Global Small Cap?

[GeoffF100]

Vanguard offers a global small cap fund, but does not currently include any small cap in its LifeStrategy funds:

https://www.vanguardinvestor.co.uk/inve ... _fund_link

Small caps appear to be about 10-12% of the global market. Here is a publication advocating 90% global large + medium cap, and 10% global small cap:

https://content.ftserussell.com/sites/d ... l-caps.pdf

[OLTB]

Vanguard offers a global small cap fund, but does not currently include any small cap in its LifeStrategy funds:

https://www.vanguardinvestor.co.uk/inve ... _fund_link

Small caps appear to be about 10-12% of the global market. Here is a publication advocating 90% global large + medium cap, and 10% global small cap:

https://content.ftserussell.com/sites/d ... l-caps.pdf

In my passive portfolio, I have two small cap selections:

SSGA World Small Cap (WOSC) 12% of my portfolio
iShares EM Small Cap (IEMS) 4% of my portfolio

Cheers, OLTB.

[mc2fool]

SSGA World Small Cap (WOSC) 12% of my portfolio

Any particular reasons for that one, OCF 0.45%pa, over iShares MSCI World Small Cap (WLDS) tracking the same index and also in GBP with an OCF of 0.35%?

[OLTB]

SSGA World Small Cap (WOSC) 12% of my portfolio

Any particular reasons for that one, OCF 0.45%pa, over iShares MSCI World Small Cap (WLDS) tracking the same index and also in GBP with an OCF of 0.35%?

Thanks mc2fool

No reason at all - I wasn't aware of the iShares one to be honest!

Even though the OCF is 0.1% cheaper, my passive portfolio is not a huge value and trading out of the current etf, reinvesting into the new one would make me worse off because of the trading fees. Looking at it again though, I would be better off after two years, so maybe I should swap as I wont be retiring for another 14 years or so.

Thanks for the 'heads-up'.

Cheers, OLTB.

[mc2fool]

No reason at all - I wasn't aware of the iShares one to be honest!

And I wasn't aware of the State Street one.

And on a bit of further digging around in their ETFs, I find the SPDR® MSCI Emerging Markets Small Cap (EMSD/EMSM USD/GBP), which at 0.55%pa is 0.19% cheaper than IEMS!

[GeoffF100]

Vanguard Global Small Cap Index Fund beats WLDS at 0.29%:

https://www.vanguardinvestor.co.uk/inve ... ion-shares

That may not be much good if you have to pay a platform fee on open ended funds though. I thought 0.29% was expensive, but I am paying 0.22% on VFEM, so maybe it is not too much. The Vanguard Global Small Cap Index Fund is restricted to developed markets.

[Hariseldon58]

Vanguard talks abouts it All World ETF VWRL in these terms;

FTSE All World Index

The index measures the market performance of large- and mid-capitalisation stocks of companies located around the world.
Includes approximately 3,900 holdings in nearly 50 countries, including both developed and emerging markets.
Covers more than 95% of the global investable market capitalisation.

I have tended to blow hot and cold on small cap, the costs are higher , not so much the management charge but the spreads and general liquidity and even if there is a return premium, some is lost in costs and at a portfolio level the return premium is pretty minimal.

FTSE All World Index

The index measures the market performance of large- and mid-capitalisation stocks of companies located around the world.
Includes approximately 3,900 holdings in nearly 50 countries, including both developed and emerging markets.
Covers more than 95% of the global investable market capitalisation.....

By combining VEVE and VFEM as opposed to VWRL you have an expense cost of around .13% pa, is small cap worth it ?

I’d personally not want to bet the ranch on it and if the desire is to improve returns then perhaps a mild degree of leverage or more equity and less bonds/ cash might be a cost effective alternate ?

[1nvest]

By combining VEVE and VFEM as opposed to VWRL you have an expense cost of around .13% pa, is small cap worth it ?

Some funds bury in their small print that the benchmark compared to is the net total return (net of dividend withholding taxes). Yet other funds, even from the same provider such as Vanguard might instead mention in the small print that ...

Basis of fund performance NAV to NAV, net of expenses, with gross income reinvested.

A assumption of a comparison figure that includes 'gross' dividends having been reinvested but where in practice gross dividends would have had withholding taxes applied at various rates in different countries?

