The New Economics of Corruption: A Survey and some New Results
| Author | Rafael di Tella,Alberto Ades |
| Published date | 01 August 1997 |
| Date | 01 August 1997 |
| DOI | http://doi.org/10.1111/1467-9248.00093 |
| Subject Matter | Article |
The New Economics of Corruption:
a Survey and some New Results
ALBERTO ADES AND RAFAEL DI TELLA
Introduction
Governments of all political colours in countries of all levels of wealth are
aected by corruption scandals with a frequency and intensity that seems to be
always on the increase. Corruption has become a `hot issue' and is now a major
topic of political competition, even in the industrialized democracies. Yet, to a
large extent, economists have remained vague about what can be done to reduce
it. A main concern is the lack of evidence in support of the main policy
alternatives. This paper reviews the state of economic knowledge on the
phenomenon of corruption, with special emphasis on the theory behind the
most common policy proposals and the evidence that supports them.
During the last thirty years or so economists from dierent ®elds have made
scattered contributions to the analysis of corruption, so that by now a certain
body of literature has emerged. The ®rst published piece on corruption that
received wide attention is Rose-Ackerman,1 though the topic was also in the
minds of people doing research in the economics of crime,2 agency theory,3
rent-seeking,4 and development economics.5 While theories abounded, a lack of
data on corruption to test the theoretical contributions allowed con¯icting
theories on the causes and consequences of corruption to coexist. As a result,
the ®eld has been unable to provide coherent policy guidelines to curb
corruption and has remained somewhat disconnected from the discussion
arena, a place mainly dominated by lawyers, businessmen and judges.
More recently, an emerging body of empirical research has begun to
appear, employing subjective indices on corruption produced for business-
related purposes. We organize the literature into two broad themes: theories
about the causes of corruption and theories of its eects. With regard to the
latter, we show how the new data has shed light on a controversy that
originated in the 1960s about the theoretical possibility that corruption may
We are grateful to Guillermo Mondino for helpful comments and suggestions. We wish to
acknowledge ®nancial support from the FundacioÂn MediterraÂnea.
1 Susan Rose-Ackerman, `The economics of corruption', Journal of Public Economics, 4(2)
(1975), 187±203.
2 Particularly Gary Becker and George Stigler, `Law enforcement, malfeasance and the
compensation of enforcers', Journal of Legal Studies, 3(1) (1974), 1±19.
3 M. Harris and A. Raviv, `Some results on incentive contracts with applications to education
and employment, health insurance, and law enforcement', American Economic Review, 68 (1978),
20±30.
4 Gordon Tullock, `The welfare costs of taris, monopolies and theft', Western Economic Journal
(now Economic Enquiry), 5 (1967), 224±32.
5 Gunnar Myrdal, Asian Drama: an Inquiry into the Poverty of Nations, vol. II (New York,
Pantheon, 1968).
# Political Studies Association 1997. Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main
Street, Malden, MA 02148, USA.
ALBERTO ADES AND RAFAEL DI TELLA
497
be bene®cial to investment and growth in countries with particularly obtru-
sive bureaucracies.
With respect to the theories on the causes of corruption, we present the
theoretical background and empirical performance of some of the most voiced
policy proposals. Broadly speaking, there are three types of policy proposals
aimed at curbing corruption, which we can call the lawyer's approach, the
businessman's approach and the economist's approach. These consist, respect-
ively, in producing tougher new laws and tougher enforcement of existing laws,6
in paying higher wages to bureaucrats7 or, ®nally, in increasing the level of
competition in the economy, both among ®rms and bureaucrats.8
Two of the least corrupt countries in the world, Singapore and Hong Kong,
are usually set forth as examples of successful applications of the lawyer's
approach, as they have quite draconian laws on corruption. As it happens, they
are also examples of countries that pay their bureaucrats exceptionally well
(especially Singapore), so they are also examples of the businessman's approach
to corruption control. However, the level of political competition and civil
liberties in these countries is quite poor, something that has allowed exceptional
levels of pay in the bureaucracy without too much political opposition, and has
often bestowed the anti-corruption agencies with sweeping powers that
amounted to a `guilty until proven innocent' principle, or even the right to
violate the privacy of individual citizens. Since these experiences cannot be
easily adapted to other countries, it becomes necessary to distinguish how much
of their clean record should be attributed to their policy of high wages in the
bureaucracy and how much to their tough approach to law enforcement. The
third proposal, the economist's approach of unleashing the forces of competi-
tion against corruption, has the least cost in terms of civil liberties. As it
happens, it is also supported by the experience of Singapore and Hong Kong, as
these countries are models of laissez-faire societies with market forces compet-
ing away rents in an extreme way. Thus, what is required is an assessment of the
relative impact of each policy option based on a wider empirical basis.
