Transcript Mushtaq Khan - The challenge of anti-corruption

The Challenge of Anti-Corruption
Conference on Corruption and Anti-Corruption: Threats, Challenges and
Opportunities
Sussex Centre for the Study of Corruption, Transparency International
Clifford Chance, London 9th September 2014.
Mushtaq H. Khan,
SOAS,
UNIVERSITY OF London
Corruption in Developing Countries
• Corruption appears to be endemic across developing countries
• Anti-corruption efforts do not appear to have lasting effects
• The relationship between corruption and economic
performance at the aggregate level appears to be relatively
weak
• One interpretation is that corruption is very difficult to fight and
the gains are uncertain.
• An alternative interpretation is that we have ignored differences
in types of corruption and our instruments do not identify or
target the most damaging types of corruption
• This is true for much of the ‘good governance’ reform policies
that target property right stability, the rule of law, political
accountability, AND anti-corruption
2
Cross-Country Evidence
Composite Good Governance Indicators and Per Capita GDP
1990/2000
Logs of per capita GDP in US$ 2000
100000
10000
1000
100
10
0
10
20
30
40
50
Average of World Bank Good Governance Indicators
in 1990 (ranges from 0 to 50)
3
Cross-Country Evidence
Composite Good Governance Indicators and Growth Rates
1990-2003
Growth Rate of Per Capita GDP 1990-2003
10
8
6
4
2
0
-2
-4
-6
-8
0
10
20
30
40
50
Average of World Bank Good Governance indicators
in 1990 (ranges from 0 to 50)
Diverging Developing Countries
4
Cross-Country Evidence
Composite Good Governance Indicators and Growth Rates
1990-2003
Growth Rate of Per Capita GDP 1990-2003
10
8
6
4
2
0
-2
-4
-6
-8
0
10
20
30
40
50
Average of World Bank Good Governance indicators
in 1990 (ranges from 0 to 50)
Advanced Countries
Diverging Developing Countries
5
Cross-Country Evidence
Composite Good Governance Indicators and Growth Rates
1990-2003
10
Growth Rate of Per Capita GDP 1990-2003
8
6
4
2
0
-2
-4
-6
-8
0
10
Advanced Countries
20
30
Converging Developing Countries
40
50
Average of World Bank Good Governance Indicators
in 1990 (ranges from 0 to 50)
Diverging Developing Countries
6
Corruption and Growth
Corruption and Growth 1990-2003
(using Knack's IRIS data)
Growth Rate of Per Capita GDP 1990-2003
10
8
6
4
2
0
-2
-4
-6
-8
-1
0
1
2
3
4
5
6
7
IRIS Corruption Index 1990
(ranges from 0 to 6)
Advanced Countries
Converging Developing Countries
Other Developing Countries
7
Growth Rates
Growth-Enhancing Governance versus Good Governance
2. Converging Developing Countries
Targeted Anti-Corruption and Governance
Capabilities trigger and sustain growth
1. Diverging Developing Countries
n
e ssi o
r
g
e
R
Line
3. Advanced Capitalist
Countries
Reforms suggested
by Good Governance
Regression Line
Good Governance Score
(Democracy, Corruption,
Stability of Property Rights)
Four types of Corruption
Corruption is always associated with state functions where a public official
is engaged in illegal activities: But the state functions themselves may be
legal and/or necessary functions, or not
Anti-corruption strategy that does not distinguish between different types of
9
corruption to target the most important and feasible types is likely to fail
Feasibility versus Impact of Anti-Corruption Strategies
State-constraining corruption is likely to be an important focus for
effective and feasible anti-corruption activities in developing countries
10
An Example of the Policy Design Problem from Bangladesh:
Regulatory Failure in the Bangladeshi Garments Industry
(Drawn from a cross-country study on The Impact of Corruption on Private Sector
Growth conducted for DFID 2014)
Another Example of State-Constraining Corruption: Customs
Corruption in the Bangladeshi Garments Industry
(Drawn from a cross-country study on The Impact of Corruption on Private Sector
Growth conducted for DFID 2014)
A Process-Analysis Approach to Anti-Corruption
• The targeting of state-constraining corruption is critically important in
developing countries
• Corruption is in general ‘over-determined’: multiple processes are
involved with firms engaging in corruption for different reasons
• In the worst cases, almost all firms are engaged in corruption, and in
these cases, everyone pays lip service to anti-corruption but standard
anti-corruption strategies have no chance
• Effective strategies require careful ‘process analysis’ to separate the
determinants of corruption for potentially compliant and potentially
non-compliant firms
• Necessary state functions have to be redesigned to remain
developmental and ensure that it becomes feasible for the potentially
compliant to become compliant and even to benefit from compliance
• Regulatory and enforcement capacities are likely to need
strengthening but since resources are limited, policy has to
incrementally target critical agencies to achieve specific objectives13