The concept of factor investing has been known and used for decades. But the introduction in recent times of factor-focused indices at low cost, coupled with investors' falling confidence in active management, has helped to usher in rapid growth in fund flows. Indeed, the growth in funds under management has been so rapid that some have begun to worry that this form of investing is getting 'too popular' and 'overcrowded'.
This is a question we often get asked, so in this paper we take a look at whether these fears are justified.
Why is overcrowding a problem?
The concept of overcrowding refers to when a particular investment or strategy becomes so popular that the resultant fund flows distort the properties of that investment. This is much akin to an investment bubble but can also happen at a smaller scale.
Overcrowding can cause two issues. The first is that the large-scale flows into an investment raise the valuation beyond what is justified by fundamentals, blunting future returns. The second is that liquidity can be reduced excessively. In other words, it can be difficult to sell out of an investment if everyone else is also selling and prices are pushed down significantly.
What is the evidence of overcrowding in factor investing?
Overcrowding is often only confirmed after the fact. However, we can use trends in investor flows and relative valuations as proxy indicators of overcrowding. Neither is perfect but does serve to give an impression.
Turning first to flows, there has certainly been rapid growth in assets under management in the factor investing arena over the past decade. One way of getting a handle on the extent of the growth is to look at flows into Exchange Traded Funds or ETFs, which is the avenue that many funds use to gain exposure to the equity market at a low cost.
According to Morningstar, total global assets under management in factor investing or ‘smart beta’ ETFs stood at $797bn at the end of 2018, having grown by 11% over the previous year and by a compound annual growth rate of 29% between 2012 and 2018. At the same time, the number of funds has also exploded over this time, with around 1,300 different funds available.
But this should be taken in the context of the passive market at large, which stood at just shy of $5trn for the global ETF market at the end of 2018. This means that assets under management for factor investing-focused funds account for less than 20% of the total – a significant portion but not yet excessive in our view.
Indeed, the real story in investing over the past decade or so is the huge shift away from active strategies (many themselves employing factors in pursuit of outperformance) into passive investing. In the US, passive funds accounted for 14% of the combined ETF and Mutual Fund market in 2005 but accounted for 37% by the end of 20171.
Each factor operates in its own cycle and, clearly, large flows into specific factors can blunt future returns. This happened with the weaker relative performance of low volatility equities during most of 2018 when the market rallied sharply. However, we do not think that these distorting effects are permanent. Equally importantly, the spread and variability of relative valuations that can be seen in the data support the case for the multifactor approach advocated by Aon.
Overall, the evidence for overcrowding is weak. From a flows standpoint, factor-based funds have certainly been popular, but this popularity has been dwarfed by the wider trend towards passive investing and away from active investing. Likewise, from a valuation standpoint, the evidence suggests variability and certainly some periods when a particular factor has become more expensive relative to history and the market cap index. But it does not suggest anything permanent or particularly concerning.
Would overcrowding hurt factor premiums?
While we think that overcrowding is not a significant issue, we should also look at whether factor premiums themselves still exist to an extent that is attractive. This is less an argument about overcrowding but rather a question about whether a confluence of factors has effectively reduced any premiums permanently.
Before we look at what has happened to factor premiums in recent times, it is important to remind ourselves of the theoretical underpinnings of the key factors – value, low volatility, momentum and quality. The table below summarises these theories, according to the emerging academic consensus.
The first thing to note is that each factor has both a risk-based and a behavioural explanation for the long-term presence of a premium compared with the market cap index. If we place more weight on the behavioural explanation, we can logically conclude that increased investor popularity, scrutiny and analysis should eventually eliminate the premium – investors will 'learn their lessons' and will not repeat their mistakes. In contrast, if we place more weight on the risk-based explanation, increased knowledge and greater investor flows are unlikely to lessen the factor’s premium.
For example, the risk-based explanation for the low volatility factor focuses on a long term premium as reward for the opportunity cost of missing out on short-term returns. Active managers seeking high returns, but restricted on the use of leverage, are unlikely to invest in the factor, meaning there are systemic reasons why this factor ought to persist.
Next, we turn to looking at the evolution of premiums over time. Have there been any changes?
