Challenging Orthodoxy on Governance and Corruption:
Lessons from Worldwide Empirics
Daniel Kaufmann, The Brookings Institution
www.govindicators.org
Presentation at the International Development
Workshop Series at the University of Pennsylvania,
Philadelphia, December 1st, 2008
Note: The research included in this presentation reflects collaboration with co-authors, notably with
Aart Kraay and Massimo Mastruzzi on the WGI, among others. Any data on governance and
investment climate is subject to margins of error and ought to be interpreted with caution. The WGI
is not used for official resource allocation decisions by the World Bank. Errors are the authors’.
1
Challenging Convention on Governance from
an Empirical Perspective
Governance Measurement
•
Evidence-based analysis & policy-making: data needed
•
But is Governance Measurable?: The WGI, the
disaggregated data, the ‘micro’, the imprecision…
Applications
•
Quality of Governance invariant in short/medium term?
•
Does Governance Matter for Growth & Development?
•
Historical vs. Current Determinants of today’s outcomes?
•
Unbundling Governance & Corruption: State Capture matters
•
Some Implications, Some Doubts…
2
Empirical Frameworks: from aggregate to
disaggregate indicators of governance
• The Macro-Aggregate Level, e.g. WGI, TI, CPIA
• The Mezzo Level, e.g. WEF, BEEPS, Doing
Business
• The Micro, in-country, in depth diagnostic, e.g.
WBI GAC diagnostics, scorecards (Bangalore),
randomized experimental projects
• Complementarities
3
Why Measure Governance?
• Governance matters for:
– Development outcomes
– Effectiveness of foreign aid
• WGI are not a “horserace”
– entry point for serious empirical analysis
drawing on a wide variety of data sources
• Data on governance:
-- empowers reformers and civil society to
advocate for change
-- enables evidence-based policymaking
4
Worldwide Governance Indicators
• Data on six dimensions of governance covering
212 countries over 1996-2007
• Synthesis of hundreds of underlying indicators
taken from 35 different data sources
• Aggregate and individual indicators available at
www.govindicators.org, largest publiclyavailable governance database in the world
• Result of longstanding research project by D.
Kaufmann & A. Kraay, featuring the
“Governance Matters” series
5
Whose Views Do the WGI Reflect?
• WGI capture views of tens of thousands of
stakeholders worldwide
– household and firm survey respondents
sharing their successful and failed interactions
with the state
– experts in NGOs, private and public sectors
sharing their international perspective on
governance successes and failures
• WGI do not reflect official view of World Bank
– almost all data sources from outside WB
– WB allocation of IDA resources based on CPIA
6
Six Dimensions of Governance
Governance as the set of traditions and institutions by
which authority in a country is exercised-- specifically:
• The process by which those in authority are selected
and replaced
– VOICE AND ACCOUNTABILITY
– POLITICAL STABILITY & ABSENCE OF
VIOLENCE/TERRORISM
• The capacity of government to formulate and
implement policies
– GOVERNMENT EFFECTIVENESS
– REGULATORY QUALITY
• The respect of citizens and state for institutions that
govern interactions among them
– RULE OF LAW
7
– CONTROL OF CORRUPTION
Individual sources for constructing the WGI
• Cross-Country Surveys of Firms: Global Competitiveness
Survey, World Competitiveness Yearbook, BEEPS
• Cross-Country Surveys of Individuals: Gallup World Poll,
Global Corruption Barometer, Latinobarometro,
Afrobarometer
• Expert Assessments from Commercial Risk Rating
Agencies: Global Insight, Political Risk Services, BERI,
Economist Intelligence Unit, Merchant International Group,
IJET Travel Consultancy, Asia Risk Consultancy
• Expert Assessments from NGOs, Think Tanks: Reporters
Without Borders, Heritage Foundation, Freedom House,
Bertelsmann Foundation, Amnesty International, IREX,
Global Integrity, Binghamton University, International
Budget Project
• Expert Assessments from Governments, Multilaterals:
World Bank CPIA, EBRD, AFDB, ADB, State Department,8
OECD, IFAD
Building Aggregate Governance Indicators
• Use Unobserved Components Model (UCM) to
construct composite governance indicators, and
margins of error for each country
• Estimate of governance: weighted average of
observed scores for each country, re-scaled to
common units
• Weights are proportional to precision of
underlying data sources
• Precision depends on how strongly individual
sources are correlated with each other
• Margins of error reflect (a) number of sources in
which a country appears, and (b) the precision 9of
those sources
Why Aggregate Indicators of Governance?
