The effects of asset purchase programs on macroeconomic variables are likely to be moderate. We reach this conclusion after simulating the impact of the Federal Reserve’s second large-scale asset purchase program (LSAP II) in a DSGE model enriched with a preferred habitat framework and estimated on U.S. data. Our simulations suggest that such a program increases GDP growth by less than half a percentage point, although the effect on the level of GDP is very persistent. The program’s marginal contribution to inflation is very small. One key reason for our findings is that we estimate a small degree of financial market segmentation. If we enrich the set of observables with a measure of long-term debt, the semi-elasticity of the risk premium to the amount of debt in private-sector hands is substantially smaller than that reported in the recent empirical literature. In this case, our baseline estimates of the effects of LSAP II on the macroeconomy decrease by at least a factor of two. Throughout the analysis, a commitment to an extended period at the zero lower bound for nominal interest rates increases the effects of asset purchase programs on GDP growth and inflation.