict healthy competition by
establishing significant and unnecessary entry barriers to particular economic activity and markets. These
restrictions can include:
- grant exclusive rights for a supplier to provide a good or service;
- affect the ability of some types of firms to participate in public procurement;
- significantly alter costs of entry or exit to a market; and/or
- create a geographic barrier to the ability of businesses to supply goods or services, invest or supply labor.
Licensing can restrict or reduce the ability of businesses to compete and innovate through:
- control or substantially influence the price at which a good or service is sold;
- altering the ability of suppliers to advertise or market their products;
- setting prescriptive standards for product/service quality that are significantly different from current
practice; and/or
- significantly altering costs of some suppliers relative to others.
Licensing reform advocates a comprehensive, top-down, and institutionalized approach. Through this approach,
reforms are driven forward by an explicit political mandate, if possible also by a quantitative target for the reform,
and by strong incentives for regulating agencies to participate constructively in the review process. Licensing reform
can only be sustained over time if other tools for regulatory management are gradually integrated in the system and
if certain institutions and capacities are created inside the administration.
Source: World Bank Group (2009)
Box 5-6: Dealing with construction permits
A building permit grants legal permission to start construction of a building project. It is usually granted by a
140
5.39. Both licenses and construction permits are potential regulatory instruments for
comprehensive review. They could also serve as platform for developing a focused, targeted strategy on
simplification with clear baselines and reduction targets. For this purpose, the use of Standard Cost
Model (SCM) seems an interesting regulatory tool that could be employed in Turkey. The SCM is a
methodology used in many developed and, increasingly, emerging countries to measure the administrative
costs – both monetary and in terms of time – required for compliance with regulations. It is a key tool for
broad administrative simplification efforts since it leads to the establishment of clear targets for the
reduction of administrative burdens. Results in the use of the SCM have shown that reduction in
administrative costs has an important impact on GDP. In the European Union, for instance, achieving the
objective of reducing administrative burdens by 25 percent could lead to an increase in the level of EU
GDP of approximately 1.4 percent or € 150 billion in the medium term.103
5.40. Turkey has started the use of SCM with its application in two pilot projects on registration
of vehicles and on establishing businesses. The use of SCM has been introduced primarily by the
Department of Administrative Development of the Prime Ministry through the MATRA project. Some
103 European Commission (2007)
municipality or a specialized local authority. The main objective of building permits is to ensure the health and
safety of the community. This has important implications for policy-makers who need to strike the right balance
between the cost imposed on industry (including the checks imposed through the building permitting process), and
the real benefits in safety and health standards.
The building permit process also plays a critical ―gate-keeper‖ role in protecting a range of other public goods such
as preventing construction close to airports, and protecting the environment or preventing potentially harmful
industries to locate in residential areas. These public goods are more fragmented, and not related to structural
properties or the risks directly associated to the structural characteristics of the building. When this ―gate-keeper‖
function in not carefully managed and coordinated with the relevant authorities, an insurmountable bureaucracy may
emerge, which is likely to discourage investment, and increase the level of informality. Carefully managing the
―gate-keeper‖ role is an important factor in the success of building permit reforms.
Reformist countries are gradually adding new policy objectives in their reform efforts, such as energy conservation
and environmental sustainability. These are important public goods, but should not translate into a more complex
process for investors. Best practice reform experience shows that new policy objectives, including those going
beyond the improvement of public safety, can be combined with effective red tape reduction programs and more
efficient and streamlined processes. In fact, building permit reforms, as observed in good practice countries, have all
generated positive impacts on processes, although streamlining procedures might not have been the original or main
focus.
