Below are summaries and links to drafts of my working papers.
We use a predictable change in the intraday volatility of index futures to identify the effect of stock returns on monetary policy. This identification approach relies on a weaker set of assumptions than required under identification through heteroskedasticity based on lower frequency data. Our identification approach also allows examining time variation in the reaction of monetary policy to the stock market. The results show a sharp increase in the response of monetary policy expectations to stock returns during recessions and bear markets. This finding is consistent with the existence of the so-called “Fed put.”
The Reagan Revolution cut deeply into the dominant Democratic coalition that began with the New Deal Democrats. This paper shows that the redistributional effect of trade across U.S. regions played an important role in the political realignment in the 1980s. While the Reagan administration promoted fair and free trade, the Democratic Party responded by supporting protectionist measures against Japanese competitors. We show that voting shares for the Republican presidential candidates decreased in the Midwest where they faced strong import competition with Japan and increased in the South and the farm belt where U.S. exports to Japan created local business opportunities.
We use a new measure of the output gap proposed by Hamilton (2017) in conjunction with Taylor’s (1979) efficiency frontier to evaluate monetary policy during the last two “Great” contractions. Our results suggest that two periods of widespread bank failure, coinciding especially with the failure of the Bank of United States in December 1930 and the failure of Lehman Brothers in September 2008, impeded the transmission of subsequent monetary policy. Our results show that in both cases the Federal Reserve failed in containing and mitigating the macroeconomic effects of the bank failures. However, our results also suggest that after the major bank failures, the Federal Reserve’s responses were substantially better during the Great Recession compared to that of the Great Depression.
Exchange Traded Commodities (ETCs) that extensively use derivatives incur high trans action costs and, due to the idiosyncratic characteristics of futures contracts, may have difficulty tracking the prices of the targeted commodity. This paper uses machine learning to select equities to form portfolios that are superior to the ETCs. The portfolios 1) more closely track the target commodity, 2) incur smaller transaction costs, and 3) can easily be implemented by retail and institutional investors.