Institut für Risikomanagement und Versicherung
print

Links und Funktionen
Sprachumschaltung

Navigationspfad


Inhaltsbereich

Aktuelle Themen

Bei Interesse an einer Masterarbeit am Institut für Risikomanagement und Versicherung kontaktieren Sie bitte Sandra Zoller. Sehen Sie bitte davon ab, einzelne Assistenten direkt auf die Übernahme einer Masterarbeit anzusprechen.

Aktuelle Themen für Masterarbeiten 

Detaillierte Themenbeschreibungen

Topic 1: Intertemporal prevention decisions with multiple risks

Supervisor: Julia Holzapfel


In the course of their life, individuals face a multitude of risks. Some of these risks are beyond their control, others can be altered by investing in prevention. Prevention refers to activities reducing either the size or the probability of a loss. Sometimes the investment in prevention and its effect on the risk are contemporaneous, sometimes the investment precedes its effect. For example, in order to prevent a burglary, one can engage a security firm (contemporaneous) or install a burglar alarm (precedes its effect). The presence of other risks affects individuals’ prevention decisions even if prevention concerns only one risk. If prevention is possible for more than one risk, the effectiveness of a given investment in prevention varies with types of losses and prevention technologies. If different prevention activities affect different risks and total resources are limited, individuals must decide how to allocate their resources.

The goal of this thesis is to analyze prevention decisions with multiple risks in an intertemporal expected utility setting. First, the literature on prevention in intertemporal expected utility settings shall be reviewed. Second, the author shall recap how the introduction of other risks affects individuals’ decision-making. Finally, existing models shall be extended to analyze and compare different configurations for the time structure of the investments in prevention and the realization of the risks.

Prior knowledge in theoretical microeconomic modeling, e.g. having attended Insurance Economics, is highly encouraged.

Topic 2: The effect of telematics on moral hazard in insurance markets

Supervisor: Elisabeth Stöckl


People tend to change their behavior as soon as they are insured. In particular, they care less about reducing the probability or size of loss than in the case without insurance (moral hazard). In-car insurance, for instance, individuals tend to drive less carefully if potential damage is borne by an insurer. Modern technologies such as telematics offer the opportunity to reduce the information asymmetry between policyholder and insurer. A Black box in the car records the driving behavior and the insurer adjusts the insurance premium according to the driving behavior by assigning rebates to careful drivers. In a theoretical approach, Shavell (1979) studies moral hazard in insurance by incorporating individuals’ level of effort to prevent a loss.

The goal of this paper is to analyze the effect of telematics tariffs on ex-ante moral hazard. Following Shavell (1979), a theoretical moral hazard model should be introduced which studies individuals’ effort level in dependence of the premium and of the probability of a loss when the effort level is observable by the insurer. It should be elaborated how the driving behavior (= level of effort) changes if it is observable by the insurer in comparison to an uninformed insurer. Finally, a comparative static analysis should be carried out to study how the effort level changes if certain determinants (loss probability, insurance coverage) change.

Topic 3: Companies' risk management strategies

Supervisor: Sandra Zoller

An essential element of risk disclosure regards the management and mitigation of the risks reported. Although identifying relevant risks already constitutes a key component of enterprise risk management (COSO 2004), adequate responses to risks are essential for reducing adverse events' likelihood and gravity. Thus, in Item 305 of regulations S-K, the U.S. Securities and Exchange Commission asks companies to provide qualitative information about how a company manages risk exposures: "Such descriptions shall include, but not be limited to, a discussion of the objectives, general strategies, and instruments, if any, used to manage those exposures." Still, previous literature mainly focuses on quantitative risk management disclosure or specific aspects of risk management (see, e.g., Friberg and Seiler (2017) and Hoyt and Liebenberg (2011)).

This master's thesis aims to understand the drivers of the risk management strategies of a company. This thesis should first provide an introduction to enterprise risk management and discuss findings of previous literature. Second, a topic modeling approach can discern risk types in the company's annual statement risk disclosure section. Third, the author should present an empirical approach to identify the company's risk management strategies concerning a specific risk type. She can then analyze potential factors that influence a risk management strategy for a risk type. These factors can be company and risk characteristics. Basic knowledge of statistical learning techniques and experience with a programming language, e.g., R or Python, are highly recommended.

Topic 4: The impact of regret on insurance fraud

Supervisor: Elisabeth Stöckl

After the (monetary) outcomes of a lottery are known, people tend to regret their initial decision if the outcome does not meet their expectation. For example, the non-conclusion of insurance can trigger regret when an individual suffered a loss in a given period. The other way around, the conclusion of insurance can also cause regret when an individual did not suffer a loss in the insured period. The scientific literature describes this phenomenon by means of the regret theory. Braun and Muermann (2004) apply the concept of regret theory on the demand for insurance. Regret may also be triggered by the (non-) commission of insurance fraud. Whereas a policyholder regrets the filing of a fraudulent claim that is detected by the insurer, he might regret the non-filing of an illegitimate claim if it had brought a financial surplus.

This master thesis aims to connect the theoretical concepts of regret theory and insurance fraud. First, the literature on regret theory and insurance fraud shall be reviewed. Second, one existing insurance fraud model should be extended with the component regret aversion and the equilibrium should be derived. Finally, the author shall analyze the impact of regret aversion on the willingness to commit fraud by means of the self-developed model.

Topic 5: The value of a statistical life in the light of the Covid-19 pandemic

Supervisor: Julia Holzapfel

During the Covid-19 pandemic, social distancing measures are adopted around the globe to pre-vent the rapid spread of the disease. Moreover, a lot of effort has been put in research activities looking for better treatments for infected patients and aiming at the development of a vaccine. The costs of these measures can often be evaluated in monetary units. The benefits, however, are saved lives and avoided morbidities. The value of a statistical life (VSL) measures the amount that individuals are willing to pay for a marginal reduction of mortality risk, per unit of risk. VSL esti-mates can be used to conduct a cost-benefit analysis of proposed policies that affect people's mortality risk. There are basically two approaches to estimate the VSL: Market-based estimates are inferred from market choices that involve implicit tradeoffs between mortality risk and money whereas contingent valuation estimates rely on surveys asking people directly to report their will-ingness to pay.

The goal of this thesis is to estimate the VSL in the context of an epidemic like Covid-19 based on the contingent valuation approach. First, the author is expected to provide an overview of the VSL literature. Based on this literature review, the applicability of VSL estimates in a cost-benefit analysis of measures taken during an epidemic should be discussed. Afterwards, the author should derive own hypotheses about the willingness to pay for mortality risk reductions during an epidemic and design an experiment to test these hypotheses. The experiment should be imple-mented as an online experiment and the outcomes should be analyzed using adequate statistical methods. Finally, the results of the experiment should be critically discussed.