Ceris-CNR, W.P. N° 5/1993
Moreover, both the stock and the bond markets are not as developed as in the US or in the UK for a set of structural reasons, including the tax treatment of interest and dividend income. This means that negative supply shocks to bank credit are likely to have powerful effects. Moreover, during the second half of the seventies and the first half of the eighties the Central Bank used quantity controls on bank portfolios to directly influence the supply of credit. The use of such controls, together with the need to keep financing substantial government budget deficits, implied that monetary contractions could generate powerful effects on the availability and cost of bank credit.
        The plan of the paper is as follows. In section 2 we describe in detail the data used in our empirical work and present some useful descriptive statistics on the basic characteristics of the Mediobanca samples of small and large firms. In particular we discuss the structure of their liabilities and assets and the composition of the sources of internal and external finance.
        Section 3 is the core of the paper and it contains three pieces of empirical evidence that throw light on the heterogeneity of firms' responses to business cycle shocks. The first piece of evidence consists of cross-correlations between sample specific indicators of demand growth, on the one hand, and the rate of growth of sales, inventories, different types of debt, short-term assets, new share issues and retentions on the other. We also discuss the cross-correlations with the mix of different types of liabilities and with the composition of the sources of funds.
        The second piece of evidence is obtained from regressing all the variables described above on contemporaneous and lagged dummy variables that capture periods of monetary tightness. A detailed description of monetary policy in Italy and of the qualitative and quantitative evidence we have used to construct the stringency dummies is contained in an Appendix to the paper* 8.
        The last piece of evidence we present consists of econometric results on the time and cross sectional variation in the excess sensitivity of inventory investment and fixed capital investment to changes in proxies for credit worthiness. We use the ratio of cash flow to interest payments to proxy for the latter and estimate both equations in an error correction form. In section 4 we summarize the main results and provide some conclusive remarks.
  as in Japan and Germany.
  8 See Römer and Römer (1989) for a seminal contribution in constructing and using qualitative proxies for m on etary stringencyfortheUS.
-4-