This post is mainly about Exchange Traded Funds, or “ETFs”. Concerns about ETFs generally, and passive index investing specifically, seem to be increasing as more and more investors opt for simply “being in the market” by investing in index-linked ETFs rather than investing in actively-managed mutual funds or buying individual stocks. Admittedly I am far from an expert on ETFs but having read more and more about how ETFs might exacerbate market volatility piqued my interest, so I did some background research on the ETF market. I am also assuming as a starting point that the reader has a basic understanding of what an ETF is. In light of my limited background, any comments or clarifications from readers would be more than welcome. My conclusions are i) as passive investment strategies gain more and more market share, alpha from active management (by definition) becomes more and more difficult, creating even more appetite for passive index-linked strategies; and ii) ETFs are not necessarily going to cause volatility to increase more than it would otherwise in a market meltdown scenario, and in fact, could prove to be stabilising.
ETFs first surfaced in the early 1990s. This new type of investment vehicle began to gain traction during the latter half of the 1990’s, as lower-cost passive investment strategies moved more into the forefront as an investment philosophy (in lieu of active portfolio management). Until ETFs burst onto the scene, many investors seeking a diverse equity portfolio would buy actively-managed mutual funds or investment trusts instead of individual stocks, acknowledging the benefits of diversification in an equity portfolio. You might recall in fact that there were a handful of actively-managed mutual funds with long histories of generating alpha and an almost cult following. Do you remember any of these from the late 1980s to the mid-2000’s - the Fidelity Magellan Fund (Peter Lynch), the Vanguard Windsor Fund (John Neff), and/or the Legg Mason Value Fund (Bill Miller)? However, as more and more money flowed into these over-achieving funds and their assets under management (“AUM”) exploded, it became more and more difficult for even these superstar portfolio managers to beat the returns on the S&P 500. As a result, these and many other actively-managed mutual funds gradually lost ground to lower-cost passive index mutual funds and the growing market for ETFs. The most common form of ETF is one that contains a basket of equities in proportion to their market-weighting in a given index, e.g. the S&P 500, the Russell 2000, the MSCI World index, the FTSE 100 index, the STOXX Europe 600, etc. Whilst ETFs can be comprised of baskets of assets other than equities and are not always passive, the large majority of ETFs are in fact passive index market-weighted equity portfolios matching a particular index like the S&P 500. They are a tailor-made product for investors which focus on macro themes rather than specific stocks. For example, even though an investor might be forgoing calls on individual stocks, he can make active calls on more macro-driven themes since ETFs exist that are focused on sectors, regions, countries, sizes of companies, etc. There are also plenty of ETFs supported by baskets of non-equity asset classes like bonds, currencies and commodities. Lastly, investors can find ETFs that incorporate leverage (generally one to three times) or can express short views by investing in ETFs with “inverse” (and often leveraged) strategies, meaning they are essentially short the underlying assets.
Stepping back into time, The Vanguard Group is often sighted as the catalyst around the birth and early evolution of low-cost passive investing, arguably the precursor to the ETF market. Vanguard, founded by the legendary John Bogle, was the first investment firm that developed and began offering a low-cost index stock mutual fund to retail investors in 1975. This fund, named the “First Index Investment Trust”, tracked the S&P 500. Mr Bogle also pioneered low-cost investing generally, setting him aside from many other mutual fund companies at the time that were focused on higher fee actively managed funds. However, investors increasingly began to question the benefit of actively managed funds, which had significantly higher running costs and often underperformed the indices. This pushed passive investing into the limelight, and this then fuelled the introduction of ETFs.
The first official ETF was launched in the U.S. in 1993 by State Street Bank, named the S&P 500 Trust ETF, or SPDR (NYSE: SPY). SPY remains one of the most popular ETF funds today. As the growth of ETFs gained steam, it was noticed of course by many other prominent asset management firms, with firms like Barclays Investment Services (1996) and Vanguard (2000) jumping into the fray. According to industry source ETFGI.com, there were nearly 7,927 ETF’s provided by 436 providers at the end of 2019, accounting for circa $6.5 trillion equivalent of AUM. Of these, 136 ETF providers were in the U.S., offering 2,062 ETF’s with AUM of $4 trillion. Today, three AM firms dominate the ETF market: Blackrock (via its mid-2009 acquisition of Barclays Investment Services and its iShares ETF business), The Vanguard Group and State Street. These three giant investment firms account for approximately 80% of total ETF assets under management today (of $4 trillion).