New Zealand and Canada have both reformed their construction building systems. The objectives of the reforms in
both countries were primarily driven by the need to increase safety, improve the standards of building practitioners,
and turn local authorities into efficient enforcement bodies. In pursuing this strategy, both countries have
considerably streamlined their processes, and made it easier for investors and developers to go through the
permitting process. Both countries have introduced a common set of reforms, i.e., pro-actively engaging private
building practitioners in the permitting process and inspections, introducing risk management, supporting innovative
one-stop-shops, consolidating pre-approval requirements, and improving appeal mechanisms for developers and
investors, etc. To address the dysfunctional relationship between central authorities and local permitting authorities
while still managing high standards of enforcement, New Zealand engaged in one of the most original and daring
reform in establishing a compulsory accreditation process for local permitting authorities. Both reform efforts are
starting to pay off with lower rates of accidents, and a faster turnaround of building permit processes. For example,
there has been a 40% reduction of accidents within the regulated industries in building sites in Ontario, and fires
have declined by 15% since the introduction and enforcement of the new law.
Source: Moullier (2009)
141
training and awareness activities have been conducted to inform the political and technical level of
different ministries about the use of SCM. In addition, in order to measure cost of sectoral licenses which
is one of the YOİKK agenda, a pilot project called ―Measurement of Administrative Burdens in Export
Procedures‖ has been conducted in cooperation with TOBB. Business associations have also participated
in dissemination activities. Some constraints faced in Turkey to introduce the use of SCM are the
financial resources and technical skills needed for its implementation.
Box 5-7: The use of SCM: International experiences
Complying with regulations usually involves costs for businesses, which can be divided into various different
categories:
- Financial costs are the result of a concrete and direct obligation to transfer a sum of money to the
government or the competent authority.
- Compliance costs are all costs to businesses of complying with regulation, with the exception of the
financial costs. Compliance costs can be divided into ‗substantive compliance costs‘ and ‗administrative
burdens‘.
- Substantive compliance costs are the costs that businesses make in order to comply with the content
obligation that legislation and regulations require of a production process or a product.
- Administrative costs are the costs imposed on businesses, when complying with information obligations
stemming from regulation.
The Standard Cost Model (SCM) is a method for measuring the administrative burdens imposed on businesses
through legislation, regulations and other requirements. The SCM has been developed to provide a simplified,
consistent method for estimating the administrative costs imposed on business by central government. It takes a
pragmatic approach to measurement and provides estimates that are consistent across policy areas. The starting point
of SCM analysis is the identification of ―information obligations‖ that businesses are required to provide to the
government and other bodies. The SCM can measure information obligations arising from different sources such as
all existing laws and regulations; a specific field of laws and regulations (like fiscal rules, the transport sector,
starting a business, employment procedures); or requirements imposed by a selected government body.
Since the 1990s, SCM has been developed and modified for use in OECD countries. The Czech Republic was
measuring the baseline of overall administrative burden. The measurement was undertaken between March 2005 and
September 2005. The baseline measurement included a measurement of all business related generally binding
regulations in 12 ministries and 10 central administrative authorities. The results of the measurement were sent to
the Department of Regulatory Reform and Central State Administration Reform by the end of September 2005.
Subsequently, the Department carried out the analytical report of overall administrative burdens on businesses
entitled ―Analysis of the administrative burdens on businesses‖, elaborated in February 2006 on the basis of the
information collected by ministries and other central state administration authorities. As part of the
recommendations of the report, a number of measures were adopted by the Government Resolution No. 759 of 11th
July 2007: i) Introduction of the obligation to assess administrative burdens ex ante in case of new legislation drafts;
and ii) Preparation of overviews enumerating the concrete legislation proposed for the purpose of the reduction. In
its latest Resolution, No. 446 of 21 April 2008, the Czech government confirmed the reduction target set in 2005;
e.g. all the legislative amendments should be accomplished by 2010 with the aim to reduce the burden on businesses
by 20 percent.
Source: www.administrative-burdens.com
5.41. ICT solutions make a valuable contribution to improve the regulatory environment of a
country. Many international experiences show that the use of ICT in relation to transactions within and
between government bodies and between government bodies and business and citizens, is a key enabler of
administrative simplification, a fundamental element of regulatory reform.
