No Margin, No Mission? A Field Experiment on Incentives for Pro-Social Tasks
Executive Summary — Organizations from large corporations to NGOs use a range of nonfinancial performance rewards to motivate their employees, and these rewards are highly valued. While theory has suggested mechanisms through which nonfinancial incentives can elicit employee effort, evidence on the mechanisms, and on their effectiveness relative to financial incentives, remains scarce. This paper helps to fill this gap by providing evidence from a collaboration with a public health organization based in Lusaka, Zambia, that recruits and trains hairdressers and barbers to sell condoms in their shops. This setting is representative of many health delivery programs in developing countries where embedded community agents are called upon to deliver services and products, but finding an effective way to motivate them remains a significant challenge. Findings show the effectiveness of financial and nonfinancial rewards for increasing sales of condoms. Agents who are offered nonfinancial rewards ("stars" in this setting) exert more effort than either those offered financial margins or those offered volunteer contracts. Key concepts include:
- Nonfinancial rewards can motivate agents in settings where there are limits to the use of financial incentives.
- Nonfinancial rewards elicit effort by leveraging the agents' pro-social motivation and by facilitating social comparisons among agents.
A substantial body of research investigates the effect of pay for performance in firms, yet less is known about the effect of non-financial rewards, especially in organizations that hire individuals to perform tasks with positive social spillovers. We conduct a field experiment in which agents recruited by a public health organization to sell condoms are randomly allocated to four groups. Agents in the control group are hired as volunteers, whereas agents in the three treatment groups receive a small monetary margin on each pack sold, a large margin, and a non-financial reward, respectively. The analysis yields three main findings. First, non-financial rewards are more effective at eliciting effort than either financial rewards or the volunteer contract, and are also the most cost-effective of the four schemes. Second, non-financial rewards leverage intrinsic motivation and, contrary to existing laboratory evidence, financial incentives do not appear to crowd it out. Third, the responses to both types of incentives are stronger when their relative value is higher. Indeed, financial rewards are effective at motivating the poorest agents, and non-financial rewards are more effective when the peer group is larger. Overall, the findings demonstrate the power of non-financial rewards to motivate agents in settings where there are limits to the use of financial incentives.