• Governance viewed as a much broader notion than
‘formal rules of the game’
• Individual data sources provide a noisy “signal” of
broader concept of governance, e.g.:
– trust in police
→ RULE OF LAW
– freedom of press →VOICE & ACCOUNTABILITY
• Benefits of Aggregation--through U.C. Method
• aggregate indicators are more informative about broad
concepts of governance
• broader country coverage (than individual indicators)
• generate explicit margins of error for country scores
10
• less likelihood of outliers
Fully Transparent and Interactive
Access to Data
• WGI are built on 35 underlying data sources
– interactive access to 6 aggregate indicators
are virtually all underlying data sources at
www.govindicators.org
• Full disclosure of limitations of data
– all governance indicators are imprecise
– WGI quantify this imprecision with “margins
of error” i.e. likely range of governance
scores for each country
– when these overlap, differences are not
significant
11
Good
Governance
Control of Corruption
2.5
Selected Countries in 2007
Margins
of Error
Governance
Level
FINLAND
DENMARK
NEW ZEALAND
SINGAPORE
UNITED STATES
CHILE
FRANCE
URUGUAY
ESTONIA
BOTSWANA
ITALY
HUNGARY
SOUTH AFRICA
GREECE
NAMIBIA
BRAZIL
MEXICO
INDIA
MAURITANIA
CHINA
KENYA
PARAGUAY
VENEZUELA
CAMBODIA
TURKMENISTAN
ZIMBABWE
Poor
Governance
HAITI
MYANMAR
-2.5
SOMALIA
0
DISCLAIMER: The data and research reported here do not reflect the official views of the World Bank, its Executive Directors, or the countries they represent.
The WGI are not used by the World Bank Group to allocate resources or for any other official purpose.
Source for data: 'Governance Matters VII: Governance Indicators for 1996-2007’, by D. Kaufmann, A. Kraay and M. Mastruzzi, June 2008,
www.govindicators.org. Colors are assigned according to the following criteria: Dark Red: country is in the bottom 10th percentile rank (‘governance crisis’);
12
Light Red: between 10th and 25th percentile rank; Orange: between 25th and 50th percentile rank; Yellow, between 50th and 75th; Light Green between 75th and
90th percentile rank; and Dark Green: between 90th and 100th percentile (exemplary governance). Estimates subject to margins of error.
13
14
15
16
17
How has the Governance Matters Research
Project Evolved Since the 1990s
• greatly expanded set of data sources (13 in 1996,
35 in 2007)– improved precision of aggregate
indicators
• extensive analytic work on governance
measurement – rejecting perception biases, halo
effects, correlated perception errors
• WGI robust to alternative methodologies
• taking margins of error seriously
– most governance and other data sources
don’t even report margins of error
– no ‘horse races’ or superficial ‘top ten’ lists 18
Application 1:
Good Governance:
a ‘Luxury Good’ that only a rich country
can ‘afford’,
and,
does governance change very little in
the short term?
19
Good Governance Can Be Found At
All Income Levels
• Over a dozen developing countries rank better
than industrialized countries like Greece or
Italy in various dimensions of governance
• Examples include Slovenia, Chile, Botswana,
Estonia, Uruguay, Czech Republic, Hungary,
Latvia, Lithuania, Mauritius, and Costa Rica
20
Quality of Rule of Law, 2006
Source for map: 'Governance Matters VI: Governance Indicators for 1996-2006’, by D. Kaufmann, A.Kraay and M. Mastruzzi, June 2007 www.govindicators.org. Colors are assigned according to the following criteria: Dark Red: country is in the bottom 10th percentile rank
(‘governance crisis’); Light Red: between 10th and 25th percentile rank; Orange: between 25th and 50th percentile rank; Yellow, between 50th and
21
75th; Light Green between 75th and 90th percentile rank; and Dark Green: between 90th and 100th percentile (exemplary governance). Estimates
subject to margins of error.
22
Governance Success Stories in Short Term
• For instance, between 1998 and 2007 significant
improvements in:
– Voice and Accountability in Ghana, Indonesia,
Liberia, and Peru
– Political Stability/Absence of Violence in
Algeria, Angola, and Rwanda
– Control of Corruption in Serbia and Liberia
• But declines in other countries too (e.g.
Venezuela, Cote d’Ivoire and Zimbabwe)
– No evidence of global trends in governance
23
2
Changes in Voice & Accountability, 1998-2007
Major
Deterioration
Insignificant Change
(selected countries)
(selected
countries)
0
Major Improvement
SIERRA LEONE
GHANA
INDONESIA
LIBERIA
PERU
UKRAINE
BRAZIL
ARGENTINA
EL SALVADOR
ECUADOR
RUSSIA
TURKMENISTAN
VENEZUELA
ZIMBABWE
-2
BELARUS
(selected countries)
Changes were calculated on the basis of the differences in country estimates from 1998 and 2007. Classification for major deteriorations and improvements
24
were based on 75% confidence interval. Source for data: 'Governance Matters VII: Governance Indicators for 1996-2007’, by D. Kaufmann, A.Kraay and M.
