Fracking and risk management

Old Milliman friend Richard Soulsby forwards his article (with Jason Kurtz and Bhavini Kamarshi) on fracking and risk management.

Fracking creates a big, tricky environmental exposure. Shooting chemical gunk into the earth at rocket speed could contaminate groundwater. As an analogue, MTBE contamination of Santa Monica’s water supply ended in a $120 million settlement. (There are air pollution concerns, too.)

From society’s P.O.V, optimal risk management practice would be safety protections – following best practices on how to drill safely, accompanied by insurance that covers the liability from an accident.

But fracking companies have cheaper solutions:

  • Some buy low limits of insurance, down to $1 million, a mere sponge of coverage in case of a major spill.
  • Others use corporate shields to wall off exposure.
  • And if there is a spill where, say, a half-dozen fracking companies operate, each will try to blame the others.

Or, of course, all of the above. To the list, I’d add that it’s pretty cheap to (legally or illegally) pay off local politicians so that proper regulations never get written. Which I’m told is pretty common in Frackland.

The guys at Milliman call for frackers to carry adequate insurance. They note (drily) that insurance companies are expected to have capital on hand to cover all risks undertaken. Fracking companies should be held to the same standard.

Right now, though, fracking is a tough risk to underwrite, so there might not be enough capacity to cover the exposure. The authors offer two solutions:

  • Government-run insurance pools. What they describe sounds a lot to me like the pool nuclear plant owners participate in. (Note the nukes appear to have a free-but-not-guaranteed government backstop above $12 billion.)
  • Casualty catastrophe bonds, about which more below.

Not too many casualty cat bonds out there. Part of the problem is the liability tail. With property coverage, the event and the ultimate payment of claims happen fairly quickly, so investors can quickly sort out whether the cat bond’s principal will be needed to pay claims. But liability claims can take years to settle, and the bondholders will not like that uncertainty.

But the Milliman team proposes bonds pegged to a casualty index. A lot of property cat bonds work this way. Language in the bond points to a triggering mechanism, usually an index like PCS in the U.S. or PERILS in Europe. These indices estimate the severity of, say, a hurricane or windstorm. If the index hits a high enough point, the bond defaults. (An example here.)

There aren’t any casualty indices like that, but some – the Milliman guys point to CatVest – are on the way. The index would assess the liability of an accident and, if necessary, cash out the bonds.

But there’s another issue, I think. The money for settlement is specifically set aside before the claims are actually settled. With property claims, this isn’t a very big deal. Claims settle quickly and in most cases, the amount they claim might settle for varies only slightly between the date the claim is reported and the date it settles. Most settlement disputes involve a few thousand dollars.

But liability claims take a long time to settle, and the amount they will settle for varies a great deal, even after the claim is reported. Plaintiffs can claim millions while insurers offer zero. That turns significant risk borne by the insurer (or whoever gets the proceeds of the cat bond), if the earliest claims eat up the bond’s proceeds.

That risk is real: One of the key items in settlement can be what the insurers themselves think the claim is worth. (D&O insurers, for example, go to great lengths to prevent plaintiffs from discovering their case reserves.) Once the casualty cat bond is triggered, the amount available for settlement is known – the information is either public, quasi-public or easily discoverable. Plaintiffs will know exactly how deep the deep pocket is. So will juries. And they will work very hard to extract every cent from that fund. When that money’s gone, the insurer is on the hook.

But maybe that’s a good thing, if there’s some recognition that once the pot of money is gone, it’s really gone.

There’s some history to this, with various asbestos funds set up to administer claims of manufacturers bankrupted by their liability. And there have been victim compensation funds, like the one set up after 9/11.

I don’t think the asbestos funds have worked too well. The 9/11 fund, by most accounts, was a success. The difference, I think, was the administration of benefits. The 9/11 fund attempted to set benefits coolly and rationally in a coordinated manner that, in the end, won great respect. The asbestos funds have been a bit more of a free-for-all.

So if I had the magic wand, I’d use cat bonds to set up an accident fund similar to what Milliman envisions, but with scheduled benefits and a government backstop out the top (above, say $5 billion or so).

Kudos to Rich and the folks at Milliman for an intriguing idea.

 

Advertisements

One thought on “Fracking and risk management

  1. Rick Soulsby says:

    Jim, thanks for your comments. You make an important point here. This is exactly the type of issue that needs to be addressed now while we are still near the beginning of this boom, in order to have in place the most effective risk management systems. We are very familiar with the 9/11 fund, which could serve as a model for claims administration. We’re likely to see quite a bit of variation in how this is approached, by regulators from different states, at the federal level, and internationally.

    Rick

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: