Too Much of a Good Thing: Reciprocity and Corruption

The typical approach that researchers have used to study corruption is through observational studies, which use cross-country data and subjective indexes of corruption.⁶ Although these studies are important for understanding the economic, social, and cultural factors which determine corruption, they do not provide us with much insight on the behavioral reasons for why officials engage in corruption. 

One relatively new approach that behavioral scientists have taken to study corruption is through controlled laboratory experiments. Experiments allow researchers to simulate the environment in which corrupt decisions are made, in order to directly observe participants’ responses to various anti-corruption interventions. This is effective if we wish to, for example, test the effects of a potential anti-corruption policy on specific behavioral responses, as experiments generate hard data that can answer very specific questions. This can all be done in a relatively cost-effective manner too, as compensating participants is usually the only significant financial investment required to run an experiment. All of this makes experiments a cheap, preliminary diagnostic tool for anti-corruption policies before they are implemented in the real world.

The power of reciprocity

Perhaps unsurprisingly, one obvious reason why officials would act unethically is because of money. If offered a substantial bribe, officials would be willing to face all the risks and the costs involved with the illicit activity in order to receive a sizeable reward.

But corruption does not just revolve around money. There are also psychological factors at play in the typical corrupt transaction. Think about it: A briber undertakes a substantial amount of risk when they approach an official for an illegal favor. The official could refuse to take the bribe, they could notify the authorities, or they could even accept the bribe and not go through with the favor. But why is it that in real life, plenty of officials are willing to not only accept bribes, but also follow through with the favor?

The answer lies in our innate desire to reciprocate the kindness of others. Reciprocity is a social norm that explains why we feel inclined to be kind to those who are kind to us (and cruel to those who are cruel to us).⁷ It can explain many phenomena that are difficult to account for using standard economic theory, which assumes people are selfish and rational—for example, why waitresses who smile more earn higher tips, or why we are willing to buy something from the salesperson who gave us a free sample.^8,9

When reciprocity fuels corruption

Reciprocity is something that is typically considered prosocial: when we reciprocate the kindness of others, we are abiding by the social norm. But our desire to reciprocate is actually a major reason behind corruption, something that is obviously not pro-social. This is because an official may perceive a bribe as a kind gesture, and by reciprocating, they would simply be behaving in a way that is deemed socially acceptable. After all, the briber was generous enough to go through the trouble of not only approaching the official, but also offering them money. 

Of course, by following through with the favor, the official is neglecting societal interest, since this favor generally has great benefits to the briber but detrimental consequences to the public. This therefore depicts a fascinating behavioral component of corruption: our desire to reciprocate seemingly overrides our greater sense of morality and causes officials to engage in corruption.

A 2002 study by Abbink, Irlenbusch, and Renner tested this idea through a pioneering experimental bribery game that simulated a typical corruption scenario, in which participants acted as either a potential briber or a public sector official.¹⁰ They found that a large proportion of officials routinely reciprocated the trust of bribers by choosing the corrupt course of action, despite having no contractual obligations to do so. Officials could have been better off materially had they simply accepted the bribe and not gone through with the corrupt action, but it was precisely because of reciprocity motivations that they chose the corrupt option instead. 

In addition, when the authors introduced negative externalities to the experiment, even though bribers and officials fully understood that their participation in a transaction had adverse effects on other participants in the experiment, corruption persisted.¹¹ Apparently, the knowledge that your actions would harm others was not enough to curb the urge to reciprocate. 

We value reciprocity over collective good

These findings capture a disheartening picture of human nature, but it is also quite unsurprising given that people act corruptly in real life despite often having knowledge of the consequences. But why do people continue to engage in corruption even though they know full well that their behavior harms many innocent people? Are people really that selfish?

Perhaps the answer lies in the fact that the effects of corruption are often felt by an unobservable group of people. Behavioral biases such as the identifiable victim effect demonstrate that people feel less empathy for large, vaguely-defined groups of people than for a specific, identifiable individual.^12,13 For example, you are likely to feel a greater sense of empathy when you hear the story of Jimmy, a 12-year-old boy in Africa who is dying of starvation, than when you hear the statistic that over 26 million children are starving in Africa. 

