(S,s) insights into the role of inventories in business cycles and high frequency fluctuations

We develop an equilibrium model to explain salient business cycle patterns involving aggregate production, sales and inventory investment alongside a distinct set of patterns at high frequencies. Our firms face idiosyncratic productivity shocks and fixed costs of ordering inputs, leading them to order infrequently and accumulate inventories. Thus, the model’s aggregate state vector includes a time-varying distribution of firms over productivities and inventories. Disciplined by data on aggregate inventories and firm-level sales and output, our model reproduces key patterns in the data at business cycle frequencies: Inventory investment is procyclical and positively correlated with final sales, GDP varies more than sales, and the inventory-to-sales ratio is countercyclical. These successes are robust to a wide range of micro-level parameters governing firms’ order costs and relative productivities, while those parameters are key to the model’s high-frequency performance. When order costs are more predictable and shifts in relative productivities are transitory, the model also performs well in key high frequency respects: The relative volatility of inventory investment rises sharply, and sales and inventory investment are negatively correlated, while both series maintain positive correlations with GDP. Despite these distinctions, our model predicts that aggregate fluctuations are surprisingly unaffected by inventories even at high frequencies.