Mastruzzi, June 2008 - www.govindicators.org
Changes in Rule of Law, 1998-2007
2
Major Deterioration
Insignificant Change
(selected countries)
(selected countries)
0
Major Improvement
LIBERIA
RWANDA
GEORGIA
SERBIA
ESTONIA
LATVIA
GHANA
MOZAMBIQUE
COLOMBIA
KAZAKHSTAN
TURKMENISTAN
BOLIVIA
VENEZUELA
ERITREA
-2
ZIMBABWE
(selected countries)
Changes were calculated on the basis of the differences in country estimates from 1998 and 2007. Classification for major deteriorations and improvements
25
were based on 75% confidence interval. Source for data: 'Governance Matters VII: Governance Indicators for 1996-2007’, by D. Kaufmann, A.Kraay and M.
Mastruzzi, June 2008 - www.govindicators.org
Application 2: Governance Matters?
• Governance Matters for Growth and
Development?
• Taking seriously the challenge of Causality
Direction
26
Governance Matters around the world:
The 300% ‘Dividend’
•
Large Development Dividend of Good Governance: a
one-standard-deviation improvement in governance
(say, in rule of law or in corruption control) raise
incomes per capita in a country by about 300% in longrun
•
But is such a decline in corruption unrealistically
large?: NO -- One S.D. is the difference from:
Somalia → Togo or Guinea-Bissau → Namibia or
Rwanda → Botswana or Portugal → Netherlands
or Sweden
•
The impact is from governance to incomes, and not
viceversa -- higher incomes alone will not
automatically do
27
Governance Matters -- The ‘Development Dividend’
Isolating Causality: From governance to income
log(Real GDP
Per Capita)
3
Kaufmann-Kraay (2002)
2
Alcala-Ciccone (2004)
Rodrik-Subramanian-Trebbi (2004)
1
Acemoglu-Johnson-Robinson (2000)
0
-3
-2.5
-2
-1.5
-1
-0.5
0
0.5
1
1.5
2
2.5
Rule of Law 2004
-1
-2
-3
28
Development Dividend From Good Rule of Law
$30,000
$3,000
$300
Low Rule of Law
Medium Rule of Law
High Rule of Law
29
Data Source for calculations: KK 2004. Y-axis measures predicted GDP per capita on the basis of Instrumental Variable (IV)
results for each of the 3 categories. Estimations based on various authors’ studies, including Kaufmann and Kraay.
Rule of Law is Associated with Lower Infant Mortality for the
Population, 2006
High
Infant Mortality (100K births)
160
Low
LBR
AGO
ZAR
120
NER
r = -0.66
GNB
GNQ
RWA
CIV TCD
CAF
GIN NGA
KHM
80
ZWE
MLI
BDI
ETH MWI
SWZ
MOZ
DJI ZMB
BFA
BEN
CMR
YEM
KEN PAK
TGO
TJK AZE
TKM
HTI MMR
SDN TMP
KGZ
UZB
PNG
LAO
KAZ
BGD
BOL
ERICOM
GMB
BWA
LSO
UGA
MRTSEN
TZA
STP MDG
GHA
GAB
NPL
MHL
GUY
BTN
IND
ZAF
KIR
NAM
PRK
40
VEN
0
-2.5
Low
-1.5
MNG
GEO
MAR
MDV
DZA
BLZ
SLB
GTM IRN
BRA
VUT
NIC
HND
SUR
IDN
ARM
TUR
CPV
LBN
DOM
PHL
CHN
EGY
SAM
SLV
JOR
MEX
ECU PER MDA
PRY
SAU
TUN
TON
FSM
PAN
TTOROM THAGRD
COL
VCTQAT
JAM
ALB
RUS LBY
ARG VNM
FJI
URY
UKR SYR
LCAATG
MKD
BIH
DMA MUS
SYC
BGR LKA
CRI
KWT
LVA
MYSOMN
BRB BHS
BLR
LTU BHRHUN
MLT USA
CHL
POL SVK
ADO
HRV
EST
CUB
ISR
KOR
LUX
AUS
AUT
NLD
NZL
CAN
CHE
ITA
GRC
FRAJPN
BEL IRL GBR
SVN CYP
PRTESP LIE
NOR
CZE
DEU
DNK
SWE
SGP
FIN
ISL
-0.5
0.5
Rule of Law
1.5
High
30
2.5
Source for Rule of Law: 'Governance Matters VI: Governance Indicators for 1996-2006’, by D. Kaufmann, A.Kraay and M. Mastruzzi, July 2007 - www.govindicators.org. Source
for infant mortality, WDI 2004.
Growth without Governance?
