Consumers have never been happy about paying more for the same from one place than from another. It has only been within the past twenty years or so that investors have gotten wiser about this practice. Financial advisors, planners, and the media have increasingly made investors more aware of how much less expensive and tax-efficient ETFs are than mutual funds using the traditional redeem-at-distributor structure. When one adds in tax efficiency, faster clearing and settlement, intraday liquidity, and zero redemption fees, there is almost no reason to pay up for new purchases of traditionally structured mutual funds.
What some find surprising is that there are cost differences, sometimes significant, between expense ratios of ETFs that are identical in terms of which securities they hold. There are at least four categories of ETFs – or ETPs popularly called ETFs – where the better-known shares are more expensive than substitutes. In all of these cases, the substitutes are large enough to be well past viability and liquid enough to satisfy the needs of most investors.
Example No. 1:
Let’s start with the oldest still-offered US ETF and by far the largest, SPDR S&P 500 ETF Trust, SPY. Part of the SPDR-branded family by State Street Global Advisors (SSgA), SPY has more than $370 Billion in Assets Under Management (AUM) although it briefly topped $400 Billion before the 2022 market declines. As the standard definer of passive index funds, SPY owns all the stocks in the S&P 500 Index weighted by float-adjusted market capitalization. Its expense ratio is 0.095% also called 9-1/2 basis points (1/100 of 1%) using industry vernacular to compare ETF fees.
There are two other very large ETFs that also track the S&P 500 Index included in this analysis:
IVV, iShares Core S&P 500 ETF from BlackRock with an expense ratio of 3 Basis points.
VOO, Vanguard 500 Index Fund also with an expense ratio of 3 basis points.
There are other key relative shortcomings that SPY has compared with the fewer ETFs. SPY was created using a Unit Investment Trust structure and its constituent changes are ordered by its Master Trustee and must be traded on the effective date of the S&P 500 Index change. The trust cannot lend securities or reinvest dividends. IVV and VOO are funded and have portfolio managers in lieu of the more constrained Master Trustee. reinvest dividends and lend stock shares to generate extra income. IVV and VOO also have more latitude on how and when the fund executed its trades regarding announced constituent changes.
The result has been an extra few basis points per year in realized return for IVV and VOO than for SPY during the past 10 years. This table shows the annualized return comparisons for 1- 3- 5- and 10-year periods ending 04/29/2022:
| ETF | Expense Ratio | Assets Under Mgmt. (AUM) | Year-to-Date | 12-monthTot. Ret. | 3-Year Ann.Tot. Ret. | 5-Year Ann.Tot. Ret. | 10-Year Ann.Tot. Ret. | Incept.MM/YYYY |
| SPY | 0.095% | $375 B | -11.86% | 1.26% | 14.30% | 13.81% | 13.67% | 01/1993 |
| IVV | 0.030% | $296 B | -11.84% | 1.31% | 14.34% | 13.89% | 13.77% | 05/2000 |
| VOO | 0.030% | $260 B | -11.84% | 1.31% | 14.34% | 13.90% | 13.79% | 09/2010 |
Although 0.125% per year may not seem enormous, after 30 years of compounding, the gross difference is 45%. This can be a substantial amount when dealing with large sums of money. The bottom line is that if your costs are lower and you own the precisely the same assets, you will make more money per year and over time the compounded difference will expand substantially.
Example No. 2:
The next comparison involves ETFs using the same and best-known small cap index, the Russell 2000 Index published by FTSE Russell Indices. This index bypasses the top 1000 US Equities ranked by float-adjusted market capitalization meeting its requirements in order to track the 1001st through 3000th such stocks.
The oldest and most popular Russell 2000 ETF is IWM, the iShares Russell 2000 ETF owned by Blackrock. Its direct competitor tracking the identical index’s holdings and weights is VTWO (“vee 2”), the Vanguard Russell 2000 ETF. VTWO has an expense ratio of 10 basis points just over half of the 19 basis points charged by IWM.
