One of the most enduring topics in financial theory is the persistence of investment risk across time. Traditional finance lacks methods for considering and hedging non-diversifiable risks. This paper is based on the general equilibrium model of Allen and Gale (1997). We extend their model in various directions: the intermediary is a firm and not a planner, financial markets are assumed to be incomplete, and the mechanism of intergenerational risk-sharing is endogenously determined. Our model allows for the analysis of optimal behavior of individuals and the intermediary together with the respective feedback processes.

We analyze the investment behavior of private clients with regard to retail structured products. To ascertain their stated and revealed preferences, we use a questionnaire and a field experiment. The real product issued in the field experiment is comparable to the hypothetical product in the questionnaire in terms of both payoff and communication. We find that a product described in simple words strongly motivates people to invest in structured products for the first time, and also eliminates gender differences. However, the real product attracts far fewer first-time buyers than we expected from the questionnaire results.

The strong autocorrelation between economic cycles demands that we analyze credit portfolio risk in a multiperiod setup. We embed a standard one-factor model in such a setup. We discuss the calibration of the model to Standard & Poor’s ratings data in detail. But because single-period risk measures cannot capture the cumulative effects of systematic shocks over several periods, we define an alternative risk measure, which we call the time-conditional expected shortfall (TES), to quantify credit portfolio risk over a multiperiod horizon.

Three channels through which the IMF rescue package may affect international lending can be distinguished: debtor-side moral hazard, creditor-side moral hazard, and debtor and creditor-side moral hazard. We show that if the rescue package fully benefits the debtor, no credit contract between him and the creditor arises. The other two kinds of moral hazard, where the creditor receives the rescue package either fully or in part, increase the scale of international lending relative to the case where no rescue package is forthcoming. The increase is larger if the creditor receives the whole rescue package than if it is shared between the creditor and the debtor. These results are based on the analysis of two sequential credit relationships, the first one between a bank and a government and the following one between the IMF and the government. Each of these credit relationships is characterized by asymmetric information and modeled by a moral hazard model. The two moral hazard models are linked by considering the different channels of the IMF rescue package.

[In a Lucas exchange economy with standard power utility, we study asset prices under learning and ambiguous information. In contrast with models featuring only learning or ambiguity, our model is successful in matching the equity premium, the interest rate, and the volatility of stock returns under empirically reasonable parameters. Our closed-form formulas also show that a severe downward bias arises in the empirical relation between stock returns and return volatility. We quantify this bias in simulations and show that our model can explain why such a relation is difficult to detect in the data.]

We analyze the empirical predictions of ambiguity aversion in intertemporal heterogenous agents economies. We examine equilibria for two tract able wealth–homothetic settings of ambiguity aversion in continuous time. Each setting is motivated by a different robust control optimization problem. We show that ambiguity aversion affects optimal portfolios in a way that is similar to an increase in risk aversion. A distinct property of our second setting of ambiguity aversion is that this increase is state dependent, highly pronounced at moderate portfolio exposures and reduces equity-market participation. In general equilibrium, ambiguity aversion raises the equity premium and lowers interest rates. A distinct feature of our second setting of ambiguity aversion is that the equity premium part due to ambiguity aversion dominates when volatility is low.

We propose a simple model of credit contagion in which we include macro- and microstructural interdependencies among the debtors within a credit portfolio. The microstructure captures interdependencies between debtors that go beyond their exposure to common factors, e.g., business or legal interdependencies. We show that even for diversified portfolios, moderate microstructural interdependencies have a significant impact on the tails of the loss distribution. This impact increases dramatically for less diversified microstructures.

Credit limit management is of paramount importance for successful short-term credit risk management, even more so when the situation in credit and financial markets is tense. We consider a continuous-time model where the credit provider and the credit taker interact within a game-theoretic framework under different information structures. The model with complete information provides decision-theoretic insights into the problem of optimal limit policies and motivates more complicated information structures. Moving to a partial information setup, incentive distortions emerge that are not in the bank’s interest. We discuss how these distortions can effectively be reduced by an incentive-compatible contract. Finally, we provide some practical implications of our theoretical results.

In a Lucas exchange economy with standard power utility, we study asset prices under learning and ambiguous information. In contrast with models featuring only learning or ambiguity, our model is successful in matching the equity premium, the interest rate, and the volatility of stock returns under empirically reasonable parameters. Our closed-form formulas also show that a severe downward bias arises in the empirical relation between stock returns and return volatility. We quantify this bias in simulations and show that our model can explain why such a relation is difficult to detect in the data.