We evaluate two leading explanations for inventories, the (S,s) and stockout avoidance motives, examining each within dynamic stochastic general equilibrium environments. We find that the (S,s) model is far more consistent with the cyclical behavior of aggregate inventories in the postwar United States when fluctuations arise from technology shocks, rather than preference shocks, whereas the converse is true for the stockout avoidance model. The (S,s) model succeeds in explaining the average magnitude of inventories in the U.S. economy and in reproducing the cyclical regularities involving inventories and other aggregate series. The stockout avoidance model does not. Even with idiosyncratic risk added to strengthen it, the stockout avoidance motive is insufficient to generate stocks near the data without destroying model performance along other important margins. Moreover, it appears incapable of sustaining inventories alongside capital. These findings suggest a fundamental flaw in reduced-form inventory models where stocks are loosely rationalized by this motive.