Berkshire Hathaway
Our Newest Insurance Business

Related Links
Discussion Boards

By Tode
March 6, 2008

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!

GEICO is "just getting started." Gen Re "is now a huge asset" and "by far our largest source of 'home-grown' float." And the "world now turns" to Ajit Jain's National Indemnity for "one-of-a-kind mammoth transactions." These are indeed three remarkable insurance operations.

But it seems to me that Berkshire has quietly created a fourth insurance operation out of thin air that has float generating economics that may put these others to shame.

This new business just got started a few years ago but has already generated premiums of $7.8 billion. It has one employee who works on a part-time basis (maybe 2% of his time), while moonlighting and performing various handyman services for other parts of Berkshire. His total salary for everything he does is $100,000 a year. That would put his salary allocated to this little side operation about $2000 a year. Not a bad expense ratio for an insurer that can produce $7.8 billion of premiums overnight.

Better yet, it produces long tail float but does not have the normal problems one associates with long tail float (judicial inflation, medical inflation, buried suits, etc.). The float seems likely to be cost-free over time. The largest line of business this company writes, in fact, has an excellent chance of producing a combined ratio of ZERO. That's right--100% of the premiums eventually may flow to the bottom line as underwriting profit.

By now you probably have figured out the identity of the insurer I am talking about--Berkshire Weapons of Mass Destruction Assurance Company.

Its largest line of business sells insurance to people who are worried what their stock portfolios will be worth 15 to 20 years from now. They so far have written premium checks that totaled $4.6 billion at the end of 2007. They don't get any protection until the maturity date arrives in 15 or 20 years. On the maturity date, if their portfolio is worth less than it was the day they bought the policy 15 or 20 years ago, we basically agree to restore them to the same portfolio value they had when they bought the policy (as I understand the deal). Meanwhile, we get to invest the $4.6 billion premium for 15 to 20 years.

Now, take note--we did not sell this insurance back in 1999 when the stock market bubble was ready to pop and lots of people "needed" it. We only started selling it recently, after our part-time employee decided that equities were, generally speaking, valued about right. If he is correct on that judgment, what are the odds they will be selling for less in 15 to 20 years than they are today? That is a math problem our guy can do in his head with ease. He's studied the correlation between GDP and market value. He expects that GDP is almost certain to be materially higher in 15 or 20 years, even though he thinks we may be in for a rough patch over the next few years. (Had his insurance buyers insisted on policies that would pay off in two or three years if markets declined, he would have walked away.)

Our guy thinks the odds are heavily in his favor that the world's stock markets will be at least a little higher 15 to 20 years from now. That's all he needs to keep the entire premium. And even if he gets very unlucky on that bet, he figures the money he earns investing the $4.6 billion over 15 to 20 years will almost certainly exceed any shortfall. His mathematical expectation is that we won't have to pay any claims, producing a combined ratio of 0%. That's good.

And if he can earn a return of say 9% on the premium, that would quadruple his $4.6 billion in 16 years.

The other line of business this company writes is insuring high yield bonds. The economics of this business also look pretty interesting. We have collected $3.2 billion in premiums. Unlike the other business, in this one we have paid some claims and expect to pay more. However, we have estimated our losses (payable in 2009 to 2013) at $1.8 billion. We likely are being conservative in our loss reserving, as always. If the loss turns out as expected, the combined ratio for this line would be 56% (1.8/3.2) That's still good. The float is not as long-tail, but still a nice business.

I am surprised this new business is getting such a ho-hum reception on this board and in the media. WEB and CM have done a remarkable job of generating insurance float in the traditional lines, now up to $59 billion. But they have commented that there are limits to the world's supply of insurance float. We already have a large share of the "good float" available. I think it is exciting news that they have figured out a way to generate this new type of low-risk, long-tail insurance float by spotting 94 mis-priced bets in the derivatives market.

I ask myself, why aren't other companies doing selling this insurance? As best I can guess, there are two reasons. They aren't willing to accept the lumpy earnings ($1 billion paper losses in a quarter is a conversation stopper in most board rooms). Also, the buyers of these policies probably are very picky about the creditworthiness of their counterparty. When you write a check for $4.6 billion as disaster insurance for your equity portfolio, you want to be certain the seller can perform its promise 20 years from now.

I also ask myself, who are the institutions who are buying these policies, especially the equity index variety? Why are buying what we are selling? I don't have a good answer for this question, but I am happy that WEB is meeting their needs.

Finally, I ask what are the prospects for growing this float this year and the next few years? My guess is the demand for this product is inversely proportional to the need for it. That is, the greater the perception of market volatility, the greater the demand for ways to hedge the risk, even though lower prices then prevailing reduce the odds of a claim being payable in 15 to 20 years. So this may prove to be another reason why a bear market in good news for Berkshire.