ETH48 Systemic Risk - Systemic Solutions
To gain in depth understanding of systemic risks, crisis and disruption in complex systems, in particular socio-economic systems, the Risk Center developed a first joined research project called "ETH48 Systemic Risk - Systemic Solutions" which started in February 2013 with eight PhD Students. The project embraces three work packages. At least two Risk Center Members, ideally from two different departments, act as supervisors for each PhD student.
Work Package 1 tries to identify the underlying mechanisms of systemic risk. This requires to develop tools that detect precursors of systemic risk and, in turn, to explore measures to improve system resilience. Specifically, in a general modeling approach - or a framework - this project wants to consider different layers and their specific contribution to the emergence of systemic risk.
Work Package 2 tries to improve our understanding of the emergence and the spread of financial crisis and explore novel, cost-effective, more robust institutional arrangements that consider decision biases of agents. In this project we design macroprudential rules that are robust and able to dampen small or medium-size aggregate shocks such that functions and structures of the financial system will continue to fulfill requirements.
Work Package 3 explores interdependencies between macro-systems, particularly resource extraction, energy production, and political stability, to improve our understanding of cross-system links. The project identifies the conditions under which limited state capacity and large-scale resource extraction are likely to trigger political instability in general, and intrastate conflict in particular.
ETH Zurich Research Grants (ETH Grants) are intended to promote highly creative and original research at ETH Zurich that may result in fundamental new knowledge or technologies. These projects involve high-risk approaches and / or novel combinations of disciplines with the potential for excellence and exciting discoveries in the field of risk. At least two Risk Center Members, ideally from two different departments, act as supervisors for each PhD student.
This project aims at improving the risk-management and risk sharing of longevity risk through annuity pensions under dependent mortality and interest rates. It will develop quantitative methods with applications in real-world situations. We take an integrative risk-management approach, treating both risks in a unified framework and allowing for an evaluation of the social benefit of risk sharing, offering innovative annuity products that may increase social welfare.
Financial markets in energy and electricity are characterized by highly specialized contracts that often run over long time horizons, with maturities up to 15 years not uncommon. Positions involving such products are exposed to a wide range of risks driven by a large set of risk factors. Market risk, credit risk, and liquidity risk are important examples. Hedging these risks is an important and difficult challenge, since suitable hedging instruments are scarce, and to the extent they exist, they often have limited liquidity. As a result, markets are both highly incomplete and illiquid, leading to a heavy reliance on models for pricing, hedging, and risk management purposes. This results in a significant exposure to model risk, especially over long time horizons. This PhD project aims to address these and related issues using recent advances in a number of areas in finance and mathematics, for example: Tractable multi-factor stochastic models, at present primarily employed in the context of fixed income, credit, and equity markets; Analysis of long term risk, such as the impact of risk factors on valuation and hedging over long time horizons; Model uncertainty, such as robustness of hedges with respect to model misspecification.
Co-Voting is a new decision-making procedure that combines advantages of representative and direct democracies. We suggest to combine votes from the elected members of the Parliament and a randomly chosen sample of the citizenry. Such an approach is less costly and decisions are less delayed then in a direct democracy. Additionally, the citizenry is not overruled. We propose that the voting process is done electronically, which reduces costs and allows for more flexibility. We analyze the strategic implications of Co-Voting--in all its variants--for politicians, parties and citizens and determine its performance. This requires mathematical modeling, as is customary in economics (in particular, in political economy). Moreover, we examine what implications an electronic voting approach has with respect to the security of a Co-voting process. This requires a precise formal definition of a voting protocol, its desired security properties and proofs that these properties hold.