On a typical day in the stock market, returns on the S&P 500 are about 4 basis points. But on other days, returns average more than four times that amount—and it’s possible to predict when it will happen in advance.
This is the conclusion of Chicago Booth’s Samuel Hartzmark and Boston College’s David H. Solomon, who find a relationship between higher dividend payouts and higher market returns. This link, they argue, illustrates why it’s important for asset pricing models to include shifts in market demand as well as supply, regardless of why these shifts occur.
Most standard-asset pricing models, based as they are on the efficient-market hypothesis, focus on fundamentals such as projections of future cash flows or company-related news. In these models, the stock price represents all of the available information about a security and the company issuing it. A Fed announcement or word of a big government stimulus may send stocks up or down, as investors adjust prices to reflect the news—but the prevailing wisdom is that fundamentals, above all else, drive asset prices.
And yet Hartzmark and Solomon argue that shifts in supply and demand, even if they are unrelated to market fundamentals, also play an important role. The researchers reviewed daily market behavior from 1926 through 2020—analyzing data from Booth’s Center for Research in Securities Prices and Compustat, as well as the website of Dartmouth’s Kenneth French—and examined how returns varied when a day had large or small dividend payments.
When investors receive dividends, they often use the cash to buy more shares of stock—and not necessarily of the company that issued the dividend. If larger payouts mean more people buying stocks, do larger payouts lead to a measurable bump in stock prices? This struck Hartzmark and Solomon as a straightforward way to test the effect of investor demand on prices—especially as purchases made in the wake of dividend payments are a predictable effect of having more cash. This predictability means that the demand can’t be explained as reflecting new fundamental information. It’s precisely the sort of thing standard asset pricing models say shouldn’t influence returns.
The demand for stocks that followed a dividend payout led to measurable, immediate market outperformance, according to the analysis. Days with low dividend payments had an average return of 2 basis points and days with high dividend payments had an average return of 8 basis points. Returns on the week with the highest dividend payments were more than 17 basis points per day. Seeing a similar pattern in 58 international markets, the researchers claim this is strong evidence that supply and demand matter even for the level of the entire stock market.
“You can’t assume away the slopes of supply and demand the way we often do in financial models,” says Hartzmark.
Recognizing the importance of supply and demand could also help explain prices in other asset classes such as cryptocurrencies, they write. Bitcoin, for example, has no future earnings or cash flow, and according to asset pricing models should have a value of $0. But investors nevertheless buy Bitcoin, and their demand for it could explain why 1 bitcoin is worth tens of thousands of dollars. Similarly, the fundamentals of electronics retailer GameStop didn’t change even as its stock price shot from $15 to $350. What did change, however, was investors’ desire to own shares of the previously little-noticed company.
This post was originally published on *this site*