There is a technical disclosure on the fungibility from one to the other. Since some of the smaller stocks may be difficult to acquire at times, both ETFs’ perspective allows for optimization and substitution so that they may not actually hold the identical securities and weights on a given day but are expected to substantially provide the same price-and-yield returns as the Russell 2000 Index. Let’s look at the comparison table:
| ETF | Expense Ratio | Assets Under Mgmt. (AUM) | Year-to-Date | 12-monthTot. Ret. | 3-Year Ann.Tot. Ret. | 5-Year Ann.Tot. Ret. | 10-Year Ann.Tot. Ret. | Incept.MM/YYYY |
| IWM | 0.190% | $55 B | -15.82% | -17.37% | 6.98% | 7.14% | 10.06% | 05/2000 |
| VTWO | 0.100% | $6 B | -15.82% | -17.27% | 7.14% | 7.26% | 10.10% | 10/2010 |
Once again, the less expensive alternative has provided higher returns.
Example No. 3:
Turning to fixed income, there are two long-term US Treasury Bond ETFs benchmarked to the Bloomberg Long-Term Treasury Index that has been in existence for more than 10 years. The older of the two is SPTL, the SPDR Portfolio Long Term Treasury ETF. The newer entry, VGLT, the Vanguard Long-Term Treasury Index ETF charges a 50% lower fee, just 4 basis points as compared with the 6 basis points charged by VGLT. There is another technical disclosure to be made herein that the holdings of SPTL and VGLT may not be 100% identical at all times. It is generally not possible to perform full replication with most bond index ETFs. Therefore, both sponsors use a combination of partial replication and optimization to produce results substantially similar to the price and yield performance of the Bloomberg Long-Term Treasury Index. At times, one or the other may be higher before expenses are taken into account. Here is the comparison:
| ETF | Expense Ratio | Assets Under Mgmt. (AUM) | Year-to-Date | 12-monthTot. Ret. | 3-Year Ann.Tot. Ret. | 5-Year Ann.Tot. Ret. | 10-Year Ann.Tot. Ret. | Incept.MM/YYYY |
| SPTL | 0.060% | $5.4 B | -17.29% | -11.30% | 0.97% | 1.89% | 2.61% | 05/2007 |
| VGLT | 0.040% | $3.8 B | -17.26% | -11.26% | 0.97% | 1.87% | 2.65% | 11/2009 |
This comparison in net total returns is not nearly as cut-and-dried as it was with the two equity ETF comparisons. Bond trading and liquidity are much less standardized and dependable than in the equity markets. In this case, the older and more expensive ETF from SPDR actually outperformed in the 5-year period and tied in the 3-year period. VGLT outperformed in the two most recent periods and the 10-year period so coupled with the lower fee, VGLT still comes out as the better buy over SPTL in this analysis based on this week’s data. For someone who owns SPTL already, however, there is probably not enough of a difference to justify making the switch.
Example No. 4:
The most dramatic example of needing to read the fact sheets and prospective deals with Exchange Traded Products (ETPs) that hold depository receipts on gold bullion. Most advisors and investors believe that GLD, SPDR Gold Trust by SSgA is the ETF they should buy if an allocation to gold needs to be added or increased in a portfolio. They are wrong on both counts!
ETFs are 40-Act mutual funds whose shares trade on the exchange and do not except daily cash redemptions except in creation-size units. Not all ETPs are ETFs.
GLD is not an ETF at all. Its structure is a grantor trust and its tax treatment when sold is at the higher level of a collectible in lieu of capital gains on a stock (the IRS considers gold a collectible and taxes capital gains on collectibles at a higher rate for many investors – you may wish to consult a tax expert).
Now, let’s get back to fees. GLD, SPDR Gold Shares, is by far the largest Exchange Traded Product to own gold with nearly $60 billion in the trust. It is also the most widely traded, a point often used to justify paying its high fee of 0.40%.
The fact is that the three major alternatives that have been in the market for more than three years:
- IAU, iShares Gold Trust by iShares (0.25%);
- GLDM, SPDR Gold Shares Mini, also by SPDR (0.18%); and
- BAR, Granite Shares Gold Trust by GraniteShares (0.17%)
A new entry from Blackrock iShares, IAUM, the iShares Gold Trust Micro, is even less expensive with an annual expense ratio of 0.15%. So now, BAR has substantially lower fees, the last two more than 50% less than GLD.
