Understanding the Inventory Cycle

Why is there inventory investment when its expected rate of return is strictly dominated by that of fixed-capital investment? Why is inventory investment procyclical at business-cycle frequencies but countercyclical at the very high frequencies (e.g., 2-3 quarters per cycle)? Why does the variance of production exceed the variance of sales at the business-cycle frequencies but not so at the high frequencies? Why is inventory investment so volatile relative to GDP at the high frequencies but not so at the business cycle frequencies? Explaining these seemingly paradoxical features of inventory behavior is of great importance because for many years economists have speculated that understanding inventory fluctuations may provide the key to understanding the business cycle. This paper provides a general equilibrium analysis on inventory cycles and their relations to the business cycle. I show that in an environment where production and fixed-capital investment cannot adjust instantaneously to respond to consumption demand shocks, firms opt to hold inventories in the short run so as to avoid stockout and to smooth production against demand uncertainty. These incentives for holding inventories in the short run have different effects on inventory cycles across different cyclical frequencies. At the high frequencies inventory fluctuations are dominated by the production-smoothing motive and at the business-cycle frequencies inventory fluctuations are dominated by the stockout-avoidance motive. Consequently, inventory investment appears to be countercyclical and volatile at the high frequencies but procyclical and relatively smooth at the business-cycle frequencies, production appears to be less volatile than sales at the high frequencies but more volatile than sales at the business-cycle frequencies. I also show that the inventory cycle and the business cycle are intimately related by sharing a common source of uncertainty--consumption demand, which leads to the phenomena that consumption Granger causes output and fixed investment, that consumption comoves with output and fixed investment, and that consumption appears to be smooth relative to output and fixed investment.