The root cause of tracking error is inherent in the nature of all indexes—which are theoretical constructs only, free of any real-world costs. They cannot be invested in directly. Managers of index funds and ETFs must deal with frictions and challenges that, if not handled well, can lead to substantial performance differences from the pure index. These involve transaction costs, sampling differences, timing of cash flows, and differences in local-market timing schedules. Asset managers who offer index products are constantly improving and refining how they reduce the impact of such frictions.

Tracking error is not how much an ETF’s performance deviates from that of its benchmark index for a given period (that’s known as excess returns). It’s about how much those excess returns oscillate around the portfolio’s benchmark during that period.

Tracking error is measured as the standard deviation of excess returns over time. It’s an indicator of how consistently close or wide an index ETF’s performance is relative to its benchmark. For investors using indexed products, any uncertainty around performance adds uncertainty costs. Depending on an investor’s investment horizon, this uncertainty cost can be even greater than the expense ratio or trading spread.

A primary driver for investors buying an index ETF is to gain exposure to a specific slice of the market. When an ETF closely tracks its benchmark, and an investor who holds it sees that the performance they experience doesn’t deviate much from that of the benchmark, then that investor can feel confident that they’ve gained the exposure they sought.

Investors using index ETFs need to have high confidence that they’ll get their desired market exposure. Any deviation from what the market is returning can add costs. If an investor allocates to an S&P 500 ETF, she is expressing a capital market view: She thinks that the slice of the market represented by this particular ETF will do well. If the fund’s tracking error is too wide, however, she could be correct on her capital market view but not have her portfolio reflect the market’s performance. 

Many equity index fund managers seek to completely replicate their underlying index’s holdings. However, thanks to the negotiated nature of the fixed income market, it’s nearly impossible for a fixed income index fund manager to do the same.

Instead, these managers must sample or optimize their portfolios to match the index as best as they can, matching key characteristics of the index such as duration, credit quality, and yield. Because of this, fixed income index ETFs are more vulnerable to tracking error than equity index ETFs are, and there can often be larger-than-expected differences in returns between the fund and its benchmark and between products that cover the same segment of the market.

Given the distinct challenge of fixed income, conducting due diligence on fund companies is a crucial step toward ensuring that you are allocating to the product with the best chance of replicating benchmark returns.

Low expense ratios are eye-catching, but if a relatively inexpensive ETF also has significant tracking error, the advantage of the low expense ratio can be eroded.

Time is also an important consideration when it comes to tracking error. In the short term, a high tracking error means higher uncertainty in performance, thus eroding the value of a tactical trade. In the long term, tracking error can be used to evaluate how consistently an index fund manager is meeting benchmark returns over time.

So, while a low expense ratio is important, a lower tracking error suggests a reduced likelihood of incurring costs from volatile performance, no matter your investment horizon.

To illustrate why tracking error can make an outsized impact on potential returns, we compare three municipal bond index ETFs, each with holdings across the broad municipal yield curve: Vanguard Tax-Exempt Bond ETF (VTEB), iShares National Muni Bond ETF (MUB), and Schwab Municipal Bond ETF (SCMB). While their inception dates greatly differ, their monthly excess returns and tracking errors over time illustrate key differences between similar products.

The excess returns measure a straight arithmetic performance deviation over the given period. The tracking error—the more meaningful metric—shows the consistency of keeping the fund’s returns in line with its benchmark. In other words, a higher tracking error means a higher risk of being out of sync with the benchmark’s performance. (Full standardized performance data for these ETFs is at the end of this article.)

The charts below show the oscillation of three ETFs’ excess returns over time—and also point to why tracking error is a more meaningful metric than excess returns.

Each investor, of course, would have a unique entry and exit point along these timelines, and these specific points will help determine whether their investment under- or outperforms the relevant benchmark.

Even so, VTEB’s returns are distributed more narrowly around those of the relevant benchmark. At the end of any given period, ETF returns clustering tightly around benchmark returns typically results in lower tracking error.

In other words, greater volatility of excess returns translates into more uncertainty around excess returns. And—again—the tighter the tracking error, the more likely it is that you’ll avoid the cost of deviating from the market’s performance, and the easier it is to formulate thoughtful asset allocation.

SCMB only came to market in October 2022; its track record is relatively limited. Still, as these charts show, its excess returns range and its tracking error both exceed those of VTEB and MUB, thus potentially nullifying the benefit of a lower expense ratio.

This case study illustrates how something that seems as mundane as tracking error can play a crucial role in a fund’s risk and return.

• Tallying the total cost of owning an ETF (article, issued December 2022)
• Case studies in the cost of ETF ownership (article, issued June 2023)
• ETF premiums and discounts, explained (article, issued August 2023)