ETFs v Investment Trusts

[chewyjoint]

Hello Thanks to the advice from posters on the Fool forum plus reading books by authors such as Tim Hale I am very much bought in to the low cost ETX / Index concept of portfolio building and management. I do have a nagging question at the back of my mind, however.

Until now I have been a big fan of Investment Trusts and have built a considerable portfolio via various shareplan schemes over a 30 year period.

Some of the larger, global, long established ITs such as Law Debenture, Witan, Scottish Mortgage, Scottish IT etc have offered very good (11%+) returns over a long period (20 years plus) despite charges that are on average 0.5% higher than ETFs.

I doubt that my global ETF portfolio will offer that level of return over the next 20 / 30 years and I wonder if switching to a bond / equity ETF mix will really perform better than a bond / IT mix?

[doug2500]

I doubt that my global ETF portfolio will offer that level of return over the next 20 / 30 years and I wonder if switching to a bond / equity ETF mix will really perform better than a bond / IT mix?

Well I doubt it too, but there will certainly be others who will say that any outperformance is just due to more risk. I think it's almost a question of faith, do you believe in active management, or passive? Or do you believe in some active management, which is quite different from all active management.

I read numerous threads on the other fool debating the merits of both approaches and I doubt anyone convinced anyone to change their opinion.

[Backache]

Hello Thanks to the advice from posters on the Fool forum plus reading books by authors such as Tim Hale I am very much bought in to the low cost ETX / Index concept of portfolio building and management. I do have a nagging question at the back of my mind, however.

Until now I have been a big fan of Investment Trusts and have built a considerable portfolio via various shareplan schemes over a 30 year period.

Some of the larger, global, long established ITs such as Law Debenture, Witan, Scottish Mortgage, Scottish IT etc have offered very good (11%+) returns over a long period (20 years plus) despite charges that are on average 0.5% higher than ETFs.

I doubt that my global ETF portfolio will offer that level of return over the next 20 / 30 years and I wonder if switching to a bond / equity ETF mix will really perform better than a bond / IT mix?

I doubt an ETF will perform as well as IT's have done historically but that is mainly because I doubt returns will be as good going forwards, we appear to be in a bit of a low interest rate low return environment.

Historically the majority of actively managed funds underperform the indexes because of charges, I see no reason why this should also not be true going forward, however low cost IT's probably have a better chance than most.
If you like the logic of the ETF route but also are wedded to some of your IT's why not pick the best looking IT's to be optimistic with and use ETF's for the bulk of your equity exposure?

[Lootman]

I doubt an ETF will perform as well as IT's have done historically but that is mainly because I doubt returns will be as good going forwards, we appear to be in a bit of a low interest rate low return environment.

If you like the logic of the ETF route but also are wedded to some of your IT's why not pick the best looking IT's to be optimistic with and use ETF's for the bulk of your equity exposure?

Lower returns will certainly diminish returns on any collective. I don't think it follows from that ITs will therefore out-perform ETFs.

I agree with your idea of a blend. You can pick ETFs for the beta of the market, and then ITs for alpha, if of course you believe you can find alpha.

[pbarne]

One aspect of the ETF vs. IT debate that you might attach importance to is regularity/predictability of income/distributions.
I've found that ITs are far more reliable in this respect.

P

[Lootman]

One aspect of the ETF vs. IT debate that you might attach importance to is regularity/predictability of income/distributions.
I've found that ITs are far more reliable in this respect.

Depends how you define "reliable"

ETFs will "reliably" pay out all dividends that they receive.

ITs will "reliably" choose to withhold up to 15% of those dividends in a reserve fund, at least partly so that they can claim to "never reduce dividends".

[pbarne]

ETFs will "reliably" pay out all dividends that they receive.

ITs will "reliably" choose to withhold up to 15% of those dividends in a reserve fund, at least partly so that they can claim to "never reduce dividends".

Hi Lootman,

I find it hard to believe that ETFs reliably pass on all the dividends they receive as distributions. I've experienced so-called "dividend aristocrat" ETFs vary by about a factor of 2 payouts for the same period in consecutive years. (I dimly remember this being reported by others previously - where the ETF managers had decided they would keep the dividends to re-invest). I'd be interested if you had a pointer to any rule that said ETFs had to pay these out?

P

[Lootman]

Hi Lootman,

I find it hard to believe that ETFs reliably pass on all the dividends they receive as distributions. I've experienced so-called "dividend aristocrat" ETFs vary by about a factor of 2 payouts for the same period in consecutive years. (I dimly remember this being reported by others previously - where the ETF managers had decided they would keep the dividends to re-invest). I'd be interested if you had a pointer to any rule that said ETFs had to pay these out?P

I don't understand what you mean by "vary by about a factor of 2 payouts for the same period in consecutive years."

