#26 | Dynamite in the Basement
Life insurance companies are supposed to be boring. They’re supposed to write predictable products backed by predictable assets that, over the long run, produce predictable returns to policyholders and/or shareholders. Financial strength, then, should be somewhat predictable. Ratings agencies and firms like ALIRT provide excellent snapshot insight into the momentary financial condition of a company. And, if things are really predictable, then momentary snapshots are the best indicator of the health of the company going forward.
The last five years should cause to throw away any notion of predictability. Life insurance companies have been systematically adding new product risk to their books for more than a decade that has arguably transformed some insurance companies from a risk pooling entities to financial counterparties. Profitability and solvency in a risk pooling entity are primarily related to underwriting. In financial counterparties, profitability and solvency has more to do with asset movements against bets. The finances of life insurance companies must reflect the products they sell. Unpredictable, risky products mean unpredictable, risky insurance companies.
Ideally, we’d want ratings agencies to tell us about the condition of a financial institution now but also over the long run. That means encompassing all of this new risk into the rating so that no one is surprised when the company bleeds if a major bet goes bad. The problem is that ratings do nothing of the sort. Ratings for a solvent, profitable company might look identical to ratings for a solvent, profitable company propped up by a massive amount of risk taking. Skeptical? One example closed the case for me. ALIRT, an independent financial research firm, rates insurance companies on a 1-100 scale where most fall at about the 50 range. In 2007, ALIRT rated John Hancock at something like a 75. In 2008, John Hancock fell to the low 40’s, a 30+ point swing. The culprit was the fact that Hancock wasn’t hedging the guarantees embedded in its VA block. Did ALIRT include this risk factor in its 2007 report? Not that I saw. Did it adjust its original rating for John Hancock because of the risk of an equity swoon? Obviously not. So if ALIRT is showing a number that we take to be a long term predictor of financial strength, how is it that the number was virtually chopped in half in the span of one year? Either the numbers are unreliable or we’re not supposed to interpret them as predictors of future results.
And this isn’t really ALIRT’s fault. It’s simply a reflection of the increasingly convoluted financial conditions of life insurers. John Hancock was far from an isolated event. Hartford was also bludgeoned by imprecise hedging on its VA block. American General’s ratings were crushed by a securities lending program to AIG’s Financial Products division – yes, that one, the one that is responsible for turning a ripple in the fabric of the financial system into a raging black hole. Companies with strong foreign parents wake up as capital-starved orphans. An unnamed company stood up in the AG 38 proceedings and declared that retrospective applicability of AG 38 to its entire block of business would force technical insolvency. Life insurers in the GUL space are relying substantially on opaque captive reinsurers with unknown assets to back their new business sales.
All of these point to the fact that life insurers are no longer predictable. Put differently, life insurers now face a bevy of non-linear risks – gunpowder in a cascade of sparks. Ratings agencies usually only tell you about an explosion when your ears are already ringing.
To be sure, analyzing carrier statutory filings has changed my view of risk at the carrier level. I used to think of carriers as generally operating according to a standard format. They write products, buy assets, pay commissions, cover overhead and produce profit. They don’t. Some carriers hedged their GMWB exposure, some didn’t. Some carriers gorged on GUL, some didn’t. Some carriers cheated on AG38, some didn’t. Some carriers use captives, some don’t. Some carriers are going long on assets to pick up yield, some aren’t. Some carriers are increasing mortgage holdings, some aren’t. For better or worse, carriers are making decisions that entail risk which will come to fruition at some point. We’ve seen a few explosions so far but an upcoming post will address the next pile of gunpowder – massive derivative holdings at the carrier level that are little more than bets on rising or falling interest rates.
Risk is intoxicating because it masquerades as return. With most things in life, we can see risk/return ratios because have historical data or, better yet, chances to watch others roll the dice in real time. Anyone who is afraid of flying should sit at Hartsfield-Jackson for a few hours to watch the hundreds of planes flow in and out of the sky. Life insurance, on the other hand, is completely different because the liabilities play out over many years – long after the folks pricing and selling them are retired. No one really knows how this will all turn out. And, more importantly, the people in the business now will likely not be on the hook for the results.
Why address this issue now? Because producers and distributors are talking about carrier strength and commitment to the life insurance market more than I’ve ever heard before. The question of carrier stability is becoming more and more paramount as producers try to stand by the products they sell and distributors sink time and effort into building business lines. These folks are asking questions about the inner workings of life insurers that were taken for granted in previous years.
And I’m going to argue that these questions start and stop with the producer. Carriers will never, under any circumstance, tell you that they have dynamite buried in the basement. Carriers will never tell you that you should stop selling their products. So, I hate to say it, but we should probably stop listening to them. Ratings agencies can only give you a single frame from a two hour movie, so we should probably not rely on them solely. I’m arguing that producers should rely on three additional things. First, diversification of coverage. Second, their own intuition and experience as to whether a company is being aggressive in underwriting, pricing and distribution. And third, their willingness to trade pricing for other benefits. Or, put differently, their willingness sell a product where the carrier can deliver consistent benefits to the client without sacrificing its own profitability or overloading on risk. It’s a return to a risk-share model and the carriers have to hold up their end of the bargain. A guarantee is a very expensive substitute for trust.