Abstract
Do crackdowns on bribery impact corrupt behavior in the long run? In this paper we observe the long-run impact of a short-term punishment institution (i.e., a crackdown) on bribery behavior in a lab setting. We conduct lab experiments in two countries with cultures that differ in corruption norms, and which experience very different levels of bribery: the US and Pakistan. Bribery is implemented in the laboratory as a repeated three-player sequential game, consisting of a firm, a government official and a citizen. The design contains three phases: pre-crackdown, crackdown, and post-crackdown. Results show that post-crackdown behavior is not significantly different from pre-crackdown behavior in either country. We conclude that short-term crackdowns may impact behavior in the short run, depending on the strength of the existing corruption norms in the country. More importantly, in our setting crackdowns are completely ineffective in the long run, as corrupt behavior rebounds to pre-crackdown levels.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
For example, Liu (1983) reports 304 cases in China over a period of just four years resulting from multiple crackdowns on corrupt practices as reported in the media. Di Tella and Schargrodsky (2003) provide an example in Argentina, and Dininio (2005) provides examples in Bolivia, Venezuela, and Italy. In recent times China has signaled intent to crack down by prosecuting a high profile politician (Bo Xilai) due to corruption (Buckley 2013).
A number of studies show that fines can crowd out intrinsic motives for refraining from “bad” behavior, with short- and long-run consequences (e.g., Gneezy and Rustichini 2000). See fn. 6.
Corruption is defined as the “use of public office for private gain” (Jain 2001). Bribery constitutes one aspect of corruption, and requires a payment by an interested party to a government official in exchange for benefits at a cost to a third party. Bribery, in and of itself, may be welfare-neutral, welfare-enhancing, or welfare-reducing, depending on the cost it imposes on parties external to the transaction (Cameron et al. 2009).
Intrinsic motives and norms are related concepts. A norm of corruption constitutes an expectation of corrupt behavior within the society. Intrinsic motivations depend on social norms: the propensity to engage in corruption is dependent upon the behavior of peers and collective reputations (Tirole 1996). As Bicchieri and Muldoon (2011) point out, policy interventions can serve to change potentially harmful norms that are prevalent in a society.
Motivation crowding theory (Frey and Jegen 2001) provides conditions under which crowding-out occurs. Interventions designed to limit behavior crowd out intrinsic motivation, while interventions designed to support behavior crowd in intrinsic motivation. This crowding out may have long term effects, making institutions that impact intrinsic motives particularly worthy as an area of inquiry. An example of the perverse effect of incentives is Gneezy and Rustichini (2000), who test the short- and long-run impact of implementing a fine on parents who arrive late to collect their children from a day-care center. They find that the fine increases the number of parents who arrive late, and its subsequent removal fails to restore the previous pattern of behavior. Implementation of the fine crowds out intrinsic motivation, and this effect persists after the fine is removed. See Gneezy et al. (2011) for an excellent review of when and how incentives work to modify behavior. Ryan and Deci (2000) discuss the same issues from a psychological perspective; see Deci et al. (1999) for a meta-analysis of 128 such studies.
See Banuri and Eckel (2014) for a more extensive discussion of sample differences in student populations in these two countries in experience with corruption.
The discussion below does not exhaust the experimental literature on corruption. For related results on bribery see also Veldhuizen (2013), Armantier and Boly (2012), Abbink (2005); and for related results on embezzlement, Azfar and Nelson (2007), Barr et al. (2009). We focus here on the work most relevant to the current study.
This design choice stems directly from the observation that petty corruption involves fixed relationships. Officials know that there is a distribution of clean and corrupt agents, and often have to signal their willingness to participate in bribe transactions before bribes are offered. Officials can signal their willingness by providing favors in the hopes of attracting bribes at a later date.
For the sake of parallelism, we asked citizens to indicate a hypothetical punishment decision, so that they were also forced to pay attention and less likely to become bored.
An alternative way to model punishment is via institutional punishment (as in Abbink et al. 2002), where detection is probabilistic, and the penalty is implemented automatically (by the experimenter) conditional on detection. See Fehr and Fischbacher (2004) for an experimental study of third-party punishment across cultures.
The current use of internet technology (for example, corruption reporting websites such as “ipaidabribe.com”) is designed to reduce the costs of reporting for the citizens.
In our model (as in the real world), bribery is risky. To overcome this, officials sometimes have relationships with specialists whose sole job is to mediate bribery transactions with citizens. An example of this is discussed in Bertrand et al. (2007), where agents exist that conduct corrupt transactions with officials in driver’s license offices in India. Having fixed partners allows subjects to accurately calibrate the risks associated with bribery.
At the Institute of Business Administration in Pakistan, all instruction is in English. Indeed, all schooling of these subjects would have been in English-based schools. Most students at the university sit for standardized British high school examinations (O and A levels). English is the language of choice for business transactions at this level.
Table 1 Sample descriptive statistics Unless otherwise indicated, all non-parametric tests between the US and Pakistan are two sample, two-tailed proportions tests. Non-parametric tests between phases are one-sample two-tailed proportions tests. Since this is a repeated game, treating each decision as independent is inappropriate. Therefore, the proportions tests are conducted using each subject as a single observation, rather than each decision as a single observation.
