Technology is dramatically driving down the cost of sequencing genetic data and increasing the quality of predictions made with this data. A standard concern is that these predictions could impair the functioning of insurance markets, either because insurers would abuse genetic information or because of adverse selection.

We make three contributions. First, we develop a methodology to measure how much selection would be created by genetic information in the market for an insurance product. Second, we extend the methodology to measure the amount of selection with future prediction technology, by combining information from heritability studies and empirical regularities from genetic epidemiology. Third, we apply the methodology to critical illness insurance using data from the UK biobank, including genetic information and National Health Service records for 800,000 UK citizens.

 The main substantive finding is that selection would make many of these markets untenable unless insurers are allowed to underwrite based on genetic information. However, there are important differences in how the results vary by the type of illness and market.