Commodity storage and the cost of capital: Evidence from Illinois grain farms

Abstract

Commodity inventories are the key state variable determining the magnitude of commodity price responses to supply and demand shocks. Many firms in commodity supply chains use storage, but we know little about which firms and why. The economic theory of storage asserts that firms in a competitive market for inventories will store based on current and expected market prices and the per-unit cost of storing. We empirically test at the firm-level the importance of one major cost of storage: the opportunity cost of capital used to value foregone revenue from deferred commodity sales. To do so, we use panel data from thousands of Illinois farms who hold inventories of corn and soybeans. Although interest rates as a measure of capital costs are unlikely to vary widely across firms in this context, we exploit variation in the weighted average cost of capital due to cross-firm differences in capital structure. Using two-way fixed effects regressions, we find a statistically significant average effect of capital costs on inventory that masks notable heterogeneity across firms. Panel quantile regressions reveal two groups of firms: one whose inventory holdings are responsive to changes in the opportunity cost of storage and another whose are not. Our results suggest some farms behave like the profit-maximizing ones from theory but substantial inframarginal commodity inventories are held by farms for other reasons.