Publications

This paper considers a risk-neutral insurer and a risk-averse individual who bargain over the terms of an insurance contract. Under asymmetric Nash bargaining, we show that the Pareto-optimal insurance contract always contains a straight deductible under linear transaction costs and that the deductible disappears if and only if the deadweight cost is zero, regardless of the insurer's bargaining power. We further find that the optimality of no insurance is consistent across all market structures. When the insured's risk preference exhibits decreasing absolute risk aversion, the optimal deductible and the insurer's expected loss decrease in the degree of the insured's risk aversion and thus increase in the insured's initial wealth. In addition, the effect of increasing the insurer's bargaining power on the optimal deductible is equivalent to a pure effect of reducing the initial wealth of the insured. Our results suggest that the well-documented preference for low deductibles could be the result of insurance bargaining.

We study how altruism networks affect the demand for formal insurance. Agents with CARA utilities are connected through a network of altruistic relationships. Incomes are subject to a common shock and to a large individual shock, generating heterogeneous damages. Agents can buy formal insurance to cover the common shock, up to a coverage cap. We find that ex-post altruistic transfers induce interdependence in ex-ante formal insurance decisions. We characterize the Nash equilibria of the insurance game and show that agents act as if they are trying to maximize the expected utility of a representative agent with average damages. Altruism thus tends to increase demand of low-damage agents and to decrease demand of high-damage agents. Its aggregate impact depends on the interplay between demand homogenization, the zero lower bound and the coverage cap. We find that aggregate demand is higher with altruism than without altruism at low prices and lower at high prices. Nash equilibria are constrained Pareto efficient.

This paper assesses whether and how setting up a sovereign wealth fund has a buffer effect against currency crises. Using an innovative dynamic logit panel model framework and a unique dataset covering 34 emerging countries over the period 1989–2019, we empirically show that sovereign wealth funds reduce the occurrence of currency crises. This result is robust to different econometric specifications, alternative definitions of sovereign wealth funds, controlling for currency crisis risk factors, and income level sampling. Our findings have important implications for financial stability and for policymakers, who could further exploit the potential of sovereign wealth funds to better manage foreign exchange risks.

Traditional mid-term electricity forecasting models rely on calendar and meteorological information such as temperature and wind speed to achieve high performance. However depending on such variables has drawbacks, as they may not be informative enough during extreme weather. While ubiquitous, textual sources of information are hardly included in prediction algorithms for time series, despite the relevant information they may contain. In this work, we propose to leverage openly accessible weather reports for electricity demand and meteorological time series prediction problems. Our experiments on French and British load data show that the considered textual sources allow to improve overall accuracy of the reference model, particularly during extreme weather events such as storms or abnormal temperatures. Additionally, we apply our approach to the problem of imputation of missing values in meteorological time series, and we show that our text-based approach beats standard methods. Furthermore, the influence of words on the time series' predictions can be interpreted for the considered encoding schemes of the text, leading to a greater confidence in our results.