Exchange-Traded Funds (ETFs) made a huge splash back in 2000 as indexing giant Barclays Global Investors rolled out just over 40 new iShares offerings, State Street came out with its StreetTracks series, and mutual fund giant Vanguard announced it would enter the fray.
2001 also proved to be an eventful year for ETFs. A number of new funds appeared both in the US and overseas; a nasty spat broke out between Standard & Poors and Vanguard; and actively managed ETFs became a hot topic. In this article, we'll take a look at the major ETF trends that emerged in 2001, as well as performance for the year, then examine some key issues to watch in 2002.
ETFs: A brief primer
ETFs are baskets of securities that are traded, like individual stocks, on an exchange. Unlike regular open-end mutual funds, ETFs can be bought and sold throughout the trading day. They can also be sold short (even on the downtick) and bought on margin -- in brief, anything you might do with a stock you can do with an ETF.
Most also charge lower annual expenses than even the least costly index mutual funds. However, as with stocks, you must pay a commission to buy and sell ETF shares, which can be a significant drawback for those who trade frequently or invest regular sums of money.
There are a number of different ETFs that currently trade in the US, including Cubes (the Nasdaq-100 Trust), SPDRs, sector SPDRs, MidCap SPDRs, iShares, Diamonds, StreetTracks, and VIPERs. They are all passively managed, tracking a wide variety of sector-specific, country-specific, and broad-market indexes. Many more trade outside the US.
ETFs' passive nature is a necessity (but as we'll see, the fund companies are trying to find a way around this limitation): the funds rely on an arbitrage mechanism to keep the prices at which they trade roughly in line with the net asset values (NAVs) of their underlying portfolios. For the mechanism to work, potential arbitrageurs need to have full, timely knowledge of a fund's holdings. Active managers, however, are loath to disclose such information more frequently than the SEC requires (which currently is twice a year).
Do they deliver?
ETFs have been aggressively marketed as having three distinct advantages over traditional index mutual funds: trading flexibility, lower expenses, and better tax-efficiency. These claims are valid for some investors and for some ETFs, but it's not quite as simple as it sounds.
Intraday trading
Trading flexibility is one area in which ETFs clearly have an advantage over index mutual funds. They trade throughout the day, whereas mutual funds do not. Investors whose strategies require intraday index trading will find this useful. We suspect, however, that apart from institutional investors using ETFs for hedging purposes, there's little point in day-trading indexes.
Costs: More expensive than they look
Many ETFs have lower annual expense ratios than equivalent index mutual funds. iShares S&P 500, for example, charges just 0.09% per year. Vanguard 500 Index, one of the cheapest index mutual funds, charges 0.18%, and also levies a USD10 annual fee on small accounts.
However, trading costs can quickly offset ETFs' expense-ratio advantage. A USD10,000 investment in the iShares S&P 500 ETF will cost investors only USD9.00 per year in annual fees. But investors making monthly investments will also incur charges of USD96.00 for their trades over the course of the year (assuming a low, USD8 per trade, brokerage commission). In contrast, Vanguard 500 would cost just USD18 per year. Even if one has a small account, the USD10 service fee tacked on by Vanguard is minuscule compared to the costs of trading the ETF.
This suggests that ETFs are only cost effective for those investors who plan to make lump-sum purchases of ETF shares and hold them long enough to allow the expense-ratio advantage to offset the commissions paid to purchase and liquidate the position. A corollary is that investors who take advantage of ETFs' trading flexibility quickly lose the benefits of ETFs' expense-ratio advantages. Investors who trade even a few times per year may well find that an ETF costs them more than a no-load index mutual fund.
Not always tax-efficient
ETFs should generally be more tax-efficient than equivalent mutual funds. Unlike mutual funds, their portfolios are insulated from investor redemptions that can force portfolio managers to sell securities. Retail investors buy and sell shares from each other on the open market, so the fund is not involved in the transactions. Further, those investors who can redeem shares from the fund company -- institutions dealing in 'creation units' (a creation unit is typically a 50,000-share block) -- must take their redemptions in the underlying portfolio securities, not in cash. This eliminates the need for the portfolio manager to raise cash to meet redemptions, and also gives the manager the opportunity to flush-out low cost-basis shares, reducing the embedded gains in the ETF.
