The deterrent effect of the death penalty? Evidence from British commutations during World War I

Abstract During World War I, the British Army relied on the death penalty to enforce strict discipline, handing down over 3000 death sentences for desertion and other offenses. Yet only around 12% of these sentences were actually carried out; the remaining 88% were quietly commuted to lesser punishments. Crucially, soldiers themselves were unaware that most death sentences would be commuted, causing them to perceive the risk of execution as uniformly high. This hidden “lottery” in the application of the death penalty provides a rare opportunity to study deterrence under conditions where the threat of capital punishment was both visible (through executions) and secretly mitigated (through commutations). I show that, overall, executing soldiers did not strongly deter subsequent desertions. However, when the executed soldier was Irish—an ethnic group often marginalized within the British Army—desertion rates in that unit actually rose. This divergence sheds light on the critical role of legitimacy in shaping compliance. Among many Irish soldiers, the British command was perceived as less legitimate, so executing an Irish comrade could breed resentment instead of deterrence. This finding underscores a fundamental argument in the literature on deterrence and compliance: punishment severity alone does not guarantee obedience. When individuals or groups already harbor doubts about the authority’s legitimacy, harsh penalties can backfire and spur further defiance. The British-Irish split thus illustrates how perceived legitimacy can magnify or negate deterrent effects—an insight that resonates in contemporary debates about the death penalty and law enforcement (JEL N44, K14, K42, P48).