5.42. Important steps have been taken to integrate the use of ICT to improve services delivery
both to citizens and businesses and to create a basis for an information society. The State Planning
142
Organization is responsible for setting the policies and coordination of implementation of e-government
in particular and information society in general under the umbrella project named e-Transformation
Turkey. Lately, an ad-hoc e-Government Group at the Prime Ministry was tasked to tackle with the
legislative and administrative barriers for selected eleven e-government projects. e-Government activities
had gained momentum after Turkey became a party to the eEurope+ Initiative, which was designed for
EU candidate countries in 2001. The State Planning Organization has been leading the e-government
work since 2003. The eTransformation Turkey Project was launched in 2003 aiming to carry out the
process of transformation into an information society. Since the inception of this Project, two action plans
have been launched and implemented successfully. An Information Society Strategy covering 2006-2010
was prepared under the coordination of State Planning Organization to enable Turkey to benefit from ICT
effectively and to identify the middle and long term strategies and targets for the realization of the
ambitious transformation. The current Information Society Strategy establishes targets and
implementation measures to transform Turkey into an information society.
5.43. The e-Government Group at the Prime Ministry is an ad-hoc group established to deal with
eleven on-going e-government projects. In addition to this task, the e-Group has prepared a draft of the
―E-Government and Information Society Act,‖ covering two main aspects: organizational (creation of an
Information Society Act and institutional architecture for e-government and information society in
Turkey) and framework articles (project management, performance evaluation, data sharing and
ownership, common e-government services, authentication and authorization, e-government portal, e-
archive, use and protection of personal data, liability and secrecy, etc).
5.44. Another dimension of transparency is the need for the government to effectively
communicate the existence and content of all regulations to the public. This means that the
regulations are available to the public at reasonable cost, in a language that can be easily understood.
Communication is also essential to achieving effective compliance. The by-law on the ―Preparation
Method and Principles of Legislation‖ (2005/9986) encourages legal drafters to use plain language in the
preparation of draft proposals. In terms of communication, Turkey has a Regulatory Information
System104
which contains all published regulations, e.g. laws, decrees and other secondary legislation. All
citizens and businesses have right to access to it freely. Amendments are consolidated daily and
republished the principal act as a single text. There is no legal basis for courts and public agencies to
accept it as ―prima facie evidence of the law‖. All regulatory instruments are published in the Official
Gazette which is available both on-line105
and in hard copy.
5.45. The quality of existing regulation is as important as the assessment of new regulations to
ensure a high quality regulatory structure. In many cases and as technology, the economy, and society
change, existing regulations often become less relevant and effective. It is therefore essential to maintain
a periodic reevaluation of whether existing regulations still constitute the best available solutions to the
problems they seek to address. A systematic approach is required to ensure that all regulations are
regularly subjected to this reassessment and several techniques exist for this purpose. In Turkey, the first
review of existing regulation took place in 1986 and included the codification of all laws, decrees having
force of law, regulations (Tüzük), and by-laws issued by the Council of Ministers. A second effort was
undertaken in 2005, when the Government identified that 13,967 by-laws were issued between 1970 and
2005 and all were reviewed. This led to a reduction to 4,510 by-laws. The most recent effort is currently
under way and comprises the review of 4,795 by-laws, in which regulatory burdens imposed to society
and businesses have been identified. The review has led the following results:
104 http://www.mevzuat.gov.tr 105 http://rega.basbakanlik.gov.tr
143
- Preparation of drafts to amend 155 by-laws;
- Removal of around 400 document and information obligations;
- Replacement of 38,000,000 obligations for submitting the copy of the ID card by declaration of
the ID number; and
- Delegation of authority to lower levels in 25 procedures.
5.46. Enforcement and compliance with regulations are two areas that remain relatively weak.
Municipalities play a role in implementing regulations and most of the inspections system remains an
issue at the central level. The impact of this institutional set up is significant in terms of the burden that is
imposed on citizens and businesses, which sometimes have to comply with procedures established by the
local levels without prior agreement at the central one. This is also a consequence of limited coordination
when preparing laws and regulations and of the duplication of responsibilities among ministries in
specific areas of action.
