Short Interest: What High Short Interest
Really Means for Stock Returns.

What Short Interest Actually Tells You

Short interest measures the total number of shares that have been sold short and not yet covered. When an investor shorts a stock, they borrow shares from a lender, sell them on the open market, and hope to buy them back later at a lower price. The aggregate level of short interest in a stock is one of the few direct measures of informed negative sentiment available to public market participants.

The critical insight most investors miss: short selling is expensive and risky. Unlike buying a stock where your maximum loss is 100% of your investment, short sellers face theoretically unlimited losses if the stock rises. They also pay ongoing borrowing fees to the share lender, face the risk of share recall at any time, and must maintain margin requirements. Despite these costs and risks, institutional short sellers consistently identify overpriced securities.

Engelberg, Reed, and Ringgenberg^[1] published a landmark study in the Journal of Finance demonstrating that stocks with more short-selling risk have systematically lower returns, less price efficiency, and less short selling overall. Their findings reveal a critical mechanism: when short selling is difficult or costly, overpriced stocks can remain overpriced for extended periods because the natural corrective force — informed short sellers — is constrained.

This creates a paradox that many retail investors fail to grasp. The stocks with the highest short interest are not stocks where short sellers are "trapped" — they are stocks where informed traders have such high conviction in overvaluation that they are willing to bear extraordinary costs to express that view. When sophisticated institutional investors who manage billions in capital collectively conclude that a stock is overpriced, the informational content of that signal is immense.

The data bears this out across decades of research. Stocks in the top decile of short interest consistently underperform stocks in the bottom decile by 1-2% per month on a risk-adjusted basis. This is not a small effect — it is one of the most robust predictive relationships in empirical finance. The short interest signal persists across market capitalizations, time periods, and geographies.

The Short Squeeze Myth

After GameStop in January 2021, a generation of retail investors internalized a dangerous heuristic: high short interest equals squeeze potential equals buy signal. This narrative is compelling, emotionally satisfying, and statistically wrong. For every GME, there are hundreds of heavily shorted stocks that continue declining as the short thesis plays out exactly as the institutional shorts predicted.

The mechanics of a short squeeze require a very specific set of conditions that rarely coexist. A true squeeze needs an extremely high short interest ratio (typically above 30% of float), a positive fundamental catalyst that contradicts the short thesis, very low available float that amplifies buying pressure, concentrated short positions among a small number of funds, and high cost to borrow that pressures shorts to cover quickly. When all five conditions align, a squeeze becomes possible. When only two or three are present — which is the typical scenario — the high short interest simply reflects accurate negative information about the company.

Autore, Boulton, and Braga-Alves^[2] studied failures-to-deliver — situations where short sellers cannot locate shares to borrow — and found that these extreme short-sale constraints lead to significant stock overvaluation. When short constraints are especially binding, they documented extreme overpricing followed by subsequent reversals as the true fundamental value eventually prevails. This is the opposite of the squeeze narrative: tight constraints do not create sustained upward pressure — they temporarily delay the inevitable decline.

The behavioral economics explanation is straightforward. Retail investors who chase high short interest stocks are engaging in a form of lottery-ticket investing. The occasional spectacular squeeze (GME, AMC) receives massive media attention, creating availability bias. The far more common outcome — gradual decline as the short thesis proves correct — receives no media coverage because it is unremarkable. This asymmetric media attention creates a distorted perception of expected outcomes.

Consider the base rates: in any given quarter, fewer than 3% of stocks with short interest above 20% of float experience anything resembling a squeeze. The other 97% either decline, trade sideways, or produce modest returns that fail to compensate for the elevated risk. Sophisticated quantitative shops have tested this extensively. The expected value of buying high-short-interest stocks is negative, even after accounting for the occasional outlier squeeze.

How Short Constraints Prop Up Overpriced Stocks

One of the most important discoveries in modern financial economics is that short-sale constraints do not just affect the stocks being shorted — they affect the efficiency of the entire market. When short selling is difficult or expensive, overpriced stocks remain overpriced for longer, and the market's ability to correct pricing anomalies is systematically impaired.

