"All there is to investing," Warren Buffett once said, "is picking good stocks at good times and remaining with them as long as they remain good companies."

If I may add to the master's aphorism: A good time to buy a good company is when other people don't like it very much. Or, as Buffett put it, "Look at [market] fluctuations as your friend rather than as your enemy."

These principles came to mind recently when I read a simple and powerful piece of research in the Bernstein Journal, a superb quarterly publication of Bernstein Investment Research and Management (www.bernstein.com) in New York.

The subject was high-quality growth stocks -- quintessential "good companies" -- and how cheap they had become. Bernstein calculated that the total market capitalization -- that is, the value according to stock price -- of the top 10 holdings in the strategic-growth portfolio it maintains for clients was $1.1 trillion at the end of 1998 and, coincidentally, $1.1 trillion at the end of 2002. But in 1998, the 10 companies generated $26 billion in net income while in 2002 they generated $60 billion.

In other words, while their prices held steady, the companies produced 129 percent higher earnings (and that doesn't count the dividends). So their valuation, in terms of price-to-earnings (P/E) ratio, dropped by more than half, down to just 16, based on projected profits for 2003.

By the way, an update finds that the stocks are up a bit -- to a market cap of $1.2 trillion and a P/E a little over 17.

Bernstein calls these attractive stocks "America's best." But are they really? Maybe they deserve their relatively low P/Es.

Hardly. "As of year-end," said Bernstein, "nine of the 10 had met or beaten Wall Street's latest earnings estimates and had stable or rising estimates for this year -- in both cases bucking the market trend. On average, they were projected to grow earnings at 16 percent a year over the next three years, versus the S&P's 10 percent. [The latest: 17 percent vs. 9 percent.] However, investors didn't seem to care."

And that's the point. The bear market has destroyed the value not only of weak companies that never deserved the support of investors, but also of strong companies that have thrived even as the economy has struggled.

Take, for example, Kohl's (KSS), a highly profitable chain of 457 specialty department stores, based in Menomonee Falls, Wis. Earnings for each share of Kohl's stock have risen in a Beautiful Line since the company went public in 1990, increasing by at least 20 percent each year and averaging a gain of more than 40 percent annually. Value Line estimates that growth will continue at 25 percent a year through 2008. Yet, based on 2003 earnings projections, the stock trades at a P/E of 24 -- the highest among Bernstein's 10 companies but still modest by historic standards.

If you look at the financial statements of Kohl's -- or those of most of the 10 Bernstein strategic-growth companies -- it's hard to tell there was a recession in 2001 and an anemic recovery since then. Kohl's, which Value Line rates "A" for financial strength, earned $372 million on $6.2 billion in sales in the boom year 2000 and $643 million on $9.1 billion in sales in the sluggish year 2002. Yet Kohl's stock is down by about one-fourth in the past 12 months.

Or consider Freddie Mac (FRE), which provides the cash that banks and mortgage lenders use to make home loans. I'll admit the company has some problems: Its implicit debt guarantee from the Treasury is under political attack, and it has disclosed some minor accounting irregularities. But this is a very solid company with an excellent reputation that has been increasing its earnings at a 17 percent pace for the past decade and is expected to continue at double-digit rates for the next three to five years. Freddie's current P/E ratio is a mere 7, and it pays a 1.8 percent dividend, its highest yield since 1995.

Then there's UnitedHealth Group (UNH), the giant managed-care company, whose revenue has risen from $1 billion to $33 billion in a decade. Cash flow is spectacular since capital requirements are so low, and Bernstein projects earnings to rise 18 percent annually for the next three years. United stock has nearly tripled since 1998, but it still trades at a P/E of 19 based on current earnings. That compares with 28 for the Dow Jones industrials and 31 for the S&P.

"We recommend this stock," writes George Rho of Value Line. "It is timely for the year ahead. As well, our projects suggest that a combination of earnings and P/E expansion will yield above-average three-to-five-year capital appreciation."

The other seven Bernstein stocks? Lowe's (LOW), home-improvement retailer; Pfizer (PFE), pharmaceuticals; Citigroup (C), financial giant; Microsoft (MSFT), software; Viacom, Class B (VIA/B), entertainment; MBNA (KRB), credit cards; and General Electric (GE).