My experience of ETFs (and I have held them for 15 years) is that, like open-ended funds, they pay out 100% of the dividends that they receive. The payout dates may fluctuate but that principle does not.

The rules are set by HMRC who naturally want revenues to be passed through on a timely basis so that taxes fall due on a timely basis. An ETF domiciled overseas would be free to not comply but, in so doing, would risk losing reporting status with HMRC, immediately making them less marketable in the UK.

So it's not so much a rule saying that ETFs cannot hoard dividends, but the UK tax authorities making it undesirable to pursue such a strategy.

[baldpaul147]

There are some ETFs that don't distribute dividends though and instead reinvest them to create an accumulation type ETF? An example is iShares core world ETF (SWDA) that has UK reporting status and is ISA and SIPP eligible. Is the rule that they either payout all dividends or reinvest them all? I too believed that ETFs paid out all dividends, but these (quite rare) accumulation type ETFs have me confused.

[Arborbridge]

ETFs will "reliably" pay out all dividends that they receive.

I don't believe this is always so.

At one time I had a holding in IAPD but was concerned that the payouts seemed to be varying in a way which made me wonder what was going on. After several emails, it turned out that somewhere up the grey-suit chain it had been decided not to pay out the dividends, but to use them as accumulation to bolster the unit price. The sales team considered it a good selling point to have a nicely rising chart at the expense of the payout, because the latter tends to be less obvious.

In the next payout, they did seem to go some way to reversing the situation, but since I didn't want an ETF where part of my pension was alterable at the whim, of the marketing department, I sold out.

Lootman has this particular objection to ITs holding back income, but I'm not bothered about that aspect, so long as it comes out in a regular stream and we eventually receive it. Yes, it's a bit nanny-ish, but I can tolerate it.

[pbarne]

ETFs will "reliably" pay out all dividends that they receive.

I don't believe this is always so.

At one time I had a holding in IAPD but was concerned that the payouts seemed to be varying in a way which made me wonder what was going on. After several emails, it turned out that somewhere up the grey-suit chain it had been decided not to pay out the dividends, but to use them as accumulation to bolster the unit price. The sales team considered it a good selling point to have a nicely rising chart at the expense of the payout, because the latter tends to be less obvious.

Hi Arborbridge,
Yes - looks like it was your post I was remembering - seems I had a similar experience.

[Lootman]

Lootman has this particular objection to ITs holding back income, but I'm not bothered about that aspect, so long as it comes out in a regular stream and we eventually receive it. Yes, it's a bit nanny-ish, but I can tolerate it.

It's not so much that I personally object to ITs retaining some of my dividends. As a practical matter it doesn't affect me much. But I think it skews comparisons with ETFs and OEICs when the basis of that comparison is dividend cuts. And renders the claim to have increased dividends every year somewhat empty. I object more to capital gains being distributed as dividends by ITs but that's another topic.

I didn't realise that an ETF had played those games you mentioned. That doesn't seem fair to me.

[hiriskpaul]

I think in theory ETFs are supposed to pay out 100% of dividends, but as always the devil lies in the detail:

- fees and other costs could be allocated 100% to dividends, 100% to capital, or somewhere in between. Clearly the more allocated to capital, the higher the ETF distribution.
- Sometimes a dividend is a capital distribution (or part capital, part from normal earnings/reserves). Should capital dividends be distributed or reinvested? I don't think there is a fixed rule here and the provider may well decide on a case by case basis.
- Similarly other corporate actions can result in cash that can be reinvested or distributed. One example is the Vodafone/Verizon split. In the Vanguard FTSE 100 ETF, they decided to distribute all (or most) of the value of the Verizon shares as part of the quarterly dividend. iShares did the opposite in their FTSE 100 ETF. FTSE themselves may have done something different again in their calculation of the actual FTSE 100 index and dividend adjustment.
- Some ETFs work by sampling companies from their underlying index instead of investing in all constituents according to the index weight. That will lead to dividend income undershoot or overshoot compared to the index.
- Some ETFs may use futures some of the time, but these are (usually) ex-div instruments. For example, before expiry the FTSE 100 future usually trades at a discount to spot value these days because it excludes dividends (as does the main FTSE 100 index itself) and short term funding rates are much lower than the FTSE dividend yield. So ETF providers needs to decide what should be done to correct for the fact that the investment in futures did not return any dividends.

All you can really say about an ETF is that the total return should track the total return of the index, less fees and costs. The ETF dividend itself does not necessarily have to track the dividend of the underlying index.

ETFs that reinvest dividends, such as most of the iShares Core series, are slightly different in that they still have to declare dividends, but they reinvest them instead of paying. Holders still need to report these dividends (as foreign dividend income), which may then be liable to income tax. They also need to separately track the reinvested dividends in order to get the capital gains calculation right. This can be quite a headache for someone not used to doing it. Completely irrelevant if held inside an ISA or SIPP of course.

[Hariseldon58]

Chewy
If I might comment on three aspects of your original post.
I have been using IT's since 1990 and with some success! I have moved almost exclusively to ETF's over the last few years with two legacy IT's I will sell when CGT allowances let me.

I am happy that my new approach is working better than my previous IT approach, which became bloated and lacked focus, I did allow my holdings to mushroom to more than 50 holdings. If ETFs did not exist then I would have moved to other forms of Index Trackers e.g. OEICs but if that option was not available then I would have remained with ITs and the results would I am sure still be satisfactory. Thus the choice for me is between best and ok.

The despondency about future returns over 20/30 years is misplaced in my opinion, it might be worse, who knows but I am sure that the investment outlook will be very different in 15 years time.

The last point is about bonds as part of a portfolio, this is about risk control not returns at the present moment. We live in uncertain times and a reserve of cash or short duration bonds may be very valuable....looking a few years out the situation may be very different and bonds may offer offer real returns.

I would diversify both globally and different types of assets and be prepared to be flexible, if you feel confident to make tactical asset allocation decisions if not then a mix of bonds/equities with possibly short bonds presently with periodic rebalancing may prove a pretty good , simple , cost effective approach.

Unexpected things happen unexpectedly and assumptions that a new 'normal' will persist is unlikely in my opinion, the more it becomes accepted then the greater chance of its failure.

[Backache]

Lower returns will certainly diminish returns on any collective. I don't think it follows from that ITs will therefore out-perform ETFs.

I didn't mean to say that they would and am sorry if it read that way.
I meant future ETF (and IT returns)are unlikely to match historic IT returns.

I am a little sceptical about IT returns beating low cost Index ETF's going forward, though they have interesting points at times when on discounts, when used for accessing illiquid markets and possibly when geared.

[GJHarney]

In response to the OP, I would say that it would depend on the type of ETF, but also the type of IT. A low cost global passive tracker ETF may well have the edge over a very large holdings global IT like F&C or a SIT over time due to the lower charges if nothing else. In comparison pick the right activist conviction global IT's and it would likely be a different story (ah, but how you pick the 'right' ones is a constant headache of course!). On the whole I'm still very happy with my IT's (even if I can't quite work out why LTI has grown so blinkin' far beyond its NAV even with the talented Mr Train at the helm, and whether I should put my buy and hold/fire and forget mentality to one side to cash the profits!). However, I think where ETF's are getting to be very interesting is in the 'smart beta' areas (dividends, value etc.). Old worrier that I am I'm more uneasy with the long term health prospects of the companies running ETF's than I am for the old-timer IT's I hold, but that's just me, probably.

[billG]

Hi,

One further point is that IT's have gearing that allow the trust to borrow money to invest. I recall seeing the graph between and IT and UT that had exactly the same manager with exactly the same investments (JPMorgan UK Smaller Companies) the only difference was the gearing and the investment returns were markedly different.

On that basis it would be possible under certain market conditions for and IT masquerading as a closet tracker with high gearing and low interest rates on its borrowing in a sharply rising market to outperform an ETF. Of course in a falling market gearing loses more money so it all depends when you do your comparisons......

BillG

[tomharmoni]

My reason for leaning towards ETF's is trading costs. If I planned to be a buy and hold investor then perhaps IT's would be the better option but as I buy and sell as the market dictates then the fact that there is no stamp duty on ETF's work strongly in their favour. My broker charges me £6 to buy and sell up to 50K GBP of an etf. To buy the same amount of an IT every time is adding another 0.5% stamp duty to the transaction. I like IT's but the trading costs put me off.

[Alaric]

You can barely tell the difference between the two investments.

Isn't that a quirk of your start point?

From the numeric table, the respective 10 year performances are 136.3% and 131.4%. However the 5 year performances are 102.1% and 71.4%

[Alaric]

I can only go back 10 years, I'm afraid. If longer term data is available, I'll willingly go back further and see what happens.

Why not draw the 5 year graph?