2.5
Causal Effect of Income on Governance
2
OLS Regression
1.5
Kaufmann-Kraay (2002)
1
MUS
BWA
Rule of Law Index, 2004
0.5
CPVNAM
ZAF
Rodrik-Rigobon (2004)
0
LSO
GHA
SEN
MWI
MDG
MLI GMB
BEN
TZA
-0.5
ZMB
STPDJI
MOZ
BFA MRT
UGA ERI
RWA
NER
ETH
-1
CMR
TGO KEN
COM
GNQ GIN
SLE
TCD
COG
GNB
AGO
CIV
CAF
NGA
BDI
ZWE -1.5
SDN
SYC
GAB
SWZ
ZAR
-2
y = 0.82x - 0.01
R2 = 0.68
-2.5
-3
-3
-2.5
-2
-1.5
-1
-0.5
0
0.5
ln(GDP Per Capita at PPP in 1996), Standardized
1
1.5
2
2
31
Growth without Governance?... When so, why?
• Well known: strong correlation between incomes and
the quality of governance across countries
• Income growth results in improved governance, or vv.?
• In Kaufmann & Kraay (2002) “Growth Without
Governance”, with WGI data, and I.V. (governance to
growth), and out of sample technique (growth to
governance), separated this correlation into: (1) a
strong positive causal effect from better governance to
higher incomes, & (2) a weak (& negative) causal effect
in the opposite direction, from incomes to governance.
• Hypothesis for explaining absence of positive feedback
from per capita incomes to governance: state capture,
where the fruits of growth accruing to elite which
benefits from low standards of public governance. 32
Application 3: State Capture Matters
• Empirics: using disaggregated/micro data is
also key – in-depth governance country
diagnostics, and enterprise surveys
• Does State Capture Matter?: various empirical
research strands
33
Unbundling Different Manifestations of Bribery, EOS 2006
% Firm Report High Bribery (1-3)
80
Utilities
Taxation
East Asia
NICs
Former
Soviet Union
Judiciary
Capture
Frequency of Bribery
Bribery in:
60
40
20
0
Low
Bribery
OECD
Latin
America
Newly
Accessed EU
34
Source: EOS firm survey, WEF2006. Questions: In your industry, how commonly firms make undocumented extra payments or bribes connected with permits /
utilities / taxation / awarding of public contracts / influence laws and regulations / judiciary? (common…never occurs).
‘Seize the State, Seize the Day’
Hellman, Jones and Kaufmann 2000
State Capture: firms shaping and affecting formulation
of the rules of the game through private payments
to public officials and politicians
Empirical Research with survey of firms (BEEPS) in
transition economies, where we found that:
• Large share of firms reported significant impact on
them by capture in many countries in the FSU (e.g.
Azerbaijan, Moldova, Russia, and Ukraine).
• Captor firms exhibited much higher output and
investment growth than non-captors (w/ controls)
• In contrast, coerced firms into paying bribes for
administrative corruption did not derive benefits
• Captor firms benefited in terms output and
investment, while the national rule of law and
35
protection of property rights was weakened
%of all Firms report negative impact of grand corruption
‘Seize the State, Seize the Day’ (Hellman, Kaufmann, Jones 2000): Differences
in Transition Countries on the Extent of State Capture
50 %
45 %
40 %
35 %
30 %
25 %
20 %
15 %
10 %
5%
0
Hungary
Estonia
Adverse Impact of ‘Purchases’ of:
Russia
Parliamentary
legislation
Azerbaijan
Decrees
Central Bank
36
Influence
‘Seize the State, Seize the Day’ (Hellman, Kaufmann, Jones 2000): Costs of
State Capture in Transition – Private Sector grows and invests less (&
Average rate of growth '97-'99
more insecure property rights)
30
Output
Investment
25
20
15
10
5
0
High capture
Countries
Low capture
countries
37
‘Capture’ literature beyond ‘Seize the State’
• Slinko, I., E. Yakovlev, and E. Zhuravskaya (2004)
“Laws for Sale: Evidence from Russia”, in American Law
and Economics Review 7, pp. 284-318
• Empirical exploration on link between capture and growth
in Russia,
• Construct measures of the political power of firms and
regional capture with micro-data on preferential
treatment of firms through regional laws and regulations
in Russia during 1992-2000.
• Findings: little evidence that overall capture affects
overall growth, but instead sizeable distributional impact:
politically powerful and well connected firms perform
better and benefit from capture, while regulatory capture
hurts the performance of politically unconnected firms.
38
Inequality of Influence, ‘Crony Bias’
•
•
•
•
•
•
In Hellman et al (2002), “The Inequality of Influence”, we developed
proxy measure of the inequality of influence (‘crony bias’ in political
system) from Business Environment and Enterprise Performance
Survey (BEEPS); 6,500 firms in 27 transition countries.
We found that inequality of influence has a strongly negative impact on
the firm’s assessments of public institutions, which in turn affects the
behavior of firms towards those institutions.
Crony bias: associated w/ negative assessment of fairness and
impartiality of courts, enforceability of court decisions, and firms
reporting prevalence of crony bias are use courts less to resolve
business disputes.
Firms affected by crony bias have less secure property rights than
influential firms. Crony bias is also associated with lower levels of tax
compliance and significantly higher levels of bribery.
