A Vectorautoregressive Investment Model (VIM) and monetary policy transmission : panel evidence from German firms

This paper proposes a new framework for studying the effects of monetary policy on business investment. Important ambiguities with the modeling of investment dynamics and interactions between real and financial decisions suggest modeling investment spending as a VAR. Based on a panel of financial statement data for 6,408 German firms (44,345 datapoints) supplemented with user costs of capital and confidential measures of creditworthiness, we generate GMM estimates of a Vectorautoregressive Investment Model (VIM). It contains investment and cash flow as endogenous variables, and the user costs of capital and sales growth as additional explanatory variables. We report four substantive findings. First, monetary policy matters, and business investment is responsive to interest rates embedded in the user cost of capital. Second, allowing real and financial decisions to interact raises the impact of monetary policy by one-third relative to simulations of an investment equation in isolation that assumes an exogenous financial policy. Third, the sensitivity of investment to cash flow shocks is raised by twothirds relative to single equation computations. Fourth, firms with poor credit ratings are "paralysed" in being unable to react to changing economic conditions as given by relative prices or demand. On the other hand and consistent with binding financing constraints, these endangered firms show a high responsiveness to cash flow shocks. Apart from these substantive conclusions, this paper demonstrate that the panel VAR approach is useful for modeling firm dynamics and real/financial interactions and for assessing monetary policy transmission.