When the pandemic hit and spread in 2020, stock markets in the European Union, Japan, and the United States plummeted up to 30 percent. The implications of the virus for public health, the global economy, and numerous aspects of everyday life were uncertain, and dire. But to financial researchers, the extent of these market reactions was still perplexing. “For the stock market to decline by 30 percent only due to revised growth expectations, the shock to future dividends needs to be large and highly persistent,” wrote Chicago Booth’s Niels Gormsen and Ralph S. J. Koijen that fall. “It would be, for instance, inconsistent with a V-shaped recovery.”
Stock markets were being volatile, yes, but they were acting more volatile than could be explained by looking at the underlying fundamentals. And there have been other examples of extreme volatility, including the stock run-ups in video-game retailer GameStop and movie-theater chain AMC. The dynamics extend to different types of markets too. A single bitcoin has no future earnings or cash flow, and should therefore be worth nothing, according to traditional models. So why pay tens of thousands of dollars for one? And why should their value swing so wildly in response to a tweet from billionaire Elon Musk?
The conventional wisdom, embodied in the efficient-market hypothesis, holds that market prices reflect the fundamental value of the underlying asset. But increasingly, research is identifying another force as being important: investor demand that may or may not be informed. Chicago Booth’s Samuel Hartzmark and Boston College’s David H. Solomon, in a study of data from 1926 to 2020, find that the stock market tended to rise more on days with heavier dividend payouts, as investors took cash hitting their accounts and reinvested it in the market. (For more, read “Dividend payouts lead to stock-price bumps.”) “There’s no reason we should see anything like we do,” Hartzmark says, unless the conventional wisdom is wrong—or missing something.
Harvard’s Xavier Gabaix and Booth’s Ralph S. J. Koijen are among those who say that it is. And their inelastic markets hypothesis lays out an argument for why financial models need to include the long-ignored forces of supply and demand. They calculate that every $1 flowing into the market pushes up aggregate prices by $5, and that these forces also amplify volatility, explaining why stock returns are more than twice as volatile as fundamental information implies they should be.
At the heart of their argument is a new description of the stock market, which has been transformed over the past few decades by the rise of index funds and other large, slow-moving investors.
“What we’re suggesting is that a large fraction of the market is restricted by mandates, therefore not necessarily reacting to new information. And in some cases, investors may be having a hard time assessing expected returns, in which case they’re also not acting much on prices,” says Koijen. With so much money essentially sitting on the sidelines, prices are more sensitive to what trading does happen. “As a result, shocks to flows and investor demand have an outsize effect on prices, leading to volatile markets.”
Gabaix and Koijen, 2021
From an efficient market to an inelastic one
The stock market, it has long been argued, reflects all available information and is driven by what matters most. The collective wisdom of investors makes the market efficient, and ultimately the price of a stock or other asset reflects its fundamental value. There are certainly price run-ups, as of GameStop stock in early 2021, but fundamentals win out and the market is self-correcting.
This view has shaped the stock market, as well as the retirement plans for millions of people. Chicago Booth’s Eugene F. Fama, a 2013 Nobel laureate, published pioneering research in the 1960s that produced the efficient-market hypothesis. The insight led to the rise of index funds, which now hold $12.5 trillion in assets, according to Morningstar Direct. After all, if the market is efficient, it’s hard to beat the average, in which case it doesn’t make sense to pay high fees to a money manager whose handpicked portfolio is unlikely to generate better returns than an index that reflects the market.
The idea of efficient, rational markets has repeatedly been challenged in the past few decades, however, such as after the dot-com bust and during the 2008–09 global financial crisis. Behavioral economists including Booth’s Richard H. Thaler, also a Nobel laureate, say investors have predictable biases, and Fama and Thaler have often debated whether the market is rational, including in The Big Question experts roundtable series.
But while they debate why investors trade, what is the impact of the trading itself? Gabaix and Koijen argue that uninformed flows may have an impact on prices. And they say their theory respects the importance of fundamentals but adds a layer to the discussion by arguing that fund flows matter, too, and help to explain deviations of an otherwise rational market.
