Corruption Cycles
| DOI | 10.1111/1467-9248.00092 |
| Date | 01 August 1997 |
| Published date | 01 August 1997 |
| Author | Cristina Bicchieri,John Duffy |
| Subject Matter | Article |
Corruption Cycles
CRISTINA BICCHIERI AND JOHN DUFFY
1. Introduction
Political corruption has been a persistent phenomenon throughout history and
across societies. It is found today in many dierent forms and degrees in all
types of political systems, in developing countries as well as in Western demo-
cracies. The phenomenon is not solely or even largely the consequence of a
moral fault or cultural backwardness. The root cause of corruption, in Europe
as well as in Japan, Latin America and the United States, is found in the mixture
of economic and political power.
We take corruption to consist in the illegitimate use of public roles and
resources for private bene®t, where `private' often refers to large groups such as
political parties. This de®nition has the advantage of subsuming many dierent
kinds of corrupt behaviour, ranging from public ocers' use of their position to
maximize personal gain by dispensing public bene®ts to the implementation of
policies that violate the common interest in favour of special interests, such as
granting large ®rms the monopoly of their services within a particular sector.
Since this de®nition of corruption is based upon legal norms, not public opinion
or social norms, it has the advantage of directing our attention to the contrast
between ocial and social norms that is present in most societies.1
The literature on corruption is a large corpus of descriptive studies, but a full
¯edged theory of corruption is yet to come. As a result, many dimensions of
corruption have been left unexplored.2 One such problem is the life-cycle of
corruption. If corruption is endemic, what are the forces that allow it to prevail?
What permits it to continue? An answer to this question cannot be separated
from an analysis of the political and economic eects of corruption. There has
been widespread disagreement among scholars studying the phenomenon
regarding the direction of its eects. The so-called Moralists maintained that
corruption is harmful, as it impedes development and erodes the legitimacy of
institutions.3 Revisionists point instead to the possible bene®ts of corruption: it
Earlier versions of this paper were presented at Caltech, Nueld College (Oxford), and at the
Conference on Political Corruption, University of Nottingham, June (1996). We wish to thank the
participants in the above seminars and Conference for many useful comments and suggestions.
1 The classic statement of this de®nition of corruption has been given by Nye. According to him,
a political act is corrupt when it `deviates from the formal duties of a public role (elective or
appointive) because of private-regarding (personal, close family, private clique) wealth or status
gains: or violates rules against the exercise of certain types of private-regarding in¯uence'. Cf. J.
Nye, `Corruption and political development: a cost-bene®t analysis', American Political Science
Review, 61 (1967), p. 416.
2 An analysis of how a corrupt political system may suddenly collapse is found in C. Bicchieri and
C. Rovelli, `Evolution and revolution: the dynamics of corruption', Rationality and Society, 7 (1995).
3 Cf. G. Myrdal, `Corruption: its Causes and Eects', in Asian Drama: an Enquiry into the
Poverty of Nations, vol. 2 (New York, Twentieth Century, 1968), and also R. Wraith and E.
Simpkins, Corruption in Developing Nations (London, Allen and Unwin, 1963).
# Political Studies Association 1997. Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main
Street, Malden, MA 02148, USA.
478
Corruption Cycles
can speed up cumbersome procedures, bypass inecient regulations, buy
political access for the excluded, thus fostering the integration of immigrant or
parochial groups, and even produce policies that are more eective than those
emerging from legitimate channels.4 Revisionists have typically focused their
attention upon non-Western nations undergoing political modernization and
development. Their belief that corruption is functional to development has
fostered the expectation that corruption will disappear once the process of
modernization is completed. However, much research in political science
indicates that corruption is endemic even in the fully developed Western
countries.5
Though it is dicult to calculate precisely the impact of corruption on social
welfare, the recent corruption scandals in Italy and Japan have brought to light
the costs and ineciencies that accompany corrupt practices. In both cases,
governments had many opportunities to grant privileges and exemptions in
return for material or political favours. Since huge pro®ts hinge upon such
privileges, companies have gone to any lengths, including bribery, to win them.
