Expense Ratios Can Be Poor Cost Indicators

October 23, 2013

USA Today’s Matt Krantz was faced this week with an investor question we often hear ourselves: Which ETFs have the highest expense ratios?

The idea here is that the more you pay for an ETF, the less you keep in your pocket. And that concern is certainly valid, particularly in an era when investors are so focused on finding sources of income given compressed interest rates and a challenging global economy.

Unsurprisingly, Krantz began his response by singling out the Market Vectors BDC Income ETF (BIZD | C-0), which comes with a reported expense ratio of 8.33 percent—the highest in the 1,509 U.S.-listed ETF market, bar none. That’s high, so, fair enough.

However, when it comes to ETFs, the headline cost we see in the form of an expense ratio is often a poor indicator of just how much it costs investors to own a given fund.

Take the U.S. large-cap equity space, for instance. There are 61 broad-based ETFs tapping into that segment, with many of them serving up very similar exposure to the space, and none comes with a lower expense ratio than the Schwab U.S. Large Cap ETF (SCHX | A-93), at 0.04 percent a year, or $4 per $10,000 invested.

Buying into that ETF, which has now accumulated more than $2 billion in assets in under three years, would give you liquid exposure to some 750 U.S. large-cap companies in a diversified portfolio. In fact, it’s fair to say that SCHX is a great choice for any investor looking for broad U.S. large-cap exposure, and for a steal of an expense ratio.

Despite trading one penny wide, because of its low share price, SCHX trades at 4-basis-point spreads on a percentage basis, which are relatively wide in the segment. The wider the spread, the more it costs investors to own the fund. In fact, once you take spreads into account, owning SCHX over the course of a year costs investors closer to 0.08 percent, or twice its expense ratio.

That added cost applies even if you are a long-term buy-and-hold type of investor because regardless of your holding period, you still have to pay half the spread when you buy into the fund, and half when you sell.

One of SCHX’s main competitors is the $12.6 billion Vanguard S&P 500 ETF (VOO | A-97), which comes with a stated expense ratio of 0.05 percent a year, or $5 per $10,000 invested—just higher than SCHX’s price tag by a tiny margin. VOO has the same one-penny spread as SCHX, but because of its higher share price, it is more efficient to trade, with an average trading spread that is nearly neglible at 1 basis point. It also has massive liquidity, trading more than $180 million per day.

VOO also does a superb job at replicating its index, underperforming its benchmark by only 0.03 percent on a median basis in the past 12 months—or less than the fund’s overall expense ratio. Looking at that median tracking difference is another vector into a fund’s actual cost of ownership.

ETFs usually underperform their indexes on a median basis by more than their stated expense ratio, and that difference translates into an added cost to investors. That’s because the actual cost to VOO investors of owning—in this case—the 500 stocks in the S&P 500 index, is ultimately the difference between the return on those stocks in the index and the return of the ETF. But since there’s a cost to buy, sell and manage that portfolio of stocks, there’s usually a gap between the ETF and the index's performance, or the so-called tracking difference.



“If the trading costs within the portfolio are minimal, management is strong and everything else equal, the fund should lag its index by its stated expense ratio,” IndexUniverse ETF Analytics’ Paul Baiocchi said. “In the real world, there are frictions that come into play, so it’s rare that a fund will lag its index by exactly its expense ratio over every one-year holding period.”

In VOO’s case, its tracking difference is actually smaller than the expense ratio, meaning the actual cost of owning the stocks in VOO’s portfolio is smaller than the 0.05 percent price tag stated in the fund’s prospectus. In the end, when you combine that tracking difference with the spread, you get a clearer picture of the full cost of ownership.

To give another example, look at another giant in the space: the iShares S&P 500 ETF (IVV | A-99), which boasts $46.6 billion in assets. Being one of only three ETFs that set out to track the S&P 500, the fund has impressive liquidity, trading, on average, more than $762 million worth of shares every day. Its trading spread is unsurprisingly narrow at 1 basis point, which, combined with the fund’s 0.07 percent expense ratio, puts its overall cost to investors in line with SCHX.

But from an efficiency perspective, the fund underperforms its index by 1 basis point beyond its stated expense ratio, meaning, on a median basis, investors are paying closer to $8 per $10,000 invested to own IVV over the course of a year, and that’s without taking into account the trading spread.

That makes VOO the cheapest ETF in the space, even though the fund does not have the lowest expense ratio.

As a final note, it would seem incomplete to look at the U.S. large-cap equity segment without mentioning the obvious behemoth in the space: the SPDR S&P 500 ETF (SPY | A-99), the world’s largest and oldest ETF, with more than $145 billion in assets, and one of the most widely traded securities in financial markets globally.

SPY costs 0.09 percent a year, or $9 per $10,000 invested, in expense ratio. But SPY brings another interesting variable to the cost structure of an ETF: Unlike its competitors, SPY is structured as a trust rather than as an investment company under the Investment Company Act of 1940, meaning its issuer can’t lend out shares to help cover costs, or reinvest dividends into the fund.

Share lending is most likely what helps funds like VOO to keep its costs low, because share lending is used to pay the costs of running the fund. The issuer earns interest on the shares it lends out, and that interest income is then used to offset things like management fees and trading costs.

In the case of SPY, that inability to lend shares and to reinvest dividends cause it to deviate further from its benchmark’s performance than competing ETFs tracking the S&P 500, which ultimately translates into higher costs for investors.

In the past 12 months, SPY has seen a median tracking difference of about -0.16 percent, meaning it’s lagged the S&P 500 Index by 7 basis points beyond its expense ratio. In other words, investors are actually paying about 0.16 percent a year, or $16 per $10,000, to hold SPY, rather than the reported expense ratio of 0.09 percent.

To go back to BIZD, the fund does come with a hefty expense ratio of 8.33 percent thanks a management fee of 0.40 percent a year and sizable acquired funds fees it’s required to pass along to investors, but that's not the whole story. A lot of these costs aren't directly absorbed by fund holders, but are instead reflected in the prices of the underlying securities. Still, at the end of the day, they are still a drag on the fund's overall gross returns overtime, as any cost would be. That might look expensive on the surface compared to other equities ETFs, but investing in business development companies outside of the ETF wrapper would probably cost even more, so that's relative.

In the end, yes, BIZD is the most expensive ETF in the market today, but expense ratios only tell part of the ETF cost tale.


Find your next ETF

Reset All