Can Speculative Trading Magnify Financial Market Co-movement?
Global financial markets tend to move together. For example, stock market movements across the globe are highly synchronized, economic data releases frequently have large spillover effects across borders, and episodes of financial turmoil often spread across countries that share no significant economic linkages. The degree of co-movement across markets often appears to be surprisingly large when compared with the strength of the underlying economic relationships. What can explain this seemingly “excessive” co-movement? This post, based on a recent research paper, argues that speculative trading may magnify financial market co-movement.
Intuitively, the price of a financial asset should be related to its fundamental value. For example, I am willing to pay more for a stock if I expect to receive high dividends in the future. In this case, the value of dividends would be one fundamental factor determining the price. One would then expect the degree to which prices of financial assets move together to be related to the correlation of their underlying fundamental factors. However, a number of empirical studies have argued that the co-movement between stock markets, bond markets, and commodity markets may be larger than that justified by underlying fundamental linkages. One aspect of this finding is the observed large cross-border impact of economic data releases. For example, a recent IMF working paper documents significant spillover effects of credit rating announcements across countries.
If fundamental factors are unable to account fully for the observed magnitude of financial market co-movement, how could speculation—the buying and selling of financial assets in anticipation of future price movements—help to explain co-movement? The key point to understand is that in contrast to long-term investors, speculators need to take into account the expectations of others. Speculators buy an asset today in the hope that they will be able to re-sell it later at a higher price. Thus, when deciding whether to purchase the asset today, speculators need to form expectations about tomorrow’s price. But tomorrow’s price will in turn depend on expectations tomorrow—averaged across investors—about prices two periods ahead. It follows then that today’s price, which depends on speculators’ demand for the asset, will be determined by the market’s expectation today of the market’s expectation tomorrow, of the market’s expectation two days ahead, and so on, of the fundamental value of the asset. In short, prices are determined by “higher order expectations,” averaged across all market participants, of “fundamentals.”
The British economist John Maynard Keynes described this situation in his well-known “beauty contest” analogy:
Professional investment may be likened to those newspaper competitions in which the competitors have to pick out the six prettiest faces from a hundred photographs, the prize being awarded to the competitor whose choice most nearly corresponds to the average preferences of the competitors as a whole; so that each competitor has to pick, not those faces which he himself finds prettiest, but those which he thinks likeliest to catch the fancy of the other competitors, all of whom are looking at the problem from the same point of view. (The General Theory of Employment, Interest and Money, London: Macmillan, 1936, p. 156)
Similarly, speculators don’t care which financial asset has the highest value in the long run, but instead care which financial asset other investors think has the highest value—because it is these assets which will go up in price in the short run. Speculators are therefore forecasting the forecasts of others, a problem first analyzed rigorously by Robert Townsend. The consequence is that prices are determined not by average expectations of fundamental asset values, but instead by average expectations of future average expectations, and so forth, of asset values. A key insight—see, for example, an article by Franklin Allen, Stephen Morris, and Hyun-Song Shin that appeared in the Review of Financial Studies—is that such average higher order expectations of fundamentals generally differ from average expectations of the fundamentals.
To see this, consider a long-term investor who evaluates the sustainability of government finances using two sources of information: a public forecast of the fiscal position that has just been published by a ratings agency, and his own private assessment. The statistically optimal forecast takes into account both the investor’s own information and the rating agency’s projection—in particular, it is a weighted average of the two, with weights that correspond to the degree of confidence the investor has in the two sources of information. Now consider a speculator who forecasts tomorrow’s price for government bonds using the same information. The speculator knows that everybody in the market has observed the projection published by the ratings agency and will thus react to it. In contrast, his own personal assessment—while possibly just as well informed as the rating agency’s forecast—is known only to him. Therefore, while both forecasts may have a similar value for assessing the government’s fiscal situation, the public forecast is more useful for predicting the market reaction, and thus tomorrow’s prices. It follows that when investors speculate on future price developments, they will place a larger weight on public news of a ratings downgrade and a smaller weight on their own private information than would be the case if they were forecasting fundamentals—in this example, the government’s fiscal position—directly. Consequently, prices will “overreact” to the ratings agency’s publication: the announcement will have a larger impact on prices than on the market’s assessment of the country’s fiscal position.
To see the implications of speculation for financial market co-movement, consider government bonds in two countries, A and B, and suppose that the countries share some economic linkages. For example, business cycles could be connected through international trade, which would imply a relationship for the dynamics of governments’ fiscal positions as well. Because of the assumed underlying fundamental link, a credit rating announcement for country A also provides information about country B’s fiscal situation—even though it is obviously a less precise signal for forecasting developments in country B. The same logic of speculation and higher order expectations applies: because the credit rating news is observed by all market participants, individual speculators will find it more useful than their own private assessment in forecasting tomorrow’s price for the government bonds of country B. Consequently, the market as a whole will place a larger weight on the public information “credit rating downgrade,” and the prices of government bonds in both A and B will “overreact” to the announcement. Thus, speculation can potentially magnify asset price co-movement and information spillovers across borders.
To conclude, economic theory suggests that speculation can magnify financial market co-movement. The key conditions for this amplification effect are that (1) there is some (possibly very weak) fundamental link across markets, which makes news releases in one market useful for forecasting fundamentals in the second market, and (2) speculators care about other investors’ assessment of asset values. What is the magnitude of this “overreaction” to public information? Because theoretical models of these effects are necessarily very stylized, bringing the model to the data is challenging. One implication of the theory, however, is that these effects should be larger in markets that are dominated by traders with short investment horizons. Furthermore, to the extent that the importance of speculation increases over time relative to the prevalence of long-term investment strategies, one would expect the magnitude of co-movement to increase.
The views expressed in this blog are those of the author(s) and do not necessarily reflect the position of the Federal Reserve Bank of New York or the Federal Reserve System. Any errors or omissions are the responsibility of the author(s).