Still, not all ETFs are tax-efficient. First, ETFs are only as tax-efficient as their underlying benchmarks. ETFs that track indexes with more turnover, for example, are likely to pay out more capital gains than ETFs that track indexes with less turnover. Generally speaking, smaller-cap indexes tend to have higher turnover than larger-cap indexes.
Further, registered investment companies in the US, including ETFs, lose the right to pass-through capital gains to shareholders if they have more than 25% of their assets in the securities of any one issuer at the end of a fiscal quarter. This means that funds that track indexes that have weightings approaching 25% in single issuers may be forced to sell down positions and make large distributions.
The latter problem has the greatest potential to appear among ETFs that track single country and sector indexes, many of which are highly concentrated. iShares MSCI Canada and iShares MSCI Sweden, for example, track indexes that are heavily weighted in Nortel Networks and LM Ericsson, respectively. In August 2000, the funds paid capital gains that amounted to 23% and 18% of their respective NAVs. Mutual funds tracking the same indexes would run into similar problems, but it's clear that ETFs are not de facto tax-efficient.
2001: A look back
Performance
For equity markets in the US and abroad, 2001 proved to be a difficult year indeed. Technology and telecom shares were down sharply, as were most growth stocks. But amid all the doom and gloom, there were a few bright spots. Value stocks and defensive areas such as real estate and precious metals held up much better than other segments of the market. Stock performance was also dramatically different across market cap ranges. As in 2000, small caps dramatically outperformed large caps in both the growth and value arenas.
ETFs clearly reflected the trends in the overall market. Among broad (non-specialty) offerings, ETFs tracking small-cap value indices were among the best performers. iShares Russell 2000 Value Index and iShares S&P Small-Cap 600/Barra Value Index both notched double-digit returns on the year, with iShares S&P Mid-Cap 400/Barra Value index close on their heels. On the flip side, the technology heavy Nasdaq-100 Trust, or QQQ, skidded to a 34% loss in 2001. Other large-cap growth ETFs were also hit hard, with the StreetTracks Dow Jones US Large Cap Growth Index and iShares Russell 1000 Growth index funds losing more than 20% each for the period.
Among specialty ETFs, the performance differential between growth and value sectors was extreme. ETF's tracking Internet shares suffered the worst losses, with the iShares Dow Jones US Internet Index and StreetTracks Morgan Stanley Internet Index funds losing more than 50% each. US investors fled Internet stocks in droves as the poor economic environment and a glut of capacity led the major players to ratchet back spending on networking gear and debt-laden Internet services companies struggled to survive.
On the flip side, funds focused on more-defensive sectors fared quite well. iShares Dow Jones US Real Estate was one of the best performers. ETFs focused on cyclicals and consumer-related stocks also delivered healthy gains. On the international front, funds tracking Mexico and emerging Asian markets such as South Korea and Taiwan were strong. iShares MSCI Japan was easily the weakest single-country ETF, as the Japanese economy continued to struggle. Europe in general was also soft, with iShares Italy and iShares Switzerland among the hardest hit single-country funds.
New funds
Although the expansion of ETFs in the US slowed somewhat in 2001, the number of ETFs available in other countries burgeoned. In the US, 22 new ETFs were offered, fewer than the 50 new additions in 2000. Barclays Global Investors rolled out by far the most new offerings, adding significantly to its iShares line-up. The new iShares included five global sector offerings tracking S&P indexes in the areas of telecommunications, technology, healthcare, financials, and energy. Barclays also brought to market new iShares offerings tracking Goldman Sachs US indexes in the technology, software, semiconductor, networking, and natural resources sectors. There were only three new US offerings from other firms: Vanguard's Total Stock Market VIPERs and Extended Market VIPERs, and StreetTracks Wilshire REIT index from State Street Global Advisors.