There is evidence that widespread bribery of public ocials to obtain public
works contracts systematically in¯ated business costs, while consumers have
been injured by monopoly pricing. Corruption has had an in¯uence on public
policies, too. Policies have been selected and implemented with a view of
generating income and political support. Firms have been `helped' with price
subsidies, regulatory exemptions, protective taris and import quotas.6
If, as we maintain, systemic corruption has extensive social and economic
costs, why is it a recurrent phenomenon? In many countries, we have witnessed
long periods of systemic corruption followed by `clean-ups' that are in turn
followed by other periods of corruption.7 An important task of a theory of
corruption is thus to identify what factors are at play in engendering such cycles.
The end of a period of corruption is not necessarily driven by the public
getting informed about the existence of corrupt activities. We have learned from
history and experience that ocials can engage in illegal activities without being
reprimanded at the polls. A typical explanation oered for this is that ocials
can trade material bene®ts in return for votes. This was certainly true of
machine politics in US cities at the turn of the century, as well as of any form of
political clientelism, where political parties channel policies and resources
towards individuals and groups and, in return, are guaranteed continual
electoral support.8 The American urban political machine, for example, dealt
4 Cf. J. Abueva, `The contribution of nepotism, spoils and graft to political development', East-
West Center Review, 3 (1966); N. Le, `Economic development through bureaucratic corruption',
American Behavioural Scientist, 8 (1964); H. D. Bayley, `The eects of corruption in a developing
nation', Western Political Quarterly, 19 (1966).
5 Cf. G. Benson, Political Corruption in America, (Lexington, 1978); J. Dobel, `The corruption of
a state', American Political Science Review, 72 (1978); S. Mamoru and H. Auerbach, `Political
corruption and social structure in Japan', Asian Survey, 17 (1977); S. Rose-Ackerman, Corruption: a
Study in Political Economy (New York, Academic, 1978).
6 In Italy, for example, by the 1970s the state sector had become a political instrument at the
hands of the Christian Democrats. Its economic functions were subordinated to the policies of
enabling the survival of that party. Cf. Donald Sassoon, Contemporary Italty (London, Longman,
1988).
7 Cf. K. Gillespie and G. Okrulik, `The political dimensions of corruption cleanups',
Comparative Politics (1991).
8 Cf. H. Gosnell, Machine Politics: Chicago Model (Chicago, University of Chicago Press, 1967).
# Political Studies Association, 1997
CRISTINA BICCHIERI AND JOHN DUFFY
479
almost exclusively in particularistic, material rewards (jobs, licenses, welfare
payments, selective non-enforcement, etc.) to maintain and extend political
control. Indeed, the costs of maintaing a clientist system or a political machine
are usually extremely high. In New York City in the 1930s, the total annual pay
for posts exempt from civil service regulations (i.e., political jobs) exceeded
seven million dollars, and it is reported that boss Tweed, in four years, raised
New York City's debt by a multiple of three while leaving both the tax rate and
assessments untouched.9
One would expect, however, that where there are no large organizations that
can provide material inducements through the multitude of explicit exchanges
that would be necessary to control elections, informed voters would be less
likely to support a corrupt candidate. To explain why corrupt incumbents get
re-elected, it has been hypothesized that there occurs an implicit exchange
between voter and candidate. According to this hypothesis, one would expect
that a corrupt candidate strategy would be to take distinct positions on things
many voters care about, and therefore arrange as many implicit trades as
possible. Indeed, there is some experimental evidence supporting the implicit
trade hypothesis.10
In this paper, we demonstrate with a formal model how, under plausible
assumptions, corruption can become cyclical. A crucial assumption of our
model is that corrupt politicians, in order to be re-elected, have to compensate
voters through material incentives. Even in the absence of grassroot clientist
organizations handing out individual bene®ts, corrupt politicians can enact
policies that bene®t particular industrial and ®nancial groups. If the bene®ted
groups control a large number of votes, a corrupt incumbent will be re-elected.
We assume that enacting such policies involves the use of resources on the part
of the...
To continue reading
Request your trial