Average dividend withholding taxes globally is something like 30%, so getting on for 1% tax/deduction for a 3% dividend yield. By comparing actual fund to net tax deducted benchmark, or comparing a funds nav with assumption of gross dividends reinvested compared to a gross index it can be made to look like a fund compared closely to the 'benchmark' index, perhaps by just the expense ratio difference, but where that might actually be >1% or more less than the actual index.

A multi-national such as Vanguard could do internal stock lending, perhaps where it incurred just a 15% withholding tax instead of 30% by lending the stock over the ex-dividend period to another office region/state. Is that flowing back to investors? Perhaps. Juniors at Vanguard earn upwards of $66,000 (average salary in UK is $100,000+). With 17,600 total employees then at that average wage = $1.75Bn/year wage bill. With $5.6 trillion assets under management and a average (https://about.vanguard.com/who-we-are/fast-facts/) 0.1% average ER = $5.6 billion. Building/office/licencing/commissions and trading ...etc. costs of approaching $4Bn/year (after wages) does seem reasonably aligned.

What are others actual experiences? Are the rewards from the likes of VWRL actually close in practice to the Russells FTSE All World Index after discounting the Expense Ratio, or are actual returns more aligned to after 30% average dividend withholding taxes having been 'lost'? Published "Past Performance" tables/charts more often show benchmark and fund having just close to the Expense Ratio difference, so either Vanguard is receiving dividends from globally very tax efficiently or the benchmark compared to is after withholding taxes. Either way that's potentially 1% lower than the 'Index'.

VWRL prospectus indicates

The OCF will not cover (to the extent not included in the Operational Fees as detailed below) non-recurring and extraordinary or exceptional costs and expenses (if any) as may arise from time to time (such as, without limitation, material litigation) and withholding taxes deducted from interest and dividend payments to the relevant Fund, stamp duties or other documentary transfer taxes, or similar duties and brokerage fees (excluding costs for research) arising on investment management activity on the purchase or sale of securities by the relevant Fund which, if they arise, will be paid out of the assets of the relevant Fund.

So the 0.22% VWRL OCF doesn't include withholding taxes that are 'taken out' (more strictly amounts not received), by the fund.

If calculations based on historic index total returns indicate that a 4% safe withdrawal rate might have been OK in say 95% of samples, but where after dividend withholding taxes of 1%, fund fees/expenses of 0.22% are added in then that pushes the figure to the equivalent of a 5.22% withdrawal rate, and that can mean significant differences in outcomes. Are customers being sold gross total return reward sales pitch by Financials, whilst actually being rewarded >1% lower actuals?

Would be nice if funds more openly published the total amounts of dividend withheld taxation figures each fund incurred - but they don't such that making informed decisions is more difficult.

[GeoffF100]

I have tended to blow hot and cold on small cap, the costs are higher , not so much the management charge but the spreads and general liquidity and even if there is a return premium, some is lost in costs and at a portfolio level the return premium is pretty minimal.

FTSE All World Index

The index measures the market performance of large- and mid-capitalisation stocks of companies located around the world.

Includes approximately 3,900 holdings in nearly 50 countries, including both developed and emerging markets.

Covers more than 95% of the global investable market capitalisation.....

By combining VEVE and VFEM as opposed to VWRL you have an expense cost of around .13% pa, is small cap worth it?

I would not want to speculate purely on small cap having higher returns than large /medium cap. I would, however, consider having a market weighting of small cap (about the same as the weighting for Emerging Markets currently). With the assumptions of the CAPM model, the optimal portfolio is a market weighted average of all the investments available. In theory, not including small cap reduces diversification and increases risk, without a compensation increased return. I am, however, discouraged by an OCF of 0.29% for the Vanguard Small Cap Index Fund. That is more than twice as much as for the large / medium cap.

[GeoffF100]

A assumption of a comparison figure that includes 'gross' dividends having been reinvested but where in practice gross dividends would have had withholding taxes applied at various rates in different countries?

If the fund management group benchmarks its fund against an index for which dividends are paid gross of withholding tax that they cannot reclaim, they will have a negative tracking error which is larger than the OCF. That would be shooting themselves in the foot. With the Vanguard funds that I have looked at, the negative tracking error closely reflects the OCF.

Overseas stocks often return money to shareholders via buy-backs rather than dividends, which reduces the withholding tax hit.

You either hold overseas stock and accept higher tax, or restrict yourself to domestic stock and accept higher risk. The choice is yours.