Inequality of influence not only damages the credibility of institutions
among weak firms, but affects the likelihood that they will use and
provide tax resources to support such institutions. By withholding tax
revenues, paying bribes, and avoiding courts, these firms perpetuate
the weakness of such state institutions, making the subject to capture
by the more influential.
The inequality of influence thus appears to generate a self-reinforcing
dynamic in which institutions are subverted further strengthening the
39
underlying political and economic inequalities
Does Capture & Undue Influence matter for success
of investment projects in emerging economies?
•
Utilizing recent data on evaluation of completed projects funded by
the World Bank, implemented by countries (IEG project
performance data): 3,340 projects during 1995-2007, in 92 countries.
Variables: Project Performance; Institutional development Impact
• For capture and undue influence, data from the annual EOS firm
survey by the World Economic Forum. Main Variables:
State Capture: Do illegal payments to influence government policies,
laws or regulations impose costs or otherwise negatively affect
your company?
Regulatory / Policy Capture: In your country, how frequently would
you estimate that firms like yours make undocumented extra
payments or bribes connected with import and export permits?
Judiciary Capture: In your country, how frequently would you
estimate that firms like yours make undocumented extra payments
or bribes connected with obtaining favorable judicial decisions?
Undue Political Influence: How much influence do you think
individuals or firms with close personal ties to political leaders
actually had on recently enacted national laws and regulation that
have a substantial impact on your business?
40
• Basic statistics, and firm-level OLS with of controls, other bribery
Summary Statistics: 3,340 projects during 1995-2007
Range
Mean
Standard
Deviation
1 – 6 (best)
4.23
1.18
World Bank IEG
0 – 1 (best)
0.78
0.41
World Bank IEG
1 – 4 (best)
2.43
0.72
World Bank IEG
1 – 7 (best)
3.72
0.74
World Economic Forum Survey
1 – 7 (best)
3.74
0.70
World Economic Forum Survey
1 – 7 (best)
3.97
0.87
World Economic Forum Survey
1 – 7 (best)
3.41
0.78
World Economic Forum Survey
1 – 7 (best)
3.77
0.91
World Economic Forum Survey
1 – 7 (best)
3.51
0.58
World Economic Forum Survey
LOG
3.47
0.35
WDI 2008
%
3.46
4.24
WDI 2008
%
1.95
1.01
WDI 2008
Variable
Project Performance Rating
Satisfactory Project
Institutional Development Impact
State Capture
Regulatory / Policy Capture
Bribery for Permits
Corruption in Procurement
Judiciary Capture
Undue Political Influence
GDP per capita (PPP)
3-year Growth Rate
Inflation Rate
Regional Dummy
Sector Dummy
Project Type Dummy
Year Dummy
Latitude
Source
0-1
…
0-1
World Bank IEG
0-1
World Bank IEG
0-1
World Bank IEG
-90 - 90
Wikipedia
41
Investment Project Performance and Capture
4.80
4.60
4.40
4.20
4.00
3.80
State Capture
3.60
Regulatory / Policy Capture
3.40
Bottom quartile
Middle-Low Quartile
Middle-High Quartile
Top quartile
Sources: World Economic Forum’s Global Competitiveness Survey, 1996-2007 and World Bank Business Warehouse. Number of projects evaluated: 3,340 for
1995-2007 for 92 countries. State Capture: Do illegal payments to influence government policies, laws or regulations impose costs or otherwise negatively
42
affect your company? Regulatory / Policy Capture: Question: In your country, how frequently would you estimate that firms like yours make undocumented
extra payments or bribes connected with import and export permits?
Project Performance and Capture –Results for
Regulatory / Policy Capture Corruption Regression
Dependent variable: Project Rating (1-6)
Independent Variables
Regulatory / Policy Capture
1
2
3
4
5
6
0.19
0.15
0.18
0.19
0.17
0.28
6.31***
4.42***
2.62***
3.31***
3.09***
4.06***
-0.03
-0.03
-0.04
0.66
0.56
0.63
Procurement Bribery
-0.03
0.49
Bribery for Permits
Inflation Rate (log)
GDP per capita (PPP, log)
Growth Rate (past 3 years)
-0.13
-0.13
-0.13
-0.13
-0.15
4.95***
5.15***
5.19***
4.95***
3.83***
0.07
0.06
0.07
0.07
-0.28
0.62
0.54
0.62
0.66
1.57
0.01
0.01
0.00
1.01
0.95
0.13
Latitude
0.40
1.21
Dummies
Y
Y
Y
Y
Y
Y
Observations
3241
3092
3098
3098
3092
1830
Adjusted R-squared
0.07
0.08
0.08
0.08
0.08
0.08
Sources: World Economic Forum’s Global Competitiveness Survey, 1996-2007 and World Bank Business Warehouse for World
43
Bank-funded projects. Each specificationincluded regional, project type, project sector and year dummies.
