Lessons of the ’90s many didn’t learn

The investment world is nothing if not optimistic — dangerously optimistic. Even after the 10% correction of the second week of this month, the average price/earnings ratio is about 30. That’s far higher than the historic average, and certainly optimistic.

Rob Arnott, managing partner of First Quadrant, recently ran into this optimism at a conference when a pension executive, boasting about the overfunded status of his plan, revealed that the assumption used in the funding ratio calculation was 10.75%. The executive thought it “stupid” to ask what the funding ratio would be with a 7% return assumption because there was no way the fund would ever earn as little as 7%

The exchange revealed that a large part of this optimism is built on a foundation of dangerous “lessons” learned during the bull market of the 1990s, according to Mr. Arnott.

The first dangerous lesson, Mr. Arnott says, is that market drops are buying opportunities. Every drop of the past 25 years has been followed by a recovery to new highs.

“This is the single most dangerous lesson of the 1980s and 1990s,” he wrote in First Quadrant’s January client letter.

He pointed out that investors who bought in 1929 had to wait 25 years to recover their losses. The buyers in 1930, after the market was already down by half, had to wait 18 years.

The second dangerous lesson is that earnings and dividends don’t matter. Dividends are a less rewarding means of delivering wealth to investors than mergers and acquisitions or internal reinvestment. And depressed earnings are acceptable as long as they are a consequence of aggressive reinvestment.

But those views are simplistic, according to Mr. Arnott. To be of value, that reinvestment ultimately must deliver profits. “Those who pay today for earnings prospects one or two years hence are, most typically, sensible; those who pay today for earnings a decade hence are, most typically, betting on the `greater fool theory,’ ” he wrote. Many or most of those bets will fail.

Another dangerous view, according to Mr. Arnott, is that a fundamental restructuring of the technological base of the entire economy will deliver higher growth rates and more reliable profitability, allowing higher valuation levels from those segments of the market that are driving the change or will benefit from it.

All this is true, at some level, he says. “But the more relevant question is whether these circumstances are already fully reflected in current market prices.”

Another dangerous lesson is that past stock market returns are a good predictor of future stock market returns. Mr. Arnott points out in his client letter that a modest, but significant, negative correlation exists between long-term past returns and subsequent future long-term returns.

don’t believe in valuation levels

He notes that the 8% real return of the past 74 years has three components — the dividend yield, real dividend growth and valuation levels. Valuation levels have more than quadrupled relative to dividends, earnings or book values.

“This last phenomenon would be very dangerous to extrapolate,” he says.

The good news is that Mr. Arnott’s over-optimistic pension executive appears to be the exception, rather than the rule.

The most recent Buck Consultants report on actuarial assumptions shows that the average discount rate used in calculating funded status in 1998, the latest year available, was 6.77% a year for corporate plans, down slightly from 7.23% in 1997. They ranged from a low of 5.25% to a high of 8.50%.

The average company had an expected long-term return for its pension fund assets of 9.11% a year. This ranged from a low of 6% to a high of 11.5%.

Most funds clustered around 9%, a far more reasonable figure than that cited by the executive. Interestingly, diversified financial services companies had the lowest expected rates of return — 8.58% a year. What do they know that the rest of us don’t?

Michael Clowes is editorial director of InvestmentNews and Pensions & Investments, where this article initially appeared.