[hiriskpaul]

Would be nice if funds more openly published the total amounts of dividend withheld taxation figures each fund incurred - but they don't such that making informed decisions is more difficult.

They do. You will find this in the annual accounts, along with revenue from stock lending. The information is not particularly helpful though as it gives the amount of WHT in the year, but the amount of outstanding shares fluctuates throughout the year so it is difficult to get an amount per share.

If you look at US listed ETFs covering the US market, eg S&P 500 ETFs, you will find that they usually compare their returns against he gross index and the long term returns. A good example is the Vanguard S&P 500 fund. The tracking is very close to the index, a few bp per year, largely explainable by the fee. European markets often benchmark against a net index, but that net figure is often before reduction of WHT by tax treaties. This is how the European version of the Vanguard S&P 500 manages to beat the index. The index assumes a 30% rate of withholding tax, but the Ireland/US treaty reduces it to 15%. In the FTSE World Index I am not sure what the overall drag due to withholding taxes is, but would guess the aggregate WHT rate closer to 15% than 30%. The US, about half the market is 15%. Other markets, such as Switzerland have higher WHT, and some have lower, or zero such as the UK, Japan, HK.

In some instances, such as with US WHT, it is possible to avoid the withholding tax entirely using swap based ETFs. But the fees are usually higher with swap based ETFs and the risk of a swap counterparty default is introduced. Personally I prefer to forgo the small extra return that you may get through swap based ETFs for the increased transparency of physical replication. Return from stock lending can also go some way to mitigating withholding taxes.

Edit: A quick look at the VWRL accounts gives for year ending 30 June 2019 gives dividends of $7,446,153, withholding tax of $877,262, which suggests withholding tax of about 12%.

[Hariseldon58]

I think Hiriskpaul has provided a great summary of the taxation question regarding ETFs, I spent a lot of time in the past coming to the conclusion that withholding taxes hover around .15% with some small other charges accounting for a few bps on the big ETFs.

Some of the off brand ETFs may have lower ocf but the real costs may be higher and in general one does well to stick with Vanguard, iShares and State Street.

[Hariseldon58]

I have tended to blow hot and cold on small cap, the costs are higher , not so much the management charge but the spreads and general liquidity and even if there is a return premium, some is lost in costs and at a portfolio level the return premium is pretty minimal.

FTSE All World Index

The index measures the market performance of large- and mid-capitalisation stocks of companies located around the world.

Includes approximately 3,900 holdings in nearly 50 countries, including both developed and emerging markets.

Covers more than 95% of the global investable market capitalisation.....

By combining VEVE and VFEM as opposed to VWRL you have an expense cost of around .13% pa, is small cap worth it?

I would not want to speculate purely on small cap having higher returns than large /medium cap. I would, however, consider having a market weighting of small cap (about the same as the weighting for Emerging Markets currently). With the assumptions of the CAPM model, the optimal portfolio is a market weighted average of all the investments available. In theory, not including small cap reduces diversification and increases risk, without a compensation increased return. I am, however, discouraged by an OCF of 0.29% for the Vanguard Small Cap Index Fund. That is more than twice as much as for the large / medium cap.

I agree in principle with a market weight whole heartedly but the extra costs and limited additional diversification in practice probably outweigh the actual benefits because the holdings firm such a small part of the whole. Ie blow hot and cold between the view that holding a market weight is sensible but the actual benefit may be illusory !

[mc2fool]

I agree in principle with a market weight whole heartedly...

Matter of interest, if you are a wholehearted believer in the principle of market weights, are you 60% in bonds and 40% equities? The global bond market is, I believe, about half again as big as the global equities market....

[GeoffF100]

Matter of interest, if you are a wholehearted believer in the principle of market weights, are you 60% in bonds and 40% equities? The global bond market is, I believe, about half again as big as the global equities market....

I will give my own answer here. In CAPM it is assumed that all investors have the same objective, risk tolerance, currency and taxes. It follows essentially from those assumptions and the efficient market hypothesis that we should all market weight our investments. In the real world, we do not all have the same risk tolerance. Most of the people here are either more risk tolerant than the market (or do not fully understand the risks that they are running), and hold more equities and less bonds than the market. Differences in currencies and taxes explain why a home bias often makes sense.

[colin]

Matter of interest, if you are a wholehearted believer in the principle of market weights, are you 60% in bonds and 40% equities? The global bond market is, I believe, about half again as big as the global equities market....

