By Guy Rubinstein
A common intuition about corruption is that individuals are more likely to engage in corruption when they witness others committing corrupt acts without facing serious consequences—in other words, a “culture of corruption” can be self-perpetuating (see here and here), and the perception or belief that corruption is widespread can itself be a cause of corruption. While compelling, this intuition has not been subjected to much empirical scrutiny. While there does seem to be some evidence of an association between individuals’ perceptions of the prevalence of domestic corruption and those individuals’ inclination to act corruptly, the research on this topic is relatively thin.
In a recent paper, a group of academics (Israel Waismel-Manor, Patricia Moy, Rico Neumann, and Moran Shechnick) weighed in, presenting the results of a controlled lab experiment that sought to assess whether news about corruption by public officials affected individuals’ incentives to behave dishonestly. The study was conducted in Israel, and participants were required first to watch a short television news segment. The treatment group’s segment revolved around an Israeli mayor suspected of certain corrupt acts, while the control group’s segment was unrelated to corruption. The participants were also given a short quiz about the segment they’d seen, and half of the participants in each group were offered a monetary reward if they answered all the questions correctly; they were told “to answer all questions from memory” and not look anything up on the internet. However, unbeknownst to the participants, one of the questions could not be answered without doing additional searches, so the researchers could use the answer to this question to identify those participants who cheated on the test. The real goal of the study (of which participants were not aware) was to see whether exposure to the corruption news story (alone or in combination with the financial incentive) affected participants’ likelihood of cheating.
Unsurprisingly, participants who were offered money for answering all questions correctly cheated far more often, regardless of which news story they watched. The study’s authors seem to have expected that those participants who watched the corruption-related story would also cheat more (holding constant whether they had financial incentives to answer questions correctly). But this did not occur: Participants who watched the news segment involving a mayor suspected of corruption did not cheat in statistically significantly higher rates than those who watched the other, unrelated-to-corruption segment. The researchers suggested that perhaps the reason was that Israelis had been inundated with so much news about official corruption around the time of the experiment (which took place in 2019), particularly in connection with the investigation and prosecution of Prime Minister Benjamin Netanyahu and other senior politicians (see here, here, and here), that the marginal impact of exposure to additional news about corruption, in the form of this one story, would not have much impact.
While that explanation is plausible, I have some other concerns about the research’s design and methodology, which make me question whether this experiment was in fact a good way to assess the “culture of corruption” hypothesis.
First, I am puzzled by the choice to expose participants to a news segment about grand corruption (defined by the researchers as “a deviation from culturally accepted norms and an abuse of power by public officials who misuse the legitimacy they receive from the public to advance their own interests”) rather than petty corruption (“acting dishonestly, unethically, or unfairly [on a smaller scale] in an effort to gain personal advantage”). Granted, it is important to study the behavioural effects of exposure to both types of corruption. But it is reasonable to assume that individuals who are facing the dilemma of whether to cheat on a test for a small financial gain would be influenced more by exposure to corrupt actions on a similar scale, taken by ordinary individuals, as opposed to grand corruption by elected officials. Bribery accusations against a mayor seem substantively different from the ethical dilemma that the participants of the study were actually facing. I therefore wonder whether the results would have been different if instead of being exposed to a segment about a senior public official suspected of grand corruption, participants had been exposed to a segment about petty corruption, like tax evasion by ordinary people.
Second, and more importantly, the researchers wanted to study the behavioural effects of exposure of individuals to corruption, but the news segment involving corruption that the researchers chose to expose participants to concerned a mayor who was “arrested for bribery, fraud, and breach of trust.” For many participants who watched this segment, the key takeaway may well have been that the mayor was arrested—in other words, the most salient lesson of the segment was not that corruption is widespread, but that corrupt individuals may face grave consequences for their illegal behaviour. If this was indeed the case, then it should not come as a surprise that participants were not encouraged by this segment to act dishonestly. Rather, it is perhaps more surprising that this segment did not have an inhibitory effect on their willingness to cheat.
To be sure, I recommend that readers of the blog check out the full article. There is little doubt that understanding the role that the media play in shaping individuals’ choices to behave unethically can be an important contribution to the global struggle against corruption. I hope that future studies in this field will take concerns like mine into account.
