Mattias Agerberg, University of Gothenburg
Making it worse: Descriptive vs injunctive norms in anti-corruption efforts
By Mattias Agerberg, Associate Professor, University of Gothenburg
"Corruption is a deadly poison." "Corruption enslaves a nation and its people."
Vivid messages like these can be read on billboards in developing countries from Kenya to Mexico to Nigeria. The billboards are part of anti-corruption campaigns funded by donors and civil society groups aiming to root out endemic corruption by engaging the public in the fight.
Tens of millions of dollars are spent yearly on such “messaging campaigns.” And for good reason. Research in the last few decades has clearly demonstrated the enormous impediment that corruption is to development and prosperity.
The billboards’ messages also align with the UN’s convention against corruption that calls on countries “to raise public awareness regarding the existence, causes and gravity of and the threat posed by corruption.” However, it is far from clear that the alarming billboards actually achieve the intended consequences.
In fact, much research today suggests that messaging strategies focusing on raising awareness about the prevalence and destructive consequences of widespread corruption have the opposite effect, making people less willing to fight corruption and more prone to engage in bribery. This may seem puzzling: why would raising awareness about corruption exacerbate the problem? And given this, what should anti-corruption practitioners and donors do about it? My research suggests that part of the answer lies in what type of social norms different messages highlight—and what behavior and values are perceived to be common in society.
Corruption and social norms
From the perspective of social norms research, the failure of messaging campaigns should perhaps not be surprising. Decades of research shows that social norms can exert a powerful influence on people’s behavior and beliefs. In particular, calling attention to so called “descriptive norms”—perceptions about what others typically do—is shown to cause people to behave more in accordance with the norm. Herein lies the problem with typical corruption messaging campaigns: saying that corruption “enslaves a nation” is also saying that corruption is common. This highlights the prevailing descriptive norm that many people engage in corruption, and according to social norms research we should expect this to make people more prone to comply with the norm, i.e., more prone to engage in bribery.
Subscribe here to receive the Corruption in Fragile State Blog's posts. We publish every three weeks—enough to keep you informed without cluttering your inboxes.
Many closely related anti-corruption efforts also highlight descriptive corruption norms more indirectly by trying to expose the public to the problem of corruption. This includes the “I paid a bribe” websites in India, Kenya, and Pakistan, which provide information on citizens’ self-reported bribe payments, or the “corruption bus tours” that show citizens the sites of corrupt exchange and the destructive results of corruption in cities like Prague. The obvious appeal of such measures in general and of messaging campaigns in particular is that the strategy potentially constitutes a very cheap way of nudging people’s behavior in a desirable direction.
Hence, finding ways to make messaging strategies a more effective anti-corruption tool without downside consequences should be a priority for researchers and practitioners. To achieve this goal, the field needs to move beyond focusing on descriptive norms.
"If people believe that essentially everyone shares their distaste for corruption, they would also be more likely to view others as potential collaborators in the fight against corruption. This would arguably increase the chance that the hard collective action problems facing anti-corruption movements could be overcome."
While the power of descriptive norms is considerable, this is not the only type of social norm worth focusing on. Psychologists early on distinguished between two main types of social norms: descriptive and injunctive norms, the latter referring to perceptions of what is approved or disapproved by others. Pioneering norms researcher Robert Cialdini has noted that in some contexts where undesirable behavior is prevalent, hoping to achieve good results by highlighting descriptive norms might be futile simply because doing so will inevitably reinforce pre-existing negative norms. Rather, in such a context focusing on highlighting injunctive norms might be a much more productive route to take.
This is an often-overlooked insight among corruption researchers. An interesting fact about people’s attitudes toward corruption make a focus on injunctive norms all the more interesting: research shows that people tend to strongly condemn corruption, even in countries where corruption is endemic. Hence, while the prevailing descriptive norm in such contexts is negative (many people engage in corruption) the injunctive norm is positive (most believe corruption is unjustifiable).
Moreover, we have reason to believe that people in contexts of endemic corruption might misperceive the prevailing injunctive norm: Simply put, it is often hard to disentangle people’s actions from their beliefs. When you hear about someone paying a bribe you might infer that the person did it because they view the action as acceptable. Psychologists call attributing intentions and beliefs from observed actions in this way “the fundamental attribution error.” When you hear that corruption is common— so common it “enslaves a nation”—you might infer that most people believe corruption to be acceptable. So while people’s personal beliefs are clear (corruption is immoral and unjustifiable), their beliefs about the prevailing injunctive norm in society are likely more uncertain.
This is a discrepancy that corruption messaging campaigns could exploit. If misperceptions about the prevailing injunctive norm in society could be partly corrected, this would likely have positive consequences: if people believe that essentially everyone shares their distaste for corruption, they would also be more likely to view others as potential collaborators in the fight against corruption. This would arguably increase the chance that the hard collective action problems facing anti-corruption movements could be overcome.
Lessons from Mexico
In a recent study in Mexico, I set out to test the basic premise that a focus on injunctive norms could be a productive route to take. First, using a question design specifically tailored to elicit honest answers, I asked about people’s own attitudes toward corruption as well as their perception of the attitudes of other people in society. The results revealed an interesting discrepancy: while 94 percent said they believed that bribery could never be justified, a substantially lower share, 64 percent, believed that other people held the same conviction. Respondents also expressed a high degree of uncertainty regarding others’ corruption attitudes.
In a second study, I randomly informed a different group of respondents about the very strong anti-corruption attitudes found in the first study. As a result, respondents reported higher trust in other Mexicans, were less prone to believe corruption to be an ingrained part of Mexican culture, and expressed lower willingness to pay a bribe to avoid a traffic ticket. Hence, this was the opposite result of the large number of studies that have tried to highlight descriptive corruption norms. In short, respondents were initially overly negative regarding the prevailing injunctive norm in society but revised their beliefs in response to credible information about Mexicans’ solid anti-corruption attitudes.
"Finding ways to make messaging strategies a more effective anti-corruption tool without downside consequences should be a priority for researchers and practitioners. To achieve this goal, the field needs to move beyond focusing on descriptive norms."
While more research is clearly needed, my study provides a framework for incorporating injunctive norms into messaging strategies that can be utilized and explored in different contexts. We know from worldwide surveys that corruption is essentially universally condemned, so the basic pattern where citizens have strong anti-corruption attitudes but don’t necessarily think that others do, is likely to be present in many, if not most, developing countries. This opens up a window to explore new approaches to corruption messaging.
Simple messaging campaigns are no quick fix for the problem of endemic corruption. But current strategies might often in fact make the problem worse. To improve efficiency, researchers and practitioners should embrace the full power of social norms by also focusing on people’s moral attitudes. Instead of communicating the fearful message that corruption enslaves the nation, billboards could be given a more hopeful tone: that citizens stand united in their condemnation of the poison that is corruption.
Mattias is an associate professor of political science and is associated with the Quality of Government institute at the University of Gothenburg. His research interests include corruption, political behavior and political psychology, social norms, armed conflict and public opinion, and political representation. He is involved in several ongoing research projects and is the principal investigator of the project "The Myth of the (Ir)rational Voter? Theoretical and Methodological Advancements for Studying Voter Rationality"