A new look at historical monetary policy (and the great inflation) through the lens of a persistence-dependent policy rule

Abstract

The origins of the Great Inflation, a central 20th-century U.S. macroeconomic event, remain contested. Prominent explanations are poor inflation forecasts or inaccurate output gap measurement. An alternative view is that the Federal Open Market Committee (FOMC) was unwilling to fight inflation, perhaps due to political pressures. Here, we sort this out via a novel econometric approach, disaggregating the real-time unemployment and inflation time series entering the FOMC historical policy reaction-function into persistence components, using one-sided Fourier filtering; this implicitly estimates the unemployment gap in actual use. We find compelling evidence for (economically interpretable) persistence-dependence in both variables. Furthermore, our results support the “unwilling to fight” view: the FOMC’s unemployment gap responses were essentially unchanged pre- and post-Volcker, while its inflation responses sharpened markedly starting with Volcker.

Finding a Stable Phillips Curve Relationship: A Persistence-Dependent Regression Model

“…Our approach differs fundamentally from the existing Phillips curve literature in that we make use of the "persistence-dependent regression" econometric methodology developed in Ashley and Verbrugge (2009) and Ashley, Tsang, and Verbrugge (2020). These methods allow us to use the data to determine the relationship between inflation and movements in the unemployment rate at different persistence levels.…”

“…13 phase plots are substantially more challenging to interpret than our approach; and in the presence of feedback, Granger describes interpretation of such plots as "difficult or impossible" (Granger, 1969). 13 Stata and RATS software to implement the decomposition is available from the authors, and an extensive description of our moving-window-based persistence decomposition algorithm can be found in the Technical Appendix to Ashley, Tsang, and Verbrugge (2020). For replicability purposes we note that (per footnote 44 on page 48 of that paper) the standard Christiano-Fitzgerald bandpass filter was iteratively used here.…”

“…Ciner (2015) finds that stock returns have a persistence-dependent relationship to inflation movements. And Ashley, Tsang, and Verbrugge (2020) find that historical FOMC policy responses to inflation and unemployment are persistence-dependent, with persistent movements in the unemployment rate, and transitory movements in inflation, being ignored by policymakers. 6 The research question motivating the present research is the following: does the natural rate distinction fully capture persistence-dependence in the Phillips relationship?…”

Is It Time to Reassess the Focal Role of Core PCE Inflation?

“…The inflation process has changed since the 1970s. Monetary policy has changed markedly (see, e.g., Ashley, Tsang, and Verbrugge (2020)) and the global economy is much more integrated. There have been major changes in available goods and services and in preferences.…”

How macroeconomists lost control of stabilization policy: towards dark ages

“…In Kalman , an electrical engineer, wrote the key paper for solving linear quadratic optimal control (linear quadratic regulator, LQR). 3 He extended the static Tinbergen (1952) principle, namely that there should be as many policy instrument as policy targets, to a dynamic setting. Kalman's (1960a) controllability definition is such that a single instrument can control for example three policy targets, but in three different periods (Aoki (1975) and a control engineer.…”