We examine the impact of large-scale asset purchases of government bonds on real GDP and the CPI in the United Kingdom and the United States with a Bayesian VAR, estimated on monthly data from 2009 M3 to 2013 M5. We identify an asset purchase shock with sign and zero restrictions. In contrast to the impulse response analysis in previous work, the reactions of real GDP and CPI are left unrestricted, so as formally to test whether these variables are affected by asset purchases. We then explore the transmission channels to the domestic economy and emerging markets. Our results suggest that asset purchases have a statistically significant effect on real GDP with a purchase of 1% of GDP leading to a .36% (.18%) rise in real GDP and a .38% (.3%) rise in CPI for the United States (United Kingdom). In the United States, this policy lowers yields on long-term government bonds and the real exchange rate. In the United Kingdom, on other hand, interest rate futures and measures of financial market uncertainty are more affected. There is also some evidence that emerging market sovereign bond and corporate bond spreads decline, with industrial production rising in response a positive asset purchase shock in either country.
Unconventional monetary policy Bayesian VAR Hierarchical prior Litterman prior