Content uploaded by Pablo Zoido
Author content
All content in this area was uploaded by Pablo Zoido on Aug 10, 2015
Content may be subject to copyright.
American Economic Review, May 1998
Regulatory Discretion and the Unofficial Economy
By SIMON JOHNSON, DANIEL KAUFMANN, AND PABLO ZOIDO-LOBATÓN *
Abstract
Johnson, Kaufmann, and Shleifer (1997) find that the share of the
unofficial economy in GDP is determined by the extent of control
rights held by politicians and bureaucrats in post-communist
economies. Exploring in more detail the role of bribes and using a
broader data set from the OECD, Latin America, and transition
economies, we find that the unofficial economy accounts for a larger
share of GDP when there is more corruption and when the rule of law
is weaker. While these findings are consistent with the earlier results
for transition economies, in the larger country sample we find it is not
necessarily the case that more regulation or higher taxes directly
increases the size of the unofficial economy. The problem appears to
be not regulation or taxation per se, but whether the state
administrative system can operate without corruption. A high level of
regulatory discretion helps create the potential for corruption and
drive firms into the unofficial economy.
* MIT, The World Bank, and The World Bank. Simon Johnson gratefully acknowledges support from the
Entrepreneurship Center at MIT. We thank Kenneth Sokoloff, Jean-Laurent Rosenthal, Andrei Shleifer and Norman
Loayza for discussions and suggestions. Authors are responsible for the paper’s views, errors and omissions; and
views expressed do not necessarily reflect those of the affiliated institutions. The presentation of indices here does not
constitute an endorsement by the authors of any individual country rating.
Politicization of economic activity means the exercise of control rights over firms by
politicians and bureaucrats. In most countries politicians maintain property rights in firms, typically
in the form of residual control rights as defined by Sanford Grossman and Oliver Hart (1986).
These control rights may have served an ideological agenda in the past, but they are often used to
further the private agenda of politicians and bureaucrats. A recent literature has established the
presence of these problems in countries as diverse as Peru, France, Russia and Ukraine (Hernando
de Soto, 1989, Andrei Shleifer and Robert Vishny 1993 and 1994, Kaufmann and Paul Siegelbaum
1997, Shleifer 1997). But how widespread are these rights and how damaging are their effects
around the world?
The usual presumption in the economics literature is that a predatory government simply
leads to lower total economic activity, but for Eastern Europe and the former Soviet Union since
1989, Johnson et al. (1997) showed that businesses have responded to politicization by going
"underground." Instead of registering their activities, managers prefer not to pay taxes and not to
benefit from key publicly provided services such as legal enforcement of contracts. For these
economies in transition from communism, there is evidence of a downward spiral, in which firms
leaving the official sector reduce state revenue, which reduces publicly provided services, and further
reduces the incentive to register in the official sector.
1
Most of the former Soviet Union has thus
ended up in a "bad" equilibrium with low tax revenue, high unofficial economy as a percentage of
GDP, and low quality of publicly provided services.
This previous work on transition economies suggests that, while formal rules may matter in
some instances, what really matters is how regulations and tax rules are actually implemented. If the
rules are fine on paper but officials have a great deal of discretion in interpretation and
implementation, this leads to a higher effective burden on business, more corruption, and a greater
incentive to move to the unofficial economy. This general idea leads to three specific propositions.
First the share of the unofficial economy in GDP should be higher when there is more regulation
and more discretion for officials regarding how the regulatory system operates. Second, the
unofficial economy should be larger when there is a bigger tax burden on firms in the official
sector, where "burden" on the firm is the outcome of how the tax system is administered as well as
what the rates are. Third, a larger unofficial economy should be correlated with weaker publicly
provided services, as measured by corruption and the "rule of law" (particularly the legal protection
provided to private-sector business investments).
This paper finds support for these propositions in a broad set of countries for which there
exist at least roughly comparable estimates of the unofficial economy in the 1990's. We have
measures for the unofficial economy for 49 countries in three regions of the world: Latin America,
the OECD, and the former Soviet bloc. A different methodology is used for each region, but the
numbers appear to be comparable; see Johnson et al. (1998) for the detailed estimates. The sample
for our regressions varies between 32 and 49 countries, depending on the coverage of right-hand-
side variables. We have not found comparable data for the unofficial economy in East Asia or for
Africa, so these countries are excluded from the regressions. We use Brazil and Russia as
illustrative regional benchmarks throughout and also report on OECD-specific countries where
relevant.
