Researchers Find Evidence

To Back Theory on Economy

By GREG IP

Staff Reporter of THE WALL STREET JOURNAL

Maybe all the new-economy hype isn't just hype after all.

Almost everyone agrees the revolution in information technology has probably played some part in the extraordinary valuations that stocks have reached this decade.

But figuring out how big a part has proved elusive. Skeptics look on "new paradigm" arguments as the sort of fuzzy-minded thinking that usually accompanies speculative bubbles in the stock market.

Now, some researchers have found compelling evidence that conventional accounting understates the earning power of today's companies -- earning power that the stock market correctly recognizes.

The research, if correct, goes a long way toward explaining how stocks, in particular of technology companies, could sensibly trade at such unprecedented multiples of earnings.

Friday, those trends were well in force. The Dow Jones Industrial Average eased 50.97 points to 11028.43. But the Nasdaq Composite Index, loaded with technology stocks, climbed 35.04 to a record 2887.06, passing its previous high of 2864.48 set on July 16. The Standard & Poor's500-stock index, which added 4 to 1351.66, now stands at a near-record 33 times trailing earnings.

But does such a high price-to-earnings ratio mean stocks are overvalued? Earnings would be higher and P/E ratios lower if companies weren't spending so heavily on "intangible assets" such as research and development, software, marketing and computer training. Intangible assets fuel future profits just as surely as would a "tangible asset" such as a piece of equipment or a factory. But intangibles are expensed against current earnings, while "tangible" assets are added to the balance sheet and gradually depreciated.

"Investment in intangible assets represents a growing proportion of our economy," Leonard Nakamura, economic adviser at the Federal Reserve Bank of Philadelphia, says in a recent paper. Corporate spending on R&D has more than doubled from 1.3% as a percentage of their revenue (after subtracting purchases from other firms) in the 1950s to 2.9% in the 1990s, the paper says (previously described in Barron's, A New Stock Valuation Model Suggests the Market May Not Be So Pricey After All, Aug. 23.)

Gillette's spending to develop its Mach 3 razor and Pfizer's to develop its Viagra impotence drug have created lucrative, intangible assets -- though they can't be found on their balance sheets, Mr. Nakamura notes.

This "helps explain the rising value of U.S. equities. That explanation, in turn, suggests that continued strong economic growth and strong profit growth in the future are not so implausible," he says.

Mr. Nakamura estimates that after treating R&D as regular investment and removing inflation's distorting impact on inventories and depreciation, the market's P/E ratio is only a little higher than in 1972, whereas unadjusted, it is 41% higher.

Join the Discussion: How can candidates truly appeal to New Economy-era voters? What will it take for the major political parties to attract the loyalty of this group?

Mr. Nakamura says other studies have found that a dollar of R&D earns $1.25 to $2 of stock-market value, while a dollar spent on plant and equipment earns just a dollar in market value. So why not spend more on R&D? For one thing, it is riskier. When a steel company builds a new mill, it can generally count on selling more steel. But a biotechnology company has a small chance of striking it rich with a wonder drug and a big chance of going bust looking for one.

This explains the extraordinary valuations commanded by "new economy" companies such as Amazon.com or Yahoo!, Michael Mauboussin, chief U.S. investment strategist at Credit Suisse First Boston, says.

Yahoo's sales tripled in 1998 to $201 million from the previous year and its "cash" earnings, or earnings before interest, taxes and depreciation, went from about zero to $48 million, and to do this it only had to spend $8 million on fixed assets, Mr. Mauboussin calculated. Prior spending on research, software and marketing has made Yahoo a primary Internet destination, and now "does not have to spend much money to serve an incremental customer." Thus, new revenue goes primarily to the bottom line.

Such incredible economies of scale are typical of new-economy companies, he says. But the flip side of Yahoo's success is that many other new-economy companies have spent enormous sums on marketing and customer acquisition and will invariably fail, leaving behind worthless brand names, inactive Web sites and poorer shareholders.

These accounting arguments don't satisfy skeptics. After all, P/E ratios are still at a record when adjusted for R&D. And impressive as Yahoo's business model may be, does it merit a trailing P/E ratio of more than 1,000?

Federal Reserve Chairman Alan Greenspan acknowledged two weeks ago that the economy's shift to "idea-based value added," where investment is expensed immediately rather than depreciated over time, has understated earnings, although that is offset by the increased use of stock options in place of wages. But he added, "It does not seem likely ... that such [accounting] adjustments can be the central explanation of the extraordinary increase in stock prices."

Mr. Nakamura says, "It could be that some proportion of what's going on now is a bubble... . It's important not to be complacent about the stock market and think it will do this forever. On the other hand, it's important to recognize we're in fact saving and investing a lot more than it appears on the surface."

The stock market also may be recognizing benefits of information technology investment that don't show up in earnings statements. Erik Brynjolfsson and Shinkyu Yang of the Massachusetts Institute of Technology looked at 1,000 companies' financial statements from 1987 to 1994 and concluded the market values a dollar invested in information technology at 10 times what it values a dollar of conventional capital.

This, they argue, is because software spending, training and other expenses likely boost spending on information technology to as much as 10 times the cost of hardware. Such expenses hurt earnings, but investors should look past that impact and see the future benefit to the company. Thus, companies that spend a lot on computers tend to have higher P/E ratios.

Mr. Brynjolfsson says, "Wal-Mart Stores has invented a whole cluster of innovations that enable it to sell products more cheaply and efficiently and get them to customers more effectively." Even if Sears installed Wal-Mart's technology, it couldn't replicate all the systems and processes Wal-Mart has developed around that technology, he says. That is why the market values Wal-Mart Stores much more highly than Sears: "It's not because they own a whole bunch of valuable land in Arkansas. It's obviously the intangibles."

The economic establishment is beginning to accept some of these arguments -- but only some. The Bureau of Economic Analysis is about to change how it calculates economic output by reclassifying software purchases as investments rather than current spending, which it estimates would have boosted the level of output in 1996 by 1.5% (although the boost to output growth would be far smaller). But for now it isn't reclassifying databases, or literary or artistic works as investments, as international guidelines suggest.