In an earlier post in my mini series on insurance fundamentals I explained the concept of moral hazard, which refers to an information asymmetry as having insurance changes unobserved behavior which increases the risk or severity of loss. There is another information asymmetry in insurance that occurs before the insurance contract is entered based on the fact that the insurance buyer knows more about their circumstances than the insurance company. 

Let’s take car insurance as an example. Suppose you were frustrated by what you perceive to be the stodgy nature of existing car insurers and so you launch NewSure. In order to compete you decide to offer lower premiums which you base on the fact that NewSure will use new awesome technology to have lower marketing and administrative cost. Everything looks great at first and NewSure signs up lots of customers. All of a sudden though it turns out that NewSure’s loss rates are much higher than industry averages, both due to higher frequency and higher amounts of claims.

What went wrong? One theory could be that you were just unlucky. Your customers just happened to have more accidents in this period. Given that you already had many thousands of customers that seems highly unlikely though. Instead you suffered from an information asymmetry known as “adverse selection.” The people who signed up for NewSure were in fact worse and/or more reckless drivers than the average driver which means that they have a higher likelihood of more severe accidents.

But why didn’t you get the average? Well, existing insurance carriers over time figure out who the better and worse drivers are and reward the former with lower rates and charge higher rates to the latter (this can be accomplished for instance through rebates). So not only do they each now know who they are but when NewSure comes on the market, NewSure’s lower premiums are more attractive to the worse drivers. And just like that NewSure suffered from adverse selection: the drivers had more information about their riskiness then NewSure and the choice of NewSure was more attractive for drivers who are worse risks.

So what is an insurer to do? The answer is that insurers try to figure out what kind of risk a customer represents as part of the so-called underwriting process (the process of figuring out whether they should insure the risk at all and what premium they should charge). For instance, in life insurance they will ask a lot of questions about your lifestyle to figure out how big a risk you represent. Interestingly, what kind of questions companies can ask is sometimes regulated by law which can limit how well an insurer can understand risk ahead of time.

The impact of adverse selection can also be reduced through more information about observed behavior after the fact. So if NewSure monitors its customers driving and then adjusts their rates upward after the fact if it observes risky behavior it can mitigate that it couldn’t distinguish risky drivers up front.

Like moral hazard, adverse selection exists outside of insurance settings as well. For instance, if you have had one big success as a venture investor you are more likely to be successful in the future. Why? Because startups will want to have a successful investor as their investor and come to you first. As a result the startups that wind up available for investment to the lesser known investors has been adversely selected from the overall pool.