2001-17

June 1, 2001

The Stock Market: What a Difference a Year Makes


Stock prices have been on an extraordinary ride in the last two and a half years, soaring to phenomenal heights and then plunging at head-spinning rates. At their peaks, the NASDAQ composite and the S&P 500 were up 579% and 233%, respectively, compared to the beginning of 1995 (see Figure 1). Since the peaks, the NASDAQ composite has fallen more than 50% and the S&P 500 has dropped 15%. Nevertheless, as of the third week of May 2001, both the NASDAQ and the S&P 500 are up 196% and 181%, respectively, since the beginning of 1995. This Economic Letter takes a closer look at the stock market: First, what happened to the NASDAQ? Second, do current stock price valuations make sense?

The gyrating NASDAQ

Although almost 6,000 stocks are traded on the NASDAQ, that market is dominated by technology companies. At the peak on March 10, 2000, the 20 largest domestic companies in the NASDAQ were in the technology sector, and together they accounted for over one-third of the total capitalization of that market. Quite amazingly, 6 of these top 20 firms were losing money as of the fourth quarter of 1999. Today the top 20 stocks in the NASDAQ include a few firms from non-tech sectors, but tech stocks still account for the lion’s share. So, in order to understand the NASDAQ, one must focus on technology stocks.

Due to advances in information technology (IT), and a little bit of Y2K fear, business investment in IT equipment and software was growing at a much faster clip than non-IT business investment, particularly during the last five years. Specifically, between 1995 and 2000, capital investment in IT equipment grew at a 22.7% compound annual rate, after adjusting for inflation, compared to a 5.1% compound annual rate for non-IT related capital spending.

As a result, revenues and earnings at technology companies in general grew very rapidly, particularly in 1999. Between 1998 and 2000, total earnings in a constant sample of 356 S&P 500 companies grew at a compound annual rate of 21%, up from what was already very fast growth by historical standards. Total earnings for the technology sector, based on a constant sample of 523 publicly traded tech companies, grew at a staggering 38% annual rate during the same time period, almost twice as fast as the S&P 500 companies.

With supranormal growth among technology companies, the increasing adoption of electronic commerce, and increasing evidence of productivity gains, widespread optimism prevailed in the stock market. In other words, we had what can be called a “virtuous cycle.” Investors were very willing to provide financing to technology companies, especially for technological innovations (from dot-coms to telecommunications-related technologies). The amount of initial public offerings of stocks shot up in 1999 and peaked in the first quarter of 2000. With abundant financing, technology start-ups and telecommunications providers invested heavily in IT. Moreover, traditional companies followed suit as a defense strategy. And this investment cycle fed on itself. Note that during this cycle, a fair amount of IT investment was made with very little regard to consumer demand.

Good times usually don’t last forever, and so it was with this virtuous cycle. Before long, the “vicious cycle” began. With decreasing marginal returns on IT investment, the Y2K threat gone, and evidence of overcapacity in certain IT infrastructures, firms scaled back their investment in IT. Consequently, revenues and earnings growth among technology companies slowed in the second half of 2000. Unable to meet Wall Street expectations, tech stocks got hammered, particularly those with very high price-earnings ratios (P/E) that were associated with very high growth expectations.

Realizing the risks involved in investing in tech companies, investors changed course. They either demanded a high rate of return for providing capital to these companies, as evidenced by the widening credit spread in the bond market, or they cut off financing altogether, as evidenced by the drying up of equity IPOs. Without infusions of capital, some firms had to scale back their investment in IT. Other firms facing cash flow problems were forced into bankruptcy, and they, in turn, liquidated their IT equipment into the secondary market, further damping the demand in the primary market. This led to a downward spiral in tech stock prices.

According to this analysis, to the extent that the current slowdown in the tech sector was caused by overinvesting in IT, lowering interest rates is unlikely to revive business investment in IT quickly. It will take time for the existing IT capacity to be absorbed at the consumer level before firms start to invest in IT again.

Do current stock valuations make sense?

Understanding what contributed to the rise and fall of stock prices in the NASDAQ is relatively straightforward. A much harder question is: Do the current valuations make sense—especially for tech stocks whose prices have fallen so much? To answer this, it is useful to look first at the relative valuation of tech stocks to non-tech stocks before looking at the valuation of the broad stock market in general.

