We use time-series methods to estimate a simple aggregate-supply-aggregate-demand model in order to analyze the comparative performance of fixed- and flexible-exchange-rate systems and test competing hypotheses designed to explain shifts between exchange-rate regimes. The paper provides a coherent explanation of the causes and consequences of the shift from the Bretton Woods System of pegged exchange rates to the post-Bretton-Woods float. The shift from fixed to floating was associated with a modest increase in the cross-country dispersion of supply shocks but not with an increase in their average magnitude. In contrast, there was little change in either the cross-country dispersion or the average magnitude of demand shocks. More important in explaining the collapse of Bretton Woods were factors that heightened the impact of shocks on the external accounts, forcing governments to respond to supply shocks with changes in demand that stabilized prices and the exchange rate at the expense of increased output volatility.