Fairness, Ambiguity and disinflation costs

The notion that much of the reduction in disinflation costs in recent decades is due to better anchoring converges with the proposition of Dow, Simonsen and Werlang (1993) that part of the sacrifice ratios of poorly anchored economies may be caused by this coordination problem, which they modeled with ambiguity. The present paper relates this effect with fairness in a labor market in which past inflation is a reference for wage readjustments, but without a materialized reduction in effort associated with reciprocity. As in Bhaskar (1990), this coordination problem is modeled with increases in the elasticity of the wage equation, which imply lower sensibility of wage readjustments to the monetary policy of disinflations – and this is used here as a measure of the studied ambiguity. Bayesian inference with US data shows that this indicator trended upward from the mid-1960s and throughout the Great Inflation, with local peaks after the oil shocks but being significantly higher during the Volcker disinflation, leading to increases in unemployment and decreases in output. Subsequently it was rapidly reduced, with further reductions during the low inflation consolidation of the early nineties, and it was quite mild in the post-COVID-19 pandemics disinflation. The fact that the effect was markedly stronger in the Volcker disinflation (a “nominal shock”) than in the sequences of the adverse oil shocks (“real shocks”) of the Great Inflation signals to the relevance of a nominal anchoring problem, with Volcker’s aggressive monetary policy being one way of dealing with this, and inflation targeting (IT) being a second-generation related treatment. Furthermore, due to their potentially higher degree of ambiguity, disinflations of very high inflation begin with the use of exchange rate anchors (a coordination mechanism similar to the Brazilian Real plan is an alternative), that should be followed, when ambiguity reaches a moderate level, by the announcement and posterior implementation of an IT regime.