The Neoclassical Theory of Firm Investment and Taxes: A Reassessment

This article corrects a 60-year history of mis-application of the neoclassical theory of investment to interpret empirical work and guide policy analysis. Empirical estimates of firm-level user cost elasticities of investment or capital have in fact studied two distinct objects, an elasticity holding output fixed that identifies the elasticity of substitution between capital and labor (Eisner and Nadiri, 1968) and an elasticity holding the wage fixed that sheds light on the revenue elasticity of capital (Coen, 1969). A review finds a consensus range for the Coen short-run user cost elasticity that translates into a priori plausible values for the revenue elasticity of capital, offering validation to the neoclassical theory. The long-run general equilibrium capital elasticity to the user cost instead holds labor fixed and depends on both firm-level parameters. This synthesis offers a framework for future work and helps to explain why policy institutions have differing views on the effects of corporate taxes.