I. Regulation and Bureaucracy
The Heritage Foundation's measure of regulation is higher, on a scale of 1 to 5, for
countries that had regulations that are worse for business in 1996 (Brian Johnson and Thomas
Sheehy, 1997). This measure includes both the formal rules and the way they are enforced. The
Czech Republic actually receives the top score; it is the only country in our sample to get a perfect
1. Most OECD countries score 2. Russia scores 4, while Brazil scores 3. Table 1 shows that a one-
point increase in this index is associated with a 14.7-percentage-point increase in the share of the
unofficial economy. Controlling for log GDP per capita reduces the coefficient on the regulation
variable to 8.1, but it remains significant.
The Global Competitiveness Survey reports results from a 1997 survey of executives on the
extent of regulatory discretion and lax enforcement of rules, on a scale of 1 to 7 (World Economic
Forum, 1997). Russia has the lowest score of 2.01, while Brazil rates better with 3.46. Most of the
OECD countries score 4.5 or higher; Switzerland has the highest score with 5.64 in our sample.
Singapore had the highest score worldwide in the survey, with 6.36. Table 1 shows that a one-point-
higher score for this index is correlated with a 9.2-percentage-point fall in the share of the unofficial
economy. However, this measure is not significant once we control for log GDP per capita.
The 1997 International Country Risk Guide of Political Risk Services measures expert
opinion of "bureaucratic quality" on a scale of 1 to 6, where a higher score means that bureaucrats
operated in a more efficient and predictable way between 1990 and 1997. Guatemala and Panama
have the lowest score of 1.44; Russia scores 3.19; and Brazil scores 4.0. The best OECD countries,
such as the United Kingdom, score 6.0. Table 1 shows that a one-point increase in this index
implies an 8.5-percentage-points decrease in the share of the unofficial economy. Controlling for
log GDP per capita reduces the coefficient only slightly to -7.7, and it remains highly significant.
Freedom House's 1995—1996 measure of economic freedom is higher for countries with
"better" regulation (i.e., more pro-business), on a scale of 0 to 16, in 1995-96 (Richard E. Messick,
1996). The United Kingdom, the United States, Denmark, Sweden, and Holland tie for top position
with a score of 16, while Azerbaijan has the lowest score of 1. Russia and Brazil score 7. Table 1
shows that a one-point increase in this scale is associated with a 2.5-percent fall in the share of the
unofficial economy, but this coefficient loses significance when we control for GDP per capita.
In summary, we find strong evidence that less regulation (i.e., a regulatory regime that is
more business-friendly and presumably represents less political control rights) is correlated with a
lower share of the unofficial economy. However, countries with a higher income level also have a
lower level of the unofficial economy, so when we control for income level two out of four
regulation variables become insignificant at the 5-percent level. The effect of bureaucratic quality
and the way regulations are administered appear to be particularly strong. This supports the idea
that regulatory discretion is an importan
t cause of unofficial activity.
II. Taxation
The 1997 Global Competitiveness Survey rates tax burden from the firm's standpoint; a
higher score was given when executives considered the tax system to be better for business, on a
scale of 1 to 7 (World Economic Forum, 1997). This measure captures not just tax rates, but also
the way the tax system is administered (e.g., if tax officials abuse higher levels of discretion, this
would likely translate into a worse score). Ukraine has the lowest score in our sample, with 1.58,
and the United Kingdom has the highest score with 4.60. Russia scores 1.80, and Brazil scores
2.22. A one-point increase in this variable reduces the share of the unofficial economy by 11.7
percentage points. Controlling for log GDP per capita reduces the coefficient to -6.5 but it remains
significant.
The Fraser Institute measure of top marginal tax rates is higher for countries that had lower
tax rates, on a scale of 1—10, in 1995 (James Gwarney and Robert Lawson, 1997). In this case the
index captures formal rates, but not the way the system is administered. The "best" tax rates are in
seemingly unlikely places: Bolivia and Uruguay both score a perfect 10.
2
The worst (i.e., highest)
tax rates are in Italy, Belgium, Sweden, Denmark, and Romania, all of which score the lowest
attainable value of 1. The United States scores 7, and the United Kingdom scores 5, while Russia
and Brazil both score 8. Chile scores 4, which is the best in Latin America. Table 1 shows that a
one-point increase in this index is actually associated with a 3.5-percentage-point increase in the
share of the unofficial economy (i.e., countries with lower marginal tax rates actually have a larger
share of the unofficial economy). Controlling for log GDP per capita reduces the coefficient on this
index to 1.9, but it remains significant.