T-stats in italics; * significant at 10% level; ** significant at 5% level; *** significant at 1% level
Proje ct Pe rformance Rating v s. State Capture
P erfo rm an ce R atin g
6
5
5
4
r = 0.48
4
3
2
3
4
5
6
State Capture
Sources: World Economic Forum’s Global Competitiveness Survey, 1996-2007 and World Bank Business Warehouse. Number of projects evaluated:44
3,340 for
1995-2007 for 92 countries. State Capture: Do illegal payments to influence government policies, laws or regulations impose costs or otherwise negatively
affect your company?
Investment Project Performance & Capture
Satis factory
4.75
Poor Governance
Medium Governance
High Governance
Project Rating
4.50
4.25
4.00
3.75
Uns atis factory State Capture
Regulatory /
Policy capture
Judiciary Capture
Undue Political
Influence
Note: Top tercile refers to good governance ratings; bottom tercile refers to poor governance ratings. Sources: World Economic Forum’s Global Competitiveness Survey, 1996-2007 and World Bank
Business Warehouse for World Bank-funded projects. Number of projects evaluated: 3,340 during 1995-2007 for 92 countries. State Capture: Do illegal payments to influence government
policies, laws or regulations impose costs or otherwise negatively affect your company? Regulatory / Policy Capture: In your country, how frequently would you estimate that firms like yours make
undocumented extra payments or bribes connected with import and export permits? Judiciary Capture: In your country, how frequently would you estimate that firms like yours make undocumented
extra payments or bribes connected with obtaining favorable judicial decisions? Undue Political Influence: How much influence do you think individuals or firms with close personal ties to political
leaders actually had on recently enacted national laws and regulation that have a substantial impact on your business? Share of Successful Projects measures the percentage of projects with a
rating of 5 or 6 (out of 6 possible ratings). Share of Failed Projects measures the percentage of projects with a rating of 1, 2 or 3 (out of 6 possible ratings). Share of Low institutional
45
Project Performance & Undue Political
Influence Corruption Variable
Dependent variable: Project Rating (1-6)
Independent Variables
Undue Political Influence
1
2
3
4
5
6
0.22
0.17
0.11
0.13
0.13
0.27
5.55***
4.00***
2.17**
2.78***
2.79***
4.27***
0.06
0.05
0.06
1.80*
1.69*
1.22
Procurement Bribery
0.07
1.95*
Bribery for Permits
Inflation Rate (log)
GDP per capita (PPP, log)
Growth Rate (past 3 years)
-0.14
-0.14
-0.13
-0.13
-0.15
5.54***
5.56***
5.24***
4.94***
3.93***
0.16
0.12
0.10
0.11
-0.26
1.44
1.04
0.92
0.94
1.48
0.01
0.01
0.01
1.38
1.35
0.80
Latitude
0.42
1.26
Dummies
Y
Y
Y
Y
Y
Y
Observations
3241
3092
3098
3098
3092
1830
Adjusted R-squared
0.07
0.08
0.08
0.08
0.08
0.08
Sources: World Economic Forum’s Global Competitiveness Survey, 1996-2007 and World Bank Business Warehouse for World
46
Bank-funded projects. Each specificationincluded regional, project type, project sector and year dummies.
T-stats in italics; * significant at 10% level; ** significant at 5% level; *** significant at 1% level
TABLE 1: Capture & Project Performance in Developing Countries
# of Projects
Percentage of
Successful
Projects
Percentage of
’Failed’ Projects
Percentage of Low
Institutional Development
Impact Projects
Top Tercile
1097
63
16
44
Middle Tercile
1134
52
23
55
Bottom tercile
1069
48
26
58
# of Projects
Percentage of
Successful
Projects
Percentage of
‘Failed’ Projects
Percentage of Low
Institutional Development
Impact Projects
Top Tercile
1125
62
16
43
Middle Tercile
999
51
23
55
Bottom tercile
1134
50
26
58
# of Projects
Percentage of
Successful
Projects
Percentage of
‘Failed‘ Projects
Percentage of Low
Institutional Development
Impact Projects
Top Tercile
1094
62
16
44
Middle Tercile
1081
52
23
55
Bottom tercile
1125
49
26
57
# of Projects
Percentage of
Successful
Projects
Percentage of
‘Failed‘ Projects
Percentage of Low
Institutional Development
Impact Projects
Top Tercile
1149
61
16
45
Middle Tercile
997
53
23
52
Bottom tercile
1112
49
27
58
State Capture
Regulatory / Policy capture
Judiciary Capture
Undue Political Influence
47
Note: Top tercile refers to good governance ratings; bottom tercile refers to poor governance ratings. Sources: World Economic Forum’s Global Competitiveness Survey, 1996-2007 and World Bank Business Warehouse for World Bank-funded projects. Number of projects evaluated: 3,340 during 1995-2007 for 92 countries. State Capture: Do illegal payments to influence government
policies, laws or regulations impose costs or otherwise negatively affect your company? Regulatory / Policy Capture: In your country, how frequently would you estimate that firms like yours make undocumented extra payments or bribes connected with import and export permits? Judiciary Capture: In your country, how frequently would you estimate that firms like yours make undocumented
extra payments or bribes connected with obtaining favorable judicial decisions? Undue Political Influence: How much influence do you think individuals or firms with close personal ties to political leaders actually had on recently enacted national laws and regulation that have a substantial impact on your business? Share of Successful Projects measures the percentage of projects with a
rating of 5 or 6 (out of 6 possible ratings). Share of Failed Projects measures the percentage of projects with a rating of 1, 2 or 3 (out of 6 possible ratings). Share of Low institutional development impact Projects measures the percentage of projects with a development impact rating of 1 or 2 (out of 4 possible ratings)
Application 4:
Revisiting Historical Origins Determinism?
