Currency areas, policy domains, and the institutionalization of fixed exchange rates

Emerging-market countries are being urged to choose between freely floating exchange rates and firmly fixed rates supported by strong institutional arrangementsûcurrency boards, monetary unions, or formal dollarization. This paper assesses the benefits and costs of institutionalizing fixed rates by synthesizing and supplementing the theory of optimum currency areas (OCA theory). It shows that (1) OCA theory and related empirical work have been excessively influenced by the special case used originally by Robert Mundell, where exogenous shocks display mirror-image asymmetry; (2) OCA theory ignores a vital difference between the domains of monetary policy under a monetary union and other institutional arrangements; (3) because it neglects the way in which a monetary union reduces the debt-creating effects of fiscal stabilizers, OCA theory understates the strength of the case for combining a monetary union with a fiscal federation. The paper also criticizes recent work by Jeffrey Frankel and Andrew Rose in which they claim to show that monetary union reduces asymmetric shocks and thus makes monetary union less costly. The paper suggests that their results may reflect the effects of monetary union on the transmission of shocks rather than their incidence.