Reverse hysteresis? Persistent effects of autonomous demand expansions

Abstract

<jats:title>Abstract</jats:title> <jats:p>Empirical works documenting highly persistent effects of negative demand shocks (‘hysteresis’) have questioned the prevailing wisdom that potential output is exogenous to aggregate demand fluctuations. We assess whether the effects of positive demand shocks also tend to persist beyond the short run. We estimate the impact of 126 aggregate demand expansions in OECD countries between 1960 and 2015 through local projections, using a dynamic two-way fixed-effects model and a propensity score-based specification. We find that demand expansions exert positive persistent effects on GDP, participation rate and capital stock. Effects on the unemployment rate and productivity are also strong and quite persistent, but evidence regarding their permanence is mixed. The effect on the inflation rate is positive but small and imprecisely estimated, and there is no sign of accelerating inflation. Our results bear relevant implications for existing models of hysteresis and for theories of demand-led growth.</jats:p>

On the Non-Inflationary effects of Long-Term Unemployment Reductions

“…A detailed list of countries in our sample can be found in Appendix 4. Notably, the selection of countries is also in line with the current literature on hysteresis, which principally involves mature economies (Ball, 2009(Ball, , 2014Blanchard et al, 2015;Girardi et al, 2020).…”

“…However, to overcome as much as possible endogeneity issues in our estimations, we will explicitly control for all the macroeconomic preconditions that turned out to be significantly different in our episodes of strong long-term unemployment reduction vis à vis the control group: practically, we will insert these (lagged) variables as additional regressors, with a view to controlling for the pre-treatment trends of these variables (cf. Girardi et al, 2020). Furthermore, it has to be recalled that in all our estimates we will make use of time-(in addition to country-) fixed effects: being our dependent variable taken in annual changes, this strategy will further mitigate the endogeneity bias, if any, inasmuch year dummies would capture the presence of potential omitted variables which would simultaneously affect both inflation and LTU shocks.…”

“…First it casts severe doubt on the legitimacy of New Consensus macroeconomics, whose main feature is confidence in monetary policy's ability to reduce output volatility and ensure stable and lasting growth in capitalist economies (Goodfriend, 2007). It also favors the revival of the theme of hysteresis, both in the unemployment rate and output level (Blanchard and Summers, 1986;Ball, 2014;Blanchard et al, 2015;Blanchard, 2018;Girardi et al, 2020). 1 Hysteresis is generally meant to imply that by increasing the actual unemployment rate significantly, a deep recession may also cause a change in the equilibrium unemployment rate or the Non-Accelerating Inflation Rate of Unemployment (NAIRU) and, consequently, in potential output.…”

“…Secondly, we present an econometric analysis aimed at verifying whether inflationary surges are likely to occur during and after episodes of strong reduction of the long-term unemployment rate and in a subset of such episodes when, according to official estimates, the unemployment rate was already below the conventionally accepted NAIRU. To do so, we employ local projections, as initially introduced in Jordà (2005) and recently used in Girardi et al (2020), to assess the dynamic impact in terms of inflation of 78 (58 in the subset) cases of sharp, long-term unemployment reductions that occurred between 1983 and 2016.…”

Negative Interest Rate Policy to Fight Secular Stagnation: Unfeasible, Ineffective, Irrelevant, or Inadequate?

“…What must be stressed further concerning fiscal policy is that its role transcends that of a simple, however important, fix for stagnation episodes. Empirical support for this position comes from the notable enquiry of Girardi, Paternesi Meloni, and Stirati (2020). The authors investigate the phenomenon of 'reverse hysteresis', whereby GDP level, capital stock and employment are all positively and lastingly affected by autonomous demand (primary public expenditure and exports) expansions.…”