5.47. The role of the judiciary is essential for regulatory quality control and better economic
performance. The effectiveness of the process arises from the ability of the judiciary to consider
regulations‟ consistency with principles of constitutionality, including notably proportionality and the
right to be heard. It also arises from courts‘ scrutiny of whether delegated legislation is fully consistent
with primary legislation. Administrative justice and judicial review are important elements for
compliance and enforcement of regulation. In Turkey, both elements remain a challenging area for further
improvement that imposes costs and delays to economic activity.
5.48. Public redress can be made directly to the administrative authority, which has officially 60
days to respond to the interested party. Given the little responsiveness in the system, most people
appeal decisions to Court. This procedure is not expensive, which makes justice accessible. However, this
creates a workload that is difficult to manage for authorities involved. In addition, the lack of specialized
courts to deal with particular ―cases‖ brought by investors reduces the effectiveness of the system. In
case the Ministry of Interior has to intervene with local authorities, delays are to be expected since the
Ministry is not equipped with the adequate resources.
5.49. The Ministry of Justice is actively participating in a number of proposals to improve
judicial review. There are current efforts to introduce new conciliation mechanisms, expanding on the
successful experience with tax system reform, and to streamline administrative and judicial review. There
is a current draft of a new General Administrative Procedure Law which will contribute to improve the
situation at the intermediary level. There is also an initiative to create an Ombudsman mechanism under
the public authority to introduce a new conflict resolution system. There are also proposals to increase the
number of courts, judges and prosecutors.
5.4 Policy Options for Regulatory Reform
5.50. Regulatory reform has the potential to be a high political priority, in particular, linking
regulatory management and reform to the current efforts of improving the investment climate. This
task, however, requires strong leadership and political back up, since it requires consolidating the current
efforts and establishing good co-ordination mechanisms between different institutions. A comprehensive
strategy for regulatory reform would make the Turkish economy more competitive and dynamic by
reducing costs on businesses, eliminating incertitude and unnecessary obligations, and providing better
services for citizens.
5.51. Turkey has adopted many good international practices in the way regulations are prepared
and implemented but still much remains to be accomplished. In particular more systemic changes that
can establish more transparent procedures, consistent consultation mechanisms, more evidence-based
144
decision making, reduced discretionary decisions and comprehensive administrative simplification efforts
could be considered. This could make Turkey a more attractive destination for investment and could
encourage better levels of compliance and enforcement, which remain low compared to international
standards. A number of actions could help Turkey to improve its regulatory management system and
strategy.
5.52. Proposal: Revise current efforts for regulatory reform to set clear priorities, objectives and
targets within a comprehensive strategy. Current efforts on regulatory reform could be combined in
order to establish clear priorities that can translate in improvements of the investment climate and
business environment. The existing work done by YOİKK Technical Committees could be a platform to
identify priority areas and start implementation efforts. In the medium term, it could be helpful to
establish a single dialogue mechanism between the Prime Ministry and YOİKK, such as regular high
level meetings, to set priorities. A decision on the sequencing of regulatory reform would also help to
better allocate resources, build capacities inside the administration and raise awareness about regulatory
reform in the public. It would be advisable to devise a strategy mapping the current efforts and priorities
to define how they could be integrated in broader regulatory reform objectives. Such a strategic document
could include not only principles of quality regulation to prepare new and amend existing regulation, but
also a clear indication to link this area to reform of the business environment and to improvement of
service delivery for citizens. A national regulatory policy could then follow a ―whole of government‖
approach that is applied by all government agencies and all levels of government. Such a document would
help (i) allocate clear responsibilities to different institutions in order to facilitate coordination and
cooperation between government agencies; (ii) establish clear objectives and targets that can be measured
over time to show progress; (iii) increase transparency and accountability in the way regulations are
prepared and implemented.