Table 2 Overall bribe and favor outcomes In the pre- and post-crackdown periods, bribery is virtually identical across the two cultures, despite major differences in corruption. There are a number of reasons why this could arise, but differences in design are likely to play a role. The experiments in Cameron et al. (2009) and Barr and Serra (2010) are one-shot, while ours are repeated (following Abbink et al. 2002). The repetition removes uncertainty associated with bribery, and population-based beliefs about others’ behavior play a limited role. Further study is needed to disentangle possible causes.
Estimates are robust to using fixed effects logit models (not shown).
We use the round numbers for each phase separately so as to compare the intercepts and slopes from the post-crackdown phase to the intercept and slope in the pre-crackdown phase directly. The online appendix presents the alternate specification under the assumption of a common trend with structural breaks. We are grateful to an anonymous referee for pointing out this alternate specification.
Results are robust to controlling for history within the triad (i.e. a variable indicating whether a favor was granted in the previous round). This is excluded from the table due to endogeneity concerns. Adding additional lags (beyond a single period) also does not alter our results.
Changes in probability were calculated using marginal effects (not reported).
What would explain the reduction in the probability of offering a bribe or favor over time in the US? This is a surprising result that has not been found in previous research. Abbink and coauthors (2002, 2004, 2006) have conducted repeated versions of bribery games, with no decline. However, a key difference between our design and theirs lies in the target of the externality. Under their framework, the negative externality of corruption is imposed on all other players of the game, and indeed, each firm/official dyad is subject to externalities generated from other players of the game. Our design implements a single target of the negative externality, which increases the salience of the externality for the subjects. Moral costs rise with cumulative externalities, and with higher moral costs, subjects find it more costly and difficult to engage in corruption. In Table A1 in the online appendix, we provide some evidence for this, as each successful bribe reduces the likelihood of a favor by 7 % in the first ten rounds \((\hbox {p}<0.05)\). We can speculate that the moral costs are more salient for the government officials (since they have the choice directly impacting the externality), and hence why the effect is observed for the favors, and not for bribes. This is suggestive evidence, as our data does not allow us to completely explore the nature of the decline. In addition, it is also not clear why culture would have a differential impact on the salience (we do not observe this downward decline in Pakistan). In the appendix we then argue that this decline may be something of an illusion, as it is not observed in a similar experiment without a “crackdown” regime change in period 11.
Since we are primarily interested in a direct comparison of the pre- and post-crackdown phases, the regression drops the crackdown phase.
One exception to the difference in firm punishment is that Pakistani subjects punish firms approximately 1.5 tokens more than US subjects when a bribe is offered, but not accepted by the official. While we have little evidence of why this occurs, we conjecture that this may be due to differences in how subjects perceive the confiscation of the bribe. US subjects may view the confiscation as punishment, while Pakistanis may not view that as punishment, and feel the need to further reprimand the firms.
We are grateful to an anonymous referee for pointing this out.
References
Abbink K (2004) Staff rotation as an anti-corruption policy: an experimental study. Eur J Polit Econ 20: 887–906
Abbink K (2005) Fair salaries and the moral costs of corruption. In: Kokinov B (ed) Advances in cogntive economics. NBU, Bulgaria
Abbink K, Hennig-Schmidt H (2006) Neutral versus loaded instructions in a bribery experiment. Exp Econ 9:103–121
Abbink K, Irlenbusch B, Renner E (2002) An experimental bribery game. J Law Econ Organ 18:428–454
Abbink K, Serra D (2012) Anticorruption policies: lessons from the lab. Res Exp Econ 15:77–115
Alatas V, Cameron L, Chaudhuri A, Erkal N, Gangadharan L (2009) Gender, culture, and corruption: insights from an experimental analysis. South Econ J 75:663–680
Armantier O, Boly A (2012) On the external validity of laboratory experiments on corruption. Res Exp Econ 15:117–144
Azfar O, Nelson W Jr (2007) Transparency, wages, and the separation of powers: an experimental analysis of corruption. Public Choice 130:471–493
Banuri S, Eckel C (2012) Experiments in culture and corruption: a review. In: Serra D, Wantchekon L (eds) New advances in experimental research on corruption, vol 15. Emerald Group, New York, pp 51–76
Banuri S, Eckel C (2014) The effects of sanctions on bribery: US versus Pakistan. Working Paper, Texas A&M University
Bardhan P (2006) The economist’s approach to the problem of corruption. World Dev 34:341–348
