Three Questions about “New Economy” Stocks

Author

Simon Kwan

FRBSF Economic Letter 2000-15 | May 12, 2000

From January 1995 through the third week of April 2000, the Dow Jones Industrial Average (DJIA) advanced a respectable 184%. Even though this venerable index has included tech firms over the last few years–most recently Intel and Microsoft in 1999–it still is taken by many to represent the industrial heavyweights of the so-called old economy.


From January 1995 through the third week of April 2000, the Dow Jones Industrial Average (DJIA) advanced a respectable 184%. Even though this venerable index has included tech firms over the last few years–most recently Intel and Microsoft in 1999–it still is taken by many to represent the industrial heavyweights of the so-called old economy. In sharp contrast to the DJIA, the S&P technology index–which represents the so-called new economy stocks–shot up 624% during the same period. As a result, the differences in valuations between the old economy stocks and the new economy stocks are striking. Even after the enormous stock market decline in the second week of April, the price of technology stocks still was averaging 62 times earnings, while the price of stocks in the DJIA was averaging 23 times earnings.

This Economic Letter addresses three questions about new economy stocks versus old economy stocks. First, how dominant are technology companies in the stock market, and how important are they in our economy beyond the stock market? Second, do the valuations of technology stocks make sense, given market expectations and underlying fundamentals? Third, how sensitive are technology companies to interest rate changes?

Dominance of technology companies

For the purposes of this analysis, the technology sector is defined to include the following industries: biotechnology, communications equipment, computer software and services, computer hardware, computer networking, computer peripherals, semiconductors, semiconductor equipment, cellular and wireless telecommunications, and Internet companies. Due to data availability, only publicly traded companies are used in the analysis.

Four measures are used to gauge the dominance of technology companies: market capitalization, total assets recorded on the balance sheet, employment, and net sales. In terms of market capitalization, the technology sector accounted for only 7% of the total stock market value back in 1990. With the big jump in market valuation of tech stocks during the last ten years, tech stocks made up 36% of total stock market capitalization as of March 2000–more than a fivefold increase.

How dominant are tech companies in the economy beyond the stock market? In terms of balance sheet assets, they play a relatively small role. Among all publicly held nonfinancial companies, the total assets controlled by technology companies was a mere 9% in March 2000 (it was 6% in 1990). The data on employment and annual sales are available only through 1998. They show that the number of employees in the technology sector accounted for only 7% of total employment among public companies in 1998, up from 6% in 1990. And in terms of sales, technology companies accounted for 10% of total sales by all publicly traded nonfinancial companies in 1998, up from 6% in 1990. These measures indicate that while technology companies account for more than one-third of the total stock market capitalization, they account for less than one-tenth of total assets, employment, and sales in the economy.

Market valuation

Quite clearly, the high valuation of tech stocks reflects investors’ willingness to pay a high premium for their future growth opportunities. The question is: How fast must technology companies grow their earnings to justify their current price-earnings (P/E) ratios, which currently average about 62? Let us assume that investors require a 15% annual return for holding technology stocks. In order to bring the P/E ratio of technology companies down to a more sustainable 30 in, say, three years, a back-of-the-envelope calculation suggests that technology companies must be able to generate average annual earnings growth of 46%. To bring the P/E ratio down to 30 in five years, technology companies still need to grow earnings at a rate of 33% per year; and even given 10 years, the required earnings growth is 24%.

How realistic are these expectations? Figures 1-3 provide some answers by displaying earnings, net sales, and operating income comparisons for tech and non-tech companies; note that the scale measuring tech performance is one-tenth the scale for non-tech performance. Figure 1 shows the aggregate earnings of all publicly traded technology and non-technology companies from 1990 to 1998. Between 1994 and 1997, when the technology sector was reporting healthy profits, the compound average annual earnings growth rate for these companies was only 7.7%. Ironically, non-technology companies reported faster earnings growth during the same period, at an average annual rate of 11.8%. In terms of sales growth, Figure 2 shows that between 1990 and 1998, technology companies’ sales growth was more than double that of non-technology companies: 11.7% for the techs versus 5% for non-techs. Turning to operating income, i.e., earnings before interest, tax, and depreciation, Figure 3 shows that operating income at technology companies grew at 10.4% between 1990 and 1998, versus 6% for non-techs. Taken together, compared to non-techs, tech stocks have very high P/E multiples, which implies extremely high earnings growth expectations by the market. In terms of fundamentals, technology firms have had slower earnings growth, but much faster sales growth, and somewhat faster growth in operating income than non-technology companies.