The contrast between the results of these two tax variables points to the importance of how
the tax and regulatory system operates, rather than the nature of the formal rules. Countries with
high marginal tax rates but a low tax burden (as evaluated by executives) actually have a low share
of the unofficial economy as a percentage of GDP (e.g., Scandinavia; see Fig. 1). Russia has
relatively low marginal tax rates but was rated with a high tax burden because of the way the tax
system operates, and thus it is associated with a relatively high share of the unofficial economy in
GDP.
III. Rule of Law and Corruption
Political Risk Services' 1996 International Country Risk Guide contains a "rule-of-law
index" which is higher where the law and order tradition was stronger during 1990—1997, on a
scale of 0—6. The United States and several other OECD countries achieve the highest level of 6.
In our sample, Colombia has the lowest score of 1.4. Russia scores 3.5, and Brazil scores 3.4.
Table 2 shows that a one-point increase in the value of this index is associated with a 10.6-
percentage-point fall in the share of the unofficial economy. In this case log GDP per capita is not
significant, and including this control variable reduces the estimated coefficient on the index only to
-9.3.
The Heritage Foundation's index of property rights is lower where property rights were
more secure, on a scale of 1—5, in 1996 (Johnson and Sheehy, 1997). The only non-OECD
country to score a perfect 1 is Chile. Four previously communist countries have the worst score of
4: Romania, Ukraine, Georgia, and Azerbaijan. Russia and Brazil score 3. Table 2 shows that a one-
point increase in this index is associated with a 13.4-percent increase in the share of the unofficial
economy. Controlling for log GDP per capita reduces the coefficient to 8.0, but it remains
significant.
In the Fraser Institute measure of "Equality of Citizens Under the Law and Access of
Citizens to a Non-discriminatory Judiciary," a higher score means a "better" legal system during
1995—a996, on a scale of 0—10 (Gwarney and Lawson, 1997). Only Belgium, Holland, Sweden,
Norway, Denmark, and Switzerland get the top score of 10. Italy, the United Kingdom and the
United States score 7.5. Russia scores 2.5, and Brazil scores 0.
3
Table 2 shows that a one-point
increase in this index implies a 3.8-percentage-point fall in the unofficial economy's share of total
GDP. Controlling for log GDP per capita reduces the coefficient to -2.3, but it remains significant.
The extended Transparency International measure of corruption, prepared by Johan G.
Lambsdorff (1998), scaled 0—10, covers 43 of the countries in our sample for 1997.
4
It is higher
for countries with less corruption. In our sample, Denmark has the highest score with 9.94 and
Bolivia has the lowest in our sample with 2.05. Russia scores 2.27 while Brazil scores 3.56. The
best Latin American country is Chile with 6.05. In Table 2 a one-point increase in this index implies
a 5.1 percentage point fall in the unofficial economy, and a 3.5-percentage-point fall when the log
GDP per capita control is included.
In the Global Competitiveness Survey measure of bribery, scaled 1—7, a higher score
means less corruption in 1997 (World Economic Forum, 1997). Among countries for which we
also have data on the unofficial economy, the highest score is Sweden with 6.61. The lowest scores
(under 3) are for several Central American countries, as well as Russia which scores 2.72. Brazil
scores 3.75. Table 2 shows that a one-point increase in this index implies a reduction in the share of
the unofficial economy by 8.0 percentage points (without the control variable) and by 3.9
percentage points (if we control for log GDP per capita).
In the Impulse index of corruption, a higher score means more corruption (Peter Neumann,
1994).
5
Russia and Brazil are both awarded 4 out of 5. The best score of 0 is awarded to the usual
OECD countries plus Lithuania. As usual, Chile is the best-ranked Latin American country,
awarded a score of 1. As Table 2 shows, a one-point increase in this index is associated with a 1.7-
percentage-point increase in the share of the unofficial economy. However, controlling for GDP per
capita reduces the coefficient by more than half and makes it significant only at the 10-percent level.
In summary, the relationship between share of the unofficial economy and rule of law
(including corruption) is strong and consistent across seven different measures. Countries with
more corruption have higher shares of the unofficial economy (see Fig. 2). This is true even when
we control for income level.
IV. Conclusion
The model of Johnson et al. (1997) has three predictions that find support in the available
cross-country data. First, countries with more regulation tend to have a higher share of the
unofficial economy in total GDP. Second, a higher tax burden, as perceived by business, leads to
more unofficial activity. Third, countries with more corruption tend to have a larger unofficial
economy.
