Hump - shape Uncertainty , Agency Costs and Aggregate Fluctuations ∗

Previously measured uncertainty shocks using the U.S. data show a hump-shape time path: Uncertainty rises for two years before its decline. Current literature on the effects uncertainty on macroeconomics, including housing, has not accounted for this observation. Consequently, the literature on uncertainty and macroeconomics is divided on the effects and the propagation mechanism of uncertainty on aggregate fluctuations. This paper shows that when uncertainty rises and falls over time, then the output displays hump-shape with short expansions that are followed by longer and persistent contractions. And because of these longer and persistent contractions in output, uncertainty is, on average, counter-cyclical. Our model builds on the literature combining uncertainty and financial constraints. We model the time path of uncertainty shocks to match empirical evidence in terms of shape, duration and magnitude. In our calibrated models, agents anticipate this hump-shape uncertainty time-path once a shock has occurred. Thereby, agents respond immediately by increasing investment (i.e. pre-cautionary savings), but face a substantial drop in investment, consumption and output as more uncertain times lie ahead. With persistent uncertain periods, both risk premia and bankruptcies increase which cause a further deterioration in investment opportunities. Besides, we show that accounting for hump-shape uncertainty measures can result in a large quantitative effect of uncertainty shock relative to previous literature.