CMStatistics 2016: Start Registration
View Submission - CMStatistics
B1228
Title: A new approach for buffering stock returns in investment-linked annuity contracts Authors:  Daniel Linders - KU Leuven (Belgium) [presenting]
Servaas van Bilsen - University of Amsterdam (Netherlands)
Abstract: The purpose is to introduce a new class of unit-linked insurance contracts providing a stream of future cash flows linked to stock market returns through the performance of a reference index. The innovation of the contract design lies in the absorption of financial shocks into future cash flows. Indeed, in order to temper the volatility in the cash flow stream, we only gradually adjust future cash flows to past financial shock. This is in contrast with the current unit-linked products where a financial shock is fully reflected in the subsequent cash flows. Our framework allows to tailor insurance contracts to the specific risk-profile of the policyholder. In order to build a realistic risk management framework, we consider a financial market with Levy distributed shocks. Whereas working in a Gaussian context might be mathematical tractable, it may be dangerous when the market exhibits extreme movements. Apart from the Gaussian distribution, our framework allows to use a broad spectrum of non-Gaussian distributions which incorporate stylized facts of stock market returns, such as heavy tails, skewness, kurtosis, etc. We show how to price and hedge the liabilities associated with our new insurance contract and provide numerical illustrations of the performance of the hedge.