We thought it would be helpful to update our perspective of exchange traded funds (ETFs) and provide our top reasons why we do not directly employ ETFs in our model. It is important to note, however, that some of our managers use ETFs in their portfolio construction to gain exposure for an asset class that may not be liquid enough to hold individual securities or for short-term exposure while cash is being deployed. We are not of the view that ETFs are “bad” investments; that said, we believe they are misunderstood by the majority of the retail investors who are using them. In this article, we provide the background of the industry, our top 12 reasons why ETFs can lead investors off-track, and the differentiator in our process.
The History of ETFs
Since 1993, the ETF industry has grown to include approximately 1,623 U.S. ETF products representing $1.86 trillion in assets as of June 30, 2014, according to Morningstar, Inc. The proliferation and obsolescence (including those merged into other funds) rate has been astounding. Of those tracked by Morningstar, approximately 1,000 funds have become obsolete or merged into another fund, which includes the 142 that have disappeared year-to-date through June 30th. This offsets the 109 new ETFs and exchange-traded notes (ETNs are securities that base their returns on the performance of a particular index) that were launched year-to-date.
The top three largest providers of ETFs are BlackRock (iShares) (39% market share), State Street (SPDR) (22%) and Vanguard (21%). Together these three firms comprise 82% of the market share of all ETFs. For the one-year period ended June 30, 2014, the international and U.S. equity categories have gathered the most assets, while year-to-date the most popular categories have been international equity, sector equity and taxable bonds.
Uses of ETFs and Current Trends
Investors and investment managers (including hedge funds) use ETFs to implement strategic core passive exposures, to make tactical adjustments, for rebalancing and manager transitions, to manage unforeseen cash flows, for liquidity and for hedging purposes. According to Morningstar, the bulk of the ETF assets are still in passively managed, long-term ETFs (87%) and passively managed sector funds (11%). To date, only 88 ETFs, representing $16 billion in assets, are considered to be “active” ETFs. However, this may change going forward since many of the “active” strategies will be obtaining their three-year history, which will catch the attention of investors who screen for at least a short-term track record. We have noticed that most of the new ETF launches are for active management strategies.
Wescott’s Top 12 Reasons Why ETFs Can Lead Investors Off-Track
1. ETFs are affected by market demand, and they are prone to arbitrage (allowing traders to capture the spread between the price and the market value of its underlying holdings (its Net Asset Value or NAV). In other words, ETFs can trade at a premium or discount to their NAV.
2. ETFs may also attract “traders” rather than “investors” that will buy and sell their shares more rapidly during a downturn. This appears to be the case since discounts are more pronounced in down markets, and there is a strong probability that “flash crashes” are accelerated by the trading of ETFs when the market has steep declines on rumors or geopolitical events. Since ETFs can trade at a premium or discount to NAV, an investor who is prone to panic may sell at an unfair price.
3. ETFs can deviate widely from their NAVs due to the time difference between the U.S. and the non-U.S. markets where the underlying securities trade. This can be particularly acute in the emerging markets which can have less overall liquidity. We studied a few of the emerging markets ETFs that are offered by the top three providers.
The Vanguard FTSE Emerging Markets ETF (VWO) holds emerging market stocks from around the world. Investors trade in the VWO while the U.S. market is open, although many of the exchanges in the emerging markets are closed due to time zone differences. If there is new and significant information, VWO’s price could differ dramatically from its NAV. In the chart below, you can see the number of days the Vanguard FTSE Emerging Markets ETF (VWO) traded at a premium or a discount and by how much of a premium or a discount. You can see that during 2013, when the MSCI Emerging Market Index was down -2.3%, the VWO traded at a discount for a majority of the time.
Vanguard FTSE Emerging Markets ETF (VWO) 2013 Discount/Premium Analysis
4. Many ETFs underperform their benchmarks. As you can see in the chart below, using the emerging markets ETFs of the top three providers, they all underperformed the MSCI Emerging Market Index for the one-year, three-year and five-year periods. Also, each ETF has a certain amount of tracking error, which is a measure of the volatility of excess returns relative to a benchmark.
5. Performance comparisons of similar-looking ETFs are tricky since they often use different benchmarks. In the chart above, the tracking error is compared to the MSCI Emerging Markets Index. The benchmarks used by the ETFs are as follows:
- iShares MSCI Emerging Markets – The MSCI Emerging Markets Index
- SPDR S&P Emerging Market – the S&P Emerging Markets BMI Index
- Vanguard Emerging Markets Stock Index ETF – The FTSE Emerging Markets Index
6. Many ETFs do not track their benchmarks closely. The FTSE Emerging Markets Index and the S&P Emerging Markets BMI Index both consider South Korea to be “developed”, which creates a high degree of tracking error for those ETFs versus the MSCI Emerging Markets Index, which includes South Korea.
