Causality in empirical analyses with emphasis on asymmetric information and risk management

We discuss the difficult question of measuring causality effects in empirical analyses, with applications to asymmetric information and risk management. It is now well documented in the economic literature that policy analysis must be causal. Hence, the measurement of its effects must also be causal. After having presented the main frameworks for causality analysis, including instrumental variable, difference-in-differences, and generalized method of moments, we analyze the following questions: Does risk management affect firm value and risk? Do we face a moral hazard problem in the insurance data? How can we separate moral hazard from adverse selection and asymmetric learning? Is liquidity creation a causal factor for reinsurance demand? We show that residual information problems are often present in different markets, while risk management may increase firm value when appropriate methodologies are applied.