Tracking Difference vs Expense Ratio: What You Really Pay

Fund pickers argue about expense ratios because the number is printed on the page. But the expense ratio is a cost input, not the result. Two funds can charge identical fees and deliver noticeably different outcomes, because what you actually receive is the index return minus everything the fund lost along the way — plus the odd things it earned. That final number has a name: the tracking difference. Once you can read it, the fee-versus-no-fee argument becomes a much more precise question about where the gap comes from.

This article defines the measurement, works a 20-year example with every assumption on the table, and explains the full wedge between an index and the fund that follows it.

The two measurements

Tracking difference (TD) is the arithmetic gap between what the index returned and what the fund returned, usually over a year: an index up 8.00% with the fund up 7.95% gives a TD of −0.05%. It is a signed number, and for most equity index funds it is negative — funds rarely beat their own benchmark. Tracking error (TE) is different: it measures the volatility of that gap, not its size. A fund can have tiny tracking error and a persistent large shortfall, or larger tracking error that nets out near zero. For a buy-and-hold investor, tracking difference is the number that compounds; tracking error matters mainly for strategies that need the fund to hug the index consistently — hedged positions, for example.

The wedge: everything that separates index from fund

The expense ratio is the most visible item, but it is rarely the whole gap:

• Fees — the management expense ratio, charged daily against assets.
• Replication style — funds holding every constituent ("full replication") track tightly; funds that sample a subset (common in bond and emerging-market funds, where some holdings are hard to buy) drift more.
• Cash drag — dividends arrive on different dates than index reinvestment assumes; new investor inflows sit in cash for days. In a rising market this shaves return.
• Trading costs — commissions and spreads when the fund rebalances or handles creations and redemptions. Bigger for thinly traded emerging-market stocks than for mega-cap U.S. names.
• Securities lending income — funds lend shares to short sellers for a fee and usually keep most of it. This is a positive contributor that partially offsets fees; on large, heavily shorted indexes it can be material.
• Futures and roll costs — for commodity funds, the "index" assumes ideal rolling of futures contracts; the fund rolls in the real market, and contango costs show up here.

This is why the fee is a floor, not a forecast: a 0.20% emerging-market fund can lag its index by 0.50% in a given year, while a 0.03% S&P 500 fund with lending income can land within a hair of zero or occasionally above it.

A worked example, twenty years long

Assume a broad index returns 8.00% every year for twenty years (chosen for arithmetic clarity, not as a forecast), and you invest $10,000 and reinvest everything:

Three things to notice. First, the fee difference is 0.17 points, but the terminal gap is about $1,100 — roughly 11% of Fund B's total gain, gone to a gap most investors never look at. Second, the wedge is not proportional to the fee: Fund B recovers 0.02 points from lending income and still trails by more than three times Fund A's shortfall. Third, at five years the gap is $87 — tracking difference is a slow tax that only compounds into real money over horizons measured in decades, which is exactly why annual reviews miss it.

One wedge, unpacked: the dividend that arrives late

Cash drag sounds abstract, so here is the smallest version of it. Suppose an index constituent pays a 2% dividend on the 15th of the month. The index mathematics assumes that money is reinvested in the whole index at that day's closing prices. The fund, however, cannot reinvest cash it has not received: the dividend pays out to the fund's account, gets pooled with creations and redemptions, and is typically deployed a few business days later. If the market rose 1% in those days, the fund bought its replacement shares 1% more expensively than the index says it did. Repeat that across a few hundred dividends a year and a few thousand constituents, and the tiny slivers add up to a fraction of a percent annually — invisible in any single statement, perfectly visible in the fund-versus-index table. Nothing was mismanaged; the gap is structural, and it grows when markets rise and shrinks when they fall, which is also why the wedge is not a fixed tax.

What typical gaps look like

Realized tracking differences cluster differently by category, and the pattern is stable enough to use as a sanity check. For large, liquid U.S. equity index funds, five-year tracking differences within 0.10% a year of the expense ratio are routine, and the cheapest funds often land within a few hundredths — occasionally above zero once lending income is counted. Broad developed international funds tend to run slightly wider. Emerging-market funds have historically shown gaps a multiple of their fee — gaps of 0.3% to 0.8% a year against headline fees of 0.1% to 0.2% were common in difficult years for frontier trading — though the cheapest full-replication EM products have narrowed this considerably. These are observed ranges, not guarantees, and any individual fund in any individual year can land outside them; the point of the check is not to predict the gap but to flag one that is inexplicably wide.

Where to find the number

Fund companies publish it, though rarely on the marketing page. The fund's annual report contains a performance table comparing the fund's return to its benchmark index over 1, 5, and 10 years; subtract one from the other and you have the realized tracking difference. Some prospectuses include a "comparable index" return precisely because the official benchmark may be narrower than what the fund actually tracks. Independent data providers and screeners also show "benchmark gap" columns. The discipline is simple: before buying, pull the 5-year table, subtract, and ask whether the gap is roughly the fee, comfortably smaller, or mysteriously larger.

When the fee is still a good shortcut

For large, liquid, full-replication indexes — U.S. total market, S&P 500, developed large-cap — the industry has converged: fees are near zero, lending income offsets sampling noise, and realized tracking differences cluster tightly around the expense ratio. There, comparing fees is a nearly complete comparison. The shortcut breaks down for:

• Emerging markets and small caps, where trading costs and sampling widen the wedge well beyond the fee.
• Bond funds, where index yields assume frictionless trading of illiquid issues.
• Commodity funds, where futures rolls make the printed index almost fictional.
• Synthetic (swap-based) funds, common in European UCITS structures, which can post near-zero tracking difference by contract — at the price of counterparty exposure that the index never had. U.S. mutual-fund law effectively prevents this structure domestically.

Taxes: the gap behind the gap

Tracking difference is measured pre-tax, and two funds with identical TDs can still land different after-tax outcomes. A traditional index mutual fund must distribute realized capital gains when its portfolio turns over or when investors redeem in size — taxable events for everyone holding it. The ETF wrapper avoids most forced distributions through in-kind creation and redemption. So for a taxable account, the question is not only "what did the fund return" but "what did the fund make me realize." The expense-ratio math and the capital gains explainer cover both halves of that calculation.

A five-minute checklist

1. Identify the fund's actual benchmark index, not the marketing category.
2. Pull the 5-year fund-versus-index table from the annual report.
3. Subtract to get the realized tracking difference; compare it to the fee.
4. If the gap is much worse than the fee, find out why (sampling? cash drag? structure?) before assuming it will improve.
5. For taxable money, check the distribution history on top of the TD.

The takeaway

The expense ratio tells you what the fund charges; the tracking difference tells you what the fund delivers. In the simplest U.S. large-cap products the two nearly coincide, which is why fee-shopping works there. Everywhere the index is harder to hold — smaller markets, bond and commodity exposures, synthetic structures — the wedge between the printed fee and the realized gap is where fortunes quietly diverge. Measured over the twenty years a retirement portfolio actually runs, a third of a percent per year, repeated, becomes the difference between a number and a noticeably smaller number. The blog index collects the rest of the fund-selection series.