Best, long-term analysis flows aff

Mehdi El Herradi and Aurélien Leroy, “Monetary policy and the top one percent: Evidence from a century of modern economic history,” Working Paper No. 632, De Nederlandsche Bank, April 2019, https://www.dnb.nl/media/jbybr2et/working-paper-no-632_tcm47-383633.pdf

5 Conclusion

This paper sought to investigate the distributional consequences of monetary policy between 1920 and 2015. The central idea that guided this paper’s argument is that the existing literature considers the distributional effects of monetary policy using data on inequality over a short period of time. However, inequalities tend to vary more in the medium-to-long run. We address this shortcoming by studying how changes in monetary policy stance over a century impacted the income distribution while controlling for the determinants of inequality. To do so, we combined two large datasets: (i) the World Inequality Database (WID) to extract data on the share of national income held by the richest 1% and the Jordà-Schularick-Taylor Macrohistory Database, which allows us to access large series of macroeconomic and financial variables.

Our empirical strategy considers two complementary approaches that conciliate between consistency and flexibility. The first approach involves a Panel VAR framework, which delivers the orthogonalized impulse response function of inequality to a monetary policy shock. The second generates dynamic responses by inequality in a Local Projections (LP) setting, wherein a rich set of macroeconomic controls is included. This framework allows us to consider a natural experiment, where exogenous perturbations in monetary policy are driven by factors unrelated to domestic economic conditions. Such exogenous perturbations enter as an instrumental variable, which traces out the impulse responses of inequality.

The results obtained from both empirical methods indicate that loose monetary conditions strongly increase the top one percent’s income and vice versa. In fact, following an expansionary monetary policy shock, the share of national income held by the richest 1 percent increases by approximately 1 to 6 percentage points, according to estimates from the Panel VAR and Local Projections (LP). This effect is statistically significant in the medium run and economically considerable. We also demonstrate that the increase in top 1 percent’s share is arguably the result of higher asset prices. The baseline results hold under a battery of robustness checks, which (i) consider an alternative inequality measure, (ii) exclude the U.S. economy from the sample, (iii) specifically focus on the post-WWII period, (iv) remove control variables and (v) test different lag numbers. Furthermore, the regime-switching version of our model indicates that our conclusions are robust, regardless of the state of the economy.