We conduct a field experiment to evaluate the effect of extrinsic rewards, both financial and non-financial, on the performance of agents recruited by a public health organization to promote HIV prevention and sell condoms. In this setting: (i) non-financial rewards are effective at improving performance; (ii) the effect of both rewards is stronger for pro-socially motivated agents; (iii) the effect of both rewards is stronger when their relative value is higher. The findings illustrate that extrinsic rewards can improve the performance of agents engaged in public service delivery, and that non-financial rewards can be effective in settings where the power of financial incentives is limited.
We posit that household decision-making over fertility is characterized by moral hazard due to the fact that most contraception can only be perfectly observed by the woman. Using an experiment in Zambia that varied whether women were given access to contraceptives alone or with their husbands, we find that women given access with their husbands were 19% less likely to seek family planning services, 25% less likely to use concealable contraception, and 27% more likely to give birth. However, women given access to contraception alone report a lower subjective well-being, suggesting a psychosocial cost of making contraceptives more concealable.
- Wartime rape is neither ubiquitous nor inevitable. The level of sexual violence differs significantly across countries, conflicts, and particularly armed groups. Some armed groups can and do prohibit sexual violence. Such variation suggests that policy interventions should also be focused on armed groups, and that commanders in effective control of their troops are legally liable for patterns of sexual violence they fail or refuse to prevent.
- Wartime rape is also not specific to certain types of conflicts or to geographic regions. It occurs in ethnic and non-ethnic wars, in Africa and elsewhere.
- State forces are more likely to be reported as perpetrators of sexual violence than rebels. States may also be more susceptible than rebels to naming and shaming campaigns around sexual violence.
- Perpetrators and victims may not be who we expect them to be. During many conflicts, those who perpetrate sexual violence are often not armed actors but civilians. Perpetrators also are not exclusively male, nor are victims exclusively female. Policymakers should not neglect nonstereotypical perpetrators and victims.
- Wartime rape need not be ordered to occur on a massive scale. Wartime rape is often not an intentional strategy of war: it is more frequently tolerated than ordered. Nonetheless, as noted, commanders in effective control of their troops are legally liable for sexual violence perpetrated by those troops.
- Much remains unknown about the patterns and causes of wartime sexual violence. In particular, existing data cannot determine conclusively whether wartime sexual violence on a global level is increasing, decreasing, or holding steady. Policymakers should instead focus on variation at lower levels of aggregation, and especially across armed groups.
Why do some armed groups commit massive wartime rape, whereas others never do? Using an original dataset, I describe the substantial variation in rape by armed actors during recent civil wars and test a series of competing causal explanations. I find evidence that the recruitment mechanism is associated with the occurrence of wartime rape. Specifically, the findings support an argument about wartime rape as a method of socialization, in which armed groups that recruit by force—through abduction or pressganging—use rape to create unit cohesion. State weakness and insurgent contraband funding are also associated with increased wartime rape by rebel groups. I examine observable implications of the argument in a brief case study of the Sierra Leone civil war. The results challenge common explanations for wartime rape, with important implications for scholars and policy makers.