Evidence of lengthy half-lives for real exchange rates in the presence of a high degree of exchange rate volatility has been considered as one of the most puzzling empirical regularities in international macroeconomics. This paper suggests that the measure of half-life used in the literature might be problematic and proposes alternative measures with desirable properties. Their focus on the cumulative effects of the shocks distinguishes them from the measures used in the literature. An empirical analysis of bilateral US dollar real exchange rates employing the alternative half-life measure produces results consistent with theory and indicates that the PPP puzzle is less pronounced than initially thought. Copyright © 2012 John Wiley & Sons, Ltd.