A few years ago three behavioural-finance researchers put a deceptively simple question* to more than 26,000 people—academic economists, ordinary households, retail investors, financial advisers and professional fund managers alike. Suppose a piece of good news about a company’s future earnings is already four weeks old. Should you expect its shares to earn higher returns from here?
Most academics said no: around 70% reckoned that month-old news was already in the price, and so told you nothing about the returns to come. Almost everyone else said yes—including about three-quarters of retail investors, two-thirds of financial advisers and, more startling, over half of professional fund managers.
The authors’ more disquieting finding was why. This muddling of a company’s fundamentals with the expectations already embedded in its price was not down to laziness, a shaky grasp of returns or some quiet faith in market inefficiency. Even when respondents were paid for accuracy, walked through the arithmetic and reminded that others had already traded on the news and moved the price, they held firm. The mistake runs deeper: a failure to see that, in equilibrium and absent mispricing, prices adjust exactly to offset the change in expected earnings, leaving no free lunch.
Nowhere is this confusion costlier, or better documented, than in foreign exchange. Ask an expert why a currency should fall and you will usually get a story about fundamentals: the economy is slowing, the central bank is cutting, the trade balance is deteriorating, the public finances are a mess. All may be true. All are also likely to be embedded in the spot rate already.
Currencies, unlike commodities**, are pure financial assets. A foreign-exchange position costs nothing to hold—it is simply an interest-bearing deposit—and shorting a currency is as easy as going long. That makes the spot rate of a major currency*** close to informationally complete: it represents what the market currently believes about the future. The forward rate merely adjusts the spot for interest-rate differentials; it carries very little information of its own.
This should change how you read market commentary. Annual outlooks, news wires, and major newspapers brim with plausible forecasts that extrapolate recent moves while ignoring what is already priced. You have undoubtedly seen some in recent days. Small wonder that Meese and Rogoff’s finding****, now more than four decades old, still stings: a naive random walk based on prevailing spot rates tends to beat structural forecasting models built from fundamentals*****, precisely because today’s rates have already done the work of incorporating the consensus, leaving only genuine surprises to move them.
Market commentary is mostly useful for telling you how participants are positioned, not what might happen next, because currencies, like shares, are repriced through revisions to the gap between expectations and reality—not by the level of the fundamentals themselves.
All of which should give today’s dollar bulls pause. Forecasters have spent the year rewriting their assumptions around American outperformance: heavy inflows into US equities, stickier inflation and a more hawkish Federal Reserve, set against gloom almost everywhere else—especially in Canada, Europe, and Japan. The dollar has duly climbed, and speculators have crowded in behind it. Yet by now that narrative is itself stale news. Near-perfection is priced in; disappointment is not. The dollar will not strengthen because America’s economy is strong—the market bought that long ago. It will move only on the gap between what is expected and what arrives. Those who mistake the fundamentals for the surprise are buying yesterday’s headline and calling it a forecast.
*’Mental Models of the Stock Market’, Andre, Schirmer, Wohlfart, 2025
**Commodities are different. Storage is expensive, shrinkage is a real problem, and shorting is difficult. So futures prices do contain genuine information about expected supply, demand and scarcity — information that isn’t in the spot price.
***Some controlled and thinly-traded currencies follow different rules, with differing demand and supply dynamics causing temporal distortions in spot rates and forward curves.
****’Empirical exchange rate models of the seventies: Do they fit out of sample?’, Meese, Rogoff, 1983
*****Note that neither is particularly effective. There’s little question that a mechanistic, layered, and tranched hedging programme outperforms forecast-driven approaches over time.