My 3 Top Tips For ETF Model Portfolios

ETFs are hard to analyse, tracking error is not the panacea and low cost does not mean quality  

Reviewed by: Jean-Rene Giraud
Edited by: Jean-Rene Giraud


Exchange traded funds are without a doubt one of the most fascinating developments the financial industry has witnessed over the last 20 years. In essence, ETFs allow investors to gain access to the performance of an index at a reasonable cost without the hurdle of investing in the individual constituents. Structured in the form of mutual funds, they offer the high level of investor protection. They can also be traded in the blink of an eye on the major stock exchanges, and are cleared and settled like the most common securities.

Delivering a delta-one exposure to an index is, however, far from being an easy task. Transaction costs, dividend treatment, tax assumptions, replication strategies, discrepancies between exchange trading times and other complexities inherent to the management of an ETF have resulted in those funds requiring a fairly complex financial, regulatory and operational set-up.

As a result, ETFs are not easy to analyse and select. Various mainstream indices are available via more than 25 ETFs in Europe, and they come in different shapes, forms, sizes and costs.

Here are our top lessons when it comes to helping very large investors using ETFs in their portfolios.

1) ETFs Are Simple By Definition But Hard To Analyse

Highlighting the key factors we believe should be considered when assessing an ETF has proven a much more difficult task than we expected.

Incorrect or out-of-date information and vastly different specifications per ETF took up a lot of our time over the past few months. We also often found that publicly available data was simply wrong. And computing garbage data is a waste of effort and resources. Therefore investors should not underestimate the risks of assessing ETFs on the sole information provided by public sources.

2) Tracking Error Is Not The Panacea

Most investors rely on tracking error as the main measure of replication quality. Tracking error is the daily standard deviation of returns between the ETF's net asset value and the index. It offers a reliable measure of the value of investor's holdings and how much of a gap in daily return might be expected.

Track Difference Track Error

But tracking error does not provide a complete picture of replication quality.

It is of utmost importance for investors to also assess the level and robustness of the tracking difference, which is the only true measure of the total cost of ownership. This measure is the difference in compounded returns over a one year period.

Besides, the tracking error fails to capture the extent to which infrequent but large differences in returns can impact your investment. Our studies shows that ETFs may have more or less limited capacity to mitigate those large differences though still exhibiting comparable tracking errors.

3) Low Expense Ratio Does Not Equal Quality

We have not identified any clear relationship between the visible expense ratio and our measures of replication quality. Selecting an ETF on the basis of management fees can therefore be seen as a naïve, if not a blind approach to ETF investing.

The following picture displays the dispersion of ETFs when it comes to tracking difference with reference to the expense ratio. It clearly demonstrates that all ETFs are not equal and that while low annual fees can certainly bring down the cost of ownership, the expense ratio cannot be considered as a determining factor for choosing an ETF. Investors will therefore have to assess those various risk and performance indicators depending on their specific circumstances.

Expense Ratio

What To Focus On

The complexity of ETFs and what is required to analyse them goes far beyond what this article contains. Our recommendation for ETF investors is to follow a number of very simple rules:

1) Understand the index you invest in and widen your universe

2) Holistically assess the funds available and look at

i. Counterparty risk rather than whether it is synthetic/physical

ii. Fund size relative to your investment amount

iii. Tracking difference and its stability over time

iv: Average daily tracking error

v: Primary and secondary market liquidity

3) Stay away from bold conclusions

4) There is no such thing as one size fits all

5) Use independent and verified sources of information

Jean-René Giraud is founder and CEO of Koris International

Koris International has been advising large institutions in the field of dynamic asset allocation and portfolio construction for over twelve years.

In response to the serious issues above, we launched a website that offers free and unbiased analysis on the most important ETFs in Europe, it can be found on