For years, the small cap premium has been under siege. Critics have argued that the well-documented tendency of smaller stocks to outperform larger ones—a cornerstone of factor investing since Rolf Banz first identified it in 1981—has faded or perhaps never truly existed in a reliable, investable form. The standard academic measure of the size factor, the Fama-French SMB (Small Minus Big), has delivered a monthly average return of just 0.17% over the past six decades, with a t-statistic too weak to be statistically convincing.

But what if the problem is not the small cap premium itself—but the way we have been defining it?

That is the central argument of a November 2025 research paper from Andrew Berkin and Christine Wang at Bridgeway Capital Management. Their insight is elegantly simple: not all stocks that are small today are the same—some shrank into the small cap universe only recently, falling from large cap status after a period of poor performance. Others are brand-new market entrants—IPOs, SPACs, and spin-offs—that have never had the chance to prove themselves. These two groups, the researchers show, are dragging down the small cap premium and masking what is a robust and persistent return opportunity.

Related:Tax-Managed Long-Short Strategies Gain Traction. Are They Worth the Risk?

Strip those stocks out—require a company to have been small not just today, but also last year—and the picture changes dramatically. In the authors’ words, the distinction is between stocks which have been small and stocks which have become small. Their paper makes the case that only the former deserves a seat in your small cap portfolio.

What the Researchers Examined

Berkin and Wang used a comprehensive database of all U.S.-listed common stocks on the NYSE, Nasdaq, and AMEX exchanges, spanning 60 years from July 1963 through June 2023. Market capitalization and return data came from CRSP; accounting variables from Compustat. The breadth and depth of the dataset is one of the study’s strengths—it encompasses the full arc of modern U.S. equity markets and predates the Russell indices, which allows the authors to show their results are not simply an artifact of index reconstitution effects.

Following the Fama-French methodology, they sorted stocks into size and value portfolios annually at the end of June. Size was split into halves (small vs. large, using NYSE median as the breakpoint), and value was categorized into low, medium, and high book-to-market terciles, creating six portfolios in their primary analysis.

The key innovation was to then ask a backward-looking question at each annual rebalance: where was this stock last year? Each small-cap stock was assigned to one of three buckets based on its prior year status:

Related:AlphaCore Puts Alternatives at the Heart of its Strategy

The researchers then measured returns, volatility, and factor exposures for each group across the full 60-year history. They also extended the lookback from one year to up to five years and tested the results using quintile-based size definitions rather than halves, providing a range of robustness checks.

Key Findings

Finding 1: The Small Cap Premium Is Alive—But Hidden by Bad Company

In the standard calculation across all small cap stocks, the annual small cap premium over large caps averages 1.29%—modest and, by many measures, statistically unconvincing. But when the authors restrict the small cap universe to only those stocks that were also small the prior year, the average annual return rises to 12.94% versus large caps at 10.87%, a premium of 2.07%. Extending the lookback to three years pushes the small cap average return to 13.79% and the implied premium to 2.86%—more than double the standard calculation.

Related:Advisors Urge Investors to Wait on SpaceX

Their adjusted SMB* factor (small minus large using only persistently small stocks) rises from 0.17% per month to 0.21% per month with a one-year lookback, and to 0.28% with a three-year lookback. Crucially, the t-statistic climbs from an insignificant 1.49 to 2.46—crossing the conventional threshold of statistical significance. The difference between SMB* and the traditional SMB is even more statistically striking, with a t-stat of 3.98 at the three-year lookback.

Finding 2: Fallen Angels Are a Serious Drag on Returns

The research makes clear that stocks that have recently tumbled from large cap to small cap status are particularly toxic to portfolio performance. Consider the small value portfolio—typically the highest-returning segment of the equity universe. When composed of stocks that were small last year, it delivered 15.44% per year. When restricted to stocks that were large cap just one year prior, the same small value bucket returned only 8.87%—a gap of more than 6.5 percentage points annually.

This dramatic underperformance of former large caps makes intuitive sense. A stock that has shrunk from large to small cap over the course of a year has, by definition, experienced significant price deterioration. It likely has poor momentum, weakening fundamentals, and carries the stigma of a business in distress. Factor analysis confirms this: fallen angels show meaningfully worse momentum, profitability (RMW), and investment (CMA) exposures than their persistently small counterparts, along with negative alpha that cannot be explained by standard risk factors.

It’s important to note that stocks that were large, but are now small, drag down returns because they not only reduce SMB exposure and have weakening fundamentals, but also their weight is more substantial. 

Finding 3: IPOs and New Entrants Are a Persistent Underperformer

Stocks that did not exist in the investable universe the prior year—a category dominated by IPOs, SPACs, and spin-offs—also drag down small cap returns consistently. In five of the six size-and-value portfolio combinations, stocks in this “new entrant” bucket underperform their counterparts among all current small caps, often by 2% to nearly 6% per year. Only small value new entrants manage to match the broader small value average, and even they lag the persistently small group.

