Asset-backed financing

This dissertation has a twofold objective: to extend the Williamson asset specificity hypothesis and to empirically test both the asset specificity hypothesis and the extension. The Williamson asset specificity hypothesis asserts that the financial leverage used by firms is a function of the specificity of the assets owned by the firm when asset specificity is defined as the readiness with which assets can be re-deployed. This results from a governance argument whereby highly specific assets can only be governed by increased equity participation. This argument is extended with the assertion that increased specificity causes operating leverage to rise and that firms counter this increased operating leverage by decreasing the financial leverage they employ. Liquidation value is employed as a proxy measure for how readily assets can be converted to cash. Data was gathered for a sample of firms who have liquidated and include firms liquidated in bankruptcy and firms liquidated voluntarily. Using these data a model is developed to estimate the liquidation value of any firm. A cross-sectional time-series formulation is employed using data gathered for thirty-six firms over a twenty-two year period. A statistically significant positive relationship was found to exist between the estimated liquidation value and financial leverage which supports the Williamson asset specificity hypothesis. Neither the cross-sectional nor time series behavior of firms provides evidence of a trade-off between interest tax shields and non-debt tax shields. No significant relationship was found to exist between the value of the non-debt tax shield and financial leverage. No evidence was found indicating a relationship between operating leverage of firms and financial leverage. However, evidence was found that firms with higher percentage changes in sales from year to year, lower probabilities of failure, higher levels of financial slack, and lower values for interest tax shields use less financial leverage. Finally, evidence was found indicating that firms employed more financial leverage in the 1980's than in the 1970's.

This dissertation examines several research questions relating to securitization by non-financial firms. Finance theories suggest securitization is most beneficial when there is high demand for liquidity. On the other hand, empirical studies have shown that firms engage in securitization to manage earnings. I find that liquidity demand, not the incentive for earnings management motivates securitization transactions by non-financial firms. I also evaluate whether earnings management in securitization is indeed undesirable from a shareholder's perspective by examining the economic consequences of the practice. Because securitization creates a large infusion of cash, one way to evaluate the economic consequences of earnings management is to examine whether securitization proceeds encourage overinvestment. I find that earnings management in securitization (i.e., recording non-zero securitization income) is unrelated to firms' suboptimal) overinvestment in the post-securitization period. Thus, it appears that earning management in securitization has no negative economic consequence in terms of generating excess securitization proceeds that encourage overinvestment. I also examine the market's valuation of securitizable assets in the accrual components of earnings and the use of securitization proceeds. Because securitizable assets can be converted into cash through securitization, I test whether the market valuation reflects the source of liquidity in securitizable assets that is similar to the cash component of earnings. I find that, for securitization firms, the market valuation of securitizable assets is similar to that of the cash component of earnings.