By Avinash Dixit (Princeton University)
How should a country fight corruption? Most people would answer that the government should make and enforce strong laws against it. But further thinking should show that this approach won’t get far. The politicians who make laws, and the officials who enforce them, all stand to benefit from the prevailing corrupt system. One cannot expect them to go against their strong self-interest. They will make weak laws with loopholes; their enforcement will be lax and itself riddled with corruption. At a minimum, formal legal avenues must be supplemented by participatory and organized efforts of the losers – citizens. Participatory because mere voting is not enough; even if a corrupt government is voted out, the new one will act with the implicit motto: “It is now our turn to eat”. Organized because any one citizen or firm is helpless when a politician or official demands a bribe, but collectively they have a lot of power. The question is how to harness it effectively.
This is a prisoners’ dilemma for consumers and businesses. If no one else is giving bribes, you improve your chances by bribery; if everyone else is complicit in bribery, you will only hurt yourself if you refrain. So bribery is the dominant strategy for all. But when everyone chooses it, in the aggregate they merely cancel one another’s actions and transfer money to politicians and officials. Social scientists have observed and theorized about numerous ways to resolve prisoners’ dilemmas using bottom-up, self-enforcing strategies. We can deploy this knowledge and experience to devise community-based action against corruption.
We know that successful collective action to resolve prisoners’ dilemmas requires: (1) a group with stable ongoing relationships, (2) common knowledge of what constitutes cooperation and cheating, (3) common knowledge of the sanctions to be imposed on cheaters, (4) good detection of cheating, (5) good communication of incidents of cheating to all participants, (6) incentives for members to take their designated action to punish a cheater.
A business community can establish a “no bribery” norm and enforce it using the sanction that anyone found violating it would be ostracized by the others, which would cut him/her off from all the interactions – contracts, supply chains, finance and so on – that any business needs to function in today’s economies.
Let us see how this meets the desiderata listed above.
(1) The community should have some organized structure such as a Chamber of Commerce, which a business is required to join in order to benefit from networking and trade relations, or be on a list qualified to bid for government business.
(2) Members must pledge not to attempt bribery to win any government contracts or licenses or to influence the decisions of regulators, legislators, and courts.
(3) Any member found violating the norm is subject to ostracism by others. This means cutting off business contacts, but can also include social ostracism if the Chamber has a social branch where the families of businesspeople meet. No one wants their spouses and children excluded from social activities of friends, so this threat can be very effective. But experience shows that sanctions should be graduated, not drastic ones triggered by small infractions; therefore violations especially by new and small members should be met first with warnings, and escalate to full ostracism only if they persist.
(4) The Chamber should have a good gossip network, and contacts with media and officials, that enable it to sniff out corruption, and a tribunal that can investigate suspicions or allegations of corruption. This can be supplemented by a more formal research unit that gives ratings to firms for their clean or corrupt behavior, similar to the Michelin star ratings for restaurants. It is extremely important to avoid false accusations, under severe penalties against anyone found making them. It is also important that the tribunal is not perceived as an insiders’ club that serves to exclude newcomers. A broad outside representation of respected senior retired businesspeople, public figures, media personnel, and academics should oversee the working of the tribunal.
(5) The name and picture of anyone convicted of violating the norms should be publicized widely, as done by the famous New York Diamond Merchants’ Club.
(6) A system like the Honor Codes against cheating that exist in some universities, where refusing to report a violation is itself a violation requiring similar sanction, can create an incentive to take part in the ostracism of a convicted briber. But more than that: If A is ostracized by everyone and invites B to deal with him, B knows that A has nothing worse to fear by cheating in their interaction, and therefore that he must give up a bigger share of the available joint profit or rent to A to keep him honest. In other words, it is more costly to deal with a convicted briber than with someone who has a clean record.
The prisoners’ dilemma view of corruption can be supplemented by a coordination game. Societies have two kinds of equilibria, one where everyone is corrupt and that is just an accepted state of affairs, and another with a clean culture where corruption is shameful and rare. How to shift from a corrupt to a clean equilibrium? Try harnessing the idealism of youth. Everywhere the young, especially the best educated and most enterprising, want their country to be modern and corruption-free. Other things reasonably equal, they prefer to work for, and buy from, firms with good governance and ethical behavior. A movement that channels these preferences into action can create an environment in which such clean firms attract the best talent, are favored by consumers, and therefore are more profitable; this builds momentum for more and more firms to eschew corruption. Indeed, such an organization is having some success in Sicily to fight the Mafia’s extortion; bureaucrats should be easier to counter.
I am not claiming that such organizations or movements will successfully eliminate corruption everywhere, or quickly, or completely. But corruption is such an obstacle to development that even a little success is worth having. Nothing else has worked at all well. Waiting for a 100% effective solution only ensures getting 0% progress.
Suggestions for further reading
Transparency International, (2016); The Benefits of Anti-Corruption and Corporate Transparency. Working Paper #01/2016.
Mungiu-Pippidi, A. (2015); The Quest for Good Governance: How Societies Develop Control of Corruption. Cambridge University Press.
Basu, K., and T. Cordella (eds). (2018); Institutions, Governance, and the Control of Corruption, Palgrave Macmillan.
Bernstein, L., (1992); “Opting out of the legal system: Extralegal contractual relations in the diamond industry.” Journal of Legal Studies, 21(1), 115–57.
Dixit, A., (2004); Lawlessness and Economics: Alternative Modes of Governance, Princeton University Press.
Dixit, A., (2017); “Fighting corruption by altering the equilibrium in an assurance game.” working paper, November 2017, available at http://www.princeton.edu/~dixitak/home/wrkps.html
Dixit, A., (2018); “Anti-Corruption Institutions: Some History and Theory.” Published in K. Basu and T. Cordella, (eds.) Institutions, Governance, and the Control of Corruption, Palgrave Macmillan, pp. 15-49.
Dixit, A. and R. Mankar (2018); “New ideas for fighting corruption in India,” LiveMint, April 23, 2018, https://www.livemint.com/Opinion/mxVdMVeQUBEfoJWmY0scRL/New-ideas-for-fighting-corruption-in-India.html .
Dugatkin, L., (1999); Cheating Monkeys and Citizen Bees: The Nature of Cooperation in Animals and Humans, Harvard University Press, 1999
Greif, A., (2006); Institutions and the Path to the Modern Economy: Lessons from Medieval Trade, Cambridge University Press.
Jacobson, P., (2014); “Addiopizzo: The Grassroots Campaign Making Life Hell for the Sicilian Mafia,” Newsweek, September 17, 2014.
Ostrom, E., (2015); Governing the Commons: Evolution of Institutions for Collective Action, Cambridge University Press, Canto Classics reissue.
Superti, C., (2009); “Addiopizzo: Can a Label Defeat the Mafia?” Journal of International Policy Solutions, 11, Spring 2009 3-11.
 This list derives from studies and meta-analyses of many prisoners’ dilemmas of collective action: common resource pool problems (most notably Ostrom 2015), contract enforcement (for example Greif 2006, and Dixit 2004), and socio-biology (for example Dugatkin, 1999), to cite just a few.
 For more detailed arguments see Dixit (2018).
 See Bernstein (1992).
 A model with supporting evidence is in Dixit (2017).
 See Superti (2009) and Jacobson (2014).
 An OpEd offers a starter attempt to implement these ideas in India: Dixit and Mankar (2018).