Who really drives stock prices
The stock market they describe is different from the market of the 1960s, transformed by the amount of money indexed and otherwise tied to benchmarks such as the S&P 500. The researchers estimate that a large portion of the market is now sitting in vehicles—including pension funds, endowments, and mutual funds—that are either tied to benchmarks or have committed to holding a certain percentage of stocks and bonds. Even if a fund isn’t formally tied to a benchmark, its manager may be compensated in part for performance relative to one and is likely to buy shares of the same underlying companies. These positions, the researchers argue, are often in large part passive, stable, and price insensitive—or inelastic. Even if news comes out that potentially affects the earnings of a company, the money there isn’t moving.
Much of, but not all of the money they point to is essentially restricted from trading by mandates. “Estimating expected returns at any point in time is very difficult,” Koijen explains. “Faced with this uncertainty, investors may respond little to movements in prices, as they are unsure how to interpret them. Such limited response can also lead to inelastic markets.” In terms of setting prices, that gives outsize influence to active traders such as hedge funds and even groups of day traders.
In 2011, as the European Union was trying to claw its way out of a sovereign debt crisis, Greece teetered on the edge of solvency. Its debt crisis was so severe that the country seriously flirted with leaving the eurozone and readopting the drachma. With such extreme measures on the table, investors fled, eschewing Greece’s high bond yields for the safety of German bonds.
A sizable spread developed between the yields on long-term bonds in Germany and those in the countries in Southern Europe. The yields indicated that investors had more confidence in Germany’s stability than in Greece’s. But bond investors also punished Italy and Portugal, which didn’t have the same fundamental problems.
Booth’s Koijen and Princeton’s Motohiro Yogo analyzed what was driving yield spreads. As all countries in the eurozone share a short-term interest rate, short-term rates obviously weren’t the issue. Comparing Germany with Greece, Italy, and Portugal, the researchers determine that macroeconomic variables accounted for 64 percent of the variance in spreads. Long-term debt levels accounted for another 14 percent. And demand had about the same weight. “Latent demand of European investors accounts for 13 percent, and latent demand of offshore financial centers accounts for 4 percent of the variation in the long-term yield spreads,” Koijen and Yogo write.
But when the researchers reconstructed what happened by country, separating the outcome for Greece from the fortunes of its neighbors, they saw investor demand at work. In Greece, fundamental problems weighed on bond yields. But in Italy and Portugal, “latent demand, which captures perceived rather than realized risk, accounts for the sharp increase in the credit spread,” the researchers write. “This finding is consistent with the narrative that Greece had a realized solvency problem, while investors perceived Italy and Portugal to be still solvent but vulnerable.” Koijen says the data indicate that the sell-off was in part driven by investors in Northern Europe.
This investor fear driven by perceived risk had real consequences. It caused a wider borrowing crisis throughout Southern Europe and forced the European Central Bank to act. The ECB cut interest rates and extended low-cost loan facilities to banks that might purchase these bonds. The ECB also purchased bonds directly.
In retrospect, if the ECB had known at the time that investors were panicking, was there an alternative? Could the crisis have played out differently? Not necessarily, Koijen says. “In ongoing research, we are developing a model that can be used for policy makers, but the current research is not there yet to speak to those interesting questions.”
Ralph S. J. Koijen and Motohiro Yogo, “Exchange Rates and Asset Prices in a Global Demand System,” Working paper, June 2020.
Gabaix and Koijen calculated the effects of fund flows on individual prices as well as on the market in aggregate. They used a variety of data sources—piecing together a picture using various public filings and reports containing information on holdings and flows in stock and bond markets, generally covering the years 1993 to 2020. Before posting their paper, they asked 102 academic researchers in economics and finance to predict what every $1 entering the stock market would do to prices. The prevailing view, held by just over half of respondents, was that it would have no price effect. Of those who said it would have some effect, only three people said it would have a multiplier effect greater than one—as in, every $1 entering the market would push up prices by $2.
But in the inelastic markets hypothesis, money that flows into the stock market leads to stronger price effects because there are essentially a set number of available shares, and many of those are not being actively traded. Pairing their theory with an empirical analysis, the researchers estimate that every $1 put into the market pushes up aggregate prices by $5.
