Model achieves higher returns compared to standard stock investment strategy

Christian Goulding’s research paper, "Disagreement of Disagreement," which focuses on disagreement among investors about the value of stocks, was awarded Best Paper in the Asset Pricing and Investments category at the recent 2024 FMA European Conference held in Turin, Italy.

Goulding co-authored the paper with Campbell Harvey, J. Paul Sticht Professor in the Fuqua School of Business at Duke University, and Hrvoje Kurtovic, lecturer on The Faculty of Business and Economics at the University of Lausanne, Switzerland.

Disagreement is ubiquitous, necessary, and varying

“Disagreement is everywhere, especially in the world of investing,” says Goulding, “and that’s a good thing because it facilitates stock trading itself. Take Tesla’s stock, for example. Some investors believe it’s underpriced and want to buy, while other investors believe it is overvalued and want to sell.”

Without this discrepancy in expectations, Goulding argues, there would be no trading.

But the level of discrepancy varies over time, notes Goulding. “Sometimes disagreement is high – buyers’ and sellers’ expectations are far apart – and other times disagreement is low.”

His research seeks to understand how variation in the level of disagreement across millions of investors shapes the prices we see in the market.

The challenge

It’s extremely difficult—impossible, Goulding says—to directly observe the level of disagreement short of going out and asking each individual investor about every financial investment they're considering and what their beliefs are.

“We need something else,” says Goulding, “some other indirect way to measure disagreement. For example, one popular measure is based on disagreement among analyst forecasts. Another measure is based on the volatility of asset returns. Another one is based on trading volume, in particular, how much short interest is out there, how much short selling is going on.”

Goulding and his colleagues document dozens of disagreement proxies proposed in top tier journals across accounting, economics, finance, and management studies. Unfortunately, these proxies tend to give conflicting indications.

“We found that there’s very low correlation between a disagreement measure used in one study and a disagreement measure used in another,” notes Goulding. “Which disagreement measure do I use as a researcher if I'm trying to figure out the impact of disagreement on prices? One proxy says disagreement is really high, and another proxy says disagreement is very low. Which one do I believe?”

The solution

The authors build a simple model of an asset market and use it to reverse engineer a new way to measure disagreement based on observable market variables.

“We now have one overarching metric – a brand new measure of disagreement that incorporates several other disagreement proxies within it, helping to temper when one is high and another's low, which one to believe,” he said.

Historical simulation, across thousands of stocks over the past 30 years, shows that applying the predictive properties of this new disagreement measure can harvest over 14% additional return compared to a standard investment strategy.

“These returns, which any investor would love to achieve year after year, make our metric worth incorporating into investment decisions,” said Goulding.

###

Learn more about Goulding's award-winning paper "Disagreement of Disagreement."

About the conference

The annual FMA European conference was hosted by the Financial Management Association International (FMA) in partnership with ESCP Business School and was attended by more than 300 academic experts and financial management professionals from around the world who came together to discuss the future of finance and share the latest in financial management industry research. Goulding's award-winning paper was among approximately 700 submissions.