Unproductive rebound: Experimental evidence from four countries

Abstract

Efficiency improvements often trigger rebound effects, but whether rebound is productive, in the sense of generating a corresponding gain in output, remains largely unknown, because existing studies observe resource use rather than the output it produces. Using a pre-registered experimental production environment across four countries (n = 4,586), we identify unproduc tive rebound, additional input use without a corresponding output gain. Following a price drop, the prevalence of unproductive rebound more than doubles, rising from 13.2% in the control condition to 31.4%, corresponding to a 2.4-fold increase. Furthermore, an experimental con dition that made the environmental cost of input use salient, by tying it to a reduction in a donation to renewable energy certificates, did not offset this response. These findings suggest that a substantial share of rebound may involve additional resource use without corresponding productivity gains, raising new questions about the welfare implications of efficiency policies.