(MONEY Magazine) – Something amazing happened on Oct. 24. For the first time in modern history, a stock closed above $100,000, as Berkshire Hathaway's A shares ended the trading day at $100,600. For the price of a single share in Warren Buffett's company, you can purchase a Mercedes SLK350 and a Lexus RX350 and have 10 grand left over. Yet the price of the average stock has remained stuck between $30 and $50 since the Hoover administration. That got me thinking: How come all stocks don't cost as much as Berkshire Hathaway's? The obvious answer is that not all companies are run by Warren Buffett. Ultimately, however, the solutions to this riddle are more subtle--and more enlightening.

GE-ometry

Back on Dec. 31, 1935, General Electric's stock closed at $38.25. On Oct. 24, 2006, when Berkshire broke the $100,000 barrier, GE closed at $35.42. So from 1935 to now, GE's share price went nowhere; in fact, it went slightly backward.

Yet over those 71 years, GE stock earned an average annual return of 12.3%. How on earth could a stock do so well for seven decades and end up with a lower price than when it started?

The answer lies in the give-and-take between dividends and stock splits. Dividends raise a stock's return without raising its price, as a company pays out excess earnings as cash or new shares to stockholders. Splits, on the other hand, reduce a stock price; typically a company replaces every old share with two new ones, each at half the former price. GE has split its stock seven times since 1935, turning each original share into 288 today; without those splits, GE shares would now be trading above $10,000.

Why do companies keep their share prices low? The usual excuse is that retail investors won't buy a stock with a high share price. But economists Shlomo Benartzi of UCLA and Roni Michaely of Cornell University point out that this makes no sense. For one thing, the typical "small" investor's stake in an individual stock exceeds $6,000. Prices for most other goods have risen roughly tenfold since the Great Depression, and no one balks at paying $25,000 for a car or $250,000 for a house. Besides, it's big institutional investors, not regular folks like you and me, that own the vast majority of stocks. Institutions pay a penny or two in commission on each share they trade, so all this splitting simply means higher costs for them; in 2005, GE investors paid more than $100 million in brokerage costs that could have been avoided if the stock had never split.

Imaginary Money

Why, then, do so many stocks trade in an almost permanent narrow range? I think it's because, even for the "experts," investing is partly about finding ways to feel good about yourself. After a stock splits, you can sell some of your new shares and end up with no fewer than you started with, thus creating what seems like found money.

Hanging on can make you feel even better. If you bought a stock at $25 and it goes to $50 and splits two-for-one, you now have two shares at $25 each--and can immediately begin dreaming that they will go back up to $50 and split again. This is a form of what psychologists call anchoring--estimating what something is worth by seizing on an obvious number. The more often a stock hits 50, then splits and goes back up to 50, the more inevitable the next doubling seems to become. By keeping stocks in a narrow trading range, companies and their investment bankers fool us into thinking the shares naturally belong at the high end of the range, regardless of how profitable the businesses may be.

The Right Question

If you focus on rising stock prices as your main hope for future wealth, you're making two mistakes. If history is any guide, instead of rising, stock prices will more or less stand still. And dividends will make you richer than rising prices alone. The Dow Jones industrial average recently broke a record of its own, rising above 12,000, but it would have broken 1 million in November if reinvested dividends were included, points out finance professor Meir Statman. So if you're not reinvesting dividends into your stocks or funds, you're eating your seed corn.

Next, remember that a stock has a price but a business has value. Let's imagine that Larry, Moe and Curly each own a profitable company and offer to sell you a stake. Larry will sell you one share for $5, Moe will sell you one for $25, and Curly will sell you one for $1,000. Would you automatically buy Larry's stock just because its price is lowest? Of course not. Instead, you would ask how successful each business is and how many other stockholders you'll divide profits with. Otherwise you have no way of knowing how much value underlies the price of each stake. Share price by itself is meaningless. Investing in a stock without doing the homework that lets you arrive at a fair estimate of the value of the business it represents is like buying a house without ever having stepped inside. You'll have no real sense of the value of what you bought, and you're likely to end up with a bad case of buyer's remorse.

Worth Cheering About?

If it weren't for splits, which don't change the underlying value of a stock, the average share price today would be close to $2,000.

$48 Average price of a stock in 1936

$47 Average 2006 price due to stock splits

$1,850 Average 2006 price without splits

SOURCES: Shlomo Benartzi, UCLA ; Roni Michaely, Cornell U.