Sunday 28 June 2015

Do ETFs Behave Differently? (Portfolio Buy)

As usual, a DIY Income Investor portfolio sale - viz. the recent Berkeley Group cash-in - leads to the difficult decision of what to buy now. This is always a challenge because, having been successful at making money (quite a lot in that particular case) you don't really want to blow it on something that will then tank.

This decision is all-the-more difficult because the strategy I have adopted in the last year or so (which involves building up the proportion of ETFs - Exchange Traded Funds) does not seem to be working very well.

Regular reader may recall that I am trying to make the portfolio much easier to manage with the idea of passing on a portfolio to someone less interested in investing than myself. I immediately excluded the option of Unit Trusts or heavily marketed 'Funds' because of the usually high fees (plus often poorly documented additional costs) or the 'black box' nature of the managers' investment choices.

One simplification option adopted by many is to go for Investment Trusts; however, I am still not convinced by them - they seem to share many of the drawbacks of Unit Trusts, although possibly to a lesser extent.

I have therefore embarked on an experiment to see if my yield-targeting investment approach - a logic that has been successful over several years for individual shares and bonds - can be adapted to  income-oriented ETFs. ETFs do have management fees and other operating costs but these are much less than Unit Trusts or Investment Trusts. Moreover, because their investment process is fairly mechanical (one of the reasons why they are cheap to operate), you know what the managers are doing.

In addition, income-oriented ETFs, quoted in London (which is my 'home' exchange) are available for both dividend shares and fixed-income securities (my two key asset classes) in different geographical areas, allowing me to diversify both geographically and currency-wise. Most importantly, the ETFs are well diversified - although the components of each specialised ETFs are, by definition, similar.

Ideal, I thought, and started buying: ETFs now account for half of the portfolio. 

However, the performance of the DIY Income Investor portfolio has continued to decline, as the strategy of ETF-ication has proceeded. Is this due to the market conditions or my strategy? The Total Return for FY 2014/15 was only 8.7%, compared with 19.5% and 29.2% in the previous Financial Years. The running Total Return for 2015 is only around 6% - roughly the estimated income yield. This means that overall, there is no capital gain!

Well, what did you expect?

First, let's recap on the basic DIY Income Investor investment approach: buy (hopefully) good-quality high-yield securities that are out-of-favour and hope to make a capital gain when the market reappraises their value. Holding costs (until the happy day you can cash in your re-valued chips) are covered by the income from the attractive yield.

By analogy, this strategy translates to: buy the (currently) highest-yield ETF, because that region or flavour (i.e. dividend share or fixed-income/bond) is out-of-favour and will - in due course - be revalued by the market. ETFs are diversified, so individual security risks are less important.

That's the theory, anyway. But how might ETFs behave differently?

First, ETFs are not fixed. The mechanical rules that dictate what is held by an ETF are usually based on a specific index. The components of that index will change over time. So what you buy on Day 1 is not necessarily what you hold a year later. This means that the price recovery - if and when it comes - might not match what you were expecting. The value/yield relationship may change gradually as the ETF components are changed. This may be a slow process but it may change the assumed basic relationship between yield and potential for price recovery.

Second, ETFs are diversified - one of their key characteristics. This means that there will be some serious averaging-out of price moves. Good, if you like stable prices - ETFs are less volatile; bad if you are hoping to make a killing.

Third, the time-frame may be different compared to individual stocks and shares. With the geographically-specialised ETFs I invest in, the higher relative yields are due to market perceptions of the future economic conditions in those regional markets. These conditions take longer to turn around than individual companies (or even countries), so a price recovery will take longer.


Certainly, the volatility of the DIY Income Investor portfolio has dropped significantly since 2013. And the ETF share of the portfolio is showing less net capital loss than the non-ETF portion (although not by a large margin). This ETF capital loss is only of the order of 4% - perhaps to be expected when investing in out-of-favour regions?

