John J., III Donohue
I discuss the empirical revolution in law and economics, and use the analysis of the deterrent impact of the death penalty to chart the tremendous advances in estimating causal effects since the mid-1970s. This story highlights how ostensibly sophisticated studies frequently generate incorrect estimates, and how difficult it is to know what studies should be believed—a difficulty open to being exploited by those (the media, think tanks, and others) who seek to promote clearly weak studies for some private agenda. I offer a hierarchy of methodologies to assist in evaluating empirical studies and some suggestions for promoting the search for truth.