UNIVERSITY PARK, Pa. — Economists have long used measures of public sentiment around the economy to forecast a wide range of key financial outcomes, like consumer spending and gross domestic product growth. Now, researchers at Penn State, Florida International University, University of Cincinnati and California State University, Fresno, have used an artificial intelligence-driven approach that analyzes media coverage data and uncovered a significant connection between the public’s perception of the economy and hedge fund returns.
Timothy Simin , professor of finance at Penn State’s Smeal College of Business, and his co-authors pulled data from the Thomson Reuters MarketPsych Indices to create a comprehensive measure of public sentiment, known as a macro sentiment index. They used the index to explore how hedge funds — actively managed funds that pool money from wealthy investors and institutions and make use of complex trading strategies — take the other side of the public's emotional swings about the economy and get compensated for the risk of doing so.
The team published their findings in the Journal of Banking & Finance .
In the following Q&A, Simin spoke about the value of these newer sentiment measures and what they reveal about both hedge funds and the risks — and potential rewards — involved in betting against public perception.
Simin: Traditional sentiment measures come from one of two places: surveys, like the University of Michigan's Consumer Sentiment Index, or financial market outcomes, like initial public offering (IPO) activity. Both are useful, but surveys are infrequent and ask a small sample of people how they feel in general, while outcome-based measures infer sentiment indirectly from prices.
Our macro sentiment index takes a different approach. Using a type of artificial intelligence called natural language processing applied to millions of articles from roughly 2,000 professional news agencies and 800 social media outlets, it measures the tone of what's actually being written about specific macroeconomic topics, like economic growth, inflation, unemployment, bond markets, politics and social disorder. We then distill these into a single index. The result is a measure that is more timely than surveys and more specific about what people are optimistic or pessimistic about while capturing the media channels through which most people form their views of the economy.
Simin: We measured how sensitive each of roughly 15,000 hedge funds was to swings in macro sentiment and then tracked how funds with different sensitivities performed. Through this analysis, we found that funds that effectively bet against public sentiment — funds whose returns move opposite to the sentiment index — outperformed funds riding sentiment by about 0.4% per month, or roughly 5% per year. With larger funds sometimes holding billions of dollars in assets, that could mean differences of tens of millions of dollars in returns.
This isn't a fluke of a few unusual funds or time periods. The result holds even after controlling for fund characteristics like size, age, fees and volatility, as well as funds' exposures to other economic risks such as inflation, default risk and various uncertainty measures. It also persists: A fund's sentiment exposure predicts its performance for about four months, which matters practically because it may extend beyond the hedge fund’s lock-up period, or the amount of time a client must keep their money in the fund.
Simin: Think of hedge funds as the market's arbitrageurs, or traders that exploit price differences for profit. When public mood about the economy swings — say, everyone becomes euphoric about growth — less sophisticated investors trade on that emotion and push prices away from what the business fundamentals like profitability and growth potential justify. Hedge fund managers who understand this tendency, and who have the tools and capital to act on it, take the other side of those trades and profit when prices eventually correct.
But this is far from free money. Sentiment can stay irrational longer than an arbitrageur can stay solvent; a contrarian position can lose money for months if the mood intensifies before it reverses, and funds facing investor withdrawals may be forced out of positions at the worst possible time.
That's the key point of our paper: The extra returns these funds earn look like compensation for bearing sentiment risk, not a free lunch. In fact, we show that macro sentiment behaves like a genuine risk factor in the models economists use to determine how much extra return investors should demand for bearing different kinds of economic risk, and the result isn't explained by manager stock-picking or timing skill.
Simin: First, sentiment expressed in news and social media isn't just noise — it contains systematic information that gets priced into markets. That should change how we think about the flood of economic commentary we all consume. It’s more than simply a reflection of the economy. Sentiment is a force that moves asset demand and, through it, prices.
It also changes how we interpret hedge fund performance. When a contrarian fund beats the market, the natural reaction is to credit a brilliant manager. Our results suggest a subtler story. Much of that outperformance is a premium for shouldering a risk that other investors don't want to hold. And practically, a fund's sentiment exposure turns out to be a useful, persistent signal that investors can use to identify funds likely to outperform over the following several months.
It’s important to note that this pattern isn't unique to hedge funds. We find the same sentiment risk premium, in weaker form, among actively managed mutual funds and even in individual stocks. That consistency across very different sets of assets is what convinces us we're looking at a genuine risk in the economy rather than a quirk of the hedge fund industry.
Finally, the effect is symmetric: Funds betting against sentiment do well whether the public mood is unusually positive or unusually negative. This goes beyond just profiting from crashes after euphoria; it's about taking the unpopular side of the market's emotional swings, in either direction, and being paid for the discomfort of doing so. An open question for future work is separating which parts of sentiment-driven demand arbitrageurs can safely trade away and which parts genuinely require this premium.
Co-authors on this paper include Mustafa Caglayan, Florida International University; Mehmet Canayaz, University of Cincinnati; and Le Zhao, California State University, Fresno.
Journal of Banking & Finance
10.1016/j.jbankfin.2026.107685
Data/statistical analysis
Not applicable
Macro sentiment and hedge fund returns
1-Jun-2026