A crowded trade is a position with a high ratio of active institutional investor involvement relative to its liquidity. Crowding is a form of endogenous market risk as it arises not from contracts’ fundamentals but from the market itself. The risk of crowding has increased in past decades due to the growing share of institutional investors in the market, particularly the activity of hedge funds. Liquidations of crowded positions can trigger price distortions and, in cases of self-reinforcing deleveraging, even systemic pressure.
Crowdedness can be measured by the total value of active institutional positioning in an asset relative to its trading volume. It indicates how long it would take institutions to exit their trades under normal market conditions. For U.S. stocks, these ratios can be calculated based on reported data. Crowding typically skews risk to the downside. This point has been proven empirically for the U.S. equity market. However, crowdedness should also command excess premia. Historically, crowded stocks have outperformed non-crowded stocks materially and with high statistical significance.
The below quotes are based on various economic papers, which are listed at the end of the post. Emphasis, cursive text, and text in brackets have been added for clarity. Also, in the below quotes the term ‘arbitrageur’ has been replaced by the generic term ‘trader’ to avoid misunderstandings for practitioners.
This post ties in with this site’s summary of endogenous market risk.
What is a crowded trade?
“Crowding occurs when the number of investors chasing a similar strategy is too large given the available liquidity or typical turnover. In recent years, crowding has been highlighted as a potential new risk consideration to investing. It can create a coordination problem that can negatively influence risk and return dynamics, making the risk of a trade endogenous to the trade itself.” [Chincarini, Lazo-Paz, and Moneta]
“The crowded-trade problem [means that] no individual trader knows exactly how much is available. This uncertainty could reflect each trader’s imperfect information as to (i) the number of other players who might be pursuing a particular trading strategy, (ii) their current capital and liquidity positions, or (iii) the nature of their alternative investment opportunities.” [Stein]
“Hedge fund positions are an important component of crowded trades. These vehicles are particularly active, take highly concentrated positions, and utilize leverage and short sales.” [Brown, Howard, and Lundblad]
The trouble with crowded trades
“An (institutional) investor cannot know how many other investors simultaneously enter the same trade. Thus, such a position or trade could be ‘crowded’ without awareness on the investor’s part. This can result in common trading pressure from investor positions relative to stock liquidity… [and] may create a negative externality and downside risk. Specifically, such an externality raises the likelihood of severe stock crashes, manifesting as downside risk.” [Conlon, Cotter and Jain]
“In the stock market, individual investors have been largely supplanted by institutions…. If one adopts the view that individuals are naive investors while institutions are rational traders, these data would seem to suggest that we are converging to a world in which the smart-money players trade intensively with one another.” [Stein]
“Between 1980 and 2021, the number of institutional investors increased 15 times from around 400 in 1980 to around 6,000 in 2021 [in the U.S.].In contrast, the number of publicly listed companies held by institutions decreased over the last 20 years, after reaching its peak of 5,756 in the late 1990s, to a total of 2,761 in 2021.” [Chincarini, Lazo-Paz, and Moneta]
“Overall global institutional ownership (in US and foreign equities) and their assets under management levels have risen over time. Institutions are major players not just in developed markets like the US. Their role is rapidly growing in emerging markets… Concern centers on self-reinforcing downward price pressure from liquidating concentrated positions by distressed institutions. The salience of this concern in the industry is evidenced by the introduction of a ‘crowding scorecard’ by MSCI.” [Conlon, Cotter and Jain]
“With imperfect information on other investors’ positions and their liquidity characteristics, investors may face a correlated shock and run for the exit at the same time contributing to a stock price crash. Many institutions have committees devoted to monitoring crowding.” [Chincarini, Lazo-Paz, and Moneta]
“Where, then, does the simple intuition about competition and market efficiency go wrong? In the most general terms, complications arise when, in the process of pursuing a given trading strategy, traders inflict negative externalities on one another… [There are] two distinct mechanisms by which such externalities are created.