• Search for fundamental antecedents to
institutional outcomes today: important
• But deeper empirical scrutiny raises questions
• Statistical Significance with worldwide data?:
Somewhat
• Magnitude Large with worldwide data?: No
• Significant and sizeable effects for emerging
economies: not really…
48
Governance and Legal Origin, World (POP>1M)
2
Governance Rating
Common Law
R = 0.10
R = 0.10
Civil Law
R = -0.04
R = 0.11
R = 0.09
R = 0.04
Political
Stability / No
violence
Rule of Law
Regulatory
Quality
Voice &
Accountability
0
-2
Control of
Corruption
Government
Effectiveness
Source for data: 'Governance Matters VI: Governance Indicators for 1996-2006’, by D. Kaufmann, A.Kraay and M. Mastruzzi, July 2007 www.govindicators.org. and Djankov et al (2005).
49
Governance and Legal Origin, Emerging Economies (POP>1M)
2
Governance Rating
Common Law
R = 0.00
R = 0.02
Civil Law
R = -0.10
R = 0.03
R = 0.00
Political
Stability / No
violence
Rule of Law
Regulatory
Quality
R = 0.01
0
-2
Control of
Corruption
Government
Effectiveness
Source for data: 'Governance Matters VI: Governance Indicators for 1996-2006’, by D. Kaufmann, A.Kraay and M. Mastruzzi, July 2007 www.govindicators.org. and Djankov et al (2005).
Voice &
Accountability
50
Rule of Law & Legal origin, Emerging Economies (POP>1M)
Today
History
Bottom Half WGI
Rule of Law 2006
Top Half in WGI Rule
of Law 2006
Common
Law Origin
swaziland jamaica liberia sierra
leone papua new guinea somalia
zambia zimbabwe nepal kenya
sudan nigeria bangladesh pakistan
trinidad and tobago gambia
botswana lesotho namibia united
arab emirates malawi sri lanka ghana
saudi arabia malaysia uganda
tanzania south africa thailand india
Civil Law
Origin
guinea-bissau albania congo central
african republic eritrea Nicaragua
libya laos togo paraguay Honduras
burundi haiti bolivia guinea chad
guatemala ecuador niger cambodia
angola cameroon Ivory coast yemen
venezuela iraq peru afghanistan
algeria colombia myanmar Congo
DR iran ethiopia indonesia
mauritius gabon kuwait oman
mauritania panama lithuania uruguay
lebanon puerto rico costa rica jordan
el salvador benin dominican republic
rwanda tunisia senegal burkina faso
mali chile madagascar syria
mozambique romania morocco
argentina turkey egypt vietnam
philippines mexico brazil
51
Source for data: 'Governance Matters VI: Governance Indicators for 1996-2006’, by D. Kaufmann, A.Kraay and M. Mastruzzi, July 2007 www.govindicators.org. and Djankov et al (2005). Full Sample included 98 countries. Top half/Common Law=16; Top Half/Civil Law:33; Bottom half/Common
Law=14; Bottom Half/Civil Law: 35 countries.
Some Implications…
• Empirics key to test theories and popular notions
• Measurement difficult, but feasible–w/ emphasis on
margins of error, triangulation, and caution
• World on average not improving on governance
• But variance in levels and in trends is very large,
and countries can improve in short term
• Governance Matters for growth & development -- but
not v.v.?
• Micro-level empirical analysis needs to complement
aggregate cross-country regressions (endogeneity &
causality direction)
• State Capture, Crony Bias & Undue Influence Matter
• Fundamental & Historical Factors do matter, but not
deterministic: proximate and current determinants
matter as well--policy/institutional change can affect
development outcomes in shorter term
52
• Transparency, civil liberties, free press
Some Conclusions – for debate
1. Governance can be measured and empirically analised
2. Not relying on one single source – WGI & others: triangulation
and aggregation, while also using disaggregated sources, &
taking margins of error seriously
3. Acknowledging & Learning from large Country and Institutional
Variation – in Africa & other regions
4. Less focus on symptoms and superficial reforms by ‘fiat’
5. Addressing capture, patronage & systemic corruption -- &
focus on deeper incentive-driven reforms: raise the cost
6. Focus on fudamentals?: Contestable Politics, Competition,
Transparency, ‘Voice’ & Free Press; Judiciary Reforms
7. Limits to Incrementalism: some shocks? …& Leadership..
8. Revisiting Role of Donors, Privates & Multinationals
9. A modicum of modesty…
53
Concretely: 10 Transparency Reform Components
1. Public Disclosure of Assets & Incomes of Candidates, Public
Officials, Politicians, Legislators - & dependents
2. Public Disclosure of Political Campaign contributions by
individuals and firms, and of campaign expenditures
3. Public Disclosure of Parliamentary Votes, w/out exceptions
4. Effective Implementation of Conflict of Interest Laws,
separating business, politics, legislation, & government
5. Publicly blacklisting firms bribing in public procurement
6. Effective Implementation of Freedom of Information Law,
with easy access to all to government information
7. Fiscal/Financial transparency: EITI, budgets, ROSCs
8. E*procurement: transparency (web) and competition
9. Media Freedoms & Media Development
10. Country Diagnostic (& Scorecard) on Governance & A-C
54
Freedom of the Press is associated with better Control of
Corruption (& civil liberties more generally is associated with better performance of
World Bank-funded projects – see WBER article 1997)
Good
Control of Corruption
1.51.0-
r = .66
0.50-0.5-1.0-1.5- _________________________________________________________________
Press Freedom
Status:
Not Free of
Partially Free
Free
Source for control of corruption: : 'Governance Matters V: Governance Indicators for 1996-2005’, D. Kaufmann, A. Kraay and M. Mastruzzi, September
55
2006 (http://www.govindicators.org/). Source for Press Freedom: 2006 Freedom House’s Press Freedom Report. Terciles divided according to Press
Freedom ratings (190 countries total). Free: 0-30 (69); Partly Free: 31-60 (54); Not Free: 61-100 (67).
Some food for thought…
“The oligarchs were so called because they had real power,
state power. They wrote laws. They appointed ministers,
often entire cabinets, and made sure that their interests
were served. They corrupted the new governing,
legislative and bureaucratic class of Russia, in the centre,
in the regions and abroad…” John Lloyd, in the
Financial Times, August 5th, 2000
Fast Forward: April 2004 in the US, a meeting in a
basement of the SEC in New York…
http://thekaufmannpost.net/capture-and-the-financial-crisis-anelephant-forcing-a-rethink-of-corruption/#more-282
Fast Rewind: The Chef of the Presidential ‘Pink
House’ “ ‘Have dinner menus here always been the same?, ’asked
a key aide to Menem’s to his chef, at the Argentinian presidential
residence. 'The menus change, the presidents change. What never
changes are the dinner guests', retorted the presidential chef,
referring to the cadre of businessmen who frequented the residence.
El Octavo Circulo, by Cerruti and Ciancaglini
In
56
Entrenched Corruption Networks:
The Case on Montesinos in Peru
Judiciary
Civil Society
International
Political Parties
Legislative
Branch
Alberto
Fujimori
1
State
(Bureaucracy)
Vladimiro
Montesinos
Media
Private
Sector
Municipal
Government
Source: “Robust Web of Corruption: Peru’s Intelligence Chief Vladimiro Montesinos,” Kennedy
School of Government Case Program, Case C14-04-1722.0, based on research by Professor Luis
Moreno Ocampo; Peru: Resource Dependency Network, 2000
Military
57
From Time Magazine, based on the WGI world map on Voice &
Accountability
58
•Source How Free is Your Country? Time Magazine, Issue of 17 September 2007, p. 16
Bribery vs. ‘Legal Corruption’
% Firms report 'corruption'
% Firms in country/region report corporate bribery vs. ‘Legal corruption’
80
Nordic Countries
60
G-7
40
East Asia
'Tigers' (NICs)
20
United States
0
Corporate Bribery
Corporate
"Legal Corruption"
59
Source: Author’s calculations, EOS 2004. http://siteresources.worldbank.org/EXTWBIGOVANTCOR/Resources/ETHICS.xls
Data for Analysis & Informing Policy Advice,
Not for Precise Rankings
Any data on Governance, Institutions, and
Investment Climate is subject to a margin of error. It
is not intended for precise country rankings, but to
highlight relative strengths and weaknesses and
draw analytical and policy lessons. The data and
indicators do not necessarily reflect official views on
rankings by the World Bank or its Board of
Directors. The collaboration and inputs from Aart
Kraay, Massimo Matruzzi, Joel Hellman and others is
appreciated. Errors are responsibility of the author.
Further materials & access to interactive data:
Biblio (SSRN):
http://papers.ssrn.com/sol3/cf_dev/AbsByAuth.cfm?per_id=163813
60
Data: www.govindicators.org