At the moment the perceived advantages of ETFs seem to outweigh the disadvantages. Fixed-income is currently under-represented in the portfolio - and I try to keep a roughly 50/50 split between dividend shares and fixed-income/bonds. The purchase decision is therefore simple: the highest-yielding fixed-income ETF that fits my investment approach is one I already hold a lot of: the iShares Emerging Markets Local Government Bond UCITS ETF (LSE:SEML). Bloomberg indicate a forecast yield of 6.2% and an Expense Ratio of 0.50%.

To quote iShares, this ETF - aims to track the performance of the Barclays Emerging Markets Local Currency Core Government Index as closely as possible. It holds lots of government debt from places like Brazil, Turkey, Poland, South Africa, Indonesia, and Thailand. Lots of country risk there, but that goes with the high yield.

This purchase brings the ETF share of the portfolio to almost exactly half. SEML is now the second-largest holding, after my big bet on Anglo Pacific (LSE:APF). Perhaps it is time to pause the process of ETF-ication until I can see more evidence that it actually works the way I want it to?

And I still have nearly 5% of the portfolio to re-invest...

[Purchase price: £43.215]

I am not a financial adviser and the information provided does not constitute financial advice. You should always do your own research on top of what you learn here to ensure that it's right for your specific circumstances.


  1. On the topic of waiting until you can see if a particular approach "works" (in quote marks as we can never truly know if something that has worked in the past will work in the future), you may have a long wait!

    I think it probably takes at least 5-years to have any sort of reasonable idea about how an investment strategy will work, and 10 years is probably a minimum period.

    There is just so much uncertainty in how different investments will perform that it takes years before you know if what you're doing is more skill (or a good process) than luck.

    I'm about 5 years into my current strategy and I'm only just beginning to think it might actually be doing okay because the basic process is sound, rather than because I've been lucky. But it will take another 5 years I think before I can be really sure.

    So in terms of your ETF move, I would say do what you're comfortable with, and perhaps 50% ETFs is a good time to pause. But don't bail out on the idea just because it's underperformed for a single year. That could just as well be down to bad luck rather than a bad strategy.


  2. It may take a little while as well to get your own mentality reconfigured along with your portfolio. Clearly, changing from a high-yield approach to an ETF one requires a lot of rejigging how you understand success and failure. You obviously understand that theoretically--as you note above--but it take a little longer to understand and be happy with it practically!

    As John says, you will have to wait and see what happens over a longer period. This will also give your mindset a bit of time to change to fit its new environment! Pausing at this stage to see how things feel going forward seems a good idea. Especially as you;re starting to have doubts about what is best. As John says, do what is comfortable!

  3. A bit different but I need some help about the Vanguard All-World High Yield ETF. According to the factsheet the dividend yield is 3,6% but according to the Vanguard’s distribution page –
    the dividend yield is just 3,2%. Why is this difference? Is it about taxation? Thanks Your answer!

  4. A bit different but I need some help about the Vanguard All-World High Yield ETF. According to the factsheet the dividend yield is 3,6% but according to the Vanguard’s distribution page –
    the dividend yield is just 3,2%. Why is this difference? Is it about taxation? Thanks Your answer!

  5. The difference is probably their costs - declared and undeclared. Interestingly, Bloomberg says 4.68% - probably currency variations?

  6. I've held ETFs in my spouse's portfolio for the same reasons as you've stated. Emotionally I tend to put up with them rather than feel fully committed to the idea. I agree as an asset class (or sub-class) they are less volatile than direct equity holdings. However the returns for each year of the four years and seven months we've held them has been less than I've generally returned in my own direct share holdings, both in terms of income and Mr Market's end of year so-called valuations. I've no idea how this comparison will work going forward, my luck on direct share holdings may well end soon.

    ETFs are less transparent than shares too (for example the confusion about actual yield in the above two comments - a question I can't answer).

    Emotionally I'd rather take the roller-coaster high volatility route or direct share holdings, but if I were to die I know my spouse is not wired the same way. For this reason ETFs are the better option in her portfolio. In the end it has to be worth sticking with a sound strategy if it is designed for good reasons.