Whether you are a long-term devotee of passive and exchange-traded funds (ETFs), or just beginning to dabble in those spaces, understanding tracking error is a worthwhile use of your time.
"Morningstar has often touted the value of investing in low-cost index ETFs as a means for investors to gain exposure to an broad basket of securities mirroring a stock index. Tracking error is one way to gauge how close an ETF mirrors the performance results of this benchmark," says Ian Tam, Morningstar Canada’s Director of Investment Research.
"A good index ETF will show very little tracking error, a sign that they have been successful in mirroring the index. For large cap indices like the S&P 500 or MSCI World, it's likely that ETF providers will have very little tracking error. However, for more esoteric indices, like those covering less liquid markets, you'd expect tracking error to be higher, but still worthwhile looking at if there are multiple providers tracking the same index."
Tracking difference and tracking error are the two most important measures of the ability of an index fund or ETF to track its benchmark. While the two terms are often used synonymously, they are not the same and should not be treated interchangeably.
What is Tracking Difference?
Tracking difference is the absolute difference between the returns of the fund you invest in and those of the benchmark at the end of a chosen period of comparison.
For instance, if you hold a passive fund for a total of five years, the tracking difference is a single measure calculated at the end of the five-year period indicating your returns relative to the benchmark you have been following. The final figure is net of fees and charges on your investment.
What is Tracking Error?
Tracking error, meanwhile, is different. That is a measure of how well your fund tracked its benchmark during the period of investment itself. Even funds that supposedly track indices perfectly will behave slightly differently to their benchmark during the course of the holding period. It’s those moments of deviation that will contribute to an analysis of tracking error. Put another way, tracking error is the annualised standard deviation of tracking difference data points for the given time period.
Overall, you might think of tracking error as a measure of volatility, then. To be clear, though, we are not talking about volatility in the fund’s underlying holdings. What we are focusing on is the deviation between the behaviour of your fund and its benchmark overall.
For example, if your index fund had a small tracking error during the five years you held it, it will have followed the benchmark very closely. A large tracking error would indicate perhaps some serious deviation from the benchmark’s trajectory.
What’s the Difference?
There’s a good way of summarising the difference between tracking difference and tracking error. Think of two rally cars going head to head in a race. Car number one is the benchmark car, while car number two is your fund.
To the onlookers observing the race and invested in the two drivers, tracking difference is the equivalent of the time difference captured when the cars cross the finish line.
Tracking error, meanwhile, is the equivalent at examining the race footage back to see how close the two cars were to each other during the race itself. Perhaps there might have been a moment where the fund car handled a hairpin bend particularly badly, and had to recover from aggressive oversteer at the very edge of the track boundary. You get the picture.
Do Tracking Error and Tracking Difference Matter?
As Morningstar’s director of passive strategies Ben Johnson explains, it is worth noting the relationship between tracking error and tracking difference.
"Tracking error can be used to develop a degree of confidence in predicting tracking difference. It is often assumed that a high amount of tracking error means that the fund has performed poorly relative to its benchmark, but it is not necessarily the case," he says.
"In fact, a high tracking error doesn’t necessarily imply anything about the magnitude of tracking difference over a given time period. At the same time, a low tracking error doesn’t necessarily imply a low tracking difference (in absolute terms)."
Whether these terms matter to you or not depends on what’s important to you. If total return is your main concern, then the tracking difference is perhaps more important. If you want your fund to stay consistent throughout your investment time horizon, tracking error may be more relevant.
Going back to the car analogy, regardless of your fund car’s hairy moment, you might well argue what matters is whether it crossed the finish line in a good time (tracking difference).
However, sometimes the journey is just as important as the destination. For passive investors who prefer to disengage and watch their investments from a distance, then, tracking error provides another opportunity to get animated, and another measure of fund behaviour to understand.
What Else Matters When Evaluating an ETF?
Apart from tracking difference and tracking error, Johnson recommends investors consider:
- Trading costs, including commissions, bid-ask spreads, and market impact ETFs’ market price relative to NAV over time (i.e. premiums and discounts);
- Counterparty risk; and
- Tax considerations.
If you’d like a deeper dive into these concepts, how they’re calculated, and how they work, you can download a paper on the subject here.
