The great technology bubble is purged
By Jeremy Siegel
Published: October 28 2002 20:23 | Last Updated: October 28 2002 20:23
One of the favourite charts of the bears on Wall Street is a historical graph of the price/earnings ratio of the stock market. Historical data permit us to calculate this all-important yardstick all the way back to 1871 for the Standard & Poor's 500 index, or a similar group of high-capitalisation stocks.
Until the past decade, the ratio fluctuated between the high single digits and the low 20s, with a mean of about 15. But during the past bull market and the recession of 2001 the ratio soared to 47 and, even with stock prices down about 40 per cent from their peak, the ratio stands today at 29 times the average reported earnings of the past 12 months of $29.94.
Valuations are higher than usual at the end of a bear market but close to those justified by economic factors
These data make it easy to convince investors that the market may have to fall substantially more before this bear market ends. But there are two counters to this pessimism.
First, p/e ratios always spike higher in recessions, since earnings are temporarily depressed. The latest recession took a severe toll on profits, particularly in the technology sector, where earnings have slumped. Over the next 12 months, reported earnings are expected to rise - using a conservative methodology - to $37.79.
Second, there is no reason why the current p/e ratio should equal the historical average, unless one believes that the economy and the financial markets are the same now as over the past 130 years. But this is clearly not the case. The sharp drop in transaction costs makes it quite easy for investors to diversify their portfolio completely and thereby reduce their risk in equities, if they choose to do so.
This action would have been quite costly and practically impossible to achieve throughout most of the 19th and early 20th centuries. Throughout history, liquid assets have always commanded higher prices than less liquid assets. Given the explosive increase in share trading volume, it cannot be denied that stocks have become much cheaper to trade than in the past.
Furthermore, extreme events that have been so detrimental in the past, such as banking panics, the Great Depression and double-digit inflation, are extremely unlikely in the future as modern central banks can both keep the financial system solvent and inflation under control.
Last, as investors become better educated about superior long-term returns on equity (which have averaged between 6.5 and 7 per cent a year after inflation in the US and are only a bit lower in the UK), shareholders should bid up stock prices to higher multiples than those that prevailed in the past. My research suggests low inflation, favourable taxes on equities and low transaction costs justify a p/e ratio in the low 20s for the S&P 500 index.
It is quite possible that from these higher levels of valuation, stockholders will receive a somewhat lower return than they have in the past. The 6.7 per cent real historical return on stocks is quite generous, given the relative long-term stability of stock returns, and is an adequate premium for investors who endure the short-term vicissitudes of the market. A forward-looking prospective real return of 5.5 per cent on equities is still about 3 percentage points above the return that is available from risk-free, inflation-indexed government bonds.
One should note that, although I have taken "reported earnings" as my yardstick, Standard and Poor's has just released estimates of "core earnings" of S&P 500 firms. Core earnings are a tough new standard that accurately reflect the profitability of a company's core businesses and should become the new standard for determining what its stock is worth. Over the past 12 months, these core earnings fell considerably below reported earnings owing primarily to the full expensing of options and the decline in pension portfolios backing defined-benefit plans.
However, if the return on stocks is a modest 7.8 per cent next year, S&P reports that pension costs will not be a negative factor in next year's earnings. Furthermore, with the decline in the equity market, stock options are becoming a far less popular form of employee compensation. Therefore, 2003 core earnings should be much closer to reported earnings than over the past 12 months.
All these factors suggest that the sharp decline in US stock prices has purged the excesses of the great technology bubble of 1999-2001. Valuations are higher than they usually are at the end of a bear market but they are close to those justified by favourable economic and market factors. From these levels, stocks should give investors long-term forward-looking returns that not only are near their historical levels but also substantially outstrip returns on fixed-income assets.
The writer is a professor of finance at the Wharton School of the University of Pennsylvania |
