Blowing Bubbles (part II)

Stock market bubbles have three other recurrent features. The first is the role of what is sometimes referred to as asymmetric information. Insiders - those concerned with the management of bubble companies - know much more than the outsiders, whom the insiders want to part from their money. Such asymmetries always exist in business, of course, but in a bubble the insiders exploit them fraudulently. The second theme is the role of crossborder capital flows. Bubbles are more likely to occur when capital flows freely from country to country. The seasoned speculator, based in a major financial centre, may lack the inside knowledge of the true insider. But he is much more likely to get his timing right - buying early and selling before the bubble bursts - than the naive first-time investor. In a bubble, in other words, not everyone is irrational; or, at least, some of the exuberant are less irrational than others. Finally, and most importantly, without easy credit creation a true bubble cannot occur. That is why so many bubbles have their origins in the sins of omission or
commission of central banks.

Nothing illustrates more clearly how hard human beings find it to learn from history than the repetitive history of stock market bubbles. Consider how readers of the magazine Business Week saw the world at two moments in time, separated by just twenty years. On 13 August 1979 , the front cover featured a crumpled share certificate in the shape of a crashed paper dart under the headline: The Death of Equities: How inflation is destroying the stock market'. Readers were left in no doubt about the magnitude of the crisis:


The masses long ago switched from stocks to investments having higher yields and more protection from inflation. Now the pension funds - the market's last hope - have won permission to quit stocks and bonds for real estate, futures, gold, and even diamonds. The death of equities looks like an almost permanent condition.

On that day, the Dow Jones Industrial Average, the longestrunning American stock market index, closed at 875, barely changed from its level ten years before, and nearly 17 per cent below its peak of 1052 in January 1973. Pessimism after a decade and half of disappointment was understandable. Yet, far from expiring, US equities were just a few years away from one of the great bull runs of modern times. Having touched bottom in August 1982 (777), the Dow proceeded to more than treble in the space of just five years, reaching a record high of 2,700 in the summer of 1987. After a short, sharp sell-off in October 1987, the index resumed its upward rise. After 1995, the pace of its ascent even quickened. On 27 September 1999, it closed at just under 10,395, meaning that the average price of a major US corporation had risen nearly twelve-fold in just twenty years. On that day, readers of Business Week read with excitement that:

Conditions don't have to get a lot better to justify Dow 36,000, say James K. Glassman and Kevin A. Hassett in Dow 36,000: The New Strategy for Profiting From the Coming Rise in the Stock Market. They
argue that the market already merits 36K, and that stock prices will advance toward that target over the next 3 to 5 years as investors come to that conclusion, too . . . The market - even at a price-to-earnings ratio of 30* - is a steal. By their estimates, a 'perfectly reasonable price' for the market... is 100 times earnings.

[NB* A ratio of stock prices divided by earnings including dividends. The long-run average (since 1871) of the price-earnings ratio in the United States is 15.5. Its maximum was reached in 1999: 32.6. It currently stands at 18.6 (figures for the Standard and Poor's 500 index, as extended back in time by Global Financial Data).]

This article was published less than four months before the collapse of the dot-com bubble, which had been based on exaggerated expectations about the future earnings of technology companies. By October 2002 the Dow was down to 7,286, a level not seen since late 1997. At the time of writing (April 2008), it is still trading at one third of the level Glassman and Hassett predicted.

The performance of the American stock market is perhaps best measured by comparing the total returns on stocks, assuming the reinvestment of all dividends, with the total returns on other financial assets such as government bonds and commercial or Treasury bills, the last of which can be taken as a proxy for any short-term instrument like a money market fund or a demand deposit at a bank. The start date, 1964, is the year of the author's birth. It will immediately be apparent that if my parents had been able to invest even a modest sum in the US stock market at that date, and to continue reinvesting the dividends they earned each year, they would have been able to increase their initial investment by a factor of nearly seventy by 2007. For example, $10,000 would have become $700,000. The alternatives of bonds or bills would have done less well. A US bond fund would have gone up by a factor of under 23; a portfolio of bills by a factor of just 12. Needless to say, such figures must be adjusted downwards to take account of the cost of living, which has risen by a factor of nearly seven in my lifetime. In real terms, stocks increased by a factor of 10.3; bonds by a factor of 3.4; bills by a factor of 1.8. Had my parents made the mistake of simply buying $10,000 in dollar
bills in 1964, the real value of their son's nest egg would have declined in real terms by 85 per cent.

No stock market has out-performed the American over the long run. One estimate of long-term real stock market returns showed an average return for the US market of 4.73 per cent per year between the 1920s and the 1990s. Sweden came next (3.71), followed by Switzerland (3.03), with Britain barely in the top ten on 2.28 per cent. Six out of the twenty-seven markets studied suffered at least one major interruption, usually as a result of war or revolution. Ten markets suffered negative long-term real returns, of which the worst were Venezuela, Peru, Colombia and, at the very bottom, Argentina (-5.36 per cent). 'Stocks for the long run' is very far from being a universally applicable nostrum. It nevertheless remains true that, in most countries for which long-run data are available, stocks have out-performed bonds - by a factor of roughly five over the twentieth century. This can scarcely surprise us. Bonds, are no more than promises by governments to pay interest and ultimately repay principal over a specified period of time. Either through default or through currency depreciation, many governments have failed to honour those promises. By contrast, a share is a portion of the capital of a profit-making corporation. If the company succeeds in its undertakings, there will not only be dividends, but also a significant probability of capital appreciation. There are of course risks, too. The returns on stocks are less predictable and more volatile than the returns on bonds and bills. There is a significantly higher probability that the average corporation will go bankrupt and cease to exist than that the average sovereign state will disappear. In the event of a corporate bankruptcy, the holders of bonds and other forms of debt will be satisfied first; the equity holders may end up with nothing. For these reasons, economists see the superior returns on stocks as capturing an 'equity risk premium' - though clearly in some cases this has been a risk well worth taking.