Even though factor premiums may be persistent, harvesting them in a consistent and efficient way is no mean feat. Advocates of factor investing argue that it has a much higher capacity than most traditional fundamental strategies. Yet , concerns about crowding have been raised.
These concerns arise from the idea that because premiums are public knowledge and have been widely documented in the academic literature, a growing number of investors will inevitably focus on a limited number of securities with attractive characteristics. So, while factor premiums may be persistent, this could lead to capacity issues for those looking to benefit from them.
Although some prominent academics argue that factor investing strategies have a very high capacity1, anecdotal evidence of overcrowding has also been reported and red flags raised. In a 2015 editorial published in The Journal of Portfolio Management, Bruce Jacobs warned that smart beta strategies are vulnerable to ‘crowding’, with increasing popularity posing a risk of overpricing and lower future returns2.
A year later, founder and chairman of Research Affiliates Rob Arnott co-authored a research paper provocatively titled ‘The incredible shrinking factor return’3. The study warned about the substantial slippage observed between theoretical and realized factor returns due to the costs of implementation. These included items that can be related to crowding, such as missed trades or bid-ask spreads.
More recently, Robeco’s Joop Huij and Georgi Kyosev showed how strategies based on popular factor indices such as the MSCI USA Minimum Volatility Index have suffered from chronic overcrowding and arbitrage. As trades are announced in advance, many market participants anticipate them, and investors who replicate these indices lose as much as 16.5 basis points per annum4.
Most of the red flags raised have to do with products based on publicly transparent factor indices
Two recent studies on low volatility by Robeco’s David Blitz, for instance, dispelled the crowding concerns associated with this popular factor. They showed that despite decades of research supporting low volatility investing, neither ETFs as a whole nor the more flexible, knowledge and opportunistic hedge funds have been positioned to exploit this factor. On the contrary5.
In fact, empirical studies suggest that when implemented properly, factor investing strategies have a very high capacity6. Some experts even argue that because of its systematic nature, factor investing has much larger capacity than most conventional active stock or bond-picking strategies, which tend to focus on small subsets of the investment universe.
In short, most of the discussion around potential capacity issues boils down to the public versus proprietary debate. While the early adopters of factor investing relied on proprietary strategies, most of the recent commercial push has come from generic approaches, so-called “smart beta” products, based on public factor indices.
These products have their merits. They are transparent and usually have very low fees. But they also have important drawbacks, in particular because – as mentioned earlier – they are prone to crowding. Most of that crowding comes from the fact that trades are publicly known in advance and that these trades are, by construction, concentrated on just a few rebalancing dates per year.
With strategies based on the MSCI Minimum Volatility indices, for instance, all trades must be carried out on the last business days of May and November. This severely limits the capacity of the strategy and explains the significant price distortions seen during the period between the announcement and effective rebalancing dates of these indices.
Active factor strategies can trade gradually, making full use of the liquidity offered by the market
Meanwhile, proprietary strategies may be less transparent and come with higher fees, but they have other advantages. One is that they can avoid overcrowding. While smart beta indices concentrate all their trades on just a handful of rebalancing dates every year, active factor strategies can trade gradually, making full use of the liquidity that is offered by the market.
In a recent simulation exercise7, Robeco researchers gradually rebalanced the MSCI Minimum Volatility indices by delaying trades. They found no loss in performance and what they characterized as a spectacular improvement in trade feasibility. Robeco researchers believe this holds true not just for the MSCI Minimum Volatility indices, but also for other smart beta indices, such as MSCI Quality and MSCI Value-weighted indices.
So, while the capacity of factor investing may be large and factor strategies may not be inherently subject to overcrowding, some products appear to us to be seriously under threat. Investors should therefore always keep in mind crowding risk and focus on strategies with proper capacity management – something public indices typically lack as they are not designed for high capacity.
To illustrate this with a concrete example, Figure 1 shows the losses incurred by investors in the widely used MSCI Minimum Volatility indices, due to price reaction before additions and deletions. As we can see, crowding and other phenomena, such as arbitrage, can be very costly for investors, and quickly wipe out the benefits of factor exposures.
This also means that choosing a product based mainly on the fee charged by the provider, without considering practical implementation aspects, may not necessarily lead to the highest net returns. Fees are important but they should not be considered in isolation. This topic will be discussed in the next article of this series.
1 See, for example, Ratcliffe, R., Miranda, P., and Ang, A., 2017, “Capacity of smart beta strategies: a transaction cost perspective”, The Journal of Indexing.
2 Jacob, B.I., 2015, “Is smart beta state of the art?”, The Journal of Portfolio Management.
3 Arnott, R., Kalesnik, V. and Wu, L, 2017, “The incredible shrinking factor return”, working paper.
4 Huij, J. and Kyosev, G., 2018, “Price Response to Factor Index Additions and Deletions”, working paper.
5 See Blitz, D.C., 2018, “Are Hedge Funds on the Other Side of the Low-Volatility Trade?”, The Journal of Alternative Investments. See also: Blitz, D.C., 2018, “Are Exchange-Traded Funds Harvesting Factor Premiums?”, Journal of Investment Consulting.
6 See for example: Li, F., Chow, T., Pickard, A. and Garg, Y., 2019, “Transaction Costs of Factor Investing Strategies”, Financial Analysts Journal.
7 Blitz, D. and Marchesini, T., 2019, “The Capacity of Factor Strategies”, The Journal of Portfolio Management.
当資料は情報提供を目的として、Robeco Institutional Asset Management B.V.が作成した英文資料、もしくはその英文資料をロベコ・ジャパン株式会社が翻訳したものです。資料中の個別の金融商品の売買の勧誘や推奨等を目的とするものではありません。記載された情報は十分信頼できるものであると考えておりますが、その正確性、完全性を保証するものではありません。意見や見通しはあくまで作成日における弊社の判断に基づくものであり、今後予告なしに変更されることがあります。運用状況、市場動向、意見等は、過去の一時点あるいは過去の一定期間についてのものであり、過去の実績は将来の運用成果を保証または示唆するものではありません。また、記載された投資方針・戦略等は全ての投資家の皆様に適合するとは限りません。当資料は法律、税務、会計面での助言の提供を意図するものではありません。
商号等： ロベコ・ジャパン株式会社 金融商品取引業者 関東財務局長（金商）第２７８０号
加入協会： 一般社団法人 日本投資顧問業協会