| Translated abstract: |
| Abstract | Language |
|---|
| Existence and efficiency of equilibria in markets with asymmetric information have been studied in great depth ever since the seminal Akerlof (1970) article. Whether asymmetric information per se has a strong impact on market equilibria thereby crucially depends on whether the asymmetry of information pertains to private or common values. When values are private, competitive equilibria exist and are efficient under fairly mild assumptions. However, in markets with common values, a competitive equilibrium might not exist at all. This is the result in the famous Rothschild and Stiglitz (1976) model on competitive insurance markets with adverse selection. Rothschild and Stiglitz (1976) show that when insurers offer contracts to consumers that have private information about their risk type, an equilibrium in pure strategies might fail to exist: If the share of low risks is high, a profitable pooling contract would be preferred by both high and low risk types over the candidate separating, zero-profit-making Rothschild-Stiglitz (RS) contracts. However, a pooling contract cannot be tendered in equilibrium as insurers would try to cream skim low risks. This equilibrium nonexistence result is not merely a theoretical oddity. In fact, equilibrium nonexistence is sometimes brought forward as an efficiency reason for regulation of insurance markets. However, equilibrium inexistence itself is not a sensible reason to call for regulation: It cannot be determined whether regulation, resulting in a particular market allocation, improves efficiency in a market where it is not even clear what the market allocation is without regulation. Rather, the equilibrium inexistence result points out that it is necessary to examine whether the RS model is the correct model to describe behavior in insurance markets and consequently to think of alternative models. Not surprisingly, the Rothschild and Stiglitz (1976) result has spurred extensive research. In chapter 2, the nonexistence problem and ensuing debate is reviewed in more detail. In terms of the resulting market allocation, the contributions following RS can with a few exceptions be classified in two broad categories: Models that yield the RS allocation, even for the case in which there is no equilibrium in the original RS model, and models that yield the so-called Wilson-Miyazaki-Spence (WMS) allocation. The crucial difference between these allocations is that the RS allocation is only second-best efficient if an equilibrium exists in the original RS model, whereas the WMS allocation is generally second-best efficient. Thus, there is an efficiency reason for regulation due to adverse selection if insurance markets are considered correctly captured in models of the first, but not the second class. However, although several modifications to the RS model have been brought forward, there is still no generally agreed upon solution. This is because proposed solutions either lack sound game-theoretic foundation or impose exogenous constraints. Chapters 3 and 4 propose solutions to the RS puzzle that tackle these problems. The third chapter introduces a dynamic model that allows insurers to withdraw contracts in reaction to their competitors. This is the logic suggested in Wilson (1977)�s �anticipatory equilibrium concept� that, in spite of departing from Nash equilibrium, to date is one of the most appealed to solutions. In our model an equilibrium with the WMS allocation always exists, thus providing a game-theoretic foundation for the Wilson (1977) equilibrium. However, jointly profit-making contracts can as well be sustained in equilibrium. We then allow for entry and show that the WMS allocation is the unique equilibrium allocation under entry. In the fourth chapter we endogenize insurer capital: Instead of being assumed to be exogenously endowed with large financial assets as in the RS model, insurers can choose their level of capital and consequently go insolvent. With this endogenous insolvency risk, an equilibrium with the WMS allocation always exists. Interestingly, solvency regulation might have unintended consequences. Whereas the consequences of asymmetric information on competition in insurance markets with adverse selection have thus been thoroughly analyzed, little is known about the impact of asymmetric information on oligopoly behavior. In the fifth chapter we therefore depart from the assumption of competition and analyze the impact of asymmetric information on the ability of insurance firms to collude. It is shown that asymmetric information destabilizes collusion, however, this is not a result of asymmetric information per se, but stems from the common value property of the market. Thus, on a general note, we identify a new factor that destabilizes collusion: payoff-relevant private information. | German |
|
![[img]](https://kups.ub.uni-koeln.de/style/images/fileicons/application_pdf.png)