Outside the US, the number of ETFs rose sharply in 2001 and early 2002. Data from State Street Global Advisors show that 96 new ETFs have been launched outside the US since the end of 2000, most of them in Europe (including the UK), with others in Asia, Canada, Australia and South Africa. By early 2002, State Street's data showed that the number of ETFs outside the US stood at 110, more than the 102 available in the US. Several US Funds are now also cross-listed on exchanges in Singapore and Hong Kong, including Diamonds, the S&P 500 SPDR, iShares Dow Jones US Technology, iShares MSCI Singapore, iShares MSCI South Korea, and iShares MSCI Japan.
Asset growth and trading volume
ETF asset growth in the US continued apace in 2001. According to data from the Investment Company Institute, US ETF assets grew by USD17bn in 2001, with USD31bn in net new issuance (the inflows were more than offset by depreciation). However, the most recent figures show that US ETFs pulled in just USD256m in net inflows in January 2002, a month when equity mutual funds garnered USD19.6bn.
ETF assets also continue to be concentrated in relatively few funds in the US. The S&P 500 SPDR has nearly USD30bn in assets, while the Nasdaq 100 Trust has just over USD20bn in assets. Together, the two funds account for about 60% of US ETF assets. No other US ETF even approaches their size: The next largest offering is the MidCap SPDR, which has just under USD5bn in assets.
Fueled by the enormous amounts of capital washing in and out of technology stocks, and the huge popularity of the Nasdaq-100 index, the Nasdaq-100 Trust (QQQ) is by far the most heavily traded ETF. On average, QQQ traded 72 million shares a day over the trailing 12 months ended March 5, 2002. The next-most-traded issue was the S&P 500 SPDR, which averaged about 15 million shares a day. Other volume leaders for the period include Diamonds (4 million shares), the Financial Select Sector SPDR (1.2 million shares), the Technology Select Sector SPDR (1.1 million shares), and the MidCap SPDR (1 million shares).
That said, trading volume and asset levels are much lower for many US ETFs, suggesting that fund companies may have overestimated investors' appetite for the vehicles. Recent Morningstar data show that 60 of 102 US ETFs have less than USD100m in assets (USD100m is considered small for a US mutual fund). Of those, 41 ETFs had less than USD50m in assets (17 of the ETFs that have less than USD100m in assets only launched in the last year, however, and they may well grow over time). The pattern is also apparent on the trading front, where Morningstar data show that 66 ETFs have traded fewer than 50,000 shares a day on average.
S&P stops VIPERs
If some companies appear to offer a surfeit of ETFs, at least one had trouble getting out of the starting gate. US indexing giant Vanguard lost a highly publicised dispute with Standard & Poors in 2001, and was prevented from creating ETFs based on S&P indexes.
Vanguard had planned to create ETF shareclasses for a number of its existing mutual funds based on S&P indexes, including its flagship Vanguard 500 fund. The firm assumed its current licensing agreements with S&P allowed it to create the ETFs without paying additional licensing fees to S&P. That would have worked nicely: The fees -- negotiated long ago -- are minuscule compared to the sums other index licensees pay.
S&P was, naturally, less than keen on the idea. It took Vanguard to court over the matter and won. Vanguard has thus been unable to create ETFs based on S&P indexes. It did, however, launch the Total Stock Market VIPER and Extended Market VIPER in 2001 (Vanguard is calling its ETFs VIPERs, for Vanguard Index Participation Equity Receipts). The former tracks the Wilshire 5000 index, and the latter tracks the Wilshire 4500 Index.
2002: What lies ahead?
Asset growth
ETF assets are still a mere drop in the bucket when compared to mutual funds. Moreover, a significant number of them have remained quite small, and seem to be having trouble gaining traction in the marketplace. Although none of the fund companies have yet shown any signs of throwing in the towel, it wouldn't be surprising to see some of these funds liquidated if the situation continues.
