RDP 9311: Agency Costs, Balance Sheets and the Business Cycle 5. Conclusions

Interest and theory concerning the effect of balance sheets on firm behaviour and economic activity appear to have come full circle. Explanations about the severity of 1930s Great Depression stressed the importance of high debt levels in causing corporate failures. This was followed by the revolution in corporate finance theory in the 1950s and 1960s that suggested that financial structure was irrelevant for investment, firm value and firm decision making. More recently, developments in the theory of incomplete information have again brought financial structure forward as an important consideration in determining the length and depth of the business cycle. Along with significant changes in the financial structure of corporations, the more recent theoretical implications of corporate balance sheets for the business cycle have attracted the attention of policy makers.

In this paper we have reviewed the principal models and mechanisms through which the structure of balance sheets can affect the cycle. Three broad mechanisms were identified. The first of these relies on the fact that the costs imposed by asymmetric information between borrowers and lenders increase as the collateral of the firm falls. Changes in these costs get translated into changes in the cost of external finance. As a result, a deterioration in firm equity leads to a higher cost of intermediated finance. The second mechanism focuses on bank balance sheets. An adverse demand shock that causes an increase in loan defaults is likely to reduce the collateral of financial intermediaries. This loss of collateral may cause portfolio re-allocations by the intermediaries and, as a result, fewer funds are made available at any given interest rate.

The third mechanism focuses on management incentives. We argue that adverse shocks that reduce the collateral of a firm may induce the management of the firm to act in a more risk averse manner. Risk aversion manifests itself in management passing up projects that it would have undertaken had the firm had greater collateral. The model developed in the paper predicts that after an adverse shock, management may wish to undergo a period of ‘reliquefication’, not only to reduce the premium on intermediated external funds, but to reduce the probability of subsequent bankruptcy and loss of their job.

In practice, empirical researchers have had difficulties in demonstrating definitive links between the nature of the business cycle and balance sheets. They have had considerably more success in showing that the structure of a firm's balance sheet can affect its decisions. Nevertheless, these micro-results, coupled with significant increases in leverage, followed by falling asset prices, has increased concern about the implications for the cycle of the corporate sector's financial health.

In the second half of the paper, we attempted to examine some of these concerns for Australia. Over the 1980s corporate indebtedness increased significantly. This was followed by some deflation of asset prices. As Mills, Morling and Tease (1993b) document, these developments have been followed by a fairly widespread attempt by the corporate sector to recapitalise. While we present no direct evidence in support of the view that this recapitalisation (or reliquefication) is an attempt to reduce the cost of external funds and reduce bankruptcy probabilities, it is certainly consistent with the predictions of the literature discussed in the paper.

The paper does, however, present direct evidence concerning the relationship among the perceived difficulty of obtaining finance, asset price inflation and corporate indebtedness. Our results show that even after controlling for expected general business conditions, both asset price inflation and an increase in corporate equity lead to firms perceiving that finance is easier to obtain. This supports the view that increases in collateral play an important role in the availability of finance. A collapse of asset prices, or an aggregate demand shock that reduces firm equity, will result in tighter credit conditions independent of the deterioration in the general business outlook. A potentially important caveat to these econometric results is a lack of degrees of freedom. While from a purely statistical point of view there are sufficient degrees of freedom, the data only covers one real business cycle. Certainly, over that cycle the evidence suggests that there was a link between corporate collateral and the conditions under which finance was extended. It is, however, difficult to assess the role that this link will play in future business cycles.

Nevertheless, the results point to a potential problem. Asset price inflation sometimes occurs for reasons unrelated to fundamentals. The collateral generated through this inflation may be used by firms to obtain additional funding. Inevitably, a correction in asset prices occurs, and this correction may well be associated with an adverse macroeconomic shock. When the correction occurs, the ‘false’ collateral disappears and firms are left with considerably higher debts and experience an increase in financial fragility. The result is a business cycle with a considerably greater amplitude than would have been the case had the corporate sector not increased its leverage.

Finally, while there is little evidence of equilibrium credit rationing, there is some evidence that the supply and demand for funds are a function of the financial structure of both the corporate and financial sectors. It is probable that the increase in leverage of the 1980s, and the more recent asset price falls, have caused a leftward shift in both the supply and demand curves for intermediated external funding.