American Economic Association Chartists, Fundamentalists, and Trading in the Foreign Exchange Market Author(s): Jeffrey A. Frankel and Kenneth A. Froot Source: The American Economic Review, Vol. 80, No. 2, Papers and Proceedings of the Hundred and Second Annual Meeting of the American Economic Association (May, 1990), pp. 181-185 Published by: American Economic Association Stable URL: http://www.jstor.org/stable/2006566 Accessed: 15-06-2016 19:28 UTC Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://about.jstor.org/terms JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. American Economic Association is collaborating with JSTOR to digitize, preserve and extend access to The American Economic Review This content downloaded from 128.103.149.52 on Wed, 15 Jun 2016 19:28:57 UTC All use subject to http://about.jstor.org/terms THE RATIONALITY OF THE FOREIGN EXCHANGE RATEt Chartists, Fundamentalists, and Trading in the Foreign Exchange Market By JEFFREY A. FRANKEL AND KENNETH A. FROOT* The overshooting theory of exchange rates Index, Mwrch 19710X 10 160 seems ideally designed to explain some im- portant aspects of the movement of the dol- 8 lar in recent years. Over the period 1981-84, 140 for example, when real interest rates in the 6 United States rose above those of its trading 120 partners (presumably due to shifts in the monetary/fiscal policy mix), the dollar ap- 2 preciated strongly. It was the higher rates of 100 + - return that made U.S. assets more attractive 0 - to international investors and caused the Lm -T R*W Al80 dollar to appreciate. The overshooting the- 2 ~i ory would say that, as of 1984 for example, 4 t60 1 1 ii E t l i iiiiitiii6 the value of the dollar was so far above its 1973 1975 1977 1979 1981 1983 1985 1987 1989 long-run equilibrium that expectations of fu- FIGURE 1. ture depreciation were sufficient to offset the THE DOLLAR AND REAL INTEREST RATES higher nominal interest rate in the minds of international investors. Figure 1 shows the Source: Peter Hooper and Catherine Mann, Federal correlation of the real interest differential Reserve Board. with the real value of the dollar, since ex- Notes: Quarterly data. The CPI-adjusted dollar is a weighted-average index of the exchange value of the qhange rates began to float in 1973. dollar against the currencies of the foreign G-10 coun- tries plus Switzerland, where nominal exchange rates I. Bubble Episodes are multiplied by relative levels of CPIs. Weights are proportional to each foreign country's share in world At times, the path of the dollar has de- exports plus imports from 1978 through 1983. The long-term real interest differential is the U.S. rate minus parted from what would be expected on the the weighted average of foreign-country rates. basis of macroeconomic fundamentals. The most dramatic episode is the period from June 1984 to February 1985. The dollar ap- preciated another 20 percent over this inter- able factors that are suggested in standard val, even though the real interest differential macroeconomic models (money growth rates, had already begun to fall. The other observ- real growth rates, the trade deficit) at this time were also moving in the wrong direc- tion to explain the dollar rise. tDiscussants: Michael Gavin, Columbia University; It is now widely accepted that standard Karen K. Lewis, New York University; Kenneth D. observable macroeconomic variables are not West, University of Wisconsin-Madison. capable of explaining, much less predicting *Department of Economics, University of California, ex ante, the majority of short-term changes Berkeley, CA 94720, and Graduate School of Business, in the exchange rate. But economists divide Harvard University, Boston MA 02163, respectively. We thank MMS International for data, and Lu Cordova into two camps on what this means. One and Joseph Mullally for valuable research assistance. view is that the unexplained short-term 181 This content downloaded from 128.103.149.52 on Wed, 15 Jun 2016 19:28:57 UTC All use subject to http://about.jstor.org/terms 182 AEA PAPERS AND PROCEEDINGS MA Y 1990 changes must be rational revisions in the heterogeneous expectations. For one thing, market's perception of the equilibrium ex- surveys of the forecasts of participants in the change rate due to shifts in "tastes and foreign exchange market show wide disper- technologies," even if the shifts are not ob- sion at any point in time. One, conducted by servable to macroeconomists in the form of the Financial Report (affiliated with the standard measurable fundamentals. A major Economist), reports a high-low range of 6- difficulty with this interpretation is that it is month forecasts that