The Bottom Line
I suspect that a hard market in the insurance business is in some ways similar to a bear market in stocks. In either case it is the price we pay for irrational exuberance. Granted, the popping of a bubble in the insurance market is not nearly as much fun as a good bear market: no Enron, no Global Crossings, no dot bombs, no congressional investigations. Just a few boring bankruptcies, lots of downgrades by Bests, and the quiet retreat of many marginal players. But in its way, the insurance business is now going through a period of enlightenment similar to what occurs at the bottom of a bear market.
Lots of insurance capital has been destroyed in the last three years. A big chunk disappeared on September 11, but the industry has also lost capital in the declining equity market, low interest rates, and operating losses in the last few years. Berkshire just announced a large increase in reserves at Gen Re for claims unrelated to 9/11.If Gen Re was under-reserving, what was everybody else doing? In reality, the capital situation in the industry may be worse than people think. The problem with insurance reserves is there is a lot of room for creative expression.
"In a given year, it is possible for an insurer to show almost any profit number it wishes, particularly if it (1) writes "long-tail" business (coverage where current costs can be only estimated, because claim payments are long delayed), (2) has been adequately reserved in the past, or (3) is growing very rapidly." – Warren Buffett 1982 annual report.
If other companies have been doing the same things that Gen Re is guilty of, then the fourth quarter is likely to bring us some nasty surprises, and lot more capital (capacity) is going to disappear as companies adjust there reserve accounts. There will be a strong tendency to dump all of the bad news into 2001. Demand is increasing, capacity is disappearing; it is the best of all possible worlds if you are a well capitalized reinsurance company. Stories from insurance industry journals in January give us strong hints. We see an insurance market in chaos, where rates were the hardest in memory, and it is basically impossible for large commercial property owners to buy enough insurance to cover their risks. Risk is being transferred out of the insurance business and back to property owners, banks, and other lenders. It is a condition that I have never witnessed in my life time.
Warren has said many times that if the prices were good he would write a lot of insurance. Well prices are good now, and he has the capital to write all the insurance he wants, at rates that should guarantee underwriting profits. It is in fact difficult to conceive of circumstances that would produce better insurance prices. Underwriting losses for the insurance industry are the result of too much capacity, and capacity is measured by capital. The great bull market of the 1990's created a lot of capital. With too much capacity, insurance rates dropped and underwriting profits disappeared. In his 1982 letter, the chairman explained how the insurance cycle works: for profits to improve, capital has to be withdrawn (destroyed).
"This contraction will not happen because of generally poor profit levels. Bad profits produce much hand-wringing and finger-pointing. But they do not lead major sources of insurance capacity to turn their backs on very large chunks of business, thereby sacrificing market share and industry significance."
"Instead, major capacity withdrawals require a shock factor such as a natural or financial "megadisaster." One might occur tomorrow—or many years from now. The insurance business—even taking investment income into account—will not be particularly profitable in the meantime.
"When supply ultimately contracts, large amounts of business will be available for the few with large capital capacity, a willingness to commit it, and an in-place distribution system. We would expect great opportunities for our insurance subsidiaries at such a time."
How much insurance business will Berkshire write in 2002?
Berkshire wrote $19 billion in insurance in 2000. ($13.3 billion of that was reinsurance.) We will not know the figures for 2001 till next month when the annual report is released, but industry wide, rates were increasing at double digit rates prior to 9/11, and then accelerated rapidly in the fourth quarter. Premiums should increase even faster 2002.
Berkshire has the capital to write a lot more business, because they write a very low level of insurance relative to their book value. (See tables.)
- Table One Book value, premium income, and net insurance profit for Berkshire Hathaway and General Re. from from 1987 to 2000. Also shown, is the same data for General Re from 1987 to the merger.
- Table Two Book value, premium income, and net insurance profit for American International Group and Merkal
For the Last fourteen years, Berkshire has written premiums that averaged 15.2% of their book value. For the 12 years prior to its purchase by Berkshire, Gen Re averaged 66.9%. Gen Re's figure is still conservative, compared to most insurance companies (second table gives figures for AIG and Merkel). If Berkshire were to write at this level in 2002, it would amount to about $40 billion in premiums.
For Berkshire, any increase in premium volume will come from two sources: 1) premium increases on old business, and 2) an increase in market share (new business). Robert Miles interviewed Ajit Jain for his excellent book, "The Warren Buffett CEO." He quotes Ajit as estimating that Berkshire Re has currently about 10% of the North American reinsurance market, but that they could increase their market share by 50%, "...at the drop of a hat."
S&P has estimated that P&C and reinsurance premiums will increase in 2002 by an average of 25% to 40%. Combine the two figures (market share increase and rate increases on old business) and there is the potential for Berkshire to increase premium volume by $10 to $15 billion. This is not a prediction. I have no idea what Buffett will do. This is meant only to be an exploration of the possibilities.
What about the bottom line?
It remains to be seen exactly how Berkshire will react, but a big jump in reinsurance premiums next year will eventually have a much larger incremental impact on the bottom line. Not only are rates increasing, but new policies are carrying much higher deductibles, much lower limits on exposure, and more limitations on what the policy will cover. So even as premiums increase, it is likely that losses will not increase as fast premiums, because the company is writing policies with less exposure.
As you can see from the GenRe's table, their average return on premiums in the 12 years prior to the merger was 22%. On this basis they were more profitable than AIG or Merkel for that period. Also I suspect they were one of the most profitable companies in the insurance industry—conditions that would help explain why Buffett found the company attractive. There is of coarse no guarantee that they can match these returns now. But market conditions are better now than they where when Gen Re was bringing in that 22% of the premiums down to the bottom line; I would guess that Buffett expects this kind of return. I further suspect he will do what he has to to realize the returns he expects. Even if it requires further personal intervention.
While Zack's is predicting earnings increases of 200% to 400% for some insurance companies in 2002, they have not published an estimate for Berkshire. But a little math applied to the above data raises some tantalizing possibilities. For instance, a $40 billion premium base at 22% would yield an $8 billion bottom line for Berkshires insurance operations. It is not my contention that this is a rational estimate for future profitability. I point this out merely because it appears to me that bad news from the insurance business for the last three years has obscured the real earnings potential of Berkshire's insurance operations, and that there is now a very real potential for some positive surprises in the next two or three years.
How long these conditions will last is the next question. But early indications from the insurance market seems to be telling us that there is a lot less capacity than people think there is. This may be because more capital has disappeared than analysts realize. If this is indeed the case, the insurance market may stay tight for a good while.