Syndicate content

Corruption, Game Theory, and Rational Irrationality

Fumiko Nagano's picture

If we had to name one reason why petty corruption is so difficult to tackle, it has to be that it makes sense for people to engage in it than not. Unlike measures such as smoking bans, seatbelt laws, and drinking and driving laws where there is a clear individual benefit to those who do the “right thing,” corruption bans are hard to enforce because there aren’t easily discernible individual benefits to those who obey them. Rather, in countries where corruption is systemic, people who do what is right and follow whatever anti-corruption law might be in place will find themselves losing out to those who don’t.

In fact, with corruption, individual opinion doesn’t seem to matter much in one’s decision whether to engage in it. In theory, most people believe that corruption is wrong. But in practice, the incentive that motivates an individual’s behavior in a corruption-prone situation is their perception of what everyone else would do in a similar situation. Would your pregnant colleague pay a bribe so that she could jump the queue and get an H1N1 vaccination when the vaccines are in limited supply? Would your neighbor, an entrepreneur, slip a few notes to a civil servant under the table to expedite the process of obtaining a business license? If the answer to each of these questions is a “yes,” then why should you bother going against the system alone? Why should you do the right thing and find yourself at a disadvantage to everyone else who will do what it takes to obtain what they need given the environment and culture in which they live?

Moreover, avoiding corruption in this setting can also lead to inefficient outcomes in addition to putting you in a disadvantaged situation. For example, if you don’t bribe, not only will your wife not get the vaccine, but also your colleague who bribed and got the vaccine might need it less than your wife does. The business license will be given not to you – the most deserving and capable entrepreneur – but to your neighbor who bribed who might not have been the most suitable candidate to get the license in the first place. Corruption hinders economic development by introducing inefficiencies in the system. And, unfortunately for those fighting corruption, it is difficult to confront because it is actually quite logical for people to succumb to the temptation to bribe or be bribed. Corruption is the Prisoner’s Dilemma in application.

In a recent New Yorker article, “Rational Irrationality,” John Cassidy explains the financial crisis with game theory which can very well be applied to corruption. He argues that the cause of the crisis was not that Wall Street C.E.O.s acted with “greed, overconfidence, and downright stupidity” as the word on the street would have you believe, but that they acted rationally, given the circumstance. He writes: “the root problem is what might be termed ‘rational irrationality’ – behavior that, on the individual level, is perfectly reasonable but that, when aggregated in the marketplace, produces calamity.”

Cassidy attributes the logic of rational irrationality to John Maynard Keynes, who, in “The General Theory of Employment, Interest, and Money,” likens investing to a game of musical chairs and newspaper competitions on pretty faces. To win either game, one’s task was to choose the outcome which you think others will select, regardless of your personal opinion. As financial markets are made up of people reacting to what others are doing, it is a version of the Prisoner’s Dilemma in which one’s actions affect everyone else’s welfare, and everyone else’s actions affect one’s welfare. This means that “attempts to act responsibly and achieve a cooperative solution cannot be sustained, because they leave you vulnerable to exploitation by others.” The similarity between the behaviors of Wall Street executives and citizens yielding to corruption is uncanny. In both cases, individuals are behaving rationally in the name of individual gain based on their speculations about how others are going to behave, with disastrous consequences to society.

So how should we fight corruption? One way is the top-down approach, with the government enforcing a ban on corruption. In countries where corruption is rife, this would be difficult to enforce for multiple reasons, but one in particular stands out as explained eloquently by John Macrae in “Underdevelopment and the Economics of Corruption: A Game Theory Approach”: “The response to a situation of this kind emerging has generally been to suggest regulation by outside authority of firm or individual behavior in an effort to ensure that private and public interests are brought into line. But this solution is of doubtful applicability in this case for it is the very behavior of these authorities that has been made an integral part of the game. The shock that the formation of the problem in the above way provides is that there may be no effective ‘independent’ agent to implement public fines or other sanctions for these agents are already engaged in taking bribes themselves.” 

Absent effective government enforcement mechanisms, we need to turn to the people. So can we expect individual citizens to fight such an endemic practice, given what we know about corruption, game theory, and rational irrationality? We can’t – at least not if people were to act single-handedly. But, as we heard at the discussions during UNCAC Conference of the State Parties last week in Doha, there is hope. People respond to the notion that they are not alone in the battle against corruption. People are willing to put up a fight if only they knew that there were others with a similar belief, the same conviction and resolve to counter demand for bribes. The important task is to communicate this critical information to the people – that they are not alone – so that they can build up the courage to come together and mobilize around this common cause. Then, public opinion can finally transform to condemn corruption at the individual level with an aggregate benefit to society as a whole.

I would like to thank Johannes Koettl and Ellen Tan for their contribution to the thinking behind this post.

