In the same week that the pound sterling and the Italian lira were forced by speculative pressures to suspend their participation in the Exchange Rate Mechanism (ERM) of the European Monetary System (EMS), a study was published which concluded that exchange market intervention was a far more potent instrument for managing exchange rates than the conventional wisdom had held (Catte, Galli and Rebecchini, I 992 a). I shall argue in this paper that, far from providing instant refutation of the new optimism regarding the potency of intervention, what we currently understand about exchange rate theory is consistent with both the tumult in the exchange markets in September I992 and the new optimism. The paper seeks to draw the implications of current knowledge for what exchange rate management can and cannot hope to accomplish. The first section of the paper reviews where exchange rate theory stands. A second section discusses what we know about the possibility of managing exchange rates and how that ties in with the theory. The final section explores how exchange rate management needs to vary, depending on whether it is intended to use the exchange rate as a nominal anchor or to limit misalignments in a more Keynesian approach to policy design.