The Dumb Money

By Daniel Kadlec

Has Warren Buffett left billions of dollars on the table by never splitting the stock of his company? It sure seems that way. In the past few weeks, dozens of firms from Internet darling eBay to Xerox to Microsoft have announced splits and watched their stocks soar. Last week athletic-shoe company K-Swiss joined the fun by announcing healthy earnings and a 2-for-1 split. Its shares jumped 23%. To the same point, when Cisco Systems on Feb. 2 posted earnings that beat Wall Street estimates but failed to declare an expected stock split, its shares dropped 2%.

Yet Buffett, the fabled investor and CEO of Berkshire Hathaway, refuses to split Berkshire's A shares, which traded last Friday for $72,500 each. If stock splits add lasting value to a company, as today's fervor suggests, it's a wonder that Buffett is held in such high regard. His reputation, though, wasn't built by pandering to passing fads.

When a company splits its stock, a share worth, say, $100 becomes two shares worth $50 each. That doesn't change the intrinsic value. But it does make the stock more affordable for small investors, who like to buy in round lots of, say, 100 shares. That's one reason stocks that split have historically got about a 5% lift between the date of the announcement and the actual split. Lately, though, the pop has been more explosive. EBay rose a quick 37%; Xerox, 10%; Microsoft, 12%. People now pay for services that alert them via pager or e-mail whenever a split is announced, so they can quickly buy the stock. It's an example of lemmings running amuck in the market--dumb money chasing any trend that doesn't require thoughtful analysis.

Even dumb money does sometimes make money. It's possible to be wrong but lucky--especially in a bull market. But luck doesn't last; fads fade. Successful investing requires research and judgment--if only to judge that you're better off buying your stocks through mutual funds.

Where is all this dumb money? Much of it is in Internet stocks, most of which command stratospheric prices despite their lack of profits and their dubious prospects. In a seminal event for these stocks' valuations, high-flying portal company Lycos last week agreed to merge with USA Networks, a real company (Home Shopping Network; Ticketmaster) with $1.5 billion in annual revenue. Lycos shares plunged 31% in two days as investors reflected on Lycos' value inside a company that must--gasp--post rising profits to boost its share price.

When America Online said last November that it would buy Netscape in a stock swap, rabid online traders drove up Netscape's price beyond what AOL had said it would pay. There was no prospect of a bidding war. The lemmings--too busy to use a calculator--were simply piling on.

Stock-split mania is another version of this greater-fool investing. Yes, studies show that stocks of companies that split their shares outperform those that don't. But that's easy to explain. Splits naturally occur in the best stocks--the ones that go up. The split signals management confidence, but the heavy lifting is done by management execution that delivers earnings. Do that, and the stock will go up whether it splits or not. Just ask Buffett--whose shares have risen, on average, 28% a year since 1985.

See time.com/personal for more on stock splits. E-mail Dan at kadlec@time.com See him on CNNfn Tuesdays at 12:45 p.m. E.T.