Financial constraints and the balance sheet channel: a re-interpretation

Abstract

Aggregate demand models extending IS/LM fixed price framework yield an enhancement mechanism of the traditional monetary transmission mechanism, the credit channel, which, according to the credit view, works through the ‘balance sheet channel’ and the ‘bank lending channel’. In this paper the augmented IS/LM model is modified assuming that investments may be financed by both internal and external sources of funds. The inclusion of internal funds in the augmented IS/LM fixed price model suggests a different interpretation of the ‘balance sheet channel’ as an enhancement mechanism amplifying monetary policy effects through the quantity rather than the cost of borrowing. Thus, changes in borrowers’ net worth over the cycle can amplify and propagate output fluctuations directly rather than indirectly as in the traditional interpretation of the balance sheet channel. The empirical analysis of the monetary transmission mechanism for Italy in the last decade accords with the interpretation of the balance sheet channel proposed in this paper.

Acknowledgement

I wish to thank Piero Alessandrini, Marco Crivellini, Giuseppe Marotta and Marco Mazzoli for their comments and suggestions to an earlier version of the paper. This paper is part of MURST's research program ‘Heterogeneity in bank and trade credit markets and in local financial system: implications for regional growth and distributional effects of monetary policy’. Financial support from MURST is gratefully acknowledged. The usual disclaimers apply.

Notes

 Given the absence of government bonds we set the opportunity cost of retained earnings equal to zero.

 The sticky dividend policies is assumed in the pecking order theory (Myers, 1984).

 A sufficiently small elasticities of loan demand to the level of goods demand is even necessary in the model of Bernanke and Blinder (1988) in order to have the interest rates moving in the same direction after a monetary policy impulse.

 The optimal lag length of the VAR was derived using a sequence of (likelihood ratio) exclusions tests.

 The introduction of a step-dummy variable from 1998 on is required to have well-behaved residuals as the real monetary base series has a level shift after the decisions of the Bank of Italy to reduce for three times in the same year both the minimum legal reserve requirements and its remuneration.

 Romer and Romer (1989), Bernanke and Blinder (1992) and Christiano et al. (1994) recent findings show that monetary policy actions are followed by movements in real output lasting for two years or more.