To provide further context, Bloomberg highlighted in an article in September 2019 that in August (2019), the amount of AUM of passive index funds and index ETFs combined ($4.271 trillion) had for the first time passed the AUM of actively managed funds ($4.246 trillion). The article goes on to say, sighting data from the Investment Company Institute (ICI factbooks are at this link), that active and passive funds and ETFs together represent only about one-third of the US market for equities, with the balance of shares held by pension funds, insurance companies, individuals and other investors. The growth of passive funds does not mean that there are not active funds / strategies that have outperformed the indices – I discuss this more below. But the growth of low-cost passive investing has put significant pressure on fund management costs, and as a result, the cost of investing across the board has fallen significantly. As you might have recently read, some brokerage firms even now offer zero-commission trading for individual shares in an effort to retain their customers.
I don’t really want to dwell on the differences between mutual funds and ETF’s in this post, as you can research this yourself if you’re not already familiar with one or both types of investment vehicles and wish to brush up. You can do this at the very informative website ETF.com. Blackrock also has a succinct summary of ETF’s on its website, found here. Comparing the two investment vehicles succinctly, ETF’s tend to be passive (but not always), low cost, more tax efficient (assuming you own them in the country in which you live), and tradeable intraday on an exchange, whilst mutual funds tend to be actively managed (but not always), higher cost (trading/churn), less tax efficient and can be bought or sold via only the sponsor based on the closing price at the end of a trading day.
With this background, I want to explore three areas involving ETFs below. 1. How have index-driven “down-the-middle” funds based on the S&P 500 performed vis-a-vis actively managed funds, and what role have fees paid in this discrepancy?
Returns: I am taking this straight from Vanguard’s website, because it says all that needs to be said:
This is not completely disheartening for active fund managers, because at least one-third still outperform passive strategies. However, investors must consider if the opportunity for outperformance is sustainable over time, and if it justifies the incremental cost of an active strategy.
Fees: According to the Investment Company Institute, the average annual fee for an actively managed mutual fund is 0.82%, and for an index (passive) ETF is 0.09%. What does that mean to you as an investor? Let’s look at an example. Assume that you:
invest $10,000 in an actively managed mutual fund and the same in an index ETF,
both provide headline (gross, pre-fee) returns of 7%/annum over 10 years, and
both charge fees equal to the averages mentioned above.
At the end of 10 years, an investor would have $19,507 in the ETF, but only $18,215 in the mutual fund. Fees would have cost you $1,292 over 10 years, a very meaningful amount of your portfolio. Another way to look at it would be to calculate the return which an investor would need in an actively managed mutual fund to offset fees, such that the mutual fund earns the same as the index EFT? The answer is 7.73%, meaning the active mutual fund manager would need to generate on average 0.73% per annum of excess return – or alpha - over 10 years. This might be possible, but it would be very difficult and thus is fairly improbable
2. How do ETFs manage the process of i) buying or selling new shares (primary), and ii) ensuring that the price of the ETF in the market is aligned with the NAV price (secondary)?
This was particularly interesting to me, because I never really looked this deeply into the mechanics of how an ETF really works. Let me try to describe the mechanism.
Each sponsor of an ETF has a list of Authorised Participants, or APs, that the sponsor uses to create the initial portfolio, create or redeem shares based on ongoing demand or lack thereof, and rebalance the portfolio from time to time to ensure that the net asset value (“NAV”) of the portfolio is the same as the market price of the ETF. ARs include banks like JP Morgan, Bank of America, Goldman Sachs, ABN Amro, SocGen and Morgan Stanley. Each fund sponsor normally has many APs with which they work to reduce their transaction risk. In fact, Blackrock – the largest sponsor of ETFs – has 42 APs (although the firm says its activity is concentrated amongst five APs). APs play an important role in the primary (creation) and secondary (liquidity) markets for ETFs.
Primary: ETF shares are created so as to meet demand for the ETF. The AP creates the ETF shares and delivers them to the ETF sponsor in exchange for the same value of the underlying stocks comprising the index. The exact opposite occurs for the redemption of ETF shares, should investors overall wish to sell instead. This occurs once per day at the closing NAV.
Secondary: In the secondary market, the ETF price should reflect the value of the underlying portfolio of assets. Transparency is excellent in the ETF market, so the underlying assets are known by investors. Normally, any deviation from NAV would quickly be arbitraged out by investors. APs of course are very active in arbitraging price differentials, because this is a profitable business for them. For example, if a given ETF price in the market is above the NAV, APs can create new ETF units and / or buy the underlying basket of shares as a form of (profitable) arbitrage to ensure that the market price and portfolio value realign.
