Summary During the 1990s, recurrent crises linked to abrupt changes in the direction of international financial flows have been observed. Both the Mexican crisis of 1994-1995 and the Asian crisis of 1997-1998 had important propagation effects and exhibited high levels of contagion that were underestimated. Observers have tried to identify the ex-ante sources of vulnerability and policy mismanagement. Once more, economists have increased the volume of knowledge arising from traumatic episodes. Among other implications, the profession is now concerned about the optimal exchange rate regime, the sustainability of large current account deficits, the vulnerability of malfunctioning financial systems, the propagation effects of highly leveraged hedge funds, and risks linked to currency mismatches and to the term structure of external liabilities. The purpose of this article is to emphasize the links between external vulnerability and excess domestic expenditure explained by private sector behaviour. The motivation is based on the feature that at the eve of their respective crises, Mexico and most of the Southeast Asian economies in which the last crisis detonated exhibited significant current account deficits while their fiscal accounts were under control. This was also the case of Chile in 1996-1997, prior to its own recession. The mix of large current account deficits and fiscal balance or fiscal surplus implies that the origin of excess domestic expenditure –and of vulnerability– was non-fiscal.What are the macroeconomic consequences of an excess of non-fiscal expenditure? What are the consequences of inconsistent policy targets? Which should be the consistent policy responses? These are the main questions tackled in this paper by means of a formal analytical model. In section A we develop the case of an open economy with no voluntary financial flows, and analyse the effects of inconsistent policy targets. Two relevant and intuitive conclusions are that fiscal policy crowds out private expenditure, and that inflation is the natural outcome of inconsistent policy targets. In section B we develop the case with an open capital account, where the conclusions are that inconsistent policy targets generate external vulnerability rather than inflation, and that private expenditure crowds out fiscal policy. Moreover, in the context of consistent targets, fiscal policy needs to accommodate any excess of private expenditure to avoid external vulnerability. That is to say, the larger the level of private expenditure funded by external financing, the larger the requirement of fiscal effort to adjust to a consistent equilibrium. In section C we argue that if fiscal policy cannot be crowded out beyond a politically feasible level of fiscal surplus, a consistent equilibrium needs additional policy instruments. We analyse the case of a Tobin tax and the case of a flexible tax as alternatives to an adaptive fiscal policy. Finally, we present a number of special cases in the Appendix which prove the robustness of our conclusions.