Adverse Selection and Liquidity: From Theory to Practice

This paper shows how to map predictions of theoretical models of market microstructure into operational empirical measures of liquidity. A meta-model implies an empirical measure of liquidity, denoted L, which describes various characteristics of trading and funding liquidity such as trading costs, bet sizes, haircuts, and capital requirements. When mapped into existingmodels of adverse selection, themeta-model also describes precisely how adverse selection shows up in pricing accuracy and resiliency. Themeta-model is consistent with models of both block trading and flow trading. It highlights a deep connection between time and adverse selection.