Why the cheaper ETF lagged more
When several ETFs track the same index, it is tempting to pick the one with the lowest expense ratio. A year later, though, the cheaper ETF sometimes sits further from the index. The stated fee is only one of several reasons an ETF moves differently from its index.
This guide calculates, using two hypothetical ETFs, what creates tracking difference, the gap between an ETF's return and its index's return. It also explains how tracking difference differs from tracking error, how to compare on the same basis, and what to look for in fund reports.
ETFs and expense ratios
Investor.gov describes an ETF as a fund holding a basket of assets whose shares trade on an exchange like stocks, with the combined holdings known as its portfolio. Most index ETFs are designed to follow a specific index, but an index is a calculation with no costs and no cash, so a real fund starts from a different place.
As Investor.gov's entry on expense ratios explains, the expense ratio is the share of fund assets taken each year for management and administration. Investors do not pay it separately; it is built into the fund's price, so it is easy to overlook. Costs outside the expense ratio, such as trading and taxes, also affect returns.
Tracking difference and tracking error
Tracking difference is the ETF's return minus the index's return over a period. If the index rose 10% in a year and the ETF rose 9.5%, the tracking difference is −0.5 percentage points. It shows how much less investors actually earned than the index, which matters most to long-term holders.
Tracking error measures how much daily or monthly return differences fluctuate. A fund can have low tracking error yet lag a little every period, building a large tracking difference; another can have high tracking error yet match the index on average. The two figures answer different questions.
Two hypothetical ETFs
The table breaks down one year of performance for hypothetical ETF A (expense ratio 0.20%) and ETF B (0.10%) tracking the same index. The index is assumed to have returned 10.00% with dividends reinvested.
| Factor | ETF A | ETF B |
|---|---|---|
| Index return | 10.00 | 10.00 |
| Expense ratio | −0.20 | −0.10 |
| Cash holdings (drag in a rising market) | −0.20 | −0.30 |
| Trading and rebalancing costs | −0.05 | −0.15 |
| Securities lending income | +0.08 | 0.00 |
| Dividend withholding tax | −0.10 | −0.05 |
| ETF return | 9.53 | 9.40 |
| Tracking difference | −0.47 | −0.60 |
ETF B charges half the fee but holds more cash and pays more to trade, so it ended further from the index. The figures illustrate the factors.
Factors beyond the fee
Cash arises from dividends and redemptions. In a rising year it earns little and causes lag; in a falling year it cushions losses. Trading costs when index constituents change, and sampling strategies that track the index with only some of its stocks, also create gaps.
Securities lending, lending held shares to other institutions for a fee, can add return and offset part of the costs. For ETFs holding foreign stocks, withholding tax on dividends varies by fund structure and country, so the dividends actually received can differ even for the same index.
Comparing on the same basis
When comparing ETFs, first make sure the index return is a total return with dividends reinvested rather than a price return, and that the ETF's return assumes distributions are reinvested. Mismatched bases create a gap the size of the dividend and distort tracking difference.
- Find the ETF's and the index's total returns for the same period in the provider's materials and calculate tracking difference yourself.
- Compare tracking difference over several periods, such as three and five years, not just one.
- Check turnover, cash weight, and replication method (full, sampled, or synthetic) alongside the expense ratio.
- See whether the fund lends securities and how much of the lending income stays in the fund.
- Add the gap between market price and net asset value, and the bid-ask spread, to your real costs.
Adding trading costs
Tracking difference arises inside the fund. When investors buy and sell, the bid-ask spread and commissions come on top, and the market price may trade at a premium or discount to net asset value. For thinly traded ETFs, these costs can exceed the fee.
As Investor.gov's material on funds explains, choosing a fund means weighing costs, risks, and objectives together. The fee is a starting point; what remains in your account is the figure after every cost.
One practical step is to look at the fund's fact sheet over several years and note the gap between its net asset value return and the index each year. A gap that is consistently close to the fee suggests efficient management; a gap well beyond it deserves a closer look.
Open questions
The returns and factors here are hypothetical. Real tracking differences vary year to year with market conditions, index changes, taxes, and fund size, and past tracking difference does not guarantee the future.
For a real ETF, check the provider's latest reports and prospectus, noting when each was last updated. This guide does not recommend buying or selling anything.