Although the SPDRs and iShares press releases at the time GLDM and IAUM were released state that these “mini” and micro products, brought out at lower prices-per-share were “designed for smaller investors”, there has been ample liquidity for purchase without causing market impact in IAU, GLDM, and BAR in three years of trading. I suspect the same will be true of IAUM. Let’s take a look at the track records of all five products:
| ETF | Expense Ratio | Total Asset Value | Year-to-Date | 12-monthTot. Ret. | 3-Year Ann.Tot. Ret. | 5-Year Ann.Tot. Ret. | 10-Year Ann.Tot. Ret. | Incept.MM/YYYY |
| GLD | 0.40% | $66 B | 3.48% | 5.28% | 13.19% | 7.92% | 0.86% | 11/2004 |
| IAU | 0.25% | $32 B | 3.52% | 5.41% | 13.38% | 8.08% | 1.01% | 01/2005 |
| GLDM | 0.18% | $5 B | 3.56% | 5.70% | 13.64% | N/A | N/A | 06/2018 |
| BAR | 0.17% | $1B | 3.70% | 6.38% | 13.48% | N/A | N/A | 08/2017 |
| IAUM | 0.15% | $1B | 3.58% | N/A | N/A | N/A | N/A | 06/2021 |
This chart makes it clear that bigger and older do not constitute better when it comes to wealth accumulation. IAU charges 15 basis points less per annum than GLD and its 10-year annualized return is 15 basis points higher. The difference is even slightly higher when the absolute numbers are larger. Across the board, there’s not a precisely proportionate correspondence but it is close. There are subtle differences in the precise nature of the gold being held and some products have gold only in London vaults and others in vaults around the world. All of the above ETPs are backed by gold bullion. BAR has the added twist of issuing shares backed by physical gold shares already in a vault but that is of little consequence to most investors.
In general, however, it is clear that IAU has systematically recorded greater rates of wealth accumulation for its owners than GLD. In 2017, the price competition became even more focused with the launch of BAR by GraniteShares and SPDRs’ counter-launch of IAUM. Now, iShares has a “Micro” ETP, IAUM that now has the lowest fee so, on the surface, seems like the best alternative. An investment adviser buying new shares of GLD for a client may have some tough questions to answer if the client eventually discovers that the same exposure could have been bought for the account at less than half the cost.
I need to mention that there are a few exceptions to the above statements. Mega-institutions such as sovereign wealth funds and huge pension plans might indeed need the liquidity present in GLD or IAU in order to avoid moving prices substantially. More commonly, hedge funds with very short-term holding periods do not care about fees and focus on mega-liquidity instead. There is also an application that some wealth managers may deem appropriate at times for their clients, a buy-write strategy, holding the gold ETP while writing options on the ETP. Only GLD and IAU currently have listed options and GLD options are much more liquid. If the advisor deems that the options income expected to be derived will be greater than the additional fee costs added to the expected tax on the options income, GLD might be the ETP of choice.
Conclusions
Please keep in mind that fees are generally not the only consideration for investors. For example, there are equity and fixed income ETFs with similar names that can have very different methodologies resulting in very different sets of holdings. In those cases, focusing on fees alone can lead to not buying the products most suitable to the needs of a given investor.
In the four examples just reviewed, however, fees are the most material differences between older ETFs and ETPs and less expensive alternatives. In fact, where subtle differences do exist, they tend to be attributable to better operational efficiencies built into the newer products. This creates a disconnection between perception and reality that can be costly to investors. The first association with buying an S&P 500 ETF might be SPY. Ask an advisor about small cap exposure and the most popular answer will probably be IWM. Advisors should take the fiduciarily responsible step of going beyond memory to review the alternatives using sources like ValuEngine, ETF.com, and ETF database in order to find out how many ETFs fit the category, what the differences are, and which fit the clients’ best interests.
Here’s a thought that may bring smiles to some faces. Anyone who still believes in the Strong Form of the Efficient Market Hypothesis should take a look at more advisors buying new shares of SPY, IWM, SPLT, and GLD for their clients rather than the less expensive and otherwise identical alternatives. When one considers that the rise of the ETF industry is built on its operational efficiency, the irony of the situation becomes noteworthy.
I own shares of VOO, BAR and IAUM.
Disclaimer: Always read the fact sheets and/or summary prospectus before buying any ETF. Past performance is not necessarily indicative of future results.