This aligns with a well-established body of prior research. Studies by Jay Ritter (1991) and Tim Loughran and Ritter (1995) documented that IPOs systematically underperform the market for three to five years after listing—a finding that the Berkin-Wang research now shows persists even at the three-year lookback horizon. Factor regressions show that new entrants have significantly lower alpha, particularly among smaller cap stocks, suggesting they bring unique risks and headwinds that standard factor models do not fully capture.

Finding 4: The Improvement Comes from Better Factor Exposures — and Something More

The performance improvement from focusing on persistently small stocks does not simply reflect a mechanical tilt toward well-known factors. Yes, portfolios constructed from stocks that were small in prior years carry higher exposures to momentum (MOM), profitability (RMW), and conservative investment (CMA)—all factors with known positive return premia. But after controlling for these exposures, there remains a statistically significant positive alpha. In other words, the definition of ‘has been small’ captures something real about a company’s quality and trajectory that goes beyond what factor models alone explain.

The three-year lookback produces the greatest improvement in all three of these factor exposures, and the alpha continues to build as the lookback lengthens—up to three years, after which the incremental benefit plateaus. This suggests that the characteristics being screened out (poor momentum from fallen angels, IPO-related headwinds for new entrants) are persistent multi-year phenomena, not just a one-year effect.

Finding 5: Results Are Robust Across Multiple Tests

The authors subjected their findings to extensive robustness checks, and the core result holds up consistently:

Their findings led Berkin and Wang to conclude: “This distinction in defining small caps doesn’t just improve returns, it also improves the characteristics of the drivers of those returns.”

Key Takeaways for Investors

1. The Small Cap Premium Is Not Dead—It Has Been Mismeasured

For investors who have reduced small cap allocations in recent years on the belief that the size premium has evaporated, this research offers a meaningful counterargument. The standard academic measure of the small cap premium is diluted by two groups of structurally weak stocks—fallen angels and new entrants—that any thoughtful portfolio manager would likely want to scrutinize carefully anyway. When those groups are excluded, the premium more than doubles and becomes statistically reliable across six decades of data.

2. Ask “Where Was This Stock Last Year?” Before Buying

The simplest practical implication of this research is a new due diligence question for any small cap investment: was this stock also small cap twelve months ago? If the answer is no—if it was a large cap last year or simply did not exist—that is a meaningful yellow flag. 

3. Be Particularly Cautious with Fallen Angels

Former large cap stocks that have shrunk into the small cap universe are the single most damaging group in this analysis. Their poor momentum, weak profitability, and negative unexplained alpha make them dangerous to hold even when other metrics might suggest value. 

4. Be Wary of IPOs, SPACs, and Spin-Offs—Especially in Small Cap

New market entrants have no track record of being small cap—by definition. The research confirms what prior literature on IPO underperformance has long suggested: these stocks tend to disappoint for years after their market debut. This is most acute in the small cap space, where new entrants make up a larger share of the universe. Investors in small cap funds should pay attention to how much new-issuance exposure their managers are taking, and whether performance is being dragged down by systematic overweighting of recent entrants.

5. A Three-Year Lookback Offers the Biggest Improvement

While even a one-year lookback screen meaningfully improves small cap returns (adding 65 basis points annually), the benefit compounds with a two- and three-year lookback, peaking at 157 basis points of additional return at three years. Beyond three years the improvement plateaus. This suggests that for managers with the ability to implement a disciplined multi-year screen, a three-year definition of “persistently small” represents the sweet spot—balancing the richness of the screen against the practical reality of shrinking the investable universe too severely.

6. The Case for a Dedicated Small Cap Allocation Remains Strong

For asset allocators who have questioned whether small cap deserves a dedicated sleeve in a diversified portfolio, this research argues firmly in favor. Small caps provide diversification benefits independent of returns. More importantly, the results here show that a well-constructed small cap allocation—one that focuses on companies with a track record of being small and avoids the structural underperformers — can deliver a meaningful return premium. For investors who also believe in mean reversion given small caps’ extended period of underperformance relative to large caps, the case for an allocation is particularly compelling at this moment.

7. Combine the Lookback Screen with Quality Factors for Maximum Impact

The most powerful version of this strategy layers the lookback screen on top of traditional factor tilts. When persistently small stocks are combined with an exclusion of the lowest-value quintile (the most speculative, “junk” end of value), the monthly size premium reaches 0.37% with robust statistical significance. This aligns with prior research showing that small cap stocks perform better when quality is controlled for (see my Alpha Architect article examining the research findings here). Investors already using a quality or profitability screen in their small cap allocation should view the lookback requirement as a complementary and additive improvement.

The Bottom Line

The intuition at the heart of this paper is memorable and useful: just as someone who has just moved to Texas wouldn’t yet be considered a true Texan, a stock that has just become small cap shouldn’t automatically be treated like one. True small cap stocks—the ones with an established track record of being small, with the characteristics and behavior that come with that history—are where the return premium actually lives.

The message for investors is pragmatic and actionable. You don’t need to build a complex new model or abandon your existing small cap framework. You simply need to ask one more question at every rebalance: was this stock already here last year? 

Reports of the small cap premium’s death, the authors suggest, have been greatly exaggerated. It is alive, robust, and statistically meaningful. It just requires a little more care in construction—and the discipline to slash the poor behavers before they drag everything else down with them.