In a previous article, I discussed the differences between the three main ways that an index ETF can copy its underlying index. Within this, I mentioned the concept of “tracking error”, that is to say a difference over time between the fund’s performance and the performance of the full index.
But why should an investor care about tracking error? For one simple reason: any investment product should act as in a Ronseal advertisement: it should “do what it says on the tin”.
Index funds should do what they say “on the tin”
If an index ETF has a significant tracking error with respect to the index it is meant to be copying, it is not delivering “what it says on the tin”. But since that is exactly why an investor bought the ETF in the first place, that can be a real problem, particularly if the index ETF is underperforming the index after management fees are accounted for.
Tracking error can make a supposed low-cost ETF rather expensive!
After all, the index fund may be cheap in terms of management fees, but may turn out to be relatively expensive for the investor if on top of the management fees there is significant underperformance that exceeds any benefit from buying a low-cost ETF.
In the example below, I show the performance of a Euro high yield corporate bond index (from IBOXX), and the performance (including dividends) of an ETF that is tracking this index using a physical stratified sampling replication method.
As you can see, while the index has gained 10% from the beginning of 2013 to mid-March, the ETF that is tracking it has only delivered an 8.2% return, some 1.8% lower. Of this, the annual management fee accounts for 0.5%, but the tracking error is responsible for the remaining 1.3% gap!