Photo Credit: Flickr user Lars Plougmann


Submitted by Zo on
I don’t entirely agree with the line of reasoning. Similar to the pregnant woman bribing for the H1N1 vaccination, consider the massive bribes made for such things as human organs. For many, the game theory approach as described above is irrelevant; they need an organ and are willing to bribe to get one. That others would do the same may help them justify their action, but it is not determining their action. Turning to the countries “rife” with corruption, I agree that the top down approach is fundamentally flawed. I am not sold on the public-opinion-transformation formula, particularly in the least developed countries. This is because (some of) the people involved see their bribe as entirely necessary, even if they consider it wrong. No bribe, can’t get through checkpoint, can’t get to work, can’t feed family. At the other end, the underpaid guard, considers it a necessary part of his salary. Basic economics needs to remain at center stage. Where things are scarce (e.g., human organs) people who can pay will pay. For regions where bribery is embedded into the economy, real incomes would have to rise to the point where enough people aren’t in such a highly exploitable survival mode. At that point the “lets stand together and end corruption” model might work.

Submitted by Christine Zarzicki on
This article points to the complexity of corruption on various levels: the individual justification of bribery, desperation as a cause of corruption, and the idea that this issue is all encompassing, as many governments perpetuate this type of behavior. But one aspect of corruption that is not highlighted in this article is that the idea that corruption, as we know it, is a Western idea based on Western values. To explain, while individuals in the United States see bribery as morally wrong (and illegal in many cases) this may not be a universal truth. In states where these “transactions” have been taking place for decades, who is to say that this is not considered "normal"? If every corrupt state existed in a vacuum, isolated from the US index of morality where the benchmark for lawful behavior was established, would they still consider this behavior to be wrong? Without the context of the international arena, perhaps this conduct would be entirely acceptable. That being said, I believe that the solution to corruption is more complex than individuals joining together to fight the cause. It is about overall economic growth that delivers to people more than can be extracted from scheming and bribery. It is about information and development and allowing these nations to realize that there are more viable options in this global generation. The benefits of a functioning economy, devoid of corruption, are enough to generate a change in sentiment. But until individuals can discern the benefits of a legitimate economy, their perceived “options” will remain the same.

Submitted by Anonymous on
You are brilliant, Fumiko. Thank you for offering this light at the end of a very dark tunnel. I now have hope that corruption can be overcome.

Submitted by S J Masty on
Here in Kabul, we spend much time thinking on this subject and Ms Nagano's article is insightful and welcome. It may be that there is no single solution, or even one single formula, to cure corruption. Here we have had numerous governments over 30 years of war, and our allies have deserted us once, abandoning Afghanistan to the Taliban in the 1990s. Western political debates on how fast they can pull out (Canadian and Dutch troops will depart in 2010) make many Afghans believe that they shall be abandoned again. This is a powerful and malign incentive for people to grab what they can in preparation for dark decades ahead - from elites to petty officials many of whom have become simply rapacious. Yet Western nations suffer economic privation at home, Western democracies debate the cost of their commitment and one sympathises with their leaders trying to support a policy that is at once expensive, strategically necessary (I think) and without much military success thus far. The Western political debate, covered by global media, undermines any sense of Afghan optimism and slashes the effectiveness of Western blood and treasure. Optimism is a variable in this version of game theory. It worked in reverse with communism, where its sense of historical inevitability was so powerful that its adherents retained a quasi-religious faith in the collectivist future despite all evidence. In Afghanistan we make gains but they have little effect in breeding optimism or changing behaviour because our allies, the country's lifeline, keep speculating loudly and persistently on how swiftly they can leave, while timetables of Western patience appear to be (but perhaps are not) clearly shorter than the obvious decade or two required to settle, develop and reform Afghanistan. Afghans seem to believe there are three elements in defeating corruption in Afghanistan but the formulae may differ from nation to nation. All three of ours fit to some degree under Ms Nagano's rubric. Law enforcement is essential. So is simplifying and clarifying government procedures internal and external. In my ministry, procedures from the monarchy, from communist rule, from the civil war, the Taliban government and post-invasion all overlap and contradict one another, leaving much room for interpretation and rent-seeking. These (the likelihood of being caught and the opportunities to cheat) affect 'the odds' for players in the dark game that Ms Nagano describes so well. But equally important is morale, sapped daily by global media - reporting a democratic process responding to public pressures - that may not be possible to change. S Masty Kabul, Afghanistan

Submitted by Anindya Bhattacharjee on
This was a thought-provoking article. We need to strike upon a mechanism to make information sharing more smooth between the common people who are supposed to unite. The bigger problem is finding some "martyrs" who would be the first movers in the war against corruption and set the example for the others to follow suit knowing that they are not alone. But how do we incentivize these "first-movers"? Or is organized,simultaneous collective action possible? There too the problem remains that someone might defect since an individual will always have an incentive to do that. Fighting corruption: How do we make it happen?

Add new comment