5.53. Proposal: Strengthen the institutionalization of regulatory reform by creating a single
oversight body for regulatory reform. Turkey has a number of institutions that deal with regulatory
reform from different perspectives. Most of the traditional roles that are conducted by oversight bodies for
regulatory reform in other OECD countries remain in hands of various institutions. At least three
institutions in the Prime Ministry deal with the quality of laws and regulations and some others are in
charge of regulatory reform issues, apart from other bodies with regulatory responsibilities across the
administration. Turkey could consider the possibility to bring together the existing expertise and entrust a
single unit with clear responsibilities for regulatory reform. The Prime Ministry could establish an
oversight body responsible for moving forward the better regulation agenda in the country, which could
incorporate the different roles and responsibilities spread across the administration, in order to implement
a single strategy for regulatory management and reform. The benefits of having a single oversight body
for regulatory reform have been documented by international experiences. Strong oversight bodies, such
as in the case of UK, USA, Canada, Mexico and Korea, have been key actors in the process of regulatory
reform, working as ―engines of reform‖, maintaining a whole-of-government approach and coordinating
inside the administration. The main role of oversight bodies is to ensure regulatory quality. This is done
through supervision, control and coordination of the regulatory program and system. Regulators are then
forced to demonstrate and justify the relevance of their regulatory actions (newly proposed and existing),
using accountability and assessment mechanisms.
5.54. Proposal: Strengthen YOİKK‟s role to improve the business environment and advocate for
regulatory reform in Turkey. YOİKK role in identifying constraints to investment climate issues in
Turkey has been very helpful. This platform has established a clear mechanism for dialogue between the
public and the private sector which could be strengthened over time. In particular, YOİKK could play a
significant role in moving forward the regulatory reform agenda, by linking the investment climate
constraints to the systemic challenges of improving the quality of regulation. YOİKK has the potential to
participate in the definition of priority areas for regulatory reform, given the existing work undertaken and
145
the different action plans already in place for policy action. The existing structure that YOİKK has built in
Turkey constitutes a solid basis for further coordination and consultation among the private and the public
sectors. It also provides a forum for identification of common challenges for different ministries. This
could be used to encourage the needed coordination that Turkey requires to implement regulatory reform.
YOİKK could also be a strong advocate for regulatory reform. In many OECD countries, advocacy and/or
advisory bodies have played a key role in shaping the regulatory reform agenda. In Japan, for instance,
the Council for the Promotion of Regulatory Reform (CPRR), composed of members from the business
sector and civil society, was established in 1994 to provide input to a 3-year action plan for regulatory
reform. In the UK, the Better Regulation Commission was a fundamental actor for regulatory reform
during its ten year existence (1997-2007). The Commission provided independent advice and challenges
to the UK government on its management and delivery of better regulation, as well as independent
scrutiny of departments' plans for regulatory simplification. With its current structure, YOİKK could play
some of these roles by setting up priorities for government action and making recommendations about
regulatory reform with a clear involvement of the private sector. In the medium and long term, YOİKK
could be a platform with an advocacy role for regulatory reform in Turkey.