The cleanest evidence comes from a natural experiment. The SEC's Regulation SHO pilot program randomly selected a subset of Russell 3000 stocks and relaxed their short-sale constraints. Chu, Hirshleifer, and Ma^[3] studied this quasi-random assignment and found a striking result: 11 well-known asset pricing anomalies became significantly weaker on the pilot stocks during the pilot period. The combined anomaly long-short portfolio returns fell by 72 basis points per month — a difference that survived risk adjustment with standard factor models.

The magnitude of this finding deserves emphasis. A 72-basis-point monthly reduction in anomaly returns translates to approximately 8.6% annualized. This means that a substantial portion of the "alpha" generated by trading on well-known market anomalies exists only because short-sale constraints prevent prices from fully reflecting negative information. Remove the constraints, and the anomalies largely disappear.

Muravyev, Pearson, and Pollet^[4] extended this research in a 2025 Journal of Finance paper, demonstrating that short-sale costs eliminate the abnormal returns on asset pricing anomaly portfolios entirely. While many anomalies persist out-of-sample before accounting for short-sale costs, they cannot be exploited with long-short strategies due to stock borrow fees. This has profound implications: much of the apparent "inefficiency" in stock prices is not actually exploitable because the costs of expressing the short side exceed the expected returns.

For individual investors, this research delivers a clear and actionable message. Stocks with high short interest are stocks where informed traders have already identified overpricing but where constraints (borrowing costs, recall risk, margin requirements) are preventing the full correction. When you buy a heavily shorted stock, you are not catching a falling knife — you are betting against sophisticated traders who have done extensive fundamental work and are willing to bear significant costs to express their negative view.

How BCR Uses Short Interest

At Blank Capital Research, short interest carries a 10% weight in our composite ranking model — the same weight as Investment and Stability. But our implementation is the inverse of what most retail investors expect: we use low short interest as a bullish confirmation signal and high short interest as a warning flag.

The logic is grounded in the academic evidence reviewed above. If short sellers are among the most informed market participants, and if high short interest predicts lower subsequent returns, then the rational response is to align with their informational signal rather than bet against it. A stock that scores well on Quality, Value, Momentum, and Investment but also has very high short interest is sending a conflicting message — and that conflict deserves weight in the final ranking.

In practice, our short interest factor operates as a negative screen. Stocks with short interest above the 90th percentile receive a significant penalty in the composite score. Stocks with short interest below the 25th percentile receive a modest boost. The asymmetry is intentional: the negative information content of high short interest is stronger than the positive information content of low short interest, because the costs and risks of short selling mean that only high-conviction negative views get expressed.

This approach has a practical benefit beyond return prediction. Stocks with very low short interest tend to have more stable price action because there is no large short base that could create forced covering (in either direction). This reduces portfolio volatility without sacrificing expected returns — a feature that matters enormously for real-money portfolio management where drawdown control is critical.

Reading Short Interest Data

Short interest data is reported by exchanges twice per month, with a settlement date approximately two business days after the reporting date. Understanding the key metrics — and their limitations — is essential for incorporating this data into any investment process.

1. 01

   Short Interest Ratio (Days to Cover)

   Calculated as total shares short divided by average daily volume. A days-to-cover ratio of 5 means it would take short sellers 5 average trading days to buy back all their borrowed shares. Higher ratios indicate more crowded shorts and greater potential for price dislocations if covering begins. Ratios above 7 days are considered elevated.

2. 02

   Short Float Percentage

   The percentage of freely tradable shares (float) that are currently sold short. This is the most commonly cited metric. A short float of 10% is modest; 20% is significant; above 30% is extreme. However, this metric alone can be misleading — a stock with 30% short float and low borrowing costs has a very different risk profile than one with 30% short float and 50% annualized borrowing fees.