Barr A, Lindelow M, Serneels P (2009) Corruption in public service delivery: an experimental analysis. J Econ Behav Organ 72:225–239
Barr A, Serra D (2010) Corruption and culture: an experimental analysis. J Public Econ 94:862–869
Basu K (2011) Why, for a class of bribes, the act of giving a bribe should be treated as legal. India Ministry of Finance Report
Becker G (1968) Crime and punishment: an economic approach. J Polit Econ 76:169–217
Bertrand M, Djankov S, Hanna R, Mullainathan S (2007) Obtaining a driver’s license in India: an experimental approach to studying corruption. Q J Econ 122:1639–1676
Bicchieri C (2006) The grammar of society: the nature and dynamics of social norms. Cambridge University Press, Cambridge
Bicchieri C, Muldoon R (2011) Social norms. In: Zalta E (ed) The stanford encyclopedia of philosophy, 2011th edn. California, Stanford
Brinkerhoff D, Kulibaba N (1996) Perspectives on participation in economic policy reform in Africa. Stud Comp Int Dev 31:123–151
Buckley C (2013) With official’s conviction, China calls attention to its crackdown on corruption. New York Times. http://www.nytimes.com/2013/09/23/world/asia/with-officials-conviction-china-calls-attention-to-graft-crackdown.html. Accessed 18 Dec 2013
Cameron L, Chaudhuri A, Erkal N, Gangadharan L (2009) Propensities to engage in and punish corrupt behavior: experimental evidence from Australia, India, Indonesia and Singapore. J Public Econ 93:843–851
Chaudhuri A (2011) Sustaining cooperation in laboratory public goods experiments: a selective survey of the literature. Exp Econ 14:47–83
Deci E, Koestner R, Ryan R (1999) A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation. Psychol Bull 125:627–668
Di Tella R, Schargrodsky E (2003) The role of wages and auditing during a crackdown on corruption in the city of Buenos Aires. J Law Econ 46:269–292
Dininio P (2005) The risks of recorruption. In: Spector B (ed) Fighting corruption in developing countries: strategies and analysis. Kumariain, Colorado
Fehr E, Fischbacher U (2004) Third party punishment and social norms. Evol Hum Behav 25:63–87
Fehr E, Gachter S (2000) Cooperation and punishment in public goods experiments. Am Econ Rev 90: 980–994
Fischbacher U (2007) Z-tree: Zürich toolbox for readymade economic experiments. Exp Econ 10:171–178
Fisman R, Miguel E (2007) Corruption, norms, and legal enforcement: evidence from diplomatic parking tickets. J Polit Econ 115:1020–1048
Frey B, Jegen R (2001) Motivation crowding theory: a survey of empirical evidence. J Econ Surv 15:589–611
Gneezy U, List J (2006) Putting behavioral economics to work: testing for gift exchange in labor markets using field experiments. Econometrica 74:1365–1384
Gneezy U, Meier S, Rey-Biel P (2011) When and why incentives (don’t) work to modify behavior. J Econ Perspect 25:191–209
Gneezy U, Rustichini A (2000) A fine is a price. J Legal Stud 29:1–17
Hauk E, Saez-Marti M (2002) On the cultural transmission of corruption. J Econ Theory 107:311–335
Herrmann B, Thoni C, Gachter S (2008) Antisocial punishment across societies. Science 319:1362–1367. doi:10.1126/science.1153808
Jain A (2001) Corruption: a review. J Econ Surv 15:71–121
Kaufmann D, Kraay A, Mastruzzi M (2005) Governance matters IV: governance indicators for 1996–2004. World bank policy research working paper 3630
Kpundeh S (1999) The fight against corruption in Sierra Leone. In: Stapenhurst R, Kpundeh S (eds) Curbing corruption: toward a model for building national integrity. World Bank, District of Columbia, pp 207–234
Licht A, Goldschmidt C, Schwartz S (2007) Culture rules: the foundations of the rule of law and other norms of governance. J Comp Econ 35:659–688
Liu A (1983) The politics of corruption in the People’s Republic of China. Am Polit Sci Rev 77:602–623
Masclet D, Noussair C, Tucker S, Villeval M (2003) Monetary and non-monetary punishment in the voluntary contributions mechanism. Am Econ Rev 93:366–380
Ryan R, Deci E (2000) Intrinsic and extrinsic motivations: classic definitions and new directions. Contemporary educational psychology 25:54–67
Tirole J (1996) A theory of collective reputations (with applications to the persistence of corruption and to firm quality). Rev Econ Stud 63:1–22
Transparency International (2011) Corruption perceptions index. http://www.transparency.org. Accessed 22 Aug 2012
Treisman D (2007) What have we learned about the causes of corruption from ten years of cross-national empirical research? Annu Rev Polit Sci 10:211–244
Van Veldhuizen R (2013) The influence of wages on public officials’ corruptibility: a laboratory investigation. J Econ Psychol 39:341–356
Acknowledgments
We are indebted to Klaus Abbink, Rachel Croson, Reuben Kline, Sherry Xin Li, Angela de Oliveira, Elizabeth Pickett, Wendy Mak, Eric McLester, Quoc Tran, participants of the Economic Science Association meetings in Tucson, AZ, and participants of the workshop on conflict experiments in Rennes, FR. The paper was substantially improved during the editorial process. Funding was provided by the National Science Foundation (NSF SES-0921884). Any errors remain our own.
Author information
Authors and Affiliations
Corresponding author
Electronic supplementary material
Below is the link to the electronic supplementary material.