"These are companies," says the article in the Bernstein Journal, "with the wherewithal to weather the current tumult, precisely because they are industry leaders in market share, customer goodwill, and earnings history and prospects."

The strategic-growth portfolio has beaten the S&P in 70 percent of the 12-month periods and 100 percent of the 10-year periods since 1980. Even if you aren't well heeled enough to be a Bernstein client, you can buy into the AllianceBernstein Premier Growth fund (APGAX) and get a similar group of stocks. At year-end (the most recent report), nine of the fund's top 10 holdings were the same as those of the strategic-growth portfolio. The difference: The fund's third-largest asset was American International Group (AIG), the huge global insurer, while GE ranked 21st.

Bernstein's analysts call growth stocks "grossly undervalued," but they also believe value stocks -- that is, companies whose earnings are growing more slowly but whose valuations are lower, too -- are "priced about one-third below our estimates of their fair value." The firm likes American Electric Power (AEP), the nation's largest generator of electricity, at a P/E, based on 2003 projections, of 11; Chubb Corp. (CB), upscale property and casualty insurance, at a P/E of 10; Hewlett-Packard (HPQ), computers, printers and other high-tech products, 13; and Safeway (SWY), supermarket chain, 7.

Whether it's growth or value, the overall theme is simple: good companies at surprisingly low prices. But why?

My guess is that many investors are suspicious of numbers that look too brilliant, so they are particularly harsh on the firms that have dazzling records. There are never any guarantees in this racket, but the truth is that some firms actually are the best.

Go back to the original list and look closely at Lowe's. Even in a retail sector that has its ups and downs, Lowe's keeps going up. Except for a small hiccup in 1995, its earnings have risen at a double-digit pace since 1991 -- with average growth since 1997 of 24 percent and projected growth through 2008 of 23 percent. Annually. This is a dazzling money machine, conservatively managed with a solid balance sheet, a lot of cash and a small but consistently rising dividend.

What's a company like Lowe's worth? Well, it's expected this year to earn $1.8 billion, and its market capitalization is $36 billion. So its earnings yield (that is, its profits divided by stock market value) is 5 percent. And if earnings rise at their expected rate, they'll be $5 billion in five years. That's an earnings yield, based on today's stock price, of 14 percent.

Now, compare those returns with a five-year Treasury note, currently yielding about 3 percent. Yes, Lowe's is riskier than the Treasury (unless inflation rises sharply), but is it all that much riskier? Does an investor require an equity risk premium of 11 percentage points by the fifth year as protection? I think not.

Bernstein notes that its research indicates that the current risk premium -- the extra return that investors demand because stocks are riskier than bonds -- is 5.5 percent, "close to the highest level posted in 50 years." That premium, it seems to me, still reflects geopolitical concerns that may well be past.

Unless, against all rationality, a high risk premium persists, then Lowe's appears enticingly inexpensive -- even if its earnings growth is half what's predicted. So does Kohl's. So does Freddie Mac. So do a number of other stocks that Bernstein likes, including Wal-Mart Stores (WMT), Intel (INTC), French drugmaker Aventis (AVE), and Japanese automakers Honda (HMC) and Nissan (NSANY).

In fact, consider the entire stock market. Using a quick-and-dirty formula, the respected Web site Economy.com plugs in assumptions periodically and comes up with a "fair value" for the S&P. The consensus of economists is that S&P earnings will grow 9 percent this year and that the 10-year T-bond yield will be 4.75 percent at the end of the next 12 months (compared with 3.95 on Thursday). Based on these figures, the Web site calculates a fair value for the S&P of 1260 -- or 41 percent higher than it was on Thursday. That's the equivalent of a Dow of 11,755.

Try even more modest assumptions: earnings growth of a mere 6 percent and a T-bond rate of 5 percent. You still get an S&P of 1181 and a Dow above 11,000.

Of course, the current economic news is, at best, mixed, and stocks can certainly fall in the short term. But, in the end, it is hard to disagree with Bernstein's analysts, who courageously resisted the excesses of the late 1990s and are known for following a tradition of caution:

"If the past is any guide to the future, the huge disconnect we're seeing now between stock price and company fundamentals will not last; stock prices should catch up." I agree, and, in taking advantage of the anomaly, I'd start with the best.

Of the stocks mentioned in this article, James K. Glassman owns Microsoft, MBNA and General Electric. His e-mail address is jglassman@aei.org.