The Economics of Death Ceilings

Incentive design involves a tradeoff between motivating agents and distorting their efforts; for many tasks the distortions are such that flat rewards may be optimal. Yet organizations of all types from school boards to the FBI increasingly utilize rewards based on observable performance metrics. We describe the effects of one such attempt, implemented in 2004, to motivate Chinese bureaucrats to reduce accidental deaths. Each province was given a set of ‘death ceilings’ that, if exceeded, would impede government officials’ promotions. For each category of accidental deaths, we observe a sharp, significant discontinuity in reported deaths at the ceiling. Provinces with safety incentives for municipal officials experienced more rapid declines in accidental rates, suggesting a complementarity between incentives at different levels of government in achieving targets. There is no effect on production volume or costs when provinces are close to their workplace death ceilings, suggesting real safety effort (through, say, additional spending on workplace safety or reduced production rates) is unlikely to account for the discontinuities we observe . Finally, we examine whether the dynamic incentives created by death ceilings affect death reporting. While realized accidental deaths predict next year’s ceiling, we observe no evidence that provinces manipulate deaths upward to avoid a ratchet effect in the setting of death ceilings. JEL classification: D73; H75