Who doesn't like lower prices? But while fee wars among ETF sponsors have gotten a great deal of attention lately, some advisors say costs are only part of the mix of factors a planner must consider.
"The lowering of fees is a good thing, both for advisors and especially for our clients," says Rick Miller of Sensible Financial Planning in Waltham, Mass. "Fees are the only predictable aspect of net performance."
Miller says he will not switch a client to a lower-fee ETF if that would result in a taxable capital gain. But expense ratios remain a major consideration. "When we're investing new money," he says, "we're going to use the lower-fee fund."
A client's needs must be considered first, of course. For a client with a strong buy-and-hold philosophy, the ETF with the lowest expense ratio is often a wise choice. But in accounts where there is frequent trading or for a client who may need to liquidate assets on short notice, a few basis points of expense ratio may be overshadowed by transaction costs, spreads and liquidity.
In short, the expense ratio can't be examined in isolation. "I'm going to use it as a consideration, but not the only consideration," says Shelley Ferro, a financial planner in Metairie, La.
Planner Ray Mignone, who practices in Little Neck, N.Y., says he'd rather place his clients in ETFs with the tightest bid-ask spreads. "In times of stress, those are the ones you want to have," he says.
Spreads of the most liquid ETFs can be as low as a penny. Often, the SPDR Gold Shares ETF (GLD) and the rival iShares Gold Trust (IAU) will have the same narrow intraday cash spread between bid and ask prices - but since the GLD price is 10 times that of IAU, its spread is much better in percentage terms. One reason: GLD, the second-largest U.S. ETF, holds about six times the assets of IAU and has the highest trading volume of any non-equity ETF. As a result, it can charge an expense ratio that is 15 basis points higher than IAU's.
Transaction costs matter, too, especially for small purchases. Morris Armstrong, an RIA in Danbury, Conn., notes that his custodian charges $16 per trade. Assuming he sells $20,000 worth of one ETF and buys the same dollar volume of another, the total transaction cost for the two trades amounts to 1.6 basis points. If he traded that amount monthly, he says, the transaction cost hits 19.2 basis points a year.
For trades of $100,000 or more in funds that are likely to be held for at least a year, Armstrong says, brokerage fees are not an issue.
Advisors also point to another potential ETF problem: tracking error. For large-cap domestic equity funds, the difference between the performance of the ETF and its underlying index should be minimal. In that case, advisors say, fees do matter: "If you're going with a large-cap fund, then I think fees are important, even if the difference is only five or 10 basis points," says John Frankola of Pittsburgh's Vista Investment Management.
Beyond plain-vanilla ETFs, some index portfolios carry significant tracking error. It can be a big problem for ETFs that specialize in relatively illiquid securities or the ones that target large indexes that must be sampled. "You do run the possibility for tracking error when the manager is not purchasing every security in the index," Frankola says.
Armstrong sees the problem in ETFs that try to replicate more exotic indexes. "If you like the concept of the underlying index, then you really have to pay attention to tracking error," he says.
TURNING TO ACTIVE FUNDS
Ferro tries to sidestep that issue by using some active funds, even though costs are higher. She cites the PIMCO Total Return Fund, managed by Bill Gross. The fund's institutional class shares (PTTRX) carry an expense ratio of 0.46%, far higher than the 0.1% of the Vanguard Total Bond Market ETF (BND) or the 0.08% of the iShares Core Total U.S. Bond Market ETF (AGG). Yet despite the higher costs, the managed PIMCO fund delivered a total return of more than 10% in 2012, while the ETFs, which are based on the Barclays U.S. Aggregate Bond Index, were returning about 4% each.
Many advisors see the PIMCO example as an anomaly, however. "I believe, and I think the data suggest very strongly, that on average you're going to do better with an index fund than you will with a managed fund," Miller says.
COMPETING ON COST
Experts widely credit Vanguard, now the third-largest U.S. ETF sponsor by assets, with powering the trend toward lower expense ratios. "They basically behave like a nonprofit, even though they're technically structured as a corporation," says Samuel Lee, a Morningstar analyst and editor of the firm's ETFInvestor newsletter. Vanguard sells ETFs at what it costs to produce and distribute them. "That puts immense pressure on the industry," he adds.
So while the increasing size of the ETF market means that some companies can increase revenues even as they lower expense ratios, there has to be a limit below which the practice becomes counterproductive.
As an example, Lee points to Scottrade's Focus Shares ETFs, which were sold with an expense ratio of five basis points. The fee structure proved to be too low to sustain the operation, and Scottrade has since exited the business. "People also have to consider the institutional stability of providers offering these cut-rate fees," Lee warns.
Another danger is that expense ratios will rise in the future if providers see they are not making enough from the current low fees. That could lock in investors with large unrealized gains on ETFs held in taxable accounts.
Yet another consideration is that a sponsor can launch a similar low-cost ETF to attract new money while keeping old money locked in at a higher rate. Take the case of two iShares products that track emerging markets. The iShares MSCI Emerging Markets Index Fund (EEM) has an expense ratio of 0.66%. Vanguard's competing Emerging Markets ETF (VWO) carries an expense ratio of 0.2% and has passed the iShares product as the category's asset leader.
To counter Vanguard, iShares started a new emerging markets product in October. The iShares Core MSCI Emerging Markets Fund (IEMG) has a net expense ratio of 0.18% and is remarkably similar to EEM.
Since IEMG is relatively new, performance comparisons are not yet helpful. But EEM is based on the MSCI Emerging Markets Index, which through Dec. 31 had a three-year annualized return of 4.66%. By contrast, IEMG is based on the MSCI Emerging Markets Investable Market Index, which had a 4.64% annualized return over the same period. The MSCI investable indexes include smaller-cap stocks, but the performance difference between the two benchmarks is negligible compared with the gap between the expense ratios of the two iShares funds.
Of course, advisors must consider a full range of factors, from expense ratios to total return, liquidity, spreads, tracking error and provider viability. But clients usually don't notice. "That's what they're paying us for," Mignone says. "They leave it up to the advisor to be smart about keeping the expenses down."
Still, every so often, a client breaks the mold. Armstrong has one who actually reads summary prospectuses - a client who retired from a state job that required her to read contracts. "She keeps me sharp," Armstrong says. "We all need a client like that."
Joseph Lisanti, a Financial Planning contributing writer in New York, is a former editor-in-chief of the Standard & Poor's weekly investment advisory newsletter,The Outlook.