Working Paper: CEPR ID: DP14595
Authors: Jan Potters; Sigrid Suetens
Abstract: We examine whether optimization incentives --- incentives to best-respond --- have an effect on behavior in finitely repeated two-player dilemma games with strategic complements. We run an experiment in which we increase optimization incentives in two different ways compared to a baseline treatment. In the first treatment, the increase in optimization incentives is created by an increase in payoffs on the best-response curve, while its slope remains unchanged. In the second treatment, the increase in optimization incentives takes the form of an increase in the slope of the best-response curve, while best-response payoffs remain unchanged. We find that the impact of optimization incentives is overshadowed by the effect of the slope of the best-response curve. Although an increase in optimization incentives leads subjects to best-respond more frequently when the best-response curve is relatively flat, it leads to more cooperative behavior if it is accompanied by an increase in the slope of the best-response function.
Keywords: optimization incentives; strategic complementarity; repeated game; cooperation; experiments
JEL Codes: C91; D01; D74
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| optimization incentives (L21) | cooperation rates (C71) |
| increase in optimization incentives and increase in slope of best-response curve (D43) | cooperation rates (C71) |
| slope of the best-response curve (C73) | cooperation rates (C71) |
| optimization incentives and increase in slope of best-response curve (D43) | cooperation rates (C71) |
| optimization incentives without increase in slope of best-response curve (D43) | cooperation rates (C71) |
| stronger strategic complementarity (F12) | positive effects of optimization incentives on cooperation (C71) |