Value investing’s pulse returns: Predictable swings in value-growth performance
A new financial study from the Cornell SC Johnson College of Business examines nearly five decades of market data and finds that the decline of value investing appears more cyclical than permanent.
David Ng, professor in the Charles H. Dyson School of Applied Economics and Management, and co-authors argue that value stocks — companies whose shares trade at relatively low prices — go through predictable cycles of outperforming and underperforming growth stocks.
The paper, “Is the Value Premium Dead? Forecasting Value–Growth Cycles with the Implied Value Premium,” published Feb. 4 in the Journal of Financial and Quantitative Analysis, introduces a new metric that can forecast the swings with surprising accuracy. Co-authors include Yan Li, professor at Temple University and Bhaskaran Swaminathan, CEO of Compassion AI.
The research team devised a forward-looking measure based on expected returns called the implied value premium (IVP). Across the 1977–2023 period, IVP proved to be the strongest predictor of whether value stocks would later outperform growth stocks. In fact, it beat every commonly used forecasting tool, including valuation spreads and indicators tied to the economic cycle.
This finding is important because many investors have questioned whether value investing is still attractive. After performing well for most of the 20th century, value stocks lagged growth stocks dramatically after 2007. Between 2008 and 2023, the classic measure of the value premium (known as HML, short for “high minus low”) fell close to zero or even negative compared to much higher historical averages.
The study shows this slump does not mean value investing has failed. Instead, the authors explain, the value’s recent troubles are simply part of a recurring cycle. Looking back to 1927, the researchers show that value and growth stocks have repeatedly taken turns outperforming each other. Value cycles typically last five to seven years, while growth cycles run two to three years. The period after the 2008 financial crisis created an unusually long stretch favoring growth, much like what happened before World War II.
“Traditional tools used to measure whether value looks cheap, such as price‑to‑book ratios, have become less reliable in an economy dominated by intangible assets like software, branding and research spending,” said Ng. “These older measures don’t capture the earnings power of companies whose value lies outside traditional accounting categories.”
IVP takes a different approach. Instead of looking backward at accounting data, it estimates the market’s implied required return for both value and growth portfolios. The difference between them, the implied value premium, acts like a barometer of how attractively priced value stocks are in the eyes of investors. When IVP is high, future returns on value stocks tend to be stronger. When IVP is low, growth stocks tend to outperform.
Across every testing horizon from six months to three years, IVP consistently predicted which group would win. No other variable, including interest‑rate spreads or default‑risk indicators, showed similar forecasting power.
One of the long-running debates in finance is whether the value premium exists because value stocks are riskier or because investors consistently misjudge them. The research team found that IVP also predicts the investment premium — the return difference between firms that invest a lot and those that invest little.
“IVP also predicts how much value stocks ‘surprise’ investors during earnings announcements,” Ng said. “When IVP is high, value stocks tend to deliver better‑than‑expected earnings relative to growth stocks. That’s a hallmark of mispricing, suggesting investors periodically become too pessimistic about value firms and too optimistic about growth firms.”
For everyday investors, the study’s message is straightforward: Value isn’t dead, but timing matters. Because value growth cycles ebb and flow, investors who can monitor the implied value premium may have a better chance of shifting their portfolios at the right moment.
“The findings also suggest that surviving in a market increasingly shaped by AI, technology and intangible assets will require more sophisticated valuation tools than those rooted in 20th-century value investing,” Ng said.
