Pol Campos-Mercade, Lund University

Threshold Incentives and Academic Performance

Abstract

Rewards conditional on attaining a given performance threshold are widespread in economic life. In education, they are especially prevalent: students are typically accepted to college or awarded scholarships only if their academic performance reaches a given standard. How do these threshold incentives affect individuals’ performance? We run a pre-registered experiment with 628 university students to answer this question. We place the threshold in the middle of the GPA distribution and explore the (potentially heterogeneous) treatment effects for students of different abilities. Only students right below the threshold improve their GPA (0.26 SD). This effect is driven by males, whose GPA increases by 0.72 SD. When incentives are removed, treated students perform worse (0.11 SD). This negative long-run effect is driven by females, whose performance decreases by 0.22 SD. Using survey evidence, we reject the hypothesis that incentives crowd out intrinsic motivation. Instead, students who are offered incentives but fail to reach the threshold seem to lose confidence in their ability. This leads them to enroll in fewer courses, thereby damaging their overall performance.

Pol Campos-Mercade is a doctoral student and does research in behavioral and expeirmental economics.

He is especially interested in understanding how goals can help people sacrifice present gratification for future happiness, as for example saving money, studying hard, eating healthy and training. He has recently received funding to study this question by using field experiments. In particular, he will study how students can be motivated to study more by using clear goals. His supervisors are Erik Wengström and Erik Mohlin.