A large body of evidence suggests that credit market frictions play an important role for firm behavior. Empirically, panel data studies find that small firms with more difficult access to credit pay fewer dividends, take on more debt, and have investment rates that are more sensitive to cash flows even after controlling for future profitability. Theoretically, numerous papers show how optimizing models of the firm with incomplete contract enforcement and asymmetric information in the lending process can rationalize the observed correlation of firm size and age with mean growth (negative) and survival rates (positive).
While the relevance of credit market frictions is well established on the microeconomic level, their macroeconomic consequences for business cycle fluctuations are less obvious. Models of financial intermediation and agency costs by Bernanke and Gertler (1989) or Kiyotaki and Moore (1997) imply that the firm’s ability to finance investment varies inversely with the value of its collateral and thus with the business cycle. This financial accelerator mechanism has the potential to generate amplified and persistent output effects in response to small shocks. Yet, simulations in a dynamic stochastic general equilibrium (DSGE) context by Kocherlakota (2000), Chari, Kehoe and McGrattan (2002), or Petrosky-Nadeau (2005) suggest that for plausible calibrations, credit market frictions of this type alone fail to generate quantitatively important business cycle fluctuations.
The lack of internal propagation can be traced back to the assumption in these models that capital from exiting firms is reallocated immediately and costlessely to new firms. In general equilibrium, this assumption implies that credit market frictions only affect the elasticity of aggregate investment with respect to average net worth of all firms. But quarterly investment as a share of fixed private non-residential capital stocks represents at most 3% in the national accounts, and the share of capital in production is about one third. It is therefore not surprising that more sensitive investment dynamics by themselves have only a very limited impact on the cyclical behavior of output.
In this paper, we investigate to what extent credit market frictions together with costly capital reallocation can generate more important business cycle fluctuations. Our investigation is motivated by firm-level observations in Ramey and Shapiro (1998), Eisfeldt and Rampini (2005a, 2005b) as well as Becker et al. (2005) who find that capital separation is an important phenomenon over and beyond depreciation, and that reallocation is a costly and time-consuming process.
The national accounts miss these reallocation flows and thus, investment as a share of the capital stock is likely to be substantially larger than reported in the aggregate data. Furthermore, the same studies report important countercyclical and procyclical variations in capital separation and reallocation rates, respectively. This implies that capital stocks used for actual production may be much more volatile than previously assumed.
Download
PDF Ebook Credit Market Frictions with Costly Capital Reallocation as a Propagation Mechanism
