Dynamic effects of Federal Reserve quantitative easing

The U.S. Federal Reserve System (the Fed) executed a series of large scale asset purchase programs, colloquially dubbed quantitative easing (QE), to support economic activity following the 2008 financial crisis. Starting in early 2010 and continuing through 2014, the Fed purchased a substantial sum of mortgage-backed securities and U.S. Treasury debt, causing its balance sheet to expand from just under one trillion USD to over four trillion USD. A significant body of empirical literature has emerged to study the impact of these financial market interventions on interest rates, asset prices, and real economic activity. These studies typically employ an event or structural vector autoregressive (SVAR) approach to model the relationships between macroeconomic variables of interest and Fed quantitative easing policy. The former method leverages high frequency data to identify the immediate, causal impact of Fed quantitative easing on financial markets, while the latter technique is able to capture the short-to-medium run dynamic relationships between QE and the macroeconomy. In this paper, we employ a novel approach that enables us to exploit high frequency data to estimate the longer-run dynamic effects of Fed quantitative easing. Specifically, we use high frequency data on Fed open market operations (OMOs) obtained from the Federal Reserve Bank of New York (FRBNY) and estimate the influence of Fed quantitative easing on interest rates, unemployment, prices, and production using a mixed-frequency vector autoregressive (MFVAR) model. This technique allows us to estimate the contemporaneous and dynamic relationships between macroeconomic variables that are released at different frequencies in a single, parsimonious framework.