The Role of Capital on Noise Shocks

An important topic in recent macroeconomic literature is the potential effects of noise, or expectational, shocks on aggregates. Most of the past analysis has used some derivative of a New Keynesian model with labor as the only input, but doing so fails to consider that some input decisions may not have their returns realized today. To capture and explore this possibility under a noise shock framework, I incorporate endogenous capital accumulation in a New Keynesian model and solve for explicit dynamics using the method of undetermined coefficients. Results show that the qualitative predictions hold between my model and that without capital, but inclusion of capital significantly dampens the size of noise shocks: excluding capital raises the initial response of output to a noise shock by more than double. As usable capital on impact is fixed, labor responds less than the no-capital equivalent because it faces diminishing returns to scale. Coupling this effect with quasi-fixed capital accumulation leads to a muted output response in comparison to labor-only models. In addition, noise shocks are transient compared to a permanent shock. This is due to noise shocks being predominantly consumption driven opposed to investment driven. These results suggest that models without capital may overstate the significance of noise shocks.