Corruption in transition: reflections & implications from governance empirics

Daniel Kaufmann, Brookings Institution

Presentation at the opening plenary session on Measurement & Consequences of Corruption at the Economic Research Forum (ERF) 18th Annual

Conference, Cairo, Egypt, March 25th, 2012

Issues in brief

• Corruption: not subject to measurement?

• Is it all-pervasive in emerging economies?

• The crucial determinant of under-development?

• Fighting corruption by fighting corruption?

• Democratic transitions lead to control of corruption?

• Does International Donor Assistance help

democratic transitions & anti-corruption?

• Insidious form of political corruption: State Capture

Rules Outcomes

Broad Specific Broad Specific

Experts Lawyers DB

Commercial Risk Rating Agencies

EIU, PRS, WMO, PRC

Non-Governmental Organizations

GII FRH/CCR GII

Governments & Multilaterals AfDB, ASD, CPIA IFD

Academics IPD

Survey Respondents Firms GCS, WCY,

WBES, ES, BEEPS

Individuals AFR, LBO, GWP, VAB WVS, GCB

Aggregate Indicators Combining Respondents

TI, WGI, IIAG

A Taxonomy of Corruption Indicators: Who (reports data) & What Type? -- rules vs outcomes-based

Whose Opinion? About What?

Measurement is possible: corruption and beyond, with caution

• Many measures of corruption exist today

• Different characteristics, virtues and challenges

• Advantages and Disadvantages of aggregating and constructing composites

• The WGI in brief

WGI: The Six Indicators of Governance Governance: Set of institutions by which authority

in a country is exercised – specifically: • Political cluster: the process by which those in authority

are selected and replaced 1. VOICE AND DEMOCRATIC ACCOUNTABILITY

2. POLITICAL STABILITY & ABSENCE OF VIOLENCE/TERRORISM

• Economic cluster: the capacity of government to implement policies and provide public services 3. GOVERNMENT EFFECTIVENESS

4. REGULATORY QUALITY

• Institutional cluster, the respect for institutions governing interactions among citizens & the state 5. RULE OF LAW

6. CONTROL OF CORRUPTION 6

Main Features of the WGI • Data on six dimensions of governance covering 213 countries

over the period 1996-2010

• Synthesis of hundreds of underlying indicators taken from over 30 different organizations & data sources

• Aggregate and individual indicators interactively available at www.govindicators.org, large database

• Unobserved Component Model (UCM): for aggregation

• Result of longstanding research project, featuring the “Governance Matters” series, plus new analytical report, by D. Kaufmann, A. Kraay & M. Mastruzzi: “Worldwide Governance Indicators: Methodology & Analytical Issues”

• Transparency & precision about imprecision: full disclosure of limitations of data: the WGI quantify this imprecision w/ “margins of error” (likely range of governance scores for each country)

7

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, Economist Intelligence Unit, IJET Travel, Asia Risk Consultancy

• Expert Assessments from NGOs, Think Tanks: Reporters Without Borders, Heritage Foundation, Freedom House, Bertelsmann Foundation, Amnesty International, IREEP, IREX, Global Integrity, Binghamton University, International Budget Project

• Expert Assessments from Governments, Multilaterals: World Bank CPIA, EBRD, AFDB, ADB, State Dept., IFAD 8

-2.5

-1.5

-0.5

0.5

1.5

2.5

SOM

ALI

A

ZIM

BA

BW

E A

NG

OLA

IR

AQ

LI

BYA

V

ENEZ

UEL

A

RU

SSIA

SY

RIA

IR

AN

EC

UA

DO

R

DO

MIN

ICA

N R

EP.

PAR

AG

UA

Y

IND

ON

ESIA

C

HIN

A

EGY

PT

IND

IA

BO

LIV

IA

AR

GEN

TIN

A

ALB

AN

IA

CO

LOM

BIA

M

EXIC

O

MO

RO

CC

O

TUN

ISIA

G

REE

CE

TUR

KEY

B

RA

ZIL

SOU

TH A

FRIC

A

SAU

DI A

RA

BIA

S.

