A central question in the empirical monetary policy literature is how do asset prices respond to an unexpected monetary policy shock. We provide empirical evidence on this issue by augmenting the VAR model specification of Uhlig (2005) with the S&P 500 Composite Index, and estimating the model on monthly US data. We use the sign restrictions put forth in Uhlig (2005) as identifying assumptions. The sign-restricted SVAR impulse responses point towards a positive asset price response to an increase in the monetary policy instrument. As it turns out, however, the resulting identified monetary policy shocks correlate only weakly with the monetary policy shock series of Romer and Romer (2004) that we view as a benchmark series for monetary policy shocks. We show that this finding matters greatly when analyzing (structural) impulse responses, and we propose to restrict attention to those (sign-restricted) specifications that yield monetary policy shocks that are highly correlated with the Romer and Romer (2004) series. Concentrating only on these (sign-restricted) specifications uncovers a mildly negative response of asset prices to an increase in the monetary policy instrument.
You can find the paper here.