*Risky Business*

The subtitle is Why Insurance Markets Fail and What To Do About It, and the authors are the highly regarded Liran Einav, Amy Finkelstein, Ray Fisman.  The level is a bit above what could make this book a bestseller, but I consider that a good thing.  The book in fact is a classic example of how to present economic research in readable, digestible form and should be regarded as such.

I do have a few qualms, but please note these are outweighed by the very high quality of the core material:

1. I think the authors underestimate how rapidly “Big Data” is shifting the information asymmetries away from consumers/policyholders.  This is related to my recent remarks on AI.

1b. For reasons stemming from #1, insurance/surveillance/control, including from employers, will rise in importance as an issue, and soon.  I don’t get a sense of that from reading this book.  We might alleviate selection problems, while creating other difficulties including ethical dilemmas.

2. I would like to see more on moral hazard.

3. I also would want to see more — much more — on the public choice reasons why government insurance markets so often fail — the authors should consider their own title!  Should the Florida government really be propping up insurance contracts and insurance markets to protect homeowners against climate change-related losses?  No matter what your view, this kind of issue is under-discussed.  How about the FDIC?  Bailout-related moral hazard issues?  Those are hardly “small potatoes.”  I get that isn’t “the book they set out to write,” but still I worry that the final picture they present is misleading when it comes to market failure vs. government failure.  Adverse selection is really just one part of insurance markets, but this book doesn’t teach you that.

3b. Isn’t excess liability through our court system another major reason why insurance markets fail?  We needed a Price-Anderson Act, where government assumes a lot of the liability, to support our nuclear power sector, even though coal alternatives were riskier and more harmful, both short run and long run.  In terms of actual importance, hasn’t this been a major, major factor?

3c. Are restrictions on “boil in oil” contracts (no matter what you think of them ethically) another factor in institutional failure here?  Maybe that is one way of making America safe for bungee jumping.  Or we can follow New Zealand, and limit liability here altogether.  The interaction of insurance and liability law is a major issue, and we have not been getting it right.

4. The authors absolutely do consider “positive selection” (e.g., it is the responsible people who buy life insurance, thus leading to a favorable customer pool), but I would give it more emphasis.  If you believe that income inequality, “deaths of despair,” and educational polarization are growing problems, this phenomenon likely is becoming more important.

4b. How about more concessions in the Obamacare analysis?  For years I read that a weaker mandate would cause the system to collapse.  Yet the Republicans significantly cut back on mandate enforcement and the system seems to be getting along OK, at least from that point of view.  (In fact, politically speaking Trump arguably saved Obamacare.)  What did everyone miss?  Did they overrate adverse selection arguments and underrate positive selection?  It seems that was a major failing of the economics profession, which if anything was more insistent on “the three legs of the stool” than policymakers were.  The authors do cover this all at length, but they can’t bring themselves to note “we got down to 7-8 percent uninsured, the whole thing actually worked out OK, and the economists didn’t quite get it right.”

5. There are plenty of cases when expected “insurance” markets do not exist, and we cannot boil those down to adverse selection.  Why don’t all those Bob Shiller proposals happen?  (Is it really inside information about gdp?  That seems doubtful.)  Why aren’t there more prediction markets?  Why have so many proposed futures contracts on exchanges failed?  These all would seem to serve insurance-like purposes, among their other functions.  Yet their supply seems skimpy, at least relative to an economist’s expectations.  Why?  Perhaps there is more to failed insurance markets than meets the eye.

I know authors can fit only so much into a book, but if I can fit this much into a blog post…I would like to see more!  And I think that would result in a more realistic policy balance as well, and draw attention to major issues other than adverse selection.


Comments for this post are closed