Jim Lynch turned me onto this paper on the insurance underwriting cycle.*
There’s a neat little graph that caught my eye:
Something big happened in 1986/1987: the United States economy started shifting serious resources into liability insurance.
In a new paper, Alex Tabarrok describes how this shift toward a more liable society decimated the small aircraft industry: companies could be sued for crash in any plane they had ever produced.
Then, incredibly, unexpectedly, he describes how Congress fixed it:
Our estimates show that the end of manufacturers’ liability for aircraft was associated with a significant (on the order of 13.6 percent) reduction in the probability of an accident. The evidence suggests that modest decreases in the amount and nature of flying were largely responsible. After GARA, for example, aircraft owners and pilots retired older aircraft, took fewer night flights, and invested more in a variety of safety procedures and precautions, such as wearing seat belts and filing flight plans. Minor and major accidents not involving mechanical failure—those more likely to be under the control of the pilot—declined notably.
We complain about the legislative process enough. It’s great to see that it can work for good, too.
As a result of this you’ll see insurance premiums for the aircraft industry fall and that means cheaper planes and cheaper flights. Think back to that graph that shows how insurance surplus has gone from 2-ish% of GDP to 4-ish%. That’s something like $300 billion locked away in low-yielding bonds.
I find it entirely plausible that the underwriting cycle is driven by one-off events. Black Swans if you will. And the big ones seem to make us permanently poorer.
*The paper develops a ‘regime switching’ model and declares it superior to an ‘autoregressive’ model of industry cycles. Without going into too much detail, I don’t really think either are very good at explaining what is going on, I don’t care what the R² score is.
Not Unreasonable 