Thursday, 3 October 2013

Is insurance a natural monopoly?

From this distance, its hard to fathom the intense hatred Republican congresspeople must have towards the new health insurance scheme in those united states. I keep thinking there must either be some complex political games in play or they are raving mad, but maybe its both.

Anyway, the plan to get lots more people into a risk pool and thereby drive down premiums has got me thinking about a question I was asked earlier in the year at one of our short courses. Is insurance a natural monopoly?

The basic economic function of insurance is to pool risk. The resulting diversification reduces risk for each insured person. It seems to follow intuitively that it might always be beneficial to have more individuals in your risk pool. (Obviously if you know for sure that someone is very high risk you'd prefer to exclude them, but these things cannot be predicted with certainty.)

Another pointer towards lower costs with a bigger risk pool is the fact that some many large firms carry their own vehicle insurance.

So in a static sense, it might well be the case that certain insurance markets are natural monopolies, meaning that the lowest cost of supply is through a single firm. That view omits some feedback effects though, such as:

  • the potential for a monopolist, free from the pressure of competition, to allow internal administrative costs to rise; and
  • whether a monopolist would feel the need to invest in claim cost management (eg helping to prevent accidents and deal with them cost effectively).

That makes the question an empirical one. The experience of Switzerland and Germany in the 90s is interesting in this regard. Von Urgen-Sternberg (pdf) (cool name!) compared comparable insurance products between regions with state-run housing insurance monopolies and those with competitive markets, and found the former had 40% cheaper premiums, invested more in fire prevention and had substantially lower damage rates.

Part of the reason was that selling costs were much lower (no advertising) but also the competitive firms were somewhat more lenient on payouts for fear of losing customers.

There is certainly a lot more to think about here, but on the principle that there is nothing so nice as an interesting question, this one is worth sharing around.

No comments:

Post a Comment