Working Papers, The Johns Hopkins University, Department of Economics 506
This paper challenges the conventional view according to which disinflations in LAC-even from low and moderate peaks-have been carried out at no cost to output. After suggesting a new methodology that allows for long-lived effects and inflation inertia when measuring costs of disinflations, large sacrifice ratios are obtained for the 1970s and 80s. Nevertheless, a new puzzle arises: disinflation costs in the 90s are negative, even with the new methodology. It is shown that an unusual combination of circumstances-i.e. capital inflows, structural reforms and the peculiar recent inflation history-can explain that fortunate result. Moreover, it is shown that LAC episodes exhibit a larger speed than G7 experiences. That speed differential explains why disinflation costs in developed nations are on average larger than LAC’s.