The researchers say that while this may seem like a big multiplier, it’s roughly in line with what other researchers find at a micro level. For example, AQR’s Andrea Frazzini and Ronen Israel and Yale’s Tobias J. Moskowitz estimated the price impact one large trader had on prices.
Jean-Philippe Bouchaud, chairman of Capital Fund Management in Paris, this past summer released his own research, in which he writes that Gabaix and Koijen’s “rather awesome recent paper” validates what he has found over two decades. He calculates that the multiplier for dollars invested can be even higher for volatile stocks—or lower for companies for which a smaller fraction of the market cap is actively traded.
“The mystery of apparently random movements of the stock market, hard to link [to] fundamentals, is replaced by the more manageable problem of understanding the determinants of flows in elastic markets,” Gabaix and Koijen write. Their work, they continue, “might lead to a more concrete understanding of the origins of financial fluctuations across markets.”
Deconstructing market moves
This logic implies that it’s possible to trace fluctuations in markets back to who caused them, which is something Koijen and his colleagues have been doing. The idea of tracing movements back to investors isn’t new, but academics abandoned it decades ago, in part because the data weren’t easily available, or available at all.
Now it’s easier to obtain measurements of holdings and fund flows, and to compare those with price data. Koijen, New York University’s Robert J. Richmond, and Princeton’s Motohiro Yogo deconstructed fund flows in the roughly $50 trillion US stock market to determine what and who is moving prices. Day traders might like meme stocks, they write, while some pension and sovereign wealth funds have mandates to invest in sustainable, more environmentally friendly companies. Many hedge funds, meanwhile, look for arbitrage opportunities. (Read “Who is driving stock prices?” as well as “The investors with the most influence over companies’ market capitalization.”)
The researchers developed a framework that starts with a simple model they applied to data to identify the characteristics that determine demand and ultimately prices. They then divided investors into eight groups according to investor size and strategy—from huge, passive investment advisers to smaller, actively managed advisers and hedge funds. Then they imagined a world in which the assets of one of those groups flows to all other institutional investors.
“For example, we ask by what percent does the average stock price move if we take BlackRock’s assets and redistribute these assets to all other institutional investors,” they write. “Because BlackRock has certain portfolio tilts, this experiment would lower the stock price along characteristics that BlackRock favors.”
Performing this portfolio thought experiment leads to many observations, among them that certain small, active investors have the largest influence on valuations. Hedge funds hold less than 5 percent of the equity market, Koijen, Richmond, and Yogo find—and yet, controlling for size, they are the most influential players in the market, more so than far larger pension funds and insurance companies.
“At the other end of the spectrum, we find that passive investment advisors (both small and large) and long-term investors have a relatively small impact on valuations,” the researchers write. That’s in line with the inelastic markets hypothesis that when many players are sitting on the sidelines, the ones on the field are moving prices and most responsible for fluctuations as well as outright volatility.
The international picture
With international holdings data, Koijen and Yogo performed a similar decomposition exercise, applying the idea to international stock, bond, and currency markets. “Global investors hold financial assets across many countries and have exchange rate exposure not only through short-term debt but also long-term debt and equity,” they write. “The portfolio decisions of these investors across countries and asset classes are important for exchange rates, long-term yields, and stock prices.”
While they’re making a case that demand matters, they’re not saying it’s the only thing that matters. Many other forces can move markets, including government and monetary policies, volatility, sovereign debt ratings, and macroeconomic conditions. To tease apart the various forces at work, they developed a model to study what moved exchange rates, long-term yields on bonds, and stock prices across 36 countries between 2002 and 2017. The International Monetary Fund’s annual Coordinated Portfolio Investment Survey provided information on the holdings of investors worldwide, and they essentially mapped investor portfolios at the country level with asset prices, estimating the demand for assets.
Global investors’ appetite for dollar-denominated assets has big implications for currency markets and the US economy.How Much Does the US Dollar’s Primacy Depend on Investor Demand?
They find that fundamentals such as macroeconomic variables, foreign exchange reserves, short-term rates, and long-term debt levels were influential. In exchange rates, fundamentals accounted for 55 percent of the price movements, and macro variables were the primary drivers of stock prices, responsible for 57 percent of the variation.
