Is VWRD the one ring to rule them all?

[Plutus]

Hi.

Reviewing this...

https://www.vanguard.co.uk/uk/portal/de ... de=EQUITY##portfoliodata

Do such figures as 0.1% in the Philippines or even 2.2% in China really do anything meaningful to the fund? Seems like it's unlikely to make any large difference in the grand scheme of things, or am I mistaken?

Hello I'm far from an expert but I would have thought that China constitutes 2% of the holdings today but in the future may be e.g. 20% depending on market capitalisation, so you wouldn't need to re-balance or do anything as Vanguard do it on your behalf.

Someone mentioned on another thread that at one time Japan overtook or was equal to the global market cap weighting of the USA, today it's about 50-60% USA and 10% Japan.

[hiriskpaul]

Excluding EM is unlikely to make a material difference to an investor unless the long term EM performance is substantially different to that of DM. Leaving out EM will certainly cost less in management charges and is quite likely to result in lower volatility.

To put some figures to it. Currently EM accounts for 8.7% of the Vanguard global ETF. If DM grew at a rate of 7% and EM at 8%, then after 10 years a developed markets tracker would give a return of 96.72%, developed+EM 98.38%, so difference of just 1.67%, but the actual final monetary amount accumulated would only be 193.38/196.72-1 = 0.85% more if EM was included. Over 20 years the difference would rise to 6.88%/1.78% and over 30 years 21.32%/2.80%. If the EM instead grew at 9%, the 10y, 20y, 30y differences in growth would be 3.48%/1.8%, 15.09%/3.9% and 49.20%/6.5% respectively. So after 30 years of compounded returns you would only have 6.5% more money if you picked a global tracker which included EM compared with one without if DM grew at 7%, EM at 9%. Even after 50 years the difference would only be 13%.

Notice how a large difference in spread can be misinterpreted and potentially misused!

Bogle advises US investors not to invest in non-US stocks. You can kind of see his point that in practical terms, US investors are unlikely to see a huge difference in final outcomes by adding non-US stocks, but will add to risk.

Sometimes we worry too much about precise portfolio composition when we don't even know what future component growth rates are going to be. I am as guilty as anyone in doing this.

[hiriskpaul]

What a fantastic chart and table 1nv35t!

To expand on my last posting about small difference in final outcomes, it is interesting to compare the equal weighted USD against the US Real (I assume cap weighted). If $100 was invested in each strategy in 1900, by 2012 the equal weighted pot would have grown to $94,826 (in 1900 dollars) and the cap weighted portfolio $89,829. A big $ difference, but just 5.6% and that is after 112 years!

It is interesting to see that the equal weighted strategy was more risky than the cap weighted one. This makes sense as there would have been a bigger tilt to small caps, which have historically carried more risk.

[paulnumbers]

Thanks people - very interesting

[hiriskpaul]

What a fantastic chart and table 1nv35t!

To expand on my last posting about small difference in final outcomes, it is interesting to compare the equal weighted USD against the US Real (I assume cap weighted). If $100 was invested in each strategy in 1900, by 2012 the equal weighted pot would have grown to $94,826 (in 1900 dollars) and the cap weighted portfolio $89,829. A big $ difference, but just 5.6% and that is after 112 years!

It is interesting to see that the equal weighted strategy was more risky than the cap weighted one. This makes sense as there would have been a bigger tilt to small caps, which have historically carried more risk.

Sorry, that posting was total garbage resulting from Excel mistyping. True figures are EW portfolio grows to $147,261. So total growth 64% more than cap weighted. So over 112 years small differences in rates of return certainly can result in a material differences in the outcome. After a more normal 30 year term with the same rates of return, the difference would only have been 14%.

[Plutus]

What a fantastic chart and table 1nv35t!

....

I agree and it'll be interesting to see how this compares in 20 or 30 years time.

[dspp]

That chart suggests that Germany has a consistently lower stock market capitalisation than the UK which is not the case when one compares GDP. Therefore - just a guess - different ownership structures may skew the capitalisation chart. I'm guessing that more of German firms are closely held/untraded whereas more of UK / USA firms are market traded and therefore easier to invest in. Ditto some other countries / times.

Not sure whether my hypothesis is correct, or what practical use can be made of the observation as a private investor if it is correct.

regards, dspp

[hiriskpaul]

With increasing globalisation and ease of moving capital around I suspect the actual country(s) a multi-national's stock is listed in will become increasingly irrelevant. As the long term returns from the global stock markets will be dominated by the returns from multi-nationals I suspect that just buying a cheap World tracker, with or without EM, is as likely to give as good/bad a long term outcome for most investors as many other non-cap weighted approaches.

We are not there yet though and the silliness of the past, for example Japanese stocks being bid to crazy valuations, can I think still be repeated.

A cap weighted global tracker is very close to being long term buy and hold is well. Small adjustments have to be made due to new issues of shares, buy backs and stocks entering/leaving the market, but turnover should be relatively low. In an accumulating ETF such as SWDA I would have thought that much of this could be handled by reinvestment of dividends, so there should be few disposals required to keep the fund in balance.

[hiriskpaul]

It is worthwhile remebering that stock markets in the early part of the 20th century were quite different to now. Levels of corruption, anticompetitive practices, insider dealing, etc. were much higher. Today's AIM market is tame compared with than any stock market in 1900 and probably a safer place to invest!

[hiriskpaul]

One aspect of accumulating funds that is a bit of work is having to monitor/report the Excess Reportable Income amount, i.e as though you were actually paid that dividend (which being accumulating was reinvested back into the fund again). As you declare and pay taxes if appropriate on that dividend, then the cost base of the holdings should be uplifted to discount that 'reinvested' amount so that you don't end up paying capital gains on too low a cost base. Not particularly difficult to work out/record, but easily overlooked.

I'm not sure about whether if you sold and repurchased across the 31st December 'distribution date' whether you might avoid having to report the excess reportable income declaration to HMRC. Buying and selling the same or similar holding again within 30 days is typically considered as not having traded (capital gains don't have to be declared, nor can capital losses be claimed/offset i.e. 30 day trading rule).

Every year you could sell SWDA at the beginning of December and buy the HSBC MSCI World ETF (HMWO). Then flip them back at the beginning of January. That would crystallise the capital gain/loss and you would not have to take the dividend. That would simplify the accounting and would be beneficial for some taxpayers with over £5k dividend income. You would of course have to pay the charges and bid/ask spread. You might also gain or lose from fluctuations the small premium/discount to NAV. This would work because of the convenient ex-div dates for HMWO (20/10/2016 and 19/01/17).

[TahiPanas]

This is a really interesting and informative thread. Many thanks to all contributors!

[GeoffF100]

I'm not sure about whether if you sold and repurchased across the 31st December 'distribution date' whether you might avoid having to report the excess reportable income declaration to HMRC.

If I remember correctly, you pay tax on Excess Reportable Income if you hold the fund at the beginning of the accounting period, not the distribution date.

[hiriskpaul]

I'm not sure about whether if you sold and repurchased across the 31st December 'distribution date' whether you might avoid having to report the excess reportable income declaration to HMRC.

If I remember correctly, you pay tax on Excess Reportable Income if you hold the fund at the beginning of the accounting period, not the distribution date.

I really hope you are wrong, as I have always considered it as income on the distribution date! I must check later. Yet to finish this years return.

[hiriskpaul]

I agree about the Excess Reportable Income date and am relieved to say I have been doing it correctly. I always buy distributing ETFs outside tax shelters which means the the amount of excess reportable income is usually peanuts, but clearly this would not be the case for an accumulating fund. So in my example for SWDA, income could be avoided by selling before the end of June (period end 30 June) and swapping to HSBC's HMWO*, then selling HMWO before its July ex-div date and buying back into SWDA. Doing what I originally said and selling/buying either side of the distribution date would not have worked.

The 30 day rule provides scope over what capital gain/loss gets booked. If someone did not want a gain or loss that tax year, simply sell HWMO and buy back SWDA within 30 days. The capital gain/loss on HWMO should then be very close to the loss/gain on SWDA.

Regarding the FX rate to apply, I always use the one on the date of the distribution. I have never really thought about this though, it just seemed obvious to me that this is what I should use. Be good to find a definitive answer - there are many things that seem obvious when it comes to tax rules that are wrong!

Regarding this:

Must admit that I thought it was proportioned, such that if you held for say 3 months in total across a reportable year you should declare 25% of the excess reportable amount as having been 'received' on the distribution date.

You are probably thinking about the way bonds work. You are liable for tax on interest accruing on a bond between payment dates. Usually this is clear because most bonds trade "clean" and the interest accrued is on the deal ticket. However, you are still liable for tax on interest if the bond trades dirty. You have to work out what this interest is yourself, which can be a bit of a nightmare.

*Vanguard's VEVE could be used instead, but a small amount of excess reportable income would arise because VEVE's accounting period is also at the end of June.

[hiriskpaul]

Excluding EM is unlikely to make a material difference to an investor unless the long term EM performance is substantially different to that of DM. Leaving out EM will certainly cost less in management charges and is quite likely to result in lower volatility.

To put some figures to it. Currently EM accounts for 8.7% of the Vanguard global ETF. If DM grew at a rate of 7% and EM at 8%, then after 10 years a developed markets tracker would give a return of 96.72%, developed+EM 98.38%, so difference of just 1.67%, but the actual final monetary amount accumulated would only be 193.38/196.72-1 = 0.85% more if EM was included. Over 20 years the difference would rise to 6.88%/1.78% and over 30 years 21.32%/2.80%. If the EM instead grew at 9%, the 10y, 20y, 30y differences in growth would be 3.48%/1.8%, 15.09%/3.9% and 49.20%/6.5% respectively. So after 30 years of compounded returns you would only have 6.5% more money if you picked a global tracker which included EM compared with one without if DM grew at 7%, EM at 9%. Even after 50 years the difference would only be 13%.

Notice how a large difference in spread can be misinterpreted and potentially misused!

Bogle advises US investors not to invest in non-US stocks. You can kind of see his point that in practical terms, US investors are unlikely to see a huge difference in final outcomes by adding non-US stocks, but will add to risk.

Sometimes we worry too much about precise portfolio composition when we don't even know what future component growth rates are going to be. I am as guilty as anyone in doing this.

This is a pedantic caveat to my previous posting. What I should have said is excluding EM is unlikely to make a material difference to an investor unless the long term EM performance is substantially different to that of DM AND the net cash flow rate into EM is not significantly different to that of DM.

Probably more easily explained by example. If DM total capital value was $90B and EM $10B, then EM would be 10% of the World Market. If there was no growth in either market between 2 dates, this 10% proportion could still rise or fall as a result of corporate actions and new stocks entering the markets. If there was $10B net inflow into new stocks entering emerging markets, but none going into DM, then the ratio would be $90B DM to $20B EM. EM would now be 20/110 = 18% of the world market. Now if activity like that continued and EM stocks also grew faster than DM then the all world tracker could significantly outperform a DM tracker over the long term.

In other words, while the proportion of EM stocks remains small, there is unlikely to be much difference between the performance of a developed+emerging markets tracker and developed markets tracker.

[GeoffF100]

You are probably thinking about the way bonds work. You are liable for tax on interest accruing on a bond between payment dates. Usually this is clear because most bonds trade "clean" and the interest accrued is on the deal ticket. However, you are still liable for tax on interest if the bond trades dirty. You have to work out what this interest is yourself, which can be a bit of a nightmare.

iWeb does not include the accrued interest on the contract note, and I once forgot to ask the dealer for the value. No problem. You can find the accrued interest, and much else besides, on the DMO website:

http://www.dmo.gov.uk/

[GeoffF100]

With regard to Excess Reportable Income, Google found me this:

https://www.triodos.co.uk/downloads/rep ... y-note.pdf

It looks like the relevant exchange rate is that on the distribution date. What time on the distribution date? I expect that there is HMRC guidance on this, but I doubt whether they will pursue us for pennies.

[GeoffF100]

You are probably thinking about the way bonds work. You are liable for tax on interest accruing on a bond between payment dates. Usually this is clear because most bonds trade "clean" and the interest accrued is on the deal ticket. However, you are still liable for tax on interest if the bond trades dirty. You have to work out what this interest is yourself, which can be a bit of a nightmare.

By the way, I think you will find that gilts trade at the dirty price, by definition. I do not know about corporate bonds. I have only traded those in an ISA.

[hiriskpaul]

You are probably thinking about the way bonds work. You are liable for tax on interest accruing on a bond between payment dates. Usually this is clear because most bonds trade "clean" and the interest accrued is on the deal ticket. However, you are still liable for tax on interest if the bond trades dirty. You have to work out what this interest is yourself, which can be a bit of a nightmare.

By the way, I think you will find that gilts trade at the dirty price, by definition. I do not know about corporate bonds. I have only traded those in an ISA.

Ok, everything trades at the dirty price, including gilts. What I meant was the convention is to quote clean prices and then incorporate accrued interest from the previous coupon to the settlement date into the deal. Most corporate bonds trade the same way.

The point I was making was that you will be liable for income tax on bond interest even if you you have not received any coupon payments. Buy a gilt one day, sell it the next and you are liable for income tax on one days interest. Even if the bond goes XD on the day you sell (so you get the whole coupon), you are still only liable for income tax on 1 days worth of interest. That contrasts with equities/preference shares, where you are not liable for income tax on a dividend unless you receive it and if you receive a dividend you are liable for tax on the whole dividend.

Sometimes a bond trades "dirty" (usually only those that have suspended coupon payments), in that prices are quoted that incorporate accrued interest, but this is not made explicit in the deal ticket. The quoted price is the dirty price instead of the clean price. However, someone buying such a bond and selling it the next day is still liable for one days income tax on the accrued interest.

[hiriskpaul]

If there was no equalisation then conceptually you might sell just prior to the end of the reporting period, repurchase a day later after the start of the new reporting period and not have to declare the excess reportable income deemed to have been paid 6 months later to HMRC, and under 30 day rules be exempt from having to declare capital gains (losses) due to such a short trade action. Which seems to be 'optimistic'.

To be pedantic, you are not exempt capital gains tax on the trade. You match the sell with the purchase a day or so later and if the buy is at a lower price than the sell, you have made a capital gain. If you buy at a higher price, you have made a capital loss.