- The first has to do with what might be termed a ‘crowded-trade’ effect. For a broad class of quantitative trading strategies, an important consideration for each individual trader is that he cannot know in real time exactly how many others are using the same model and taking the same position as him. This inability of traders to condition their behavior on current market-wide arbitrage capacity creates a coordination problem and… in some cases can result in prices being pushed further away from fundamentals.
- A second way in which traders inflict externalities on one another is through their leverage decisions. If two traders follow the same set of signals and buy the same stocks using leverage, then if one is hit with a negative shock—say, losses in an unrelated part of his portfolio—he will be forced to liquidate some of the commonly held stocks to meet margin calls, potentially creating a fire-sale effect in prices.” [Stein]
“During the week of August 6, 2007, many popular quantitative strategies simultaneously experienced enormous negative returns—in several cases, the daily movements were on the order of 10 or more standard deviations relative to historical norms… Given the rapid growth in this sector, quant managers grossly underestimated the total amount of money invested in their favorite strategies and then compounded this error by leveraging their positions to a degree that, at least in hindsight, seems excessive.” [Stein]
“Competition to exploit less crowded markets for risk diversification benefits can erode their risk diversification potential. This is due to the consequent increase in their crowding which makes them more risky… [There is] increased likelihood of severe stock crashes (as encapsulated by our firm downside risk measures) associated with crowding in non-US markets.” [Conlon, Cotter and Jain]
How to measure the crowdedness of trades?
“Theory suggests elevated crowding levels are likely associated with high institutional ownership, low share turnover, and low market liquidity…
- A crowding measure called DaysADV [is] defined as the total value invested in a security relative to the security’s average daily volume (over the past quarter). They posit the interplay between the magnitude of institutional ownership with the illiquidity of the positions is collectively important (as opposed to illiquidity in isolation). The two-dimensional focus on ownership and (il)liquidity reinforces the idea that during an emergency, the time required to evacuate depends on the number of people in the room and the size of the door… The DaysADV measure can be interpreted as how many days, under typical trading volume activity, it would take the selected investors universe to exit its collective position. From the measure, we can infer a security becomes more crowded if the security’s illiquidity increases and/or if the institutional universe increases their position…
- Alternatively, [one can] use a similar measure called activity ratio as the ratio of the percentage of share i held by an investor at the end of the quarter (t − 2) divided by the average share turnover of the stock i during the quarter (t − 1)… Higher values [of the activity ratio] stand for a more crowded position in a given stock…
These crowding measures are log transformed to reduce the effect of their outliers, their skewed distributions and their large magnitudes on the estimated coefficients, in accordance with the literature… Data on firm-level global institutional ownership is sourced from FactSet. The coverage of quarterly filings of institutional holdings ranges from the first quarter of 1999” [Conlon, Cotter and Jain]
“High values of Days-ADV and activity indicate more crowded positions in a given stock… Both measures attempt to measure the excess ownership in a security that given its typical trading volume might cause price distortions or demand-supply imbalances.” [Chincarini, Lazo-Paz, and Moneta]
“Our preferred measure of crowdedness calculates hedge fund shareholdings as a percentage of average daily trading volume (Days-ADV); this can be interpreted as how many days, under typical trading volume activity, it would take the hedge fund industry to exit its collective position. Days-ADV captures crowdedness both in terms of the size of hedge fund holdings as well as the underlying liquidity of the stock.8 Specifically, this two-dimensional focus on ownership and (il)liquidity reinforces the idea that the time required to evacuate a building depends on the number of people in the room and the size of the door.” [Brown, Howard, and Lundblad]
Crowdedness and downside risk
“Many investors are increasingly concerned about crowding, an investment risk that arises when too much capital is pursuing the same investment strategies. Our previous research demonstrated that crowding increases the downside risk of stocks and factor strategies.” [MSCI]
