Pitfalls in the Use of Time Series Methods to Study Deterrence and Capital Punishment

Abstract

Evaluate the use of various time series methods to measure the deterrence effect of capital punishment. The analysis of the time series approach to deterrence is conducted at two levels. First, the mathematical foundations of time series methods are described and the link between the time series properties of aggregate homicide and execution series and individual decision making is developed. Second, individual studies are examined for logical consistency. The analysis concludes that time series methods used to study the deterrence effects of capital punishment suffer from fundamental limitations and fail to provide credible evidence. The common limitation of these studies is their lack of attention to identification problems. Suggestions are made as to directions for future work that may be able to mitigate the weaknesses of the current literature. Time series studies of capital punishment suffer from sufficiently serious identification problems that existing empirical findings are compatible with either the presence or the absence of a deterrent effect.