Harvey, Irma and the TBTF implicit subsidy problem

September 8th, 2017 at 12:44 pm

On my way in this AM, I heard an interesting interview by the great Bloomberg Surveillance team of Tom Keene and David Gura (and I’m not just saying that because they occasionally invite me on; those guys get good guests and give them the time to answer good questions). They were talking to an insightful guy from the re-insurance industry about the costs to the industry of Harvey and Irma. To be clear, what follows abstracts from the human costs of these disasters, which I take extremely seriously. In fact, both the implicit subsidy point I stress below and the more elaborate argument I make here are intended to link the economics to the too-often tragic human outcomes.

Their discussion emphasized the large share of residential and commercial real estate vulnerable to flooding but without flood insurance (I think they put this share at 50% in South FL.). Moreover, as I stressed in the WaPo piece linked above, the federal flood insurance program has long underpriced the risks against which they insure, leading to subsidy #1.

Subsidy #2, though, is the one currently in play in DC, and discussed in the interview: those who build in harm’s way can trust that there’s an implicit subsidy as the gov’t (mostly federal) will pony up serious money for their relief and rebuild efforts. That is as it should be. When Americans are laid low by natural disasters, we pitch in, whether as taxpayers or as revealed by the behavior of many good people in TX. And when we’re talking about “black swans”–very low probability events with very high, unforeseen costs–the federal sector, which can borrow at favorable rates and run deficits long-term, is uniquely positioned to help.

And yet, there are problems with this model. Subsidy #2 is one of those implicit subsidies, not unlike the one tapped by some of the big banks or Fannie and Freddie after the financial crash. The bailouts that saved these firms were of course premised on “too big to fail” (TBTF), in that their collapse would generate nationwide damage. To the extent that this is true, there’s a huge moral hazard problem, which will, and has, led to systemically underpriced risk.

I understand that mapping these observations onto big places like Houston and South Florida is tricky. These are big chunks of America and American commerce and thus they too are legitimately classified as TBTF. But that doesn’t mean we ignore the inherent problems this creates, anymore than we ignored the financial crisis, which I assure you as a former insider was a huge motivator for the Obama administration, leading ultimately to the Dodd-Frank reforms.

Instead, we must work to both mitigate and adapt to the impacts of climate change, particularly the intensity of storms and flood surges. We must update our priors around alleged 500-year floods that appear to be showing up a lot more frequently than that. And we must generate more accurate price signals, particularly in the face of implicit TBTF subsidies.

Print Friendly, PDF & Email

3 comments in reply to "Harvey, Irma and the TBTF implicit subsidy problem"

  1. D. C. Sessions says:

    Pity, isn’t it, that places such as (randomly chosen) New Mexico [1], which has a very low rate of “black swan” subsidy, can’t get compensating help for its own infrastructure needs, notably education.

    If it makes you fell better, you could probably replace NM with West Virginia or maybe Minnesota.

  2. Smith says:

    In Florida, a swing state, any Democrat should have an easier time winning in 2020, simply by claiming the intensity and frequency of this year’s storms might have been caused in part by global warming. That’s a bigger deal than flood zone subsidies.
    Also, there should be a distinction made between insuring people who rebuild or repair in a flood zone that they lived in all their lives, and new construction where builders (invariably very rich) are benefiting from the government’s subsidies and stupidity.

  3. elkern says:

    Just cap FEMA payouts at $100,000 per owner for residential units. The poor schmucks get a new Trailer; the rich schmucks will get the signal and price in the (somewhat) more accurate cost of risk.

    Of course, Devil’s in the details, as the only Ferrengi to ever run for President said.

    But am I missing any big econ/math/moral hazard aspects with this kind of solution? (Ignoring for now the fact that the rich schmucks have a whole lot more political leverage than the poor ones). The papyrus my Econ BA was written on is starting to crumble, but I’ve tried to keep up with things in the intervening millennia.