But in terms of exchange-rate variation, investors’ latent demand (the component of investors’ demand that cannot be explained by prices and observable fundamentals) accounted for the balance, the researchers write, and this demand was geographically concentrated in large investor countries, with the US and European countries alone driving 16 percent of the variation.
This approach also offers a way to interpret global economic events, presenting a view of why Greece’s debt crisis spread to its neighbors in 2011, for example—and showing how fund flows can have major market and political implications. (Read “Understanding the European sovereign debt crisis.”)
A theory with big implications
The inelastic markets hypothesis raises questions, one of which is: If flows have a larger impact on prices than standard theories allow, how many of those flows are still made on the basis of fundamentals? Gabaix and Koijen haven’t sought to answer that.
Still, the conversation about the hypothesis has leapt pretty quickly from academic workshops to a public discussion. The theory has been described in Bloomberg Businessweek, the Wall Street Journal, the Economist, and the Financial Times, among other outlets. “The 2020s will probably be a tough decade, but I think the market is being driven right now by inflows,” Amy Raskin, chief investment officer of the adviser Chevy Chase Trust, said on CNBC’s Halftime Report this past July. “I’m a complete believer in the inelastic markets hypothesis, which posits that there are fewer sellers due to fixed allocations, and when you get huge inflows like this, they have a multiplier effect on the market.”
Yale’s Moskowitz was on the committee that awarded Gabaix and Koijen’s paper the AQR Insight Award, an annual prize from the investment management firm. “It’s a provocative paper that I think will spawn more research in this direction,” Moskowitz says, noting that the field of academic finance moves slowly and deliberately, and it could take years for the hypothesis to be fully vetted.
The stakes are high, as the models it has the potential to influence are used to decide everything from what an investment fund’s next trade or longer outlook should be to how the Fed and other central banks can best support the economy in a crisis. For example, as Gabaix and Koijen note, central banks have bought trillions of dollars of bonds to support markets. If it turns out that the multiplier effect exists, and is stronger in stock markets than in bond markets, this could imply that a central bank seeking to influence the economy should buy stocks instead of bonds. Indeed, in August 1998, the Hong Kong government did just that, buying up 6 percent of the stock market, a move that sent returns soaring 24 percent, consistent with the researchers’ theory. The Bank of Japan owned 5 percent of the market in March 2018, and the Chinese government indirectly owned 5 percent of China’s market in 2020.
If the inelastic markets hypothesis sticks around, it may also have implications for corporate buybacks. By how much do those affect prices? While Gabaix and Koijen clarify conceptually when share buybacks can impact prices, they haven’t estimated key inputs to quantify the effects.
“There has been a standard way of thinking about finance roughly since the 1980s that has really been the dominant paradigm,” says Booth’s Hartzmark. Now, “we’re in a bit of a moment where no one is quite sure where the future is.”
The idea that market flows matter has skeptics and fans, and perhaps could be the future. Koijen says the new model provides a framework to start exploring the elasticity of the aggregate market and related questions. “We’re suggesting that flows from different groups of investors can have a significant impact on prices,” he says. “That’s how a lot of people intuitively thought about things, but there was no theory to back that up.”
- Jean-Philippe Bouchaud, “The Inelastic Market Hypothesis: A Microstructural Interpretation,” Working paper, August 2021.
- Andrea Frazzini, Ronen Israel, and Tobias J. Moskowitz, “Trading Costs,” Working paper, August 2018.
- Xavier Gabaix and Ralph S. J. Koijen, “In Search of the Origins of Financial Fluctuations: The Inelastic Markets Hypothesis,” Working paper, June 2021.
- Samuel Hartzmark and David H. Solomon, “Predictable Price Pressure,” Working paper, October 2021.
- Ralph S. J. Koijen, Robert Richmond, and Motohiro Yogo, “Which Investors Matter for Equity Valuations and Expected Returns?” Working paper, August 2021.
- Ralph S. J. Koijen and Motohiro Yogo, “Exchange Rates and Asset Prices in a Global Demand System,” Working paper, June 2020.
More from Chicago Booth Review
We want to demonstrate our commitment to your privacy. Please review Chicago Booth's privacy notice, which provides information explaining how and why we collect particular information when you visit our website.