Regulatory arbitrageurs of the day, insurance edition

By Felix Salmon
September 12, 2013

Well done to Benjamin Lawsky, who is getting serious about the regulatory arbitrage which is endemic in the insurance industry — or, as he put it in a powerfully-worded letter yesterday, “the gamesmanship and abuses associated with the setting of reserves”.

The topic here is decidedly gnarly, but also extremely important. Insurance companies, just like banks, have both assets and liabilities. Their assets are generally financial investments (stocks, bonds, private equity limited partnerships, that kind of thing) — and therefore pretty easy to calculate. Their liabilities, on the other hand, are fuzzier: at some unknown point in the future, they’re going to have to pay out some unknown sum of money to an unknown number of insureds claiming on an unknown number of events.

Take life insurance, as an example: everybody dies. But that doesn’t mean all life insurance policies pay out: many of them expire before their holders do. And in general, the longer that you live, the more money that your life insurer will be able to make before having to pay out on your policy. And as with most other forms of insurance, life insurance is prone to black-swan events. US life insurers, for instance, dodged a bullet during the AIDS crisis: they were incredibly lucky that the disease hit populations — intravenous drug users, gay men — who tended not to hold much if any life insurance. If AIDS had hit young professionals with families instead, then it’s entirely possible we would have seen a mass of insurers going bust.

A money manager once told me that all insurers go bankrupt eventually; it’s just a question of when. If you insure property then a hurricane or earthquake will come along; if you insure cars then they’ll get hit by a freak hailstorm. And of course all insurers are exposed to market risk: if the value of their investments plunges sharply, then they’re suddenly much less able to meet their obligations going forwards. He might have been exaggerating a little for effect, but his point is well taken — and helps to underscore why it is crucial that the insurance industry be closely and independently regulated.

Which, needless to say, it isn’t. Mary Williams Walsh has the background to the latest war of words, which is taking place between Lawsky, who is New York State’s top financial services regulator, and his counterparts in other states. In a sense, you don’t even need to know the substance: all you need to know is that if you live in a country which has 50 different insurance regulators, and no real federal oversight of what they’re doing, then it’s inevitable that some states will have laxer regulation than others — and that the big insurance companies, no matter where they’re physically located, will manage to come up with a way of doing lots of business in those states.

Which is exactly what has happened. An insurer in New York, or Connecticut, can set up what’s known as a “captive”: it can reinsure its own risks (which is normally a good thing) — except it owns the company it’s moving those risks to. Naturally, these captive reinsurers are all located in jurisdictions with lackadaisical regulatory oversight.

As Lawsky points out in his letter, this is just like the SIV nightmare we all went through five years ago: financial institutions are nominally moving risk off their own books, even though, when push comes to shove, they’re still ultimately responsible for it.

The parallels don’t end there. There were two main types of regulatory arbitrage which helped to cause the financial crisis. One was the SIVs; the other was Basel II, which allowed banks to basically determine for themselves how much capital they should hold. (Surprise: the answer turned out to be “not very much”.) In the wake of Basel II, you’d think that such outsourced regulatory regimes were a thing of the past — and you’d be wrong. The insurance-industry version of Basel II is called “principles-based reserving“, or PBR. In practice, it has done exactly what Basel II did: it has allowed insurers to reduce the amount of capital they need.

Lawsky is fed up with all this, and is withdrawing from all attempts to implement PBR; good for him. But so long as the fractured regulatory regime remains in place, Lawsky’s actions will not have much effect on the insurance industry more generally. We desperately need a national insurance regulator, and we need one with some spine. Otherwise, we’ll continue to live in a world with the unedifying spectacle of companies like MetLife saying with a perfectly straight face that they need to set up captive reinsurance companies, rather than simply hold more capital, because “using equity could reduce returns to levels below those required by investors”.

If life insurers dodged a bullet during the AIDS crisis, then all of America — indeed, the world — dodged a bullet during the financial crisis, when any mark-to-market analysis of insurers’ balance sheets would have showed them to be mostly insolvent. Luckily for all of us, the markets rebounded, and the insurers are now, financially, much healthier than they were. The problem is that their financial health will inevitably leach away into their shareholders’ pockets, unless strong regulators prevent that from happening. Lawsky is leading the way, here. If we ever do get a national insurance regulator (and I’m not holding my breath) then he’d be a good person to lead it.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
Comments
8 comments so far

You don’t need equity per se; you need liabilities that are subordinate to the insurance claims. If MetLife can’t, given a reasonable amount of time, raise capital through issuance of equity, preferred equity, and subordinated debt at returns “required by investors”, they shouldn’t be in business.

