We focus on the determinants of investment decisions at …rm level allowing for heterogeneous capital goods. Our data, on more than 1500 private Italian …rms, from 1982 to1995, allow us to distinguish between equipment and structures. The standard average Q model being based on stock market evaluation, is generally unsuitable for our problem of heterogeneous capital goods. Moreover our data come from unquoted companies, mostly small and medium-sized. We follow the marginal q approach and estimate investment using a two-stage procedure. 1) we estimate a VAR model for each type of capital good and calculate a ”q” for equipment and structures; 2) we estimate the investment equations by GMM, including the fundamental q’s obtained in the …rst stage. In order to capture the possible existence of nonconvexities in adjustment costs and irreversibilities, we specify a richer-than-usual functional form. This appears to be particularly relevant in the case of structures. We show how aggregation tends to hide these important di¤erences.
A Q Model of Heterogeneous Capital Goods: an Empirical Investigation on a Panel of Non-Quoted Italian Firms
BONTEMPI, Maria Elena;
2001
Abstract
We focus on the determinants of investment decisions at …rm level allowing for heterogeneous capital goods. Our data, on more than 1500 private Italian …rms, from 1982 to1995, allow us to distinguish between equipment and structures. The standard average Q model being based on stock market evaluation, is generally unsuitable for our problem of heterogeneous capital goods. Moreover our data come from unquoted companies, mostly small and medium-sized. We follow the marginal q approach and estimate investment using a two-stage procedure. 1) we estimate a VAR model for each type of capital good and calculate a ”q” for equipment and structures; 2) we estimate the investment equations by GMM, including the fundamental q’s obtained in the …rst stage. In order to capture the possible existence of nonconvexities in adjustment costs and irreversibilities, we specify a richer-than-usual functional form. This appears to be particularly relevant in the case of structures. We show how aggregation tends to hide these important di¤erences.I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