5.55. Proposal: Design a comprehensive administrative simplification strategy with clear
objectives, targets and review criteria for lower level of regulation to improve the business
environment. A particular area with great potential that has not been fully explored in Turkey is the
design and implementation of a comprehensive administrative simplification strategy for lower level of
regulation. Such a strategy could be relevant for Turkey for several reasons. First, it could establish a
clear link between the importance of improving the quality of regulation and the improvement of the
business environment, which is in fact one of the central purposes of regulatory reform. By looking at
lower level of regulation, such as licenses or administrative decrees, the Government could produce
important economic gains that would translate into a better economic performance. Second, such a
strategy could also be the impulse that Turkey needs to fully embark in the implementation of a better
regulation agenda. Third, such a strategy would be an excellent opportunity to strengthen public and
private dialogue by establishing clear reduction targets of administrative burdens. This could strengthen
existing consultation mechanisms, encouraging the private sector to play a leading role in the
implementation of the strategy. This would also facilitate setting up monitoring and evaluation
mechanisms to measure progress. Fourth, this strategy could make it possible to sequence regulatory
reform in a more strategic way. For instance, it would facilitate the implementation of a number of
regulatory tools that have been tested in pilot projects in Turkey, such as the use of Standard Cost Model
and Regulatory Impact Analysis, which require time and capacity building to be embedded across the
administration. Building on the work carried out by YOİKK and in particular by the Technical Committee
on Licenses, an action plan could consider the following steps:
- Identify a political and technical platform (Committee, unit and/or oversight body for regulatory
reform) to lead the administrative simplification and reform program, supported at the highest
political level
- Prepare and/or review an inventory of selected lower regulation with input from all authorities
involved
- Based on the inventory, select a number of lower level regulation subject to measurement with
the Standard Cost Model
- Use working groups to accompany the process and validate data
- Based on measurement results and impact on the economy, select lower level regulations to be
reviewed in a first review process (―quick wins‖)
- Establish clear criteria for review and discuss provisions of lower level regulation in working
groups with public and private participants
- Working groups make recommendations to simplify, streamline or eliminate procedures to
Committee at political level
146
- Committee revises and makes recommendations to be taken at high political level
- Working groups work on the rest of licenses
- After recommendations taken, list of ―clean‖ lower level regulation is put on an e-registry for the
particular lower level regulation
- Once the stock has been ―cleaned‖, Regulatory Impact Analysis is used to review the new
regulations
5.56. Proposal: Improve coordination mechanisms inside the administration when preparing
laws and regulations. Since a number of institutions are responsible for regulatory issues and there is no
high level authority imposing a single strategy, it is difficult to ensure coordination among various
institutions. Several countries have improved coordination at technical level by nominating ―regulatory
reform experts or champions‖ in each ministry or regulatory agency. The creation of networks of experts
dealing with regulatory reform facilitates dialogue among institutions and ensures that regular meetings
are the platform for discussions and sharing experiences. Turkey could design a network of experts on
regulatory reform to build capacities, discuss priorities and communicate in a strategic way what could be
implemented. In the medium and long term, the establishment of an oversight body for regulatory reform,
with a clear mandate about its responsibilities, could also help improve coordination among institutions.
Turkey has already relevant institutions with skills and expertise in improving the quality of regulation,
there would be no need to create new institutions, but to reengineer the existing mechanisms.
5.57. Proposal: Strengthen coordination and cooperation among levels of government. Being
provinces and municipalities the main levels responsible for implementing regulations, it is essential to
establish formal mechanisms to discuss the way their implementation and enforcement can be improved.
In particular, provinces and municipalities could be systematically consulted in the preparation of laws
that can have a direct impact in their roles. Turkey could reinforce the current consultation mechanisms
with provinces and municipalities and strengthen the participation of lower levels of government in the
preparation and design of laws and regulations. Consultation could be undertaken in a formal and
systematic way, using approaches already proved in other countries, such as permanent roundtables or
discussion meetings. The Government could also encourage the development of capacities at local level
to improve implementation. Lower levels of government could be encouraged to respect the obligations
imposed on businesses and citizens when delivering permits or licenses. The central government could
also establish mechanisms to supervise accurate implementation of national directives. For instance, a
revision of the current inspections system could be introduced, in order to improve enforcement and
compliance. Inspection services, which are currently cumbersome and lack sufficient skilled staff, could
be implemented based on assessed risk and the likelihood of violations, increasing transparency in the
process and accountability of the concerned agencies.
5.58. Proposal: Make consultation with stakeholders compulsory for the preparation of new and
amended laws and regulations. Consultation in the preparation of laws and regulations has improved
over time in Turkey since consultation among institutions was made compulsory. Nonetheless, many
gains could be obtained from making it compulsory for government to consult with external stakeholders.
In particular, consultation would help obtain a better understanding of the way regulations affect citizens
and businesses. This would require the Government to establish clear guidelines for consultation,
clarifying the steps to follow, such as deadlines for comments, and ensuring that regulators are
accountable for the comments received. Training regulators on consultation techniques could also be
encouraged, as well as making use of the data gathered during the process.