7. ETFs differ by the number of holdings and whether they employ full replication or stratified sampling techniques (an optimization technique that permits a high degree of replication of an index across factors without including all the securities in the index). The inclusion or exclusion of non-index securities is one of the causes of tracking error.
8. Size and liquidity are also factors when looking at ETFs. The iShares and Vanguard have assets of $41 billion and $48 billion respectively, which give them more liquidity; in contrast the SPDR portfolio has only $235 million, which limits its liquidity.
9. Although ETFs are thought to be lower-cost options, the costs to trade ETFs include commissions, operating expenses, and “bid-ask” spreads. For actively managed ETFs, add in management fees. The bid (price you pay for it)/ask (the price you sell it for) spread may be the most impactful, since it depends on how many “market makers” keep inventory in the ETF to provide market liquidity when there are more sellers than buyers. The spread can be more than 1% when there are too few market makers; the spread narrows when there are enough market makers to compete over price. The smaller the transaction and/or the portfolio, the greater the concern should be about total costs.
10. We have found that many ETFs track esoteric non-mainstream indices in areas in which we would not make a strategic allocation such as in specific currencies, countries and/or industries. At Wescott, our allocation focuses on a higher strategic level and we rely on our managers to select the countries, currencies and industries in which to invest, based upon current market conditions and opportunities.
11. Leveraged ETFs and Inverse ETFs have a high degree of risk. Leveraged ETFs use borrowed money to magnify their bets. Inverse ETFs, which are designed to move in the opposite direction of a specific index, have the risk that they will not move exactly inversely from the direction of the index. Moreover, many of these ETFs use total return swaps, which bear counter-party risk. We would not subject our clients to these types of risks.
12. Bond ETFs typically sell at a premium because they are priced at the mid-point of their bid-ask spreads, while the underlying bonds are priced at the bid price. Premiums and discounts can also arise for fixed-income ETFs due to the fact that many bonds do not trade on a daily basis. The more selling pressure there is on a bond ETF the more it will lower the bid-ask price and its midpoint. In that situation the spread between the ETF price and its NAV could pivot from trading at a premium to trading at a discount.
Current Trends: Active ETFs and Our Use of DFA Strategies
The market has focused a great deal on the new “smart beta” products. Smart beta products seek to enhance returns or minimize risk relative to a traditional market capitalization-weighted benchmark. Other products seek to address some of the drawbacks of market-capitalization products, which include concentration risk during bubbles from overvalued securities that dominate the Index. Many of these new products have specific investment rules around different factors such as size, value, momentum and low volatility. Some strategies combine factors and others use a single factor such as low volatility.
Many of the ETF smart beta strategies have very short histories. To properly understand the behavior of a smart index in different environments, one should analyze performance over long historical periods that include multiple economic cycles; factors that are currently driving smart beta index returns may not persist in the future.
We have chosen Dimensional Fund Advisors (DFA) to manage our passive allocation; their investment process has similarities to “smart beta” strategies. DFA provides comprehensive exposure to the U.S. and non-U.S. equity and bond markets using a multi-factor approach including size, value and profitability. DFA’s process has academic roots in the work of Professor Kenneth French and Nobel Prize winner Professor Gene Fama, and others of the firm. DFA has been managing actual portfolios for more than 30 years and it has $378 billion under management. Its research employs the CRSP indices, which date back to 1928.
There are strategies that are less robust than those of DFA; they typically employ simple “tilts” toward factors like dividends, valuations or sectors. In an ETF form, these “smart beta” strategies can be inefficient from a turnover perspective, or can have concentrated exposure to a few sectors, creating unexpected and significant risks. For example, pre-financial crisis, some ETFs that focused only on high dividend payers had their largest weights in financials, which were most impacted by the credit crisis.
Are ETFs useful in some situations? Yes. However, we believe that our approach is better able to capture the desired exposure to diversified markets with less tracking error, greater certainty about price, and with an emphasis on company fundamentals.
An original article authored by the Investment Research Group of Wescott Financial Advisory Group LLC
© Copyright 2014, Wescott Financial Advisory Group LLC All Rights Reserved