Four Essays on Debt Securitization and Entrepreneurial Finance

Abstract

Chapter 1 is joint work with Günter Franke, University of Konstanz, and Markus Herrmann, HSBC London, and addresses the question how collateralized debt obligation (CDO-) transactions are designed. An important issue is the information asymmetry about the quality of securitized debt between originator and investors. Investors insist on credit enhancements which mitigate potential problems of information asymmetry. First Loss Positions (FLPs) are the most important enhancement. We analyze the optimal size of the FLP in a model and the actual size in a set of European CDOs. We find that the asset pool quality, measured by the weighted average default probability (WADP) and the diversity score, plays a predominant role for the transaction design. Characteristics of the originator play a minor role. A lower asset pool quality induces the originator to take a higher FLP. Assuming a lognormal distribution for the loss rate of the underlying portfolio we find that the FLP bears on average 86% of the expected default losses. This loss share is largely independent of the asset pool quality. However, it is significantly higher for bond transactions. The loss share and the asset pool quality strongly affect the rating and the credit spread of the lowest rated tranche.Chapter 2 analyzes the premia of tranches in synthetic CDO-transactions. These premia are credit spreads on top of the risk free interest rate and compensate investors for the expected annualized default loss and for the risk of default losses. Focusing on the second component, the pure risk aversion premium, the paper estimates the relative risk aversion (RRA) implied by the credit spreads and the loss distributions of 215 differently rated tranches of 59 European CDO transactions. A pricing kernel approach is applied, using the aggregate loss rate as the systematic risk factor. Estimated RRA is significantly lower for mezzanine tranches which are more exposed to default risks. This may be caused by market segmentation. A higher WADP of the portfolio induces higher average tranche risk and lowers the estimated RRA. This effect is strongest for the senior tranches and smaller for the mezzanine tranches. Better diversification allows the originator to transfer more risk to investors at relatively moderate risk premia. Controlling for expected losses of tranches, better portfolio diversification reduces the risk premia for the lowest two rated tranches of a transaction. Higher tranches are relatively unaffected by variations in portfolio diversification as they are well protected against information asymmetries by the first loss piece and several junior tranches. Controlling for tranche rating, the RRA estimates are higher for the lowest rated tranche of a given transaction. This hints at additional compensation for the lowest rated tranches, presumably for the associated information problems.Chapter 3 is joint work with Julia Hein, University of Konstanz, that was motivated by the US mortgage market crisis. Banks' lending standards have been criticized to have contributed to the crisis. Especially problematic appears the use of adjustable rate mortgages that start with fairly low interest rates which re later replaced by higher rates. The interest rate reset poses a payment shock to debtors and causes additional defaults, in particular if house prices have already declined. Several policy options have been discussed to mitigate the current mortgage crisis. We analyze an interest rate freeze on adjustable rate mortgages as one possible reaction. We study the implications on tranches of residential mortgage backed securities (RMBS) sold briefly after origination, taking into consideration feedback effects on house prices. For investors of rated tranches we show that relatively moderate positive effects of a rate freeze on foreclosures and house prices can outweigh the negative effect of lower interest income. The owners of the FLP need much stronger feedbacks to assure a net benefit.Chapter 4 is joint work with Oliver Fabel, University of Vienna, on industrial organization. In an economy, innovative projects can either be carried out by small entrepreneurial partnerships or within large industrial firms. The existence of an incubator organization, such as the Silicon Valley network, can improve the matching of individuals in teams and, thus, foster the entrepreneurial sector. However, the industrial sector loses these professional elites and there is less risk sharing in the economy. We investigate the effects of incubators on welfare and capital input and derive ambiguous results: only societies whose members exhibit relatively low degrees of risk-aversion are likely to benefit from the improved matching. We argue that this helps to explain why technology or science parks require persistent public support in Europe whereas they emerge as efficient institutions in the US.