This evidence suggests, although it does not prove, that the extent of regulatory discretion is
a key determinant of underground activity. Lax regulations in settings with undisciplined
bureaucracies and weak rule of law allow officials to decide individual cases without effective
supervision. This creates conditions ripe for corruption (see Kaufmann and Jeffrey Sachs, 1998).
Under such circumstances, many firms choose to operate underground.
REFERENCES
Political Risk Services, International Country Risk Guide. New York, 1997.
de Soto, Hernando. The other path. New York: Harper and Row, 1989.
Grossman, Sanford J. and Hart, Oliver D. "The Costs and Benefits of Ownership: A Theory of
Vertical and Lateral Integration." Journal of Political Economy, August 1986, 94 (4), pp. 691-719.
Gwarney, James and Lawson, Robert, eds. Economic freedom of the world, 1997 annual
report. Vancouver, BC: Fraser Institute, 1997.
Johnson, Bryan and Sheehy, Thomas. Index of economic freedom 1997. Washington, DC:
Heritage Foundation, 1997.
Johnson, Simon; Kaufmann, Daniel and Shleifer, Andrei. "The Unofficial Economy in
Transition." Brookings Papers on Economic Activity, Fall 1997, (2), pp. 159-239.
Johnson, Simon: Kaufmann, Daniel and Zoido-Lobaton, Pablo "Corruption and the
Unofficial Economy." Unpublished manuscript, World Bank, Washington, DC, 1998.
Kaufmann, Daniel; and Sachs, Jeffrey "Determinants of Corruption." Unpublished manuscript,
Harvard University, Cambridge, MA., May 1998.
Kaufmann, Daniel and Siegelbaum, Paul. "Privatization and Corruption in Transition
Economies." Journal of International Affairs, Winter 1997, 50(2), pp. 419-58.
Lambsdorff, Johan G. "Corruption in Comparative Perception" in: The Economics of
Corruption, A.K. Jain ed., Kluwer Academics, 1998.
Loayza, Norman V. "The Economics of the Informal Sector: A Simple Model and Some
Empirical Evidence from Latin America." Carnegie-Rochester Conference Series on Public Policy,
December 1996, 45, pp. 129-62.
Messick, Richard E. World survey of economic freedom, 1995-1996. New York: Freedom
House, 1996.
Neumann, Peter. "Bose: Fast Alle Bestechen" ["Flaunting the Rules: Almost Everybody"].
Imouise. 4 January 1994, 4, pp. 5-14.
Shleifer, Andrei. "Schumpeter Lecture: Government in Transition." European Economic Review,
April 1997, 41(3-5), pp. 385-410.
Shleifer, Andrei and Vishny, Robert W. "Corruption" Quarterly Journal of Economics, August
1993, 198(3), pp. 599-617.
________. "Politicians and Firms." Quarterly Journal of Economics, November 1994, 109(4), pp.
995-1025.
World Economic Forum. Executive survey. Global competitiveness report. Geneva, Switzerland:
World Economic Forum, 1997.
* MIT, The World Bank, and The World Bank. Simon Johnson gratefully acknowledges support
from the Entrepreneurship Center at MIT. We thank Kenneth Sokoloff, Jean-Laurent Rosenthal,
Andrei Shleifer and Norman Loayza for discussions and suggestions. Authors are responsible for
the paper's views, errors and omissions; and views expressed do not necessarily reflect those of the
affiliated institutions. The presentation of indices here does not constitute an endorsement by the
authors of any individual country rating.
1
Norman Loayza (1995) has similar theoretical results for Latin America. In his model
unregistered firms use but do not pay for public services, thus leading to congestion costs for
public goods, such as roads, and lower growth.
2
Bolivia's recent tax reform is presumably reflected in this rating.
3
In most Asian countries, this index is highly correlated with measures of corruption. Thus Hong
Kong and Korea score 7.5 on this Fraser Institute measure, while Thailand, Malaysia, and
Indonesia score 2.5. Singapore is again an anomaly because it scores 0 on this measure, despite
having very little corruption.
4
This index requires that countries have had only 2 (rather than 4) surveys. Even in the extended
sample, apart from Hong Kong and Singapore, all the other countries that score above 6.5 are long
standing democracies.
5
Among the 103 countries surveyed, the worst score is awarded to Bangladesh, Myanmar (Burma),
Indonesia, Iran, Nigeria, Pakistan, the Philippines, and Thailand.
