Here is a recent Fortune article...but I know Warren has warned about pension fund estimates before this.
http://money.cnn.com/2008/02/29/news/companies/berkshire_annual_report.fortune/index.htmPension problems ahead
Natural disasters aren't the only worry on Buffett's mind. He is also concerned about the bill that will come due when U.S. companies are forced to tell shareholders they have been pumping up their earnings by under funding their pension plans.
Most big companies have been vastly overestimating the kind of returns their pension plans can realistically expect to earn, Buffett writes. He writes that a survey of S&P 500 companies with pension plans shows the companies on average expect their pension assets to earn an annual return of 8%. With more than a quarter of those assets invested in cash and bonds, Buffett writes, the implicit expected annual stock investment return is 9.2% - and that's after fees. "How realistic is this expectation?" Buffett asks.
Not very, he finds. He writes that the Dow Jones Industrial Average surged from 66 to 11,497 during the 20th century. That is a huge rise - yet it averages out to just 5.3% compounded annually, Buffett writes. What's more, were the DJIA to repeat that 5.3% average annual gain throughout the 21st century, its value on Dec. 31, 2099, would approach 2 million.
"It's amusing that commentators regularly hyperventilate at the prospect of the Dow crossing an even number of thousands," he writes. "If they keep reacting that way, a 5.3% annual gain for the century will mean they experience at least 1,986 seizures during the next 92 years. While anything is possible, does anyone really believe this is the most likely outcome?"
...
With all that in mind, why are companies making investment assumptions that could be vastly overstated? Buffett believes it's a case of CEOs manufacturing earnings growth now, at the expense of problems that will manifest themselves only later. Assuming higher returns means having to set aside less money now to build their pension funds. "What is no puzzle ... is why CEOs opt for a high investment assumption," he writes. "It lets them report higher earnings. And if they are wrong, as I believe they are, the chickens won't come home to roost until long after they retire."