Collateral, Risk Management, and the Distribution of Debt Capacity

In our paper, Collateral, Risk Management, and the Distribution of Debt Capacity, which is forthcoming in the Journal of Finance, we provide a dynamic model of collateralized financing in which collateral constraints are endogenously derived based on limited enforcement. In the model, firms have access to complete markets, subject to collateral constraints, and thus are able to engage in risk management. We show that there is an important connection between firm financing and risk management since both involve promises to pay by the firm, which are limited by collateral.

Our model predicts that firms with low net worth exhaust their debt capacity and hedge less, since financing needs override hedging concerns, consistent with the empirical evidence. In contrast, this evidence is considered a puzzle from the vantage point of the standard theory of risk management, which takes investment as given. Rampini and Viswanathan (2009) study an infinite horizon model and show that the same trade-off between financing and risk management obtains generally.

The cost of conserving debt capacity is the opportunity cost of foregone investment. This cost is higher for firms with low net worth since they operate at smaller scale and are hence more productive at the margin. When firms differ in their productivity, more productive firms are more constrained and hence exhaust their debt capacity rather than keeping financial slack to take advantage of future investment opportunities. This has important implications for the cross-sectional distribution of debt capacity, which is endogenous in our model. In downturns, when cash flows are low but investment opportunities arise because the price of capital is low, more productive and less well-capitalized firms may hence not be able to seize these opportunities because their debt capacity is exhausted. Indeed, due to their lack of financial slack, they may be forced to scale down investment in such times. More productive and less well-capitalized firms are hence likely more vulnerable to economic downturns since they optimally keep less financial slack. As a result, capital may be less productively deployed in such times and aggregate productivity shocks may be amplified due to these distributional effects.

Higher collateralizability allows firms to borrow more ex ante and thus increases leverage, but leaves them with less net worth ex post. When capital is more collateralizable, firms which exhaust their debt capacity may hence be forced to scale down investment by more. Thus, the amount of capital deployed by more productive and low net worth firms may be more volatile in that case. If collateralizability increases over time, as arguably it has recently, the effects stressed in this paper become even more important.

We think that similar considerations apply to the risk management by households. Received theory would predict that less well-off households insure more, which seems counter to anecdotal evidence and the evidence in the insurance literature. In contrast, the prediction of our model, reinterpreted in terms of household finance, is that less well-off and hence likely more constrained households insure less and are more vulnerable to economic downturns.

The full paper is available for download here.