Adding All-Country-Equal-Weight tilt to portfolio

[San_Francisco]

@Grog

I don’t think Switzerland has lots of tax treaties with frontier EM

…which generally wouldn’t matter anyway, as elevant ETFs or other funds aren’t domiciled in Switzerland. Rather in the U.S., Ireland or Luxembourg, maybe the U.K. So in effect, there would be no withholding tax when distributing to Swiss investors.

With regard to the source countries, they should, on average, be no worse than the average European ETF that holds U.S. stock:

Argentina: 7%
Brazil: 0%
China: 10%
Colombia: 7.5% (for profits taxed at corporate level)
India: 0%
Indonesia: 20%
Korea (ROK): 22%
Malaysia: 0%
Mexico: 10%
Nigeria: 10%
Pakistan: 15%
Philippines: 30%
Poland: 19%
Russia: 15%
South Africa: 20%
Taiwan: 21%
Thailand: 10%
Turkey: 15%
Vietnam: 0%

These are non-treaty rates, so rates higher than 15% would in most cases be reduced to 15% or less. Though as I understand, dividends wouldn’t actually