Using vector autoregressions on U.S. time series for 1957-1979 and 1983-2004, we find government spending shocks to have stronger effects on output, consumption, and wages in the earlier sample. We try to account for this observation within a DSGE model featuring price rigidities and limited asset market participation. Specifically, we estimate the structural parameters of the model for both samples by matching impulse responses. Model-based counterfactual experiments suggest that increased asset market participation accounts for some of the changes in fiscal transmission. However, the key quantitative factor appears to be the more active monetary policy of the Volcker-Greenspan period.