This could be a potential explanation for why officials are undeterred by the negative externalities of corruption. Even though officials know their actions have harmful consequences, the effects are felt by a large, unobservable group of people, while the officials are unlikely to feel the burden themselves. This behavior can be likened to other harmful actions such as littering: we all know that littering has negative effects on the environment, but because we do not feel the burden ourselves or see the impacts directly, we do not feel the necessity to stop.

So far, we have discussed how lab experiments can simulate the corruption decision environment to uncover certain aspects of corrupt behavior, but experiments can also effectively test anti-corruption interventions. 

We can classify the potential solutions for corruption into two broad categories: extrinsic motivations and intrinsic motivations. Behavior that is extrinsically motivated is driven by external incentives, such as money or power; meanwhile, intrinsically motivated behavior is done for its own sake.¹⁴ 

Extrinsic motivation: A different kind of bribe

Most experiments to date have investigated the extrinsic motivations behind corruption, and have focused on altering external incentives to reduce it. These studies have proposed solutions such as introducing harsher penalties for those caught in corruption,¹⁵ increasing monitoring (for example, through increased audits),¹⁶ raising the wages that officials earn,^17,18 rotating staff more frequently, and requiring that multiple officials be involved when making important decisions.²⁰

Reframing corruption

One fundamental drawback to these extrinsic motivation solutions is that they are often costly to implement. This may make them relatively unattractive for policymakers. On the other hand, focusing on intrinsic motivation as an anti-corruption strategy could be a more cost-effective solution. Since corruption is generally considered morally unacceptable, researchers have also examined ways in which we can induce feelings of guilt or shame in order to reduce corrupt behavior from occurring.

 These studies primarily involve framing corruption in an immoral light (for example, by using the loaded phrase “private payment” in experiments instead of the neutral “transfer”) and/or making the negative externalities of corruption more salient (or noticeable) to try and generate an emotional response nudge people away from corruption.^21,22,23

By changing the language around corruption, participants may become more aware of their own dishonesty and subsequently refrain from behaving fraudulently. Although this approach is still in its infancy, these studies can provide insights into how policymakers should frame their messages, and what information they should emphasize, when designing anti-corruption awareness campaigns.

Using reciprocity: Building trust

Other approaches that have been suggested in this domain include promoting trust between employers and public officials. Companies and governments often force workers to sign compliance forms and sit through anti-corruption workshops, all of which can create a spirit of distrust. As we know, people have an innate desire to reciprocate, so perhaps the solution may be for leaders to have more faith in their workers and treat them better.²⁴ This way, workers may feel intrinsically obliged to reciprocate the employer’s kindness and always act honestly.

You may argue that lab experiments cannot capture real-world corruption accurately, since the fact that participants know they are being watched may bias their decision-making (see the observer expectancy effect).²⁵ The external validity of economic laboratory experiments is a common critique of experimental economics in general, where critics often argue that lab experiments may not be reflective of real-world decisions which typically depend on many unobservable factors.²⁶ For the topic of corruption, however, there has been evidence to suggest the results from the lab experiments do have real-world validity. For example, one group of researchers managed to replicate their lab experiment findings in a field experiment where participants were unaware that they were participating in an experiment.²⁷ This research provided evidence that lab experiments can capture some of the key motives behind real-life corruption behavior and have external validity.

Behavioral science has grown to become extremely influential in the development of policies in recent years, and these experimental corruption studies provide another example of their usefulness. Through the innovative methods researchers have developed, we are now able to study a complex, unobservable phenomenon in a powerful and cost-effective way. And despite this being a relatively new field of study, several effective solutions have already appeared in the literature (with many of which not mentioned in this article). Of course, behavioral approaches are not, and should not, be the only approach we should use to analyze corruption, but they are another tool that we can utilize to understand this phenomenon better.