“[There are two] firm downside risk measures… The first is the negative skewness of daily stock returns, which we refer to as [negative coefficient of skewness]. The second is the natural logarithm of the ratio of the standard deviation of daily returns in down periods, to the standard deviation in up periods, which we refer to as [downside volatility ratio]… We primarily make use of these idiosyncratic measures of risk, rather than traditional tail risk measures (such as Value-at-Risk), as we are interested in the impact of our variable of interest on risk that is purged of the effect of the market.” [Conlon, Cotter and Jain]
“We assess [the] relations [of crowdedness] with firm downside risk using a comprehensive international dataset. A firm’s [stock] future downside risk has a strong and robust positive association with the crowding of its trades. In particular… A one standard deviation increase in crowding raises a firm’s negative coefficient of skewness by 22.65%. A one standard deviation increment in crowding corresponds to a 5.36% increase in a firms downside volatility ratio… The crowding effect is statistically significant at a 1% level. It holds after controlling for other firm fundamentals and characteristics. These include size, return on assets, market-to-book ratio, leverage, risk, and institutional ownership. The effect is strongest for hedge funds and weakest for investment advisers respectively. This suggests crowding’s impact is stronger for the ‘smart money’ of hedge funds relative to investment advisers… Hedge fund crowdedness exposures are significant and also explain downside risk outside the US.” [Conlon, Cotter and Jain]
“We examine the relation between crowdedness and [hedge] fund downside risk… We find that funds with higher average investment weights in stocks present in the most crowded Days-ADV portfolio experienced more severe drawdowns during the financial crisis. We also find the incidence of fund disappearance from the database (delisting) is significantly higher for high crowdedness funds.” [Brown, Howard, and Lundblad]
Crowdedness and expected returns
“Crowding is positively associated with stock expected returns… A trading strategy that invests in the most crowded stocks and sells the least crowded stocks delivers a large and significant alpha… This relationship is more pronounced in a set of well-known asset pricing anomalies. Intuitively, investment strategies based on stock market anomalies are good candidates to become crowded as investors are aware of their existence once they are published… Anomaly risk-adjusted returns are only realized among the most crowded stocks.” [Chincarini, Lazo-Paz, and Moneta]
“Why would hedge fund managers enter into levered positions en masse? One explanation is that hedge funds (or a subset of funds) have an informational advantage and can identify certain stocks with higher return potential. Thus, they are willing to bear the possible risk associated with adverse left-tail outcomes in certain states in order to earn higher expected returns in other states.” [Brown, Howard, and Lundblad]
“Every quarter we sort stocks into quintile portfolios based on the crowding variable and then proceed to build long and short portfolios selecting the top and bottom quintiles as those most and least crowded, respectively. Next, we examine the returns of these portfolios in the quarter after portfolio formation. In this single sorting approach and using Days-ADV as the crowding variable, we find that a value-weighted quintile portfolio of the most crowded stocks delivers a Fama and French 3-factor monthly alpha of 0.54%, whereas the alpha of the lowest crowding quintile is -0.90%, and both alphas are highly significant.” [Chincarini, Lazo-Paz, and Moneta]
“We examine whether crowdedness is important for explaining the observable variation in reported [hedge] fund returns… We regress hedge fund returns on our crowdedness measure and other traditional risk factors. We find that the Days-ADV spread portfolio is statistically significant at the 10% level for about 35% of the hedge fund universe, and certain subsets of hedge fund strategies are highly exposed to crowdedness…. These results indicate that crowdedness is important at the hedge fund portfolio level as well as the individual security holding level.” [Brown, Howard, and Lundblad]
Main papers
Brown, Gregory. and Howard, Philip and Lundblad, Christian (2019), “Crowded Trades and Tail Risk”, June 2, 2019
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3326802
Chincarini, Ludwig, Lazo-Paz, Renato and Moneta, Fabio (2024), “Crowded Spaces and Anomalies”. April 2014
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4618248
Conlon, Thomas, Cotter, John, and Jain, Kushagra, (2004), “Crowding and Downside Risk: International Evidence”, pre-print submitted to Elsevier, November 26, 2024
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5036389
Stein, Jeremy (2009), „Sophisticated Investors and Market Efficiency”, Journal of Finance
https://scholar.harvard.edu/stein/files/presidential-address-jf-final.pdf
