COMMENT & ANALYSIS: Warren Buffett's big mistake: Risk assessment may be a science but the unexpected will always be just that, as the great investor has discovered
Financial Times; Mar 12, 2002
By PETER MARTIN
globalarchive.ft.com
Warren Buffett's annual letter to the shareholders of his holding company, Berkshire Hathaway, often contains rueful admissions of mistakes.
Like his equally traditional confession of baseball incompetence, these are intended, I suspect, as much to amuse as to apologise. Past failures include overestimating the importance of Sunday opening to shoppers in Boise, Idaho, and selling out of successful investments too early.
The error to which he admitted in his latest shareholders' letter does not fall into that category, however. It was expensive and painful and Mr Buffett pulled no punches in his description of the mistake. Indeed, he gave a preview of the apology last November.
Although Berkshire Hathaway has a wide range of folksy operating businesses and a portfolio of blue-chip securities, its main business is insurance. It is here that the problems occurred.
In 1998, Berkshire Hathaway bought General Re, one of the world's biggest reinsurance companies. At the time, Mr Buffett had nothing but praise for the company, singling out its disciplined attitude to pricing and underwriting risk.
Berkshire could add absolutely nothing to the skills of General Re's managers, he said. "Nevertheless we believe that Berkshire's ownership will benefit General Re in important ways and that its earnings a decade from now will materially exceed those that would have been attainable absent the merger." Access to Berkshire Hathaway's capital would do the trick.
Since then, the tone of Mr Buffett's comments about General Re has got noticeably cooler. He has three rules of sensible insurance: accepting only risks that can be sensibly evaluated and priced; avoiding the danger of aggregated losses from a single event; and steering clear of moral risk ("trying to write good contracts with bad people doesn't work").
As he said in November: "during the past three years, all of those rules were broken at General Re". Alas, he gives no details of any bad people the company may have been doing business with. But he goes into a lot of detail about the other breaches of his rules.
General Re appears to have been historically underpricing risk across a number of its businesses. It has also been under-reserving for losses - that is, underestimating the ultimate costs when insured-against risks occur. And, of course, it failed to assess correctly the risks of terrorism on US soil, so tragically revealed on September 11.
In the last respect, it was no worse than the rest of the industry. But Mr Buffett feels particularly bad about this because he did spot the potential danger but did not convert thought into action. "I violated the Noah rule: predicting rain doesn't count; building arks does."
Investors appear forgiving: in the six months since September 11, Berkshire Hathaway's shares have performed pretty much exactly in line with the market. But Mr Buffett's remarks deserve further scrutiny.
He still believes that the purchase of General Re will pay off healthily. Loyalty is a Buffett virtue - and occasionally an expensive one: he stuck with the US textile and shoemaking industries long after their problems were clear.
The issues that have hurt General Re's profitability clearly range far wider than September 11. It is by no means a troubled company. But it operates on a scale and in market sectors that go well outside Berkshire Hathaway's past insurance experience, substantial though that is.
Will the General Re purchase ultimately be justified? A second comment by Mr Buffett bears on that. Stressing Berkshire Hathaway's appetite for large risks, he says: "We are perfectly willing to lose Dollars 2bn to Dollars 2.5bn (Pounds 1.4bn-Pounds 1.75bn) in a single event (as we did on September 11) if we have been paid properly for assuming the risk that caused the loss (which on that occasion we weren't)."
The trouble with this statement is its lack of operational content. How will he ever know that he has been paid properly for assuming a risk? If the risk does not occur, any price looks good. But if the risk does occur, it tells us little about the pricing of that particular policy. Only if the industry was in effect giving the cover free - as for terrorism - can we definitively say the risk was underpriced.
Mr Buffett may have been influenced by the record of his other reinsurance business, headed by Ajit Jain, which provides "super-catastrophe" cover for very rare but very costly events. This unit's consistent profits, despite occasional huge pay-outs, provides strong evidence that it is pricing risk properly. At least, it tells us that Mr Jain's general level of pricing is correct. But it does not tell us whether any individual risk has been properly priced.
It is likely that a specialist "super-cat" business, like the one Mr Jain runs, does not face the same relentless price pressure as General Re. If that is so, General Re's managers will be faced with a difficult decision.
Either they turn down business on price grounds, and thus lose market share, or they continue to accept risks in line with their competitors and hope that the industry as a whole is coming to its senses. Both these possible outcomes cast some doubt on whether General Re can be the "huge asset for Berkshire" that Mr Buffett still predicts.
That is enough Schadenfreude. Now think for a moment of the wider consequences. If an investor as canny as Warren Buffett, with decades of experience in analysing the insurance business, must admit to such mistakes, what of the rest of us? In particular, what risks are we unwittingly assuming - and underpricing? You do not have to be an insurance company to acquire unexpected exposure. Inventory risk, counterparty risk, credit risk, computer risk, currency risk, employee risk, accounting risk: the list is endless.
We cannot be sure that we have properly evaluated each of these risks and agreed an appropriate price for absorbing or transferring it. We shall know only when the unexpected happens. As Peter Bernstein says in his book on risk* "Wars, depressions, stock-market booms and crashes and ethnic massacres come and go but they always seem to arrive as surprises." As Mr Buffett has discovered, there is plenty of scope for retrospective apology.
* Against the Gods: the remarkable story of risk. Wiley, Pounds 9.99 peter.martin@ft.com .. Reports and text of the Buffett letter at www.ft.com/buffett |