Figure 2 shows the weighted average P/E ratio of all the technology companies in the S&P 500, the weighted average P/E ratio of all the non-technology growth companies in the S&P Barra Growth 500, and the P/E ratio of the S&P 500 Index. In calculating the average P/E ratio for technology companies and growth companies, only stocks that have positive earnings are used, because the P/E ratio is not meaningful for companies that have no earnings. Before 1996, the average P/E ratio for tech stocks closely tracked the market average, measured by the S&P 500, while growth stocks were selling at a premium due to their growth potential. In 1997, tech stocks started to command a premium and on average were selling at a multiple that was similar to non-tech growth stocks. Beginning in mid-1998, the run-up in tech stocks pushed their P/E ratio to more than double that of non-tech growth stocks at the market peak. Currently, the P/E ratio of tech stocks is below the P/E ratio of non-tech growth stocks but is still higher than that of the S&P 500.

The reason that the market awards a higher P/E ratio to both technology companies and non-tech growth firms is because of their supranormal growth potential. To delve deeper into the quality of earnings between tech and non-tech growth companies, it is instructive to compare their earnings growth and earnings volatility.

The weighted average five-year compound annual growth rate of earnings for technology and non-technology growth companies is calculated for each of the past six years. The five-year growth rate in tech earnings, averaging 42% between 1995 and 2000, is significantly faster than that of non-tech growth companies, which averaged only 17% during the same time period. Thus, other things being equal, tech stocks should command a higher P/E than non-tech growth stocks.

But other things are not equal. Between 1995 and 2000, the weighted average five-year earnings volatility for technology companies is about 50% higher than the earnings volatility for non-tech growth companies, although the gap seems to be narrowing over time. Taken together, the supranormal growth in tech earnings is associated with higher risk, as we have been painfully witnessing in the current downturn.

Thus, relative to the S&P 500 and non-tech growth stocks, the current level of tech stock prices do not seem to be out of line. However, on an absolute basis, the current P/E ratios for the S&P 500, the techs, and the non-tech growth companies are all still quite high compared to the historical averages. So, the more important question is: Does the valuation of the market as a whole look reasonable?

To answer this question, Figure 3 depicts the S&P 500 earnings-price ratio (E/P) and the real AAA-rated corporate bond yield. The E/P ratio is simply the inverse of the P/E ratio, and it measures the earnings yield from holding stocks; the benchmark for comparison in this case is the inflation-adjusted yield from holding high-grade corporate bonds. Before 1997, the S&P 500 earnings yield had almost always been higher than the real corporate bond yield, on average by 160 basis points over this period. This reflects the risk premium from holding stocks over bonds. However, since 1998, the run-up in stock prices pushed the earnings yield well below the real AAA bond yield, suggesting that there was a risk discount rather than a risk premium for holding stocks. Currently, the earnings yield and the bond yield are about the same. One may argue that with corporate earnings growing so fast and the inflation rate so low, the equity risk premium ought to come down—but it’s still somewhat difficult to imagine that it will come all the way down to zero. Hence, to justify the current valuation, corporate profits must be able to grow at a fairly rapid rate in the future, which remains the biggest uncertainty given current economic conditions.

Short answers

This Economic Letter began with two questions. First, what happened to the NASDAQ? The sharp rise and fall of the NASDAQ composite has been mainly a tech phenomenon. The rapid ascent in tech stock prices was fueled by “exuberant” growth expectations. As the vastly inflated expectations ultimately could not be met, tech stock prices plummeted.

The second question was: Do current stock price valuations make sense? At the current level, tech stock prices do not look unreasonable, relative to the broad stock market and in particular to non-tech growth stocks. However, in terms of the broad stock market, it is somewhat difficult to judge whether future earnings growth will be fast enough to justify the current valuation.

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 Sam Zuckerman and Anita Todd. Permission to reprint must be obtained in writing.

Please send editorial comments and requests for reprint permission to

Research Library
Attn: Research publications, MS 1140
Federal Reserve Bank of San Francisco
P.O. Box 7702
San Francisco, CA 94120