This paper begins with the observation that the volatility of factor input growth is insufficient to explain the volatility in the growth rate of output, and explores the empirical plausibility of the hypothesis that this fact is due to the presence of productive externalities and increasing returns to scale. We construct a quantitative equilibrium macroeconomic model which incorporates these features, and allows for demand shocks operating at the level of the consumer. We employ the method of Hall (1986) and Parkin (1988) to measure these demand shocks, and use these measured disturbances to conduct stochastic simulations of the model. We find that the model with increasing returns, when driven by measured demand shocks, generates time series which replicate the basic stylized facts of U.S. business cycles, although with lower amplitude. However, in the absence of increasing returns the measured demand shocks do not produce a characteristic business cycle response. When preference shocks are combined with productivity shocks, we find that both the increasing returns and the constant returns models correctly predict a weak correlation between hours and wages, while the predictions of the increasing returns model provide the better overall match with the data.