It is also important to recognise that different APs might be better in certain asset classes than others. Aside from straight-forward S&P 500 index tracking for example, there are more specialised asset classes such as emerging markets bonds or equities and high yield bonds, amongst others. Blackrock has a good description of the role of APs in the primary issuance of ETF’s (albeit slightly dated), and it can be found here. 3. Why is there so much concern with ETF’s today, as they become a more significant part of the overall market?
There are plenty of articles out there painting a negative picture of ETF’s, or more broadly, passive investment strategies (whether ETF’s or index mutual funds). Many of these concerns have to do with the theoretical case of a potentially sharp move down in asset prices, which in turn would cause passive investment vehicles to unload shares to meet redemptions (mutual funds) or rebalance ETF prices with NAV. This could create a negative feedback loop: share prices fall, investment vehicles unload shares to meet redemptions (funds) or rebalance (ETFs), this puts more downward pressure on prices which fall further, again funds / ETFs sell into this to meet liquidity needs, and so on. Let’s dig deeper.
Firstly, in the case of ETFs, the sponsor would not have to sell the underlying basket to meet redemptions, because these are exchange-traded vehicles and sponsors have no obligations to provide liquidity directly. (This is different of course for open-ended index mutual funds.) As the market deteriorates, the change in the price of the ETF should reflect the value of the underlying basket of assets. As discussed in 2 above, arbitrage should ensure that this naturally occurs. The difficulty though might be with finding prices for and / or selling the underlying assets in a market with few buyers. However, I find this no different than investors all rushing for the doors at the same time in any market crisis – prices will inevitably gap down as dealers try to find the right price for the assets in a one-way market and bid-ask spreads will widen considerably. But eventually, the markets will find levels at which to transact. Although there could be a period of price dislocation, I believe that APs will almost certainly find the right levels to arbitrage the value of the underlying basket to the ETF price.
There is an exception to this however, and I believe this would be around illiquid assets. Stocks - especially those that are in indices - tend to be the most liquid assets, and even in difficult markets generally find pricing levels quickly. But it is a different story when you move to more illiquid assets classes like bonds, especially high yield or emerging markets bonds, and leveraged loans. Or even into ETNs (exchange traded notes) which are tied to, but not technically secured by, the underlying assets, like commodity ETNs. In these cases, there are much fewer counterparties and price discovery is more difficult even in good times, and in some cases the collateral is physical. You can imagine in a market sell-off in order to exploit the arbitrage, an investor or AP has to first find the assets (certainly fewer counterparties to start with), and then find an appropriate price. This might not be easy and could take time, because some assets – take leveraged loans for example – might see their market completely collapse and remain dormant for a period of time. Again though, the reality is that this is not so much to do with the ETF but more about the illiquidity in the underlying assets. Yes, the dislocation will travel to the ETF, almost certainly creating a difference in NAV and the underlying value of the basket, and this might persist for some period of time until the underlying assets can find pricing levels. However, I do not necessarily see how ETFs make this situation worse, and I would even argue that ETFs in this case might create a stabilising force that accelerates the discovery of appropriate pricing. I remain unconvinced that ETFs are destabilising in this scenario, because the performance of illiquid assets in a downturn, when the assets become out-of-favour, will be poor, and the underlying market will be highly illiquid until the imbalance is sorted, no matter what.
In closing out this section, I would say that the growth of passive investment strategies, and specifically ETFs, have likely increased the short-term volatility in the equity markets, but probably have not materially affected long-term volatility. Empirically, I would guess that some academic has looked at this, but this is too much for this post! My instinct – and this is instinct only – is that short-term volatility can be driven by straight pricing arbitrage which is identified by and acted upon almost instantaneously in program trading. And we have seen that program trading can occasionally cause bizarre price gyrations, even if infrequent and very short-term, in the equity markets.
In conclusion, ETFs - found now in an assortment of flavours - have proliferated, riding the reality that passive investment strategies generally have outperformed active investment strategies, at least as far as the most liquid assets classes like mid/large cap equities and government bonds. Active management still outperforms one-third of the time, although fees can quickly erode this advantage. As you move towards less liquid asset classes like emerging markets stocks and bonds, high yield bonds and leveraged loans, active management probably better justifies its cost, as the underlying assets tend to be of lower quality, the market is less liquid, and the price is more difficult to pinpoint. In a downturn, it is the ETFs supported by less liquid assets that will probably see their price drift away from NAV (as best as can be determined) for a longer period of time.