The relentless growth of passive investing has been the most defining theme of the asset management industry over the past decade. Indeed, the relative market share of passive funds has grown in almost every asset class. On average, about 30% of global assets are managed passively. In more mature markets, such as US and Japanese equity, passive market share has reached 45% and 76%, respectively.
In 2014 both ASML and Imtech stock were part of the Dutch AEX index. A passive investor simply allocated his money to all companies comprising the index, without thinking about the prospects of individual companies. A thorough analysis could have revealed that ASML’s perspectives were significantly better than Imtech’s. We know what happened afterwards: the share price of the chip machine manufacturer doubled, while Imtech went in the opposite direction: the company went bankrupt in 2015.
This example illustrates the main argument in the intensifying debate against too much passive investing: the primary function of the capital markets, i.e. effectively allocating capital to companies, could be impeded by too many passive investors. In a well-functioning system the share price of promising and well-governed companies rise, so that they can make investments and expand. A direct effect of a well-functioning capital market is that companies are able to raise capital more easily in a new equity offering. An indirect effect of a higher share price is the increased ability to raise debt. On the other hand, less promising or poorly governed companies go bankrupt so that resources can be allocated to more productive purposes.
Passive investors spread their capital over all available companies without any specific analysis on the prospects of those companies. The majority of passive products track value-weighted indices, which leads to the largest company today still being the largest company tomorrow in terms of market capitalization. Small but promising companies that are not yet included in the followed indices do not benefit from a passive investor’s capital. Hence, it is only logical that passive investing is cheaper: no costs are incurred to separate the wheat from the chaff in large indices as companies not included in an index are simply disregarded.
Of course, passive investing also has many positive aspects for investors. On top of being inexpensive, passive products are also highly liquid, easy to understand and a convenient diversifier for an average investor. In the current debate, one might also argue that the rise of passive investing is actually favorable to active investors.
The reasoning is that, as passive managers are gaining market share at the expensive of active managers, there is less room for active management and the better active managers should remain.
The future of active investing
At what point will we reach equilibrium between active and passive? Alternatively, should we expect cyclicality in the market share of active and passive management?
For active management to stage a comeback outperformance would probably be required. The rise of passive could actually bring a lot of opportunities for active managers. The hypothesis goes as follows: passive investors do not analyze which securities to buy and which not, by definition. This creates a first-mover advantage for active managers. Indeed, if active managers correctly identify winning and losing securities through fundamental analysis, then they would profit substantially once the markets confirm their investment thesis and reflect the true prices.
To test this hypothesis, we looked at the performance of the best actively managed funds over the last 15 years. We selected a small sample of funds in each asset class that had sufficient return history as well as significant outperformance. We have used our quarterly fund rankings as a shortlist for selecting the funds for the analysis.
We wanted to understand the impact of the rise of passive investing on the returns of active funds. For the purposes of this research, we use a basic version of the Capital Asset Pricing Model (CAPM) with the market factor and extended it with an additional variable to indicate the relative change in passive share. Conceptually, the model looks as following
We ran a panel regression for the 5 largest equity asset classes where the market and risk-free return as well as passive share was asset class specific, meaning we took the benchmark and the investment fund universe for each asset class individually. The above methodology resulted in α, β and γ estimates for each asset class, giving us opportunity to address and discuss the differences between markets.
The estimate of γ indicates the sensitivity of active fund’s return to the monthly change in market share of passively managed funds. Positive and significant estimates indicate that the larger the relative market share occupied by passive managers, the higher the return delivered by active managers. Insignificance indicates no impact of change in passive market share on the returns of active managers. Finally, significantly negative coefficient estimate would indicate that the increase in passive market share actually has negative impact on the return of actively managed funds.
The table above reports market share of passively managed funds as of end of June 2017 and the estimates for α, β and γ per asset class based on the panel data regression with monthly frequency for the period spanning between July 2002 and June 2017.
Firstly, the passive share differs per asset class. Global, European and emerging market equity are still dominated by active managers. The passive share in these asset classes is currently between 32 and 34%. On the other hand of the spectrum is Japan Large Cap Equity with more than 75% passive share. US large cap equity has passive share at 45%.
The estimate of α, the amount of outperformance generated by the subset of active managers, varies between 6 and 49 bps, and is insignificant only for Japan. The estimate of β, measuring the level of systemic risk employed by the active managers, is between 0.87 and 1.08 and always significant at 1%.
Finally, the estimate of γ varies across asset classes. Interestingly, it is significantly negative for four of the studied asset classes while it is insignificant for Japanese equities.
The regression result for γ might seem a bit striking at first but could be very well explained in the following way. An increase in passive market share in the category global large cap equity of 1% is predicted to have a negative impact of 109 bps on the return of the best performing active funds. This is logical since relative increase in the passive market share means that more assets are held according to the market index and active managers (that deviate from the benchmark in order to deliver outperformance) show underperformance. Hence, in the periods of large increases in relative passive market share it is very difficult for active managers to deliver returns in excess of their benchmark.
In addition, when the regression was run using only the market factor, i.e. a standard CAPM, the estimate of α was much lower. Hence, it seems that active managers have more skill in delivering outperformance when accounting for the negative impact of a rising passive market share.
A significantly negative gamma coefficient may also indicate that active funds are truly active in the sense of substantial deviation from the benchmark. Growth in the passive share has immediate negative impact on the returns of active managers, as there is a large difference in returns caused by differences in holdings between the active portfolios and the index. The insignificant results for Japan could mean that the sample of active managers in this asset class is composed only of index huggers. Additionally, no change in relative passive market share in the remaining four asset classes would result in an outperformance between 25 and 49 bps for an active fund in the sample.
What do these results mean for active managers? We would argue that the impact of the rise of passive investing is two-fold. Firstly, the steady shift from active to passive products had (and will continue to have) negative impact on performance of actively managed funds. Indeed, as more assets are managed passively in line with market indices, the prices of securities included in these indices increase as they are driven by the difference in flow. Active managers find it more difficult to deliver outperformance as their portfolios that deviate deviating from their benchmarks do not profit from these price increases as much as the market indices.
Secondly, it seems that after reaching a certain point in split between active and passive management, the negative impact of increases in passive market share on returns of active funds disappears. Looking at the case of Japan large cap equity, one can see that there is no significant impact of changes in the relative size of passive market on the returns of actively managed funds. However, the absence of this negative impact of changes in passive market share does not necessarily mean that the active management will come back or survive. As was the case in Japan, inflows into passive funds did not slow down or reverse after the negative impact on active managers disappeared.
Our research clearly points out that active managers are impeded in their search for alpha by the rise of passive investing. According to our figures, all active managers in our sample would have delivered positive alpha when accounting for the negative effects of the growing passive market share into the equation. Therefore, it seems that active managers are not to be blamed for lack of outperformance over the last few years. Given the statistical limitations, the lack of inclusion of other factors and small sample size of this study, further research is certainly warranted.
Download the article here.