TABLE 1—REGRESSIONS OF UNOFFICIAL ECONOMY (AS % OF GDP)
ON MEASURES OF REGULATION
Independent Variable Dependent Variable: Unofficial Economy
Log GDP per capita -7.4* -7.4* -1.0 -7.4* -7.3* -7.0*
[1.6] [2.3] [2.9] [2.0] [1.5] [1.6]
Measures of Regulation
Regulation
b
14.7* 8.1*
[2.5] [2.6]
Regulatory Discretion
a
-9.2* -2.9
[1.7] [2.5]
Bureaucratic Quality
a
-8.5* -7.7*
[1.0] [2.3]
Economic Freedom
a
-2.5* -0.8
[0.5] [0.6]
Measures of Taxation
Tax Burden
a
-11.7* -6.5*
[2.4] [2.1]
Tax Rates
a
3.5* 1.9*
[0.7] [0.7]
R-Squared 0.43 0.62 0.47 0.60 0.65 0.65 0.38 0.54 0.43 0.68 0.37 0.57
Number of Observations 47 47 34 34 39 39 42 42 34 34 42 42
Notes: Standard errors are in parentheses; * denotes significant at 5% level and ** denotes significant at 10% level. a)
A higher value for this variable stands for a better score for private business. b) A higher value for this variable
stands for a worse score for private business.
TABLE 2—REGRESSION OF UNOFFICIAL ECONOMY
ON LEGAL ENVIRONMENT AND CORRUPTION
Independent Variable Dependent Variable: Unofficial Economy
Log GDP per capita -1.9 -4.8** -5.2* -4.0* -5.8* -6.5*
[1.7] [2.6] [1.9] [2.3] [2.5] [1.9]
Legal Environment
ICRG Rule of -10.6* -9.3*
Law Index 90-97
a
[1.0] [1.5]
Property Rights
b
13.4* 8.0*
[1.8] [3.4]
Equality of Citizens -3.8* -2.3*
Before the Law
a
[0.6] [0.8]
Corruption
Transparency -5.1* -3.5*
International (extended)
a
[0.7] [1.1]
World Economic Forum
a
-8.0* -3.9*
[1.3] [2.1]
Impulse's Exporter 1.7* 0.8**
Bribery Index
b
[0.4] [0.4]
R-squared 0.77 0.78 0.55 0.58 0.53 0.60 0.57 0.60 0.55 0.62 0.36 0.50
Number of Observations 39 39 47 47 43 43 43 43 34 34 44 44
Notes: Standard errors are in parentheses; * denotes significant at 5% level and ** denotes significant at 10% level. a)
A higher value for this variable stands for a better score for private business. b) A higher value for this variable
stands for a worse score for private business.
FIGURE 1. UNOFFICIAL ECONOMY AND CORRUPTION
Notes: Unofficial Economy estimates from Johnson, Kaufmann and Zoido-Lobatón (1998), and Corruption Index
from Lambsdorff (1998).
FIGURE 2. UNOFFICIAL ECONOMY AND TAX BURDEN ON INDIVIDUAL FIRMS
Notes: Unofficial Economy estimates from Johnson, Kaufmann and Zoido-Lobatón (1998), and Tax Burden on
Individual Firms from World Economic Forum (1997).
VE N
URY
PER
PAN
MEX
HND
GTM
ECU
CRI
CO L
CHL
BRA
BO
L
ARG
USA
SWE
PRT
NOR
NLD
JPN
ITA
IR
L
GRC
GBR
FRA
ESP
DNK
DEU
CH E
CAN
BEL
AUT
UKR
SVK
RUS
RO M
POL
LV A
HUN
EST
CZE
BLR
BG R
0%
10%
20%
30%
40%
50%
60%
70%
Corruption inde
x
Unofficial Economy (% GDP)
LO W CORRUPTION HIGH CORRUPTION
VEN
PER
MEX
HND
GTM
CRI
COL
CHL
BRA
AR G
US A
SWE
PRT
NOR
NLD
JPN
ITA
IRL
GRC
GBR
FRA
ESP
DNK
DEU
CHE
CAN
BEL
AU T
UKR
SVK
RUS
PO L
HUN
CZE
0%
10%
20%
30%
40%
50%
60%
70%
Tax Burden on Individual Firms
Unofficial Economy (% GDP)
H
IGH BURDEN
L
OW BUR DEN