averages 15.2 percent. difficult to believe that there could have been Data in a survey conducted by MMS Inter- an increase in the world demand for U.S. national show a dispersion of opinion (as goods (or in U.S. productivity) sufficient to measured by the standard deviation across increase the equilibrium real exchange rate respondents) at the 1-month horizon that by more than 20 percent over a 9-month averaged 2.2 percent for the yen/dollar rate. period, and that such a shift would then be The dispersion was slightly higher for the reversed over the subsequent 9 months. mark, pound, and Swiss franc rates. This brings us to the second view: that the appreciation may have been an example of a II. Trading in the Foreign Exchange Market speculative bubble-that it was not deter- mined by fundamentals, but rather was the The tremendous volume of foreign ex- outcome of self-confirming market expecta- change trading is another piece of evidence tions. The dollar in 1985 "overshot the over- that reinforces the idea of heterogeneous ex- shooting equilibrium." Some have suggested pectations, since it takes differences among that the appreciation of 1988-89, on a market participants to explain why they smaller scale, may also have been of this trade. The Federal Reserve Bank of New nature. York has released its 3-yearly count of trans- There exist elegant theories of rational actions in the U.S. foreign exchange market. speculative bubbles, in which all participants It showed that in April 1989, foreign ex- know the correct model. Some observers have change trading (adjusted for double-count- suggested that 1984-85 may be best de- ing) totaled $128.9 billion a day, an increase scribed as a bubble that was not character- of 120 percent from March 1986. Simultane- ized by rational expectations.' We have sug- ous counts in London and Tokyo reported gested earlier that such episodes may best be $187 billion and $115 billion a day, respec- described by models of bubbles in which tively. Thus the worldwide total is over $430 market participants do not agree on the billion of foreign exchange trading a day. model for forecasting the exchange rate (see Interestingly, the banks in the New York our forthcoming paper). Federal Reserve Bank census reported that While the conventional approach in the only 4.9 percent of their trading was with a literature, theoretical as well as empirical, is nonfinancial firm, and the nonbanks only 4.4 to assume that there is such a thing as " the" percent; in other words, 95 percent of the market expectation of the future exchange trading takes place among the banks and rate, there is evidence that investors have other financial firms, rather than with cus- tomers such as importers and exporters. Clearly, trading among themselves is a major 1Paul Krugman (1985) was one of the first to suggest economic activity for banks. that the market did not appear to realize the extent to What is the importance of trading volume which the appreciation of the dollar was not sustain- (beyond motivating the importance of het- able. Charles Engel and James Hamilton (forthcoming) find that long-term swings are a general characteristic of erogeneous expectations)? There are three exchange rates, and that they are not adequately re- possible hypotheses, with regard to implica- flected in the forward market. Such findings of pre- tions for movements in the market price. dictable excess returns are standardly interpreted as risk 1) The higher the liquidity or "depth" of the premiums. But evidence from survey data on expecta- tions of market participants suggests that the prediction markets, the more efficiently is news regard- errors of the forward market are not due to risk premi- ing economic fundamentals processed and ums (see our 1989 paper). the smaller is "unnecessary volatility" in the This content downloaded from 128.103.149.52 on Wed, 15 Jun 2016 19:28:57 UTC All use subject to http://about.jstor.org/terms VOL. 80 NO. 2 RA TIONALITY OF THE FOREIGN EXCHANGE RA TE 183 exchange rate. 2) The foreign exchange mar- 111. The Rising Importance of Chartists ket is already perfectly efficient, so that trad- ing volume is irrelevant to price movements We now turn to the question of how the and therefore uninteresting. 3) Much trading existence of different forecasting techniques is based on "noise" rather than "news," and might lead to "excess volatility." It has long leads to excessive volatility. been remarked that if there exist traders who Choosing convincingly among these three tend to forecast by extrapolating recent hypotheses may be too large a task to ac- trends (i.e., who have "bandwagon expecta- complish here.