How have factor premiums performed over time?
The short answer is that some factor premiums have been lower in the past few years compared with the longer-term history. This is one of the reasons we have advised our clients not to expect the same degree of outperformance suggested in back tests. The longer answer is that, while premiums have come down for the value and momentum factors in recent years, the same cannot be said for the low volatility or quality factors, which have held up well.
Additionally, we must keep in mind that premiums can and do vary over time. The chart below shows the comparison of the premiums of the major factors over time. This data is based on the well-known database of factor returns maintained by Kenneth French. It shows how the premiums (on a three-year rolling average basis) fluctuate hugely over time and it also shows that the premiums for the four factors rarely move in unison (ie, they are not closely correlated).
But the data also shows something with the potential to cause concern. The premiums on the value and momentum factors have been declining in recent decades compared with their very long run history (back to 1927). Indeed, the momentum factor has actually delivered a negative premium in the past decade.
Why have some premiums fallen away?
Why have premiums fallen for the value and momentum factors lately?
One probable reason for the value premium lies in the financial sector’s woes during the global financial crisis and the subsequent years of relatively weak performance in the equity markets. The financial sector has a large weighting in the value indices in general – 21.6% in the Edhec value sub-component index versus 16.1% in the standard market cap MSCI World index. So the difficulty the value factor faces in achieving any kind of out performance makes sense.
Another possible driver of the declining value factor premium is the increasing dominance of the technology sector in society. Technology stocks are many things, but they are not value stocks in the main. So, value indices have not kept pace for this reason too.
As for the momentum factor, when the whole market cap index becomes increasingly momentum-driven, it becomes difficult for the momentum factor to outperform. This has been the case especially over the past decade since the global financial crisis, in which time market cap indices have kept up a relentless pace. The momentum factor premium would likely be restored in periods when market performance is much more dispersed. This way, not all of the market is trending strongly upwards, which gives room for the momentum factor to outperform the market cap index.
So, should we be worried about the decline in these premiums? We think that we should keep this trend under review. If these trends persist and do not reverse, it calls into question the viability of certain factors for inclusion in multifactor indices.
However, we do not think that this is the case just yet. Despite the decline in the value factor premium, it still exists. Meanwhile, the negative premium for the momentum factor since 2008 can be explained by the rarefied circumstances of the past decade. Our view is that the market will become much more dispersed over the coming decade compared with the past one, and this would be an environment in which the momentum factor will perform better.
We cannot be complacent, however, and we continue to keep this under review as part of our commitment to continued research within the Aon Factor Service.
Our conclusions and view
Factor investing has undoubtedly been a success story over the past decade and the industry has grown rapidly from a low base. However, this is still small compared with the fund flows out of active management into passive investment strategies.
Meanwhile, relative valuations move in cycles and are not commonly closely linked across all the factors, so the evidence of overcrowding is limited and based more on perception than reality.
The gradual erosion of the premiums in value and momentum factors requires attention. However, there are valid theories as to why this has occurred, and it is quite possible for both premiums to make a comeback over the coming decade. Close, ongoing scrutiny is required to confirm these theories, and this is something that we do as part of the Aon Factor Service. At the same time, these trends, along with the varying popularity of the factors over time, all support the argument for taking a multifactor approach to investing in this area.
1 The Shift from Active to Passive Investing: Potential Risks to Financial Stability, Federal Reserve Bank of Boston, Risk and Policy Analysis Unit, Working Paper, August 27, 2018
Aon plc (NYSE:AON) is a leading global professional services firm providing a broad range of risk, retirement and health solutions. Our 50,000 colleagues in 120 countries empower results for clients by using proprietary data and analytics to deliver insights that reduce volatility and improve performance.
For further information on our capabilities and to learn how we empower results for clients, please visit http://aon.mediaroom.com.
© Aon plc 2020. All rights reserved.
Aon Hewitt Limited is authorised and regulated by the
Financial Conduct Authority. Registered in England & Wales.
Registered No: 4396810.
The Aon Centre
The Leadenhall Building
122 Leadenhall Street
London EC3V 4AN
Copyright © 2020 Aon plc