You need to know the full picture before you invest your hard-earned cash into any investment. And with exchange traded funds (ETFs), one of the key elements is tracking difference - the difference between the performance of an ETF and the index that it aims to replicate.
If tracking difference is not disclosed on the ETF's factsheet, the sensible thing to do is to get your hands on some credible third-party data. This should help you get an unbiased snapshot of funds so that you can see if there's anything you need to know that isn't obvious from the marketing bumf.
But, unfortunately, we have some bad news. According to ETF providers, even the figures from big-name providers such as Morningstar, Lipper and Bloomberg can't be trusted 100 per cent of the time as they are prone to mistakes, which presents another obstacle between you being able to distinguish good funds from the bad.
They say there are a number of deep-set problems associated with ETF information from these data providers that cause severe inaccuracies and can distort tracking difference and tracking error, the main indicator of funds' performance and cost, by up to 100 basis points, or 1 per cent in the worst cases.
Poor 'scrubbing' of data, failing to take 'fund holidays' into account and rounding numbers to too few decimal places are the most commonly reported problems with third-party data, and can make funds look much better or worse than they really are.
Most ETF providers have a long way to go to improve transparency, but some have been keen to tell the IC about the failings of external data.
Michael John Lytle, chief development officer at Source, told us that data providers have supplied the national press with "wildly incorrect" performance data on its ETFs on a number of occasions. For example, last month the IC ran some Lipper data showing that the two-year tracking difference on Source's FTSE 100 ETF was -2.19. But Source claims this is incorrect and that the real tracking difference is -0.91, almost half that shown by Lipper.
He says this is down to Lipper taking snapshots of the data on a day when the fund was closed, but the index was still tracking, meaning the difference between the two was much greater than usual. These are called 'fund holidays' and occur on bank holidays or during a national crisis when stock markets close as a result.
Detlef Glow, head of EMEA research at Lipper, admitted issues such as fund holidays were a problem and caused tension with ETF providers. "They are always complaining about data," he said. "I can understand why they get annoyed, but we have to do something to fill the gaps."
He added that an ETF provider had recently tried to take Lipper to court over claims they had used the wrong benchmark and made their funds look like they weren't performing as well as they were.
But Ben Johnson, head of passive research at Morningstar, said he and his research team were working towards improving the accuracy of its data and making it as "investor friendly" as possible.
Source also claims Bloomberg's 'NAV Track. Error' measure, which it says has no definition as to its calculation, period or methodology, has also significantly distorted data on its funds. As shown in Image 1, below, Bloomberg states the tracking error on the S&P 500 Source ETF is 0.692. But Source claims its tracking error is 0.0178 per cent (calculated by a standard deviation of daily performance differences, annualised over the full calendar year).
A spokesperson from Bloomberg said: "We support daily, weekly and monthly tracking error calculations on the Bloomberg terminal. We decided to post a weekly frequency as the default for our ETF description pages because we want our clients to see a consistent snapshot of fund performance across all ETF securities. However, Bloomberg's tracking error calculation is configurable, so a client can adjust it depending on the method they deem most appropriate."
But a spokesperson from Source said: "We would be concerned if people started drawing conclusions from this headline number." But they also admitted that while the measure is not one that investors use, they had not "had time" to deconstruct their calculations.
Mr Lytle said some data providers' inexperience may be to blame for their mistakes: "Some of these data providers have been producing ETF data for only a year or so. Their history is with actively managed funds - giving investors information to help them distinguish between quality and average performance."
"They need to be much more rigorous when scrubbing their data, but to do this would be a huge undertaking, as there are 1,700 exchange traded funds on the market coming from 39 providers."
And Ursula Marchioni, director of Emea investment strategy at iShares, said mistakes created by data providers are very common due to "IT problems". But she added that sometimes ETF providers too can make mistakes, which can also cause discrepancies with third-party data.
The European Securities and Markets Authority (ESMA), the European regulating body in charge of the UK's Financial Services Authority (FSA), has called for a single, universal platform where investors can access consistent ETF data.
But providers say it will be years before such a platform emerges. In the meantime, the best you can do is research scrupulously and check stats out with providers if you aren't sure about them. There are some excellent products on the market that could make profitable investments, so double check before you let duff data stand in the way of you and ETFs.
