Because ETF investors have become more cost conscious, a fund’s expense ratio and trading costs are always top of mind. Given the focus on due diligence, tracking error—which in some circumstances may increase costs—is becoming more heavily scrutinized.
What is tracking error?
Tracking error is a measure of how consistent a portfolio’s return is with that of its benchmark. The term “tracking error” gets thrown around a lot in the industry, but it can refer to two different measurements, which creates confusion among investors during the due diligence process.
The two most commonly used measures of tracking error when analyzing passively managed strategies are:
- Tracking Return Difference: This is the difference between the index and an ETF’s net asset value (NAV) return. This figure can be negative or positive, and it measures the extent to which an ETF’s return differs from the index it seeks to track over a particular time period. The lower the absolute number, the better. However, tracking return difference is typically a negative number due to the impact an expense ratio has on returns. While an index’s return does not factor in expenses, an ETF’s does.
- Tracking Return Volatility: This measures the variability of an ETF’s excess return, and it is a way to gauge the consistency of a fund’s performance. It is calculated by taking the annualized standard deviation of the difference in the ETF NAV and index returns. A lower figure indicates a more consistent fund performance. This measure is more commonly used in the due diligence process for active management as a way to examine how much risk an active manager is taking versus an index.
Tracking error: More than meets the eye
A large tracking error between a passively managed ETF and the index it tracks may be a red flag. It could signal excessive trading costs or issues relating to fund management.
But tracking error is an inherent feature of investing. To begin with, a passively managed strategy will likely lag behind its benchmark because of fees associated with the fund. Another culprit? Cash drag. If a fund has a cash position of 1% and the market rises 10%, that cash will not benefit from the market jump—resulting in tracking error. While passively managed ETF cash levels tend to be low, they can fluctuate as a result of dividend payments. In these days of record-setting markets, even a low cash position can have an impact on tracking error.
Beyond that, a passively managed ETF’s replication method, weighting methodology and rebalancing schedule play a role in tracking error. For instance, an ETF that uses full replication—holding all the securities at the same weight as its underlying index—typically has a lower tracking error than an optimized portfolio that closely matches the characteristics of the underlying index but doesn’t hold the exact same securities. While constraints are used in the optimization process to mitigate tracking error, they cannot eliminate this tracking difference.
Tracking error can also be higher for strategies that hold less liquid securities, such as high yield bonds or emerging market debt, because transacting in these securities tends to involve higher costs.
To assess tracking error, look under the hood
Assessing tracking error requires an in-depth analysis, especially when the error appears to be large. It’s important to consider multiple time periods and data points. By isolating—or “cherry picking”—a particular period, you can be left with incomplete or misleading information. When an analysis is expanded beyond a single period to instead focus on rolling periods, trends may be revealed, like whether a fund’s tracking error is improving or deteriorating—or whether an aberration caused a sudden shift.
An example of this can be seen with the SPDR® Bloomberg Barclays High Yield Bond ETF (JNK). JNK’s annualized performance over the past three years has been 6.81%, or 50bps less than its benchmark.1 But in early 2016, State Street Global Advisors made changes to the portfolio management process to address this tracking error. Since the changes were fully implemented, JNK has delivered an annualized tracking error (return difference) of 45 basis points from March 31, 2016 until June 28 of this year.2
If an investor studied just one time period, they could have overlooked the trend of improvement and been misled as to JNK’s current management process and potential for tracking error. Similar to measuring an ETF’s performance, capturing trends in tracking error are far more important than looking at just one single period.
Digging deeper with ETF tracking error
When gauging an ETF’s performance, it’s important to not only look at its expense ratio, but also to understand that there are other variables that add to the total cost of ETF ownership. Different approaches to managing an ETF dictate how closely a fund may track its index—and how well a fund may suit your particular portfolio.
1Bloomberg Finance L.P., as of 06/29/2019.
2Bloomberg Finance L.P., as of 06/29/2019.
The amount of cash that a fund holds at a specific point in time.
The portion of your investment that the fund charges on an annual basis for management fees—from trading and marketing expenses to custodial and index licensing fees.
Net Asset Value
The price of a share determined by the total value of the securities in the underlying portfolio, less any liabilities.
Measures the historical dispersion of a security, fund or index around an average. Investors use standard deviation to measure expected risk or volatility, and a higher standard deviation means the security has tended to show higher volatility or price swings in the past.
Tracking error is a measure of how consistent a portfolio’s return is with that of its benchmark. In reality, no indexing strategy can perfectly match the performance of the index or benchmark, and the tracking error quantifies the degree to which the strategy differed from the index or benchmark, by measuring the standard deviation between the two values, annualized.