Prime candidates might include some of the smaller sector and single-country offerings. iShares Dow Jones US Chemical, StreetTracks Morgan Stanley Internet, iShares MSCI Belgium, and iShares S&P/TSE 60 recently had less than USD10m in assets each, as did a handful of other offerings.
The industry's growth may also be limited by the commodity-like nature of its offerings. Put simply, once there's an ETF that tracks a specific index, there's not much point in other firms bringing out ETFs that track the same index. For mutual funds, multiple sales channels effectively provide different markets for funds tracking the same indexes, but ETFs are available to all equity market participants. Further, given their already compressed expense ratios, there appears to be limited room for price competition. Given that most money flows to relatively few, high-profile indexes, it may become increasingly difficult for new ETFs to attract significant capital. ETFs' are also presently shut out of the market for actively managed funds. That portion of the market is huge, as many investors continue to believe in the virtues of active management.
Actively managed ETFs
Perhaps aware of the limits of indexing, some ETF companies are working to develop actively managed ETFs. In doing so, they're confronted by at least two problems. First, an extremely high level of portfolio disclosure is the lynch-pin of the ETFs' ability to keep their market prices close to their NAVs, but such frequent disclosure has the potential to increase the market impact of actively managed funds. Second, like all ETFs, actively managed offerings would require exemptive relief from the US Securities and Exchange Commission (SEC). Before granting such relief, the SEC must find that it would be in the public interest and consistent with the protection of investors and the Investment Company Act of 1940.
Only investors dealing in the large blocks of shares called 'creation units' can purchase or redeem ETF shares directly from the ETF at NAV. All other investors buy or sell ETF shares from other investors -- at market prices, not NAV -- over the designated stock exchange. The idea is that if the market price of an ETF's shares diverges from its NAV, large investors will use their opportunity to buy or sell creation units at NAV to earn riskless profits. In so doing, they will either buy or sell enough ETF shares on the market to drive its market price back into line with its NAV. To accomplish this, however, those dealing in creation units must have relatively full and timely disclosure of portfolio holdings, something that could crimp an actively managed fund's ability to implement its strategy effectively.
The SEC is clearly concerned about this issue. In a recent concept release asking for public input, it noted that if an actively managed ETF disclosed its portfolio less frequently, it " . . . could have a less efficient arbitrage mechanism than index-based ETFs, which could lead to more significant premiums or discounts [relative to NAV] in the market price of its shares". The commission also noted that actively managed ETFs might not be as tax-efficient or have expenses as low as index-based ETFs.
The latter is an important point. ETFs have been sold largely as cost-effective, tax-efficient alternatives to mutual funds. As we've seen, the cost argument doesn't always hold true, because it fails to take into account brokerage commissions. But actively managed ETFs would likely have steeper annual expense ratios than indexed ETFs. They would also presumably trade their holdings more often, on average, than index ETFs, eroding their tax-efficiency.
Fund companies have responded to the SEC's queries in various ways. Nuveen proposes 'self-indexing' funds (a concept which the firm is apparently trying to patent) that would essentially define the index being 'tracked' as whatever securities the fund happens to own each day. The index's daily closing value would be the ETF's NAV. Self-indexing funds would report daily portfolios, according to Nuveen's response to the SEC. Yet, whatever one calls it, this doesn't appear to account for the possibility of intraday changes to the portfolio. The more actively managed such an ETF were, i.e. the more trades it made each day, the likelier its market price would be to deviate from its NAV.
State Street's response to the SEC acknowledges that the market prices of active funds might deviate more from their NAVs than those of indexed ETFs', but argues that the SEC should let market forces determine if such offerings are desirable. Investors, argues State Street, are less likely to use ETFs whose market prices do not track their NAVs closely. Barclays Global Investors, which runs index funds almost exclusively, sounds the most negative note among the various responses to the SEC's queries. The firm says that retail investors, the primary audience for actively managed funds, might "derive little benefit from the ability to trade such shares intraday, and, under certain circumstance, may be disadvantaged".
