How high is high? Every investor has to be wondering if, tomorrow, this glorious stock-market genie will crawl back into its bottle.

Even Jeremy Siegel, the fearless defender of most of today's high stock valuations, says he's uncomfortable. By his measure, some popular growth stocks have definitely overshot.

Siegel, a professor of finance at the Wharton School in Philadelphia, wrote the well-regarded book, "Stocks for the Long Run." He's a growth-stock fan. These are companies whose earnings have grown, historically, at a faster-than-average pace.

He argues his case for growth stocks by analyzing the so-called Nifty Fifty. They're the 50 high-fliers most beloved of institutional investors, and for which they're willing to pay almost any price.

The Fifty are thought to be "one decision stocks" buy now and hold forever. Even when purchased at high prices, they're expected to deliver superior returns.

Of course, they might not. The Nifty Fifty of the last growth-stock mania, in 1972, plunged during the vicious bear market of 1973-74, leaving believers in the dust.

But as it turns out, those stocks although at very high prices weren't overvalued by much, Siegel says. Between year-end 1972 and August 1998, they underperformed the general market by only 0.5 percent a year, as measured by Standard & Poor's 500-stock index.

At first blush, this looks like a pretty strong argument for skipping growth stocks and buying an index mutual fund that copies the performance of the S & P.;

But there's another way of looking at high-fliers, Siegel says.

He divided the Nifty Fifty of 1972 in half, according to their price/earnings (P/E) ratios. That's the relationship of their stock price to their earnings per share for the past 12 months. A stock with a 30 P/E is selling at 30 times earnings.

The 25 growth stocks with the lower ratios gained 14 percent a year between 1972 and 1998, outperforming the 12.7 percent racked up annually by the S & P.; That's the group Siegel likes best.

As for the stocks with the highest valuations, none subsequently matched or beat the market if they started with a P/E ratio higher than 57.

What of the Nifty Fifty of 1999? Many are selling for more than 57 times earnings, including Airtouch Communications, America Online, Charles Schwab Co., Cisco Systems, Dell Computer, EMC Corp., Lucent Technologies, Medtronic and Microsoft.

The Fifty today are growing faster than the Fifty of 1972, Siegel says, so it's OK for them to carry somewhat higher valuations. Even so, "they're getting pricey," he says.

As a group, he thinks they'll be worth their current prices "over the long run." But this merely means that many, many years from now they'll have performed as well, or as poorly, as the market as a whole. Big deal.

Market analyst Steven Leuthold of the well-regarded Leuthold Group in Minneapolis takes a more skeptical view. He checked the performance of 1972's Nifty Fifty through March 1999, and concluded that they had lagged the market by 1 percent a year. (His methodology is a little different from Siegel's.)

One percent doesn't sound like much. But if you'd started with $1,000 in each of the 50 popular stocks, and reinvested dividends, your portfolio today would be worth 22 percent less than if you'd put that money into an S & P; index fund.

Leuthold also looked at the 25 priciest of the original Fifty stocks, which he calls the Religion Stocks (because they inspire blind faith).

A handful performed fantastically well (Merck, Schering Plough, Coca-Cola, Disney and Eli Lilly). But most of them didn't. A $1,000 investment in each of the top 25 would have yielded 36 percent less than an investment in an S & P; fund.

Siegel continues to believe that big growth stocks are worth high P/E ratios, as long as they're not too high and provided that you'll hold them for a long, long time.

That's the rub, Leuthold says. If your growth stocks plunge 57 percent, as the Religion Stocks did in 1973-74, would you steadfastly hold on? Or would you weep and sell?

Long-term investors shouldn't even consider the hottest New Era stocks. At this writing, America Online sells for 444 times earnings. Says Siegel, stocks in that range will eventually have to produce huge and fast-growing earnings or else their prices will collapse.

Social Security outlook

Here's something that will surprise you about the supposedly gloom-and-doom outlook for Social Security. By one official estimate, there's no Social Security problem at all.

The forecast you hear most often says that the Social Security trust fund will run out of money in 2034. That comes from Social Security's trustees, who make annual projections about the system's health. But Social Security's trustees make three projections, each one based on a different assumption about future economic growth.

In public discussion, you always hear the "intermediate" projection, which warns that the trust fund will be used up by 2034. This scenario assumes that economic growth will slow from 3.9 percent last year to 1.2 percent in 2075.

The second projection assumes super-slow growth. If that happened, the trust fund would be gone by 2024.

The third projection is the interesting one. It assumes the economy slows to a growth rate of 2.1 percent in 2075. That's almost a full percentage point better than the growth assumed in the intermediate forecast. If the economy reaches that higher mark, Social Security's trust fund will be large enough to pay every dime in benefits that has been promised today.

Over the past 75 years, economic growth in America has averaged 3.1 percent annually. So, you might ask, why are Social Security's trustees forecasting slower growth than that? The reason is that America's population growth is slowing down. The smaller the number of workers in the years ahead, the lower the growth in production and consumption.

Even with fewer workers, however, Social Security could remain whole if the economy achieved the highest of the three growth projections above.

To be prudent, the country shouldn't rely on the highest growth the trustees can foresee today. On the other hand, how drastic a change in Social Security do you want to make, knowing how hypothetical the projected outcomes are?

Congress does indeed have to prepare for the possibility that the trust fund will run out. That means shaving benefits or approving a tax hike. But maybe these changes should be phased in over the next 25 years, so they can be canceled if it turns out that business does better than anyone thought.

Syndicated columnist Jane Bryant Quinn can be reached in care of the Washington Post Writers Group, 1150 15th St., Washington D.C. 20071-9200.

For reprint and licensing requests for this article, CLICK HERE.