Interest rate sensitivity

How will monetary policy, and hence changes in interest rates, affect tech stocks? Some market commentators suggest that the tightening of monetary policy would do little harm to tech stocks while damaging non-tech stocks disproportionately. In particular, they assert that technology companies are less sensitive to interest rate hikes because these companies tend to rely much more heavily on equity financing than on debt financing. To shed light on this, it would be useful first to examine the capital structure of tech and non-tech companies and then to discuss the effect of changes in interest rates on the value of a capital asset using the discount cash-flow model.

Figure 4 shows that, in the aggregate, the debt-to-asset ratio for technology companies is much lower than for non-tech, nonfinancial companies, and that the gap has been widening. In 1990, the aggregate debt-to-asset ratio for techs was about half that of non-techs. By 1998, only 7% of tech firms’ assets were financed by debt, as compared to 27% for non-techs. This confirms that technology companies tend to have much less debt in their capital structure than do non-techs, suggesting that, other things being equal, technology companies would be less sensitive to interest rate changes than non-technology companies.

But other things are not equal. As discussed earlier, tech stocks tend to rely on more distant future cash flows to support their current stock prices than do non-tech stocks. Using the discount cash-flow model, an increase in interest rates will lower the present value of future cash flows. More importantly, for a given interest rate increase, the more distant the future cash flow, the bigger the drop in its present value. This implies that tech stocks still could be sensitive to interest rates if they rely on more distant future earnings to support their current stock prices.

To sort out the net effect of changes in interest rates on tech stocks, the standard two-factor model that is well known in academic research is estimated. Specifically, individual stock returns are regressed against the market return and the percentage change in the 10-year Treasury bond yield, using daily data from 1980 to 1998. The results show that after controlling for systematic stock market movements, technology stocks in general are not sensitive to interest rate changes. Some non-technology stocks, however, such as financials and utilities, are quite sensitive to interest rate changes even after controlling for the market effect. The results suggest that increases in interest rates will drag down the broad stock market but are not expected to have any additional effects on tech stocks. That is, the effects of higher interest rates on tech stocks are expected to be about the same as those for the overall market.

Short answers on “new economy” stocks

To sum up this exercise, I will review the three questions and answers addressed in this Economic Letter.

Q: How dominant are technology companies?

A: Technology companies dominate in terms of market capitalization, but not in terms of tangible assets, employment, and sales.

Q: Do the valuations of technology companies make sense?

A: Tech stocks look very richly priced, especially in light of their demonstrated fundamentals. While there is no question that many technology companies are high achievers, the kind of future earnings growth expected by the market is still quite daunting.

Q: How sensitive are tech stocks to interest rate changes?

A: Research suggests that technology stocks are not sensitive to interest rate changes after controlling for systematic stock market movements.

With tech stocks accounting for a large share of the stock market, at sky-high valuations, any wealth effects stemming from the stock market can be largely attributable to technology companies. However, can monetary policy be used to bring down these high fliers? The answer appears to be “no”–not without bringing down the entire market. This suggests that targeting stock prices may not be an effective policy approach. Rather, targeting inflation with an eye on the wealth effect seems to be the better policy to pursue.

Simon Kwan
Senior Economist

Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System. This publication is edited by Anita Todd and Karen Barnes. Permission to reprint portions of articles or whole articles must be obtained in writing. Please send editorial comments and requests for reprint permission to research.library@sf.frb.org