Date created: 28 November 1996
Last modified: 18 November 1997

Maintained by: John Quiggin
John Quiggin

Will rational bubbles fall on the infallible markets

Australian Financial Review

24 Jun

The recent rise in long-term market interest rates and the government's response raise questions that go to the heart of the case for financial deregulation. Can financial markets be systematically wrong ? That is, can they price assets in a way that is inconsistent with fundamental considerations apparent to any observer. If the answer to this question is "Yes", it is likely that government can improve the allocation of resources by imposing quantitative and qualitative controls on financial markets.

The question is surprisingly difficult to answer. Prices often fluctuate wildly in the absence of any obvious change in fundamentals. In the aftermath of a market crash, it is easy to point to assets given a high value on the basis of arguments that seem, in retrospect, to be utterly implausible. History affords some stunning cases. For example, during the Dutch tulip mania of the seventeenth century, individual tulip bulbs changed hands for tens of thousands of dollars. In the South Sea Bubble of the eighteenth Century, large sums of money were raised on a prospectus that promised 'an enterprise of great value, but nobody to know what it is.' The same patterns can be seen at work in our own century. In every stockmarket boom, entrepreneurs rediscover the fact that, in a rising market, leverage is the key to massive profits, and are hailed as financial geniuses for their discovery. Investors flock to follow the coattails of these magicians and rediscover for themselves the other side of the equation - a leveraged portfolio is wiped out by the first drop in prices.

All of this would seem to support the assumptions that governed financial policy through the fifties and sixties - that speculative markets are inherently irrational and have little contact with reality. In Keynes' words, it would be a foolish government that entrusted the management of investment to the players in a casino.

Despite all of this apparent evidence of irrationality, the majority of economists studying financial markets adhere to one or other form of the efficient markets hypothesis - the claim that financial markets price assets in line with the expected present value of future earnings based on all information available at the time. Critics of the profession would be tempted to laugh this off, as evidence of dogmatic adherence to theory in the face of reality. But financial markets generate huge volumes of data that can be subject to statistical analysis, and the results of this analysis have been surprisingly favorable to the efficient markets hypothesis.

In their first foray, beginning in the sixties, the efficient market theorists demolished the pretensions of technical analysts or 'chartists'. Beneath the elaborate vocabulary of trends, support levels, head-and-shoulders patterns and so forth lies the basic claim that the past behavior of stock prices can be used to predict their future movements. Statistical analysis demonstrated, on the contrary, that nothing in the past behavior of prices (except of course the current price itself) is of any value in forecasting prices. This is the so-called weak efficient markets hypothesis. (Cynics might argue that the inefficiency of financial markets is demonstrated by the fact that so many chartists still manage to make a living there, decades after their uselessness has been demonstrated by efficient market theorists).

The really interesting issue however, is the so-called strong efficient markets hypothesis - that the current price incorporates all available information. This does not mean that markets can never be mistaken. Market participants may reasonably expect, for example, that inflation is going to rise, but an unexpected change in policy or an unanticipated shock to the real economy may prove them wrong in the event. But the strong efficient markets hypothesis denies the possibility of a situation where financial market prices imply a rise in the rate of inflation while any reasonable assessment of the situation implies that inflation will remain low.

Under appropriate assumptions, market booms and crashes are consistent with the strong efficient markets hypothesis. Many papers have argued this case in relation to the crash of 1987. In an excess of zeal, one Chicago economist even published an article purporting to prove that the Dutch tulip mania showed efficient markets at work.

As these examples show, the strong efficient markets hypothesis is much harder to test than the weak version, and tests in the seventies were largely inconclusive. During the eighties, however, the strong hypothesis came under sustained empirical and theoretical challenge. The main empirical challenge came from the observation that stock market prices were much more volatile than the theory predicted. The theoretical challenge came from the development of 'rational bubble' models, in which prices diverged from fundamental value even though all market participants are rational.

The theoretical status of rational bubble models remains unclear. Given market participants who integrate information perfectly and always maximise profits, it is not clear whether rational bubbles can emerge. A more robust conclusion is that, in highly speculative markets, a little bit of irrationality, or a small deviation from profit-maximisation, can go a long way in separating prices from fundamental values.

Consider for example the present situation in the bond market. The fundamentals in this market are determined by the willingness of individuals, corporations and governments to enter into long-term debt contracts stretching into next century. The profitability of such contracts will depend on the rate of inflation and the real return to capital over the nest ten years. Yet the vast majority of market participants have no interest in any of these issues. (The volume of speculative transactions in Australian dollar financial markets exceeds the volume of underlying real transactions by a factor of between twenty to one and a hundred to one.) Nearly all bond market traders are concerned purely with the bond price that will prevail next week. By the time the long-term outcome of the bond contracts is known, their current positions will have been liquidated and many of them will have moved on to new employers and new markets. For these traders the cost of being wrong is small, provided they are wrong with the crowd. The efficient market theory maintains that this vast mass of short-term speculators is forced to focus on fundamental values by the small minority willing, if necessary, to take positions in the ten-year bond market and hold them until the redemption date in 2004. The alternative view is that, in situations where the volume of speculative trade greatly outweighs the underlying real market.

The potential for self-fulfilling irrationality becomes apparent when we consider the government's response to the current decline in bond prices. The government is a large issuer of bonds - on Budget forecasts, its total outstanding debt will approach $100 bn in 1995. A number of ministers have expressed the view that prices are much below what is justified on the fundamentals. If that view is genuinely held, the Treasury should now be instructed to buy large quantities of long-term debt, and replace it with short-term debt, not with the objective of stabilising the market, but simply to reduce the cost burden of public debt interest. But it seems likely that the power of the belief that the market cannot be wrong will lead the government to confine itself to verbal criticism of the foolishness of the market, rather than putting our money where its mouth is.

John Quiggin is Professor of Economics at James Cook University and author of Great Expectations: Microeconomic reform and Australia, published by Allen & Unwin.

Read more articles from John Quiggin's home page

Go to John Quiggin's Weblog

Read more news articles from 1994 and earlier