E CONOMICS IS ABOUT supply and demand—just not in financial markets. A building block of asset-pricing theory is that the value of stocks and bonds is determined by their expected future payoffs rather than by ignorant trades. If an investor unthinkingly throws money at, say, shares in Apple, opportunistic short-sellers are supposed to line up to take the other side of the bet, keeping the share price anchored to where it ought to be, given Apple’s likely profits. Free money gets picked up and dumb money gets picked off. Markets are efficient, in that prices come to reflect genuine information about the future.
At this point your columnist may be in danger of provoking guffaws. It has been a bad year for the textbooks. Retail investors have driven up the prices of meme stocks such as GameStop and AMC. Cryptocurrencies, the fundamentals of which cannot easily be analysed, have soared too. Even America’s bond market is a puzzle: the ten-year Treasury yield is only 1.4% even though annual consumer-price inflation has reached 5.4%. So-called “technical” explanations for market movements—“where you put things that you can’t quite explain”, according to Jerome Powell, the chairman of the Federal Reserve—are in fashion. So it is apt that an emerging theory of how markets work says that even random financial flows may matter a great deal to asset prices.
In a recent working paper Xavier Gabaix of Harvard University and Ralph Koijen of the University of Chicago study how the aggregate value of America’s stockmarket responds to buying and selling. Researchers have studied flows before, typically finding noticeable effects as investors sell one stock and buy another. Messrs Gabaix and Koijen are interested in whether this finding scales up to move the market as a whole—a thesis that is consistent with the smaller-scale findings, but more provocative.