Researchers still lack consensus about the size of the user cost elasticity of capital. The divergence in prior estimates may have arisen because one of the two strands of research has neglected cointegration among capital, its user costs, and sales. Using German firm-level panel data, I show that estimating a distributed lag model, prevalent in prior literature, leads to low estimates of the user cost elasticity. Properly accounting for cointegration in an error correction model yields a much larger point estimate of the price elasticity of capital (close to −1). Non-random sample attrition is found to be relatively unimportant.