Transcript Growth and Corruption: What are the stylized facts in

Growth and Corruption:
What are the stylized facts in
ECA’s transition countries?
Pradeep Mitra
Chief Economist
Europe and Central Asia Region
Prepared for a Panel Discussion on “The Growth-Corruption Paradox”, in a Core
Course on Public Sector Governance & Anti-Corruption, February 14-16, 2006
I thank Jim Anderson and Alexander Murevyev for helping put this presentation
together. Views expressed are mine and do not necessarily represent those of the
World Bank.
Growth in GDP per capita from 1998 to 2003 owes more to
productivity growth than improved labor utilization
(Growth in GDP/POP) = (Growth in GDP/EMPL) + (Growth in EMPL/POP)
15%
GDP/EMP
EMP/POP
GDP/POP
10%
5%
0%
-5%
Turkey
Macedonia, FYR
Czech Republic
Poland
Slovak Republic
Slovenia
Croatia
Kyrgyz Republic
Romania
Moldova
Hungary
Bulgaria
Lithuania
Belarus
Estonia
Georgia
Ukraine
Albania
Latvia
Armenia
Kazakhstan
Azerbaijan
-15%
Russian Federation
-10%
2
Firm entry and exit are comparatively more important
for productivity growth in transition countries
3
While the business environment has been improving steadily
in the transition countries, it is in general more difficult in
2005 than in the cohesion countries of Western Europe
3.5
3.0
2.5
2.0
1.5
1.0
Regulation
Labour
Taxation
1999
2002
Institutions
2005
Infrastructure
Cohesion countries 2005
Note: 1 represents no obstacle; 4 represents a major obstacle
Finance
Macro
instability
4
Unbundling by ownership category reveals that de novo firms find
the business environment in 2005 to be more difficult than
privatized and state firms in areas such as regulation, taxation and
institutions and property rights . . . . .
3.0
2.5
2.0
1.5
1.0
de novo
state & privatized
de novo
Regulation
state & privatized
de novo
Taxation
Cohesion countries
New EU member states
state & privatized
Institutions
SEE
CIS
5
. . . . particularly with respect to regulatory constraints such
as customs and trade regulations and business licensing and
permits, but note this is true in the cohesion countries too
2.5
2.0
1.5
1.0
de novo
state & privatized
Customs and trade regulations
Cohesion countries
de novo
state & privatized
Business licensing and permits
New EU member states
SEE
CIS
6
. . . . as well as institutional constraints such as the
judiciary and corruption
3.0
2.5
2.0
1.5
1.0
de novo
state & privatized
de novo
state & privatized
Judiciary
Cohesion countries
Corruption
New EU member states
SEE
CIS
7
Moving from entrant firms to potential exiters, while the
fraction of firms with arrears and subsidies has been falling in
the transition countries, indeed lower than the cohesion
countries for subsidies . . .
0.30
0.20
0.10
0.00
1999
2002
2005
2002
2005
Subsidies
Cohesion countries
Arrears
New EU member states
SEE
CIS
8
. . . . subsidies to a higher fraction of state and
privatized firms retard their exit
0.5
0.4
0.3
0.2
0.1
0
Cohesion countries
New EU member states
state & privatized
SEE
de novo
CIS
9
. . . . and arrears by state and privatized firms on
utility payments and taxes as well as to employees
and suppliers also retard exit
0.3
0.2
0.1
0.0
Cohesion countries
New EU member states
state & privatized
SEE
de novo
CIS
10
Unsurprisingly, various measures of administrative corruption
are related to changes in the business environment and show
improvement, e.g. bribes as a proportion of sales, the more
rapid the improvement where corruption was higher . . . .
1.0
LowIncome
CIS
Southeast
Europe
Bribe Tax (log)
0.5
MiddleIncome
CIS
2002
2005
0.0
EU-8
-0.5
-1.0
EU Cohesion
-1.5
5
6
7
8
9
10
11
GDP per capita (log)
Source: BEEPS 2002, 2005.
Notes: Low-Income CIS icnludes Arm, Aze, Geo, Kyr, Mol, Taj, Uzb; Middle-Income CIS includes Bel, Kaz, Rus, Ukr; Southeast Europe includes Alb,
BiH, Bul, Mac, Rom, SAM; EU-8 includes Cze, Est, Hun, Lat, Lit, Pol, Slk, Sln; EU Cohesion includes Gre, Ire, Por, Esp. Income levels based on
previous full year (2001, and 2004).
Source for this and next two slides: Anticorruption in Transition 3 (forthcoming)
11
. . . . but progress is more mixed when corruption is
20%
15%
Transition Countries:
decrease between
2002 and 2005
Transition Countries:
level in 2005
10%
Transition Countries:
increase between 2002 and
2005
5%
Comparator Countries:
level in 2005/2004
Source: BEEPS 2002, BEEPS 2005
Gov. Contracts
Taxes
Bus. Licenses
Customs
Courts
Fire and Blg Insp.
Occ. Health and
Safety Insp.
Env. Insp.
Utilities
0%
Influence Laws
average percent of firms indicating type of bribery is frequent
unbundled, although the deterioration for courts and public
procurement is not statistically significant.
12
Frequency of bribes in dealing with courts is increasing in all
sub-regions but not significantly so. But the gap is
significant between EU-8 and cohesion countries
Frequency of Bribery at Courts
2.3
2.0
2002
Southeast
Europe
2005
1.8
LowIncome
CIS
1.5
MiddleIncome
CIS
EU Cohesion
EU-8
1.3
5
6
7
8
9
10
11
GDP per capita (log)
Source: BEEPS 2002, 2005.
Notes: Low-Income CIS icnludes Arm, Aze, Geo, Kyr, Mol, Taj, Uzb; Middle-Income CIS includes Bel, Kaz, Rus, Ukr; Southeast Europe includes Alb,
BiH, Bul, Mac, Rom, SAM; EU-8 includes Cze, Est, Hun, Lat, Lit, Pol, Slk, Sln; EU Cohesion includes Gre, Ire, Por, Esp. Income levels based on
previous full year (2001, and 2004).
13
Conclusions
Rapid growth in the ECA transition countries since
1998 owes more to productivity growth than
improved labor utilization.
Firm entry and exit play a more important role in
productivity growth in the transition countries
compared to other emerging and OECD countries.
Notwithstanding improvements in the business
environment throughout the region, de novo firms in
transition and cohesion countries confront a tilted
playing field in areas such as regulation, taxation, the
judiciary and corruption, while exit is retarded by
softer budget constraints for state and privatized firms
compared to de novo firms.
14
Conclusions
While aggregate measures of corruption show
improvement, unbundling confirms the same trend
for most measures but no significant improvement in
the frequency of bribery in government procurement
and in dealing with courts.
An operationally-oriented discussion of growth and
corruption requires both to be unbundled, e.g. (i)
what are the sources of productivity growth, and (ii)
along which dimensions is corruption the most
serious, so as to guide policy.
15