3. 03

   Cost to Borrow

   The annualized fee that short sellers pay to borrow shares. General collateral (GC) stocks have borrowing costs under 1% per year. 'Hard to borrow' or 'special' stocks can carry costs of 10-50%+ annually. High borrowing costs are the strongest indication that short demand substantially exceeds available supply — a direct measure of negative conviction intensity.

4. 04

   Failures to Deliver (FTDs)

   When a short seller cannot locate shares to borrow and fails to deliver shares to the buyer on settlement date. High FTDs indicate extreme short-sale constraints and, per the research of Autore et al., are associated with significant overpricing and subsequent reversals. FTD data is published by the SEC with a delay.

For a live view of short interest data across our coverage universe, visit our dedicated Short Interest Terminal. The terminal displays current short float percentages, days-to-cover ratios, and historical trend data updated with each exchange reporting cycle.

When Short Squeezes Actually Happen

Despite everything above, short squeezes do occasionally occur. Dismissing them entirely would be intellectually dishonest. The question is not whether squeezes happen — they do — but under what conditions, and whether the expected value of betting on them is positive for a systematic investor.

A genuine short squeeze requires the simultaneous presence of multiple extreme conditions. First, the short interest must be exceptionally high relative to float — typically above 30-40% of float, with days to cover exceeding 10. Second, the cost to borrow must be elevated (above 20% annualized), indicating that supply is genuinely constrained and shorts are under financial pressure to cover. Third, a positive fundamental catalyst must emerge that contradicts the short thesis — an earnings beat, FDA approval, acquisition offer, or other material event that forces short sellers to reassess their position.

Fourth, the float must be genuinely low, meaning a small number of tradable shares amplifies any buying pressure. And fifth, the short positions should be concentrated among a relatively small number of funds, because concentrated positions create correlated covering pressure when any single fund decides to exit.

GameStop in January 2021 met all five conditions. Short interest exceeded 140% of float (due to re-borrowing of already-borrowed shares), borrowing costs were extreme, the float was low, positions were concentrated among a handful of hedge funds, and a combination of Ryan Cohen's activist involvement and coordinated retail buying provided the catalyst. This was not merely an unusual event — it was a statistical near-impossibility that required the simultaneous alignment of five independent extreme conditions.

The danger of the GME narrative is that it teaches investors to search for the next GME without understanding how extraordinary the conditions were. Stocks with 15-20% short interest — which retail investors frequently identify as "squeeze candidates" — rarely experience squeezes because they fail to meet the threshold conditions. The short sellers in those positions have adequate liquidity, reasonable borrowing costs, and no catalyst forcing them to cover. They can maintain their positions indefinitely, and the short thesis usually plays out as the fundamental deterioration they identified becomes apparent to the broader market.

For the systematic investor, the correct framework is probabilistic. If short squeezes occur in fewer than 3% of high-short-interest situations, and the other 97% result in flat-to-negative returns, then the expected value of a squeeze-hunting strategy is negative even accounting for the occasional outsized win. This is the same mathematical logic that explains why buying lottery tickets is a negative-expected-value activity despite the existence of occasional jackpot winners.

A Practical Framework for Using Short Interest

Based on the academic evidence and our own implementation experience, here is a practical framework for incorporating short interest into an investment process. This framework treats short interest as what it is — a valuable informational signal from informed traders — rather than what retail culture wants it to be — a contrarian buy indicator.

The key principle is alignment, not opposition. Align your positioning with the informational signal embedded in short interest data. When short sellers — who face unlimited risk and ongoing costs — choose to bet against a stock, they are providing you with free intelligence. The correct response is to incorporate that intelligence into your investment process, not to assume they are wrong and you are right.

Short interest is not a standalone factor. It is most powerful when combined with other quantitative signals. A stock with high short interest and deteriorating momentum is a strong avoid. A stock with low short interest and strong quality metrics is a strong candidate. The BCR composite model captures these interactions automatically through its multi-factor weighting system.