The purpose of this paper is, therefore, twofold: ®rst, to describe the main
topics in the economics of corruption, including the present directions of
research; and second, to place special emphasis on the theory and evidence
behind the policy alternatives in restraining corruption.
The New Data on Corruption
In general, recent empirical studies of corruption have used data from three
dierent sources. All three data sets have been created for business related
purposes, to be consumed by banks, institutional investors, or multi-national
®rms. The ®rst two data sets are bought at quite high prices.
The ®rst data set comes from Business International (BI), a subsidiary of The
Economist Intelligence Unit. Data is available for the period 1980±83 and covers
6 One such policy proposal is by Italy's judge Antonio Di Pietro, who argues that a way to curb
corruption is to reform the judiciary system to allow for plea bargaining and to give judges the
ability to grant immunity from prosecution to those that cooperate with the investigations.
7 See, for example, Becker and Stigler, `Law enforcement, malfeasance and the compensation of
enforcers'.
8 See Susan Rose-Ackerman, Corruption: a Study in Political Economy (New York, Academic,
1978) and Alberto Ades and Rafael Di Tella, `Competition and corruption', Applied Economics
Discussion Paper Series No. 169, Oxford University (1995) and Christopher Bliss and Rafael Di
Tella, `Does competition kill corruption?', Journal of Political Economy (1997), forthcoming.
# Political Studies Association, 1997
498
The New Economics of Corruption
close to 70 countries. The corruption measure in this data set is quite general
and is provided by BI's network of correspondents who must grade on a scale to
10 `the degree to which business transactions involve corrupt payments' in each of
the countries covered. All correspondents use the same methodology and their
reports are further checked for comparability at the regional level and at BI's
headquarters.
A second data set comes from the World Competitiveness Report (WCR), a
business publication produced by the World Economic Forum in Switzerland,
and consists of a survey of top and middle managers in the most dynamic ®rms
in each of the countries covered. The surveys have included a question on
corruption since 1989, cover a minimum of 32 countries and usually involve
over a thousand executives. The question asked is `the degree to which improper
practices (like corruption) prevail in the public sphere'. An advantage of the
WCR over the BI data is that it covers people with an intimate knowledge of
business practices in each of the countries covered. But the apparent lack of a
centralized oce to consolidate the answers of those surveyed by the WCR
could be a drawback of the WCR data in a cross-section study, as it calls into
question the comparability of the answers between countries. The fact that the
companies to which the survey respondents belong are successful and
internationally oriented is only a partial answer to that concern.
The third data set was gathered by Peter Neuman and his collaborators at
Impulse, a German business publication. It also consists of a survey, this time of
German businessmen who normally conduct business with each of the countries
covered (typically exporters). On average, 10 persons where interviewed for each
country, and an eort to have a minimum of three respondents per country was
made. An important advantage of this data set is that there is less subjectivity
involved as respondents must simply provide an estimate of the kickback per
deal (as a percentage of the deal's value) that would have to be paid in order to
conduct business in each country. The data was published in 1994 and covers
103 countries. Another advantage is that it originated in a very homogeneous
group of people (German exporters), with practical business experience in each
country covered and who answer a quantitative question.
Traditionally, empirical economists expect to use hard data in their research.
In...
To continue reading
Request your trial