Are Women Targeted for Corruption? Evidence from Randomized Conjoint Experiments on Traffic Police in Nigeria

Policy documents from organizations such as Transparency International and the United Nations Office on Drugs and Crime have long claimed that women are more likely than men to be targeted for bribes by street-level bureaucrats, particularly in contexts of petty corruption (UNDP 2008; Matsheza et al. 2012). These claims are often grounded in qualitative evidence and lived experience, and typically focus on extortion—defined by Lambsdorff (2007) as an exchange in which “the agent uses her power to extract money or other benefits from the client,” who may be legally entitled to receive the service without payment (p. 18). One common explanation is that women are perceived as easier targets—less confrontational and less likely to retaliate. Studies have also shown that women, particularly female entrepreneurs, may be more susceptible to bribery demands due to a lack of political connections and limited access to networks of influence (Wellalage, Locke, and Samujh 2019; Nguyen, Nguyen, and Pham 2021). However, cross-national surveys often show that women are less likely than men to report paying bribes (Mocan 2008; Goetz 2007; Swamy et al. 2001). Several explanations have been proposed: women may have fewer interactions with the state that would expose them to corruption, may be more honest and less corrupt (Dollar et al. 2001; Esarey and Schwindt-Bayer 2018), or may fear retaliation if they report corrupt officials (Asomah et al. 2023). Yet these studies typically rely on cross-sectional data and fail to distinguish between extortion and collusive forms of petty corruption—transactions that benefit both the bribe-payer and the official (Romero 2024). This conflation, combined with unmeasured confounding variables, limits our ability to assess whether gender itself drives targeting decisions by corrupt officials. This project addresses these gaps through a set of randomized conjoint experiments designed to isolate the effect of a potential bribe-payer’s gender on the likelihood of being solicited for a bribe. The experiments are conducted with approximately 600 traffic police officers in Kano City, Nigeria—a setting marked by high levels of petty corruption, frequent bribe solicitation by police, and widespread reports of violence both by and against officers (Alemika and Chukwuma 2000; Ike et al. 2021). Conjoint experiments are well suited to estimate causal effects in observationally messy environments, as they allow researchers to systematically vary specific attributes—such as gender—while holding others constant. In addition to estimating the direction and strength of the gender effect, the project investigates seven potential mechanisms that may mediate or moderate this relationship. These include benevolent sexism (the belief that women should be protected), hostile sexism, and perceptions of the individual’s ability to pay, social status, political connections, formal officeholding, legal knowledge, and susceptibility to violence. Prior work shows that street-level officials tend to extort individuals perceived to have low capacity to retaliate, whether due to lack of political power or economic means (Fried, Lagunes, and Venkataramani 2010; Robinson and Seim 2018). In patriarchal societies, women often possess less institutional power, which could make them more vulnerable—though norms of protection may simultaneously dissuade targeting. By distinguishing between extortion and collusion and leveraging experimental methods, this study offers a more precise understanding of how gender shapes experiences with corruption in everyday state–citizen interactions.