5.59. Proposal: Continue implementation of Regulatory Impact Analysis (RIA). Turkey has made
important steps in the introduction of Regulatory Impact Analysis (RIA). The work done so far by the
Regulatory Reform Group has raised awareness of the importance to make wiser decisions based on
evidence. The training program on RIA has proved to be useful in developing skills inside the
147
administration, but it is certainly only an initial step to a deeper use of this tool. The Government could
ensure that the work done by the Regulatory Reform Group is continued and supported at high political
level. The expertise gained in the introduction of RIA could be transferred to a future oversight body in
charge of regulatory reform, which could be entrusted with the responsibility to challenge those RIAs that
do not comply with the quality criteria established by the government. This function is essential to
improve the quality of the Turkish regulation in the future. Capacity building in RIA would also be
necessary with training programs linked to the use of other tools, such as consultation techniques and the
use of Standard Cost Model. It would be important to revise the current guidelines to update them with
the most recent lessons learned in the use of RIA in Turkey.
5.60. Proposal: Use existing e-government strategies to support regulatory reform and
simplification efforts. Turkey has made improvements in the application of e-government strategies to
deliver better services for its citizens. One of the main goals is to reach an automatization of procedures
that could facilitate entrepreneurship. The Government is committed to ensure that every document is
available to citizens in soft copy and transactions can be done electronically. This is in line with good
international practice, but it requires adequate resources, skills and ensuring that all society takes
advantage of the use of ICT mechanisms. It is essential for Turkey to link these efforts to a broader
regulatory strategy. There are already good examples in Turkey in the way ICT can be used to improve
the quality of the regulation. The Regulatory Information System is an initial step that could be further
developed. In the same way, a broad administrative simplification effort could rely in the use of ICT tools
to be implemented.
References
Botero, Juan, Simeon Djankov, Rafael La Porta, Florencio Lopez-de-Salinas, and Andrei Shleifer (2004), ―The
Regulation of Labor,‖ The Quarterly Journal of Economics
Department of Administrative Development. 2009. Standard Cost Model and Administrative Simplification, Ankara,
September
European Commission. 2007. Action Programme for Reducing Administrative Burdens in the European Union,
COM(2007) 23 final, Brussels
Government of Australia. 2006. Rethinking Regulation, Taskforce on Reducing Regulatory Burdens on Business,
Canberra
Kaufmann, D., Kraay, A. and Zoido-Lobatón, P. 2002. ‗Governance Matters II: Updated indicators for 2000/01‘
Policy Research Working Paper, World Bank, Washington
Kox, Henk. 2005. Intra-EU differences in regulation caused administrative burden for Companies. CPB
Memorandum Number, No. 136. Rev.1
Nijland, Jeroen. 2008. The Dutch Approach, EIPASCOPE, No. 2
Moullier, Thomas. 2009. Reforming Building Permits: Why It Is Important and What Can IFC Really Do?, IFC
Advisory Services, Cairo
OECD. 2001. Regulatory Reform in Turkey, Paris
OECD. 2005a. OECD Principles for Regulatory Quality and Performance, Paris
148
OECD. 2005b. ―The benefits of liberalizing product markets and reducing barriers to international trade and
investment: the case of the United States and the European Union‖, Economics Department Working Paper, No.