Posted by dWj | Report as abusive

I work in this field–a couple of notes that may be useful.

Life insurance is different from all the other sorts in one way; there’s a table for risk (mortality) that is agreed, by everyone in the industry and 49 of the 50 state regulators, to be far too high for underwritten life insurance. Some insurers like it anyway–either because it reduces tax liabilities or because they think it hurts their competition more than it hurts them–but no one thinks it’s accurate. It’s not like property insurance, where the risks are iffy–how many people die annually is a well-studied issue.

Second, to dWj’s point, the whole quarrel is over the use of subordinated debt in one form or another. The reason you use a captive is to be able to have debt that is subordinated to a defined insurance risk (and the reason you want that is that everyone knows that reserving mortality is too high, so captive debt is much lower-interest than general corporate debt.)

Posted by SamChevre | Report as abusive

I work in this field–a couple of notes that may be useful.

Life insurance is different from all the other sorts in one way; there’s a table for risk (mortality) that is agreed, by everyone in the industry and 49 of the 50 state regulators, to be far too high for underwritten life insurance. Some insurers like it anyway–either because it reduces tax liabilities or because they think it hurts their competition more than it hurts them–but no one thinks it’s accurate. It’s not like property insurance, where the risks are iffy–how many people die annually is a well-studied issue.

Second, to dWj’s point, the whole quarrel is over the use of subordinated debt in one form or another. The reason you use a captive is to be able to have debt that is subordinated to a defined insurance risk (and the reason you want that is that everyone knows that reserving mortality is too high, so captive debt is much lower-interest than general corporate debt.)

Posted by SamChevre | Report as abusive

I work in this field–a couple of notes that may be useful.

Life insurance is different from all the other sorts in one way; there’s a table for risk (mortality) that is agreed, by everyone in the industry and 49 of the 50 state regulators, to be far too high for underwritten life insurance. Some insurers like it anyway–either because it reduces tax liabilities or because they think it hurts their competition more than it hurts them–but no one thinks it’s accurate. It’s not like property insurance, where the risks are iffy–how many people die annually is a well-studied issue.

Second, to dWj’s point, the whole quarrel is over the use of subordinated debt in one form or another. The reason you use a captive is to be able to have debt that is subordinated to a defined insurance risk (and the reason you want that is that everyone knows that reserving mortality is too high, so captive debt is much lower-interest than general corporate debt.)

Posted by SamChevre | Report as abusive

“We desperately need a national insurance regulator”

Felix, a national insurance regulator would be promptly gamed by the industry and self neutered via regulatory capture and you’d see the same BS as you now have with the SEC, the OCC, FINRA, etc.

Better to keep insurance regulation at the state level, where its too difficult to do so.

Posted by crocodilechuck | Report as abusive

“A money manager once told me that all insurers go bankrupt eventually; it’s just a question of when.”

I wouldn’t say that this statement is merely “exaggerated a little for effect”. I’d say that it’s such an overstatement as to be moronic. Relatively few large insurers have ended up insolvent – and the most notable one (AIG) did so for reasons having nothing to do with its traditional insurance businesses.

Posted by realist50 | Report as abusive

” I’d say that it’s such an overstatement as to be moronic”

It’s surely just a (presumably humorous) restatement of the gambler’s ruin theorem. All insurers, apart from odd runoff stories with more capital than liability, have positive probability of ruin, and something with positive probability of ruin will be ruined in finite time with probability 1.

Posted by dsquared | Report as abusive

dsquared – that logic applies to every company, whether an insurer, some other type of financial firm, or a non-financial manufacturing or service company. The probability of any company becoming bankrupt in a given timeframe is north of zero, so over a long enough period of time it approaches 100%. Not sure if that tells us much beyond placing value on dividends and remembering the uncertainty of long-term forecasts.

Posted by realist50 | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/