I will give my own answer here. In CAPM it is assumed that all investors have the same objective, risk tolerance, currency and taxes. It follows essentially from those assumptions and the efficient market hypothesis that we should all market weight our investments. In the real world, we do not all have the same risk tolerance. Most of the people here are either more risk tolerant than the market (or do not fully understand the risks that they are running), and hold more equities and less bonds than the market. Differences in currencies and taxes explain why a home bias often makes sense.

Anyone who believes the above missunderstamds the CAPM . The CAPM does not assume that all investors have the same risk tolerance, it just assumes that investors will price undiversifiable risk (beta) to provide a higher return than diversifiable risk.

[GeoffF100]

Matter of interest, if you are a wholehearted believer in the principle of market weights, are you 60% in bonds and 40% equities? The global bond market is, I believe, about half again as big as the global equities market....

I will give my own answer here. In CAPM it is assumed that all investors have the same objective, risk tolerance, currency and taxes. It follows essentially from those assumptions and the efficient market hypothesis that we should all market weight our investments. In the real world, we do not all have the same risk tolerance. Most of the people here are either more risk tolerant than the market (or do not fully understand the risks that they are running), and hold more equities and less bonds than the market. Differences in currencies and taxes explain why a home bias often makes sense.

Anyone who believes the above missunderstamds the CAPM . The CAPM does not assume that all investors have the same risk tolerance, it just assumes that investors will price undiversifiable risk (beta) to provide a higher return than diversifiable risk.

My answer was not accurate as it could have been. Irrespective of your ratio of equities to bonds, you can clearly change your risk level by holding more cash or by borrowing. If you want to compensate to having "too high" a percentage of bonds, you can borrow money to increase your risk. We will have to pay a premium interest rate to do that, so it is not an attractive option. Simply holding less bonds is a much more attractive option. Nonetheless, if too little money goes into bonds, they will become under-priced with respect to equities.

More precisely, you will find the mathematics here:

http://wps.pearsoned.co.uk/ema_ge_berk_ ... index.html

15. Three main assumptions underlie the Capital Asset Pricing Model (CAPM):

a. Investors trade securities at competitive market prices (without incurring taxes or transaction costs) and can borrow and lend at the risk-free rate.

b. Investors choose efficient portfolios.

c. Investors have homogeneous expectations regarding the volatilities, correlations, and expected returns of securities.

16. Because the supply of securities must equal the demand for securities, the CAPM implies that the market portfolio of all risky securities is the efficient portfolio.

[colin]

Your understanding of the CAPM is inadequate without the concept of beta which implies that high beta stocks are expected to have a higher than market average rate of return to compensate for the undiversifiable risk, excess return over and above that predicted by beta is supposed to be the result of fund manager skill or alpha. The advanced mathematics are irrelevant to non accademics . In any case a study published by Fama & French on all stocks traded in the US between 1963-1990 showed that there was no relationshiip between beta and return for individual stocks or managed funds. In other words they found that higher than market risk investments did not poduce higher returns, which the CAPM implied. Fama and French went on to produce the ' 3 factor model' and resurected beta as a usefull concept by claiming they found it could be used to provide higher risk adjusted returns than a broad matket index when combined with other factors , an approach utilised by Dimensional Fund advisors to construct client portfolios to provide the highest expected return commensurate with their fee paying clients risk tolerance. Jack Bogle expressed cynism about the independance of Fama & French's research due to their ascociation with Dimensional Fund Advisors.

[GeoffF100]

Of course, I understand beta. Beta is fundamental to CAPM. The mathematics of CAPM is not particularly advanced and it is relevant. If you do not understand the mathematics, you can not fully understand the model. Nonetheless, that does not matter here. What matters for this thread are the assumptions underlying CAPM and the conclusion that the optimal portfolio is the market portfolio. Those assumptions are only approximations to the real world. Where they do not apply the conclusions do not apply. I have said that I do not see any good reason not to market weight small cap, except for the increased costs. I have also given two examples where I think a deviation from market weight is reasonable. Other people hold differing opinions.

[colin]

Not only is the mathmatics used to calculate beta advanced it has been demonstrated to be irelevant to investors because as I have already stated the Eugene Fama and Kenneth French study published in 1992 found no evidence that the predictions of the CAPM were true.