432, OECD, Paris
OECD. 2009. Indicators of Regulatory Management Systems. Update and Preliminary Results, Working Party for
Regulatory Management and Reform, GOV/PGC/REG(2009)2, Paris
Prime Ministry. 2007. Circular Letter 2007/6. Regulatory Impact Analysis Guide, Directorate General of Personnel
and Principles, Ankara
Rodrigo, D., L. Allio and P. Andres-Amo. 2009. "Multi-Level Regulatory Governance: Policies, Institutions and
Tools for Regulatory Quality and Policy Coherence", OECD Working Papers on Public Governance, No. 13, Paris
State Planning Organization. 2006. Information Society Strategy, Ankara
World Bank. 2007. Turkey: Investment Climate Assessment, Washington
World Bank. 2008. Regulatory Transformation in the Republic of Korea, Case Studies on Reform Implementation,
FIAS, Washington
World Bank Group. 2009. Policy Framework Paper on Business Licensing Reform and Simplification, Investment
Climate Advisory Services, Washington
YOİKK. 2009. Activities of the Coordination Council for the Improvement of the Investment Environment and
Technical Committees Action Plans, Ankara
Annex 5-1. OECD Principles of Regulatory Quality and Performance
The 2005 OECD Guiding Principles for Regulatory Quality and Performance capture the dynamic and on-going
whole-of-government approach to implementation of regulatory quality. These principles state that governments
should:
- Adopt at the political level broad programs of regulatory reform that establish clear objectives and
frameworks for implementation.
- Assess impacts and review regulations systematically to ensure that they meet their intended objectives
efficiently and effectively in a changing and complex economic and social environment.
- Ensure that regulations, regulatory institutions charged with implementation, and regulatory processes are
transparent and non-discriminatory
- Review and strengthen where necessary the scope, effectiveness and enforcement of competition policy.
- Design economic regulations in all sectors to stimulate competition and efficiency, and eliminate them
except where clear evidence demonstrates that they are the best way to serve broad public interests.
- Eliminate unnecessary regulatory barriers to trade and investment through continued liberalization and
enhance the consideration and better integration of market openness throughout the regulatory process, thus
strengthening economic efficiency and competitiveness.
- Identify important linkages with other policy objectives and develop policies to achieve those objectives in
ways that support reform.
Source: OECD (2005a)
149
Annex 5-2. The Sources of Law in Turkey
The Turkish legal system is composed of the following sources of law:
Constitution. The Constitution is the highest legal rule which binds all state organs and individuals. The provisions
of the Constitution are fundamental legal rules binding upon legislative, executive and judicial organs, and
administrative authorities and other institutions and individuals
Statutes. Statutes lay down principles, but leave to the judge the problem of interpreting and applying these
principles to concrete facts. Statutes written down in a systematic fashion to regulate specific areas of law are given
the title of "Codes". Statutes enacted by the Parliament must comply with the Constitution.
International Agreements. International law is also a direct source of Turkish legal system. In line with art. 90 of
the Turkish constitution, approval of international agreements signed with foreign countries and international
organizations are valid hence they are approved with a law enacted by the National Assembly.
Decrees having force of statutes. According to art. 91 of the Turkish constitution, the National Assembly can
authorize the cabinet for enacting a decree law. This authorization determines the aim, scope, principles, using time
of this authority and if more than one decree law can be enacted. Decree laws are in force when they are published
in official gazette, but another date can be shown in decree law as an enforcement date. Decree laws are submitted to
the National Assembly on the same day they are published in the official gazette.
Regulation (Tüzük). In accordance with art. 115 of the Turkish constitution, the cabinet can make regulations to
specify the matters in which it is obliged by laws. Regulations should not be contrary to laws and they should be
examined by the Council of State
By-Law. In accordance with art. 124 of the Turkish constitution, ministries and public administrations prepare by-
laws to implement laws and regulations related to their sphere of duties. The Cabinet can make by-laws on the basis
of general regulation authority.
Administrative regulation. The Cabinet and other administrative institutions can issue administrative regulation in
accordance with the upper regulations such as laws, regulations (tüzüks), etc. These instruments can be called
decree, circular, principles, command, etc. Cabinet decree is a common form used by authorities, in particular when
the laws gives the authority to regulate certain area. Even if there is not a specification about its authority to make
legal arrangements in laws, cabinet can make regulations to implement the laws based on its general regulation
authority. The subject of cabinet decisions can be general regulations as a matter of public concern like tax rates or
changing the tax amounts, or can be personal proceedings like appointment decisions.
Source: General Directorate for Laws and Decrees