KO

REA

ES

TON

IA

BO

TSW

AN

A

UA

E U

SA

UK

C

HIL

E Q

ATA

R

DEN

MA

RK

Source: “WGI: A Summary of Data, Methodology and Analytical Issues,” by D. Kaufmann, A.Kraay and M. Mastruzzi, October 2011 – www.govindicators.org; Dark red indicates countries in the bottom 10th percenti8le rank; light red between 10th and 25th percentile rank; orange between 25th and 50th; yellow between 50th and 75th; light green between 75th and 90th and Dark green between 90th and 100th percentile (exemplary governance)

Governance Level

Margin of Error

9

Good Corruption Control

Control of Corruption in Select Countries, 2010

Source: WGI: A Summary of Data, Methodology and Analytical Issues,” by D. Kaufmann, A.Kraay and M. Mastruzzi, October 2011 – www.govindicators.org 90th-100th percentile

75th-90th percentile

50th-75th percentile

25h-50th percentile

10th-25th percentile

0th-10th percentile 10

Figure 3. Control of Corruption in the Arab world, 2010

Does Governance & Corruption Matter?

12

The 300% Development Dividend From Improving Governance & Controlling Corruption

High Corruption Medium Corruption Low Corruption

$300

$3,000

$30,000

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.

Control of Corruption and WEF GCR Competitiveness, 2010

Source: EOS firm survey, WEF Global Competitiveness Survey 2011 & '“Worldwide Governance Indicators,” by D. Kaufmann, A.Kraay and M. Mastruzzi, October 2011 – www.govindicators.org

BRA

CHL

CRI

DOM

SLV

JAM

0

20

40

60

80

100

0 20 40 60 80 100

Glo

bal

Co

mp

etit

ive

ne

ss In

dex

, 20

11

Control of Corruption, 2010

r = 0.79

High

High Low

Low

Corruption: fundamental driver, symptom, or proximate cause?

16

Voice and Accountability

Source: WGI: A Summary of Data, Methodology and Analytical Issues,” by D. Kaufmann, A.Kraay and M. Mastruzzi, October 2011 – www.govindicators.org.

90th-100th percentile

75th-90th percentile

50th-75th percentile

25h-50th percentile

10th-25th percentile

0th-10th percentile

18

Freedom of the Press associated w/ better Control of Corruption (& civil liberties more generally is associated with better performance of World Bank-funded

projects – see WBER article 1997)

Source: WGI: A Summary of Data, Methodology and Analytical Issues,” by D. Kaufmann, A.Kraay and M. Mastruzzi, October 2011 – www.govindicators.org. & Freedom House, “Press Freedom”, 2011. Terciles according the Press Freedom criteria – free press (rating of 30 or below), partly free (ratings between 30 and 60) and not free (rating above 60).

-1.0

-0.5

0.0

0.5

1.0

Not Free Partly Free Free

WG

I Co

ntr

ol o

f C

orr

up

tio

n, 2

01

0

Good

Effect of Media Freedom on Control of Corruption (when Low Rule of Law)

-2.0

-1.0

0.0

1.0

2.0

Low Press Freedom/Low Rule of Law High Press Freedom/Low Rule of Law

Sources: Kaufmann D., A. Kraay, M. Mastruzzi (2010), The Worldwide Governance Indicators: Methodology and Analytical Issues , September 2010; Freedom House, Freedom of the Press, 2005-2009 editions

Effect of Media Freedom and Rule of Law on Control of Corruption

-2.0

-1.0

0.0

1.0

2.0

Low Press Freedom/Low Rule

of Law

High Press Freedom/Low Rule

of Law

Low Press Freedom/Hight Rule

of Law

High Press Freedom/High Rule

of Law

Sources: Kaufmann D., A. Kraay, M. Mastruzzi (2010), The Worldwide Governance Indicators: Methodology and Analytical Issues , September 2010; Freedom House, Freedom of the Press, 2005-2009 editions

Democratic Transitions, and Corruption

‘Net Democracy’ Scores Pre-Transition & during Initial Transition Average Polity Scores for 91 countries

-10

-5

0

5

10

T-5

T-4

T-3

T-2

T-1

T

Source: Polity IV Project: Political Regime Characteristics and Transitions, 1800-2010. In T-5 the sample consists of 92 countries.

Transition Period

‘Net Democracy’ Scores Pre-, During & Post-Transition Average Polity Scores for 77 countries

-10

-5

0

5

10

T-5

T-4

T-3

T-2

T-1

T

T+1

T+2

T+3

T+4

T+5

T+6

T+7

T+8

T+9

T+1

0

T+1

1

T+1

2

T+1

3

T+1

4

T+1

5

T+1

6

T+1

7

Source: Polity IV Project: Political Regime Characteristics and Transitions, 1800-2010. In T-5 the sample consists of 92 countries.

Transition Period

‘Net Democracy’ Scores for Transitioning Countries, by Democratic Performance Groups (Polity)

Source: Polity IV Project: Political Regime Characteristics and Transitions, 1800-2010. There are 44 countries in the improving/performing category; 31 in the stagnating category and 17 in the deteriorating category

-10

-5

0

5

10

T-5

T-4

T-3

T-2

T-1

T

T+1

T+2

T+3

T+4

T+5

T+6

T+7

T+8

T+9

T+1

0

T+1

1

T+1

2

T+1

3

T+1

4

T+1

5

T+1

6

T+1

7

Improving/ Performing (44)

Stagnating (31)

Deteriorating (17)

T

T+7

T+15

Transition Period

Institutional Performance: Select Countries from Improving/Performing Group, 1985-2010

-10

-5

0

5

10

1985 1990 1995 2000 2005 2010

Chile

Slovenia

Liberia

Source: Polity IV Project: Political Regime Characteristics and Transitions, 1800-2010.

Institutional Performance: Select Countries from Stagnating Group, 1985-2010

-10

-5

0

5

10

1985 1990 1995 2000 2005 2010

Russia

Ethiopia

Tanzania

Togo

Source: Polity IV Project: Political Regime Characteristics and Transitions, 1800-2010.

Institutional Performance: Select Countries from Deteriorating Group, 1985-2010

Source: Polity IV Project: Political Regime Characteristics and Transitions, 1800-2010.

-10

-5

0

5

10

1985 1990 1995 2000 2005 2010

Venezuela

Congo (Brzz)

Iran

How Well Do the ‘Democratizing’ Transition do in Rule of Law & Control of Corruption (WGI 2010, by Polity Performance Category)

0

20

40

60

80

100

Deteriorating Stagnating Improving/Performing

Rule of Law Control of Corruption

Polity Group::

Sources: Polity IV Project: Political Regime Characteristics and Transitions, 1800-2010; WGI: A Summary of Data, Methodology and Analytical Issues,” by D. Kaufmann, A.Kraay and M. Mastruzzi, October 2011 – www.govindicators.org

Probability of (Relatively) Successful Control of Corruption performance in 2010, by Polity-Democratic Performance Category

0%

20%

40%

60%

80%

100%

Deteriorating Stagnating Improving/Performing

Control of Corruption

Sources: Polity IV Project: Political Regime Characteristics and Transitions, 1800-2010; WGI: A Summary of Data, Methodology and Analytical Issues,” by D. Kaufmann, A.Kraay and M. Mastruzzi, October 2011 – www.govindicators.org

The International Community, Aid Effectiveness, and Transition

Voice & Accountability in the Arab World, 2010

Source: WGI: A Summary of Data, Methodology and Analytical Issues,” by D. Kaufmann, A.Kraay and M. Mastruzzi, October 2011 – www.govindicators.org 90th-100th percentile

75th-90th percentile

50th-75th percentile

25h-50th percentile

10th-25th percentile

0th-10th percentile

0 10 20 30 40 50 60 70 80 90 100

LIBYA

SYRIA

SAUDI ARABIA

IRAN

TUNISIA

YEMEN

EGYPT

DJIBOUTI

OMAN

ALGERIA

QATAR

UAE

JORDAN

MOROCCO

KUWAIT

LEBANON

32

Voice & Accountability in Arab World: 2000 (bottom bar) & 2010 (top bar)

Source: WGI: A Summary of Data, Methodology and Analytical Issues,” by D. Kaufmann, A.Kraay and M. Mastruzzi, October 2011 – www.govindicators.org 90th-100th percentile

75th-90th percentile

50th-75th percentile

25h-50th percentile

10th-25th percentile

0th-10th percentile

0 10 20 30 40 50 60 70 80 90 100

LIBYA

SYRIA

SAUDI ARABIA

IRAN

TUNISIA

YEMEN

EGYPT

DJIBOUTI

OMAN

ALGERIA

QATAR

UAE

JORDAN

MOROCCO

KUWAIT

LEBANON

33

0

20

40

60

80

100

0

2

4

6

8

10

2000 2002 2003 2004 2005 2006 2007 2008 2009 2010

WG

I VA

(P

erc

en

tile

Ran

k)

Ne

t O

DA

(B

illio

ns

USD

)

Arab World ODA (excl. Iraq) Arab World V&A

Figure 1. Evolution of Development Assistance versus Voice and Democratic Accountability in the Arab region, 2000-2010

Note: WGI = Worldwide Governance Indicators; V&A: Voice and (democratic) Accountability; ODA = official development assistance. Iraq is excluded, representing a special case. Sources: D. Kaufmann, A. Kray and M. Mastruzzi, “The Worldwide Governance Indicators: Methodology and Analytical Issues,” 2011, at www.govindicators.org; Organization for Economic Cooperation and Development, Development Assistance Committee Database.

Share of Aid to Recipients, By Recipient Category (1960-2010)

Source: Polity IV Project: Political Regime Characteristics and Transitions, 1800-2010; OECD, DAC, 2012

0%

20%

40%

60%

80%

100%

Fully Democratic Aid

Stagnating-Democratic Aid

Non-Democratic Aid

35

Aid per Capita To Transitioning Countries, by Performance Group (3-Year Moving Averages)

Source: Polity IV Project: Political Regime Characteristics and Transitions, 1800-2010; OECD, DAC, 2012

0

5

10

15

20

25

30

t-5

t-

4

t-3

t-

2

t-1

t t+

1

t+2

t+

3

t+4

t+

5

t+6

t+

7

t+8

t+

9

t+1

0

t+1

1

t+1

2

t+1

3

t+1

4

t+1

5

t+1

6

t+1

7

t+1

8

Aid

pe

r C

apit

a (U

SD)

Deteriorating

Stagnating

Improving/ Progressing

No Transition (All Years)

No Transition (2010)

36

State Capture & Legal Corruption: A Worldwide Challenge

Bribery vs. ‘Legal Corruption’ (& soft Capture), 2004

0

20

40

60

80

Corporate Bribery Corporate "Legal Corruption"

% F

irm

s re

po

rt 'c

orr

up

tio

n'

Nordic Countries

G-7

East Asia 'Tigers' (NICs)

United States

Source: Author’s calculations based on EOS 2004.

% Firms report ‘corruption’

Bribery vs. ‘Legal Corruption’ (& soft Capture), 2004

0

20

40

60

80

Corporate Bribery Corporate "Legal Corruption"

% F

irm

s re

po

rt 'c

orr

up

tio

n'

Nordic Countries

G-7

East Asia 'Tigers' (NICs)

United States

Source: Author’s calculations based on EOS 2004.

% Firms report ‘corruption’

40

Six Implications - as contributions to debate

1. Governance can be measured: triangulation; transparency

about imprecision, & interpretative caution

2. Governance and Corruption Matters

3. Aid strategies ought to support democratic governance

4. Transitions can succeed or fail, and controlling corruption is

key, and difficult

5. But one does not ‘fight corruption by fighting corruption’:

voice & democratic accountability, gender rights,

transparency, rule of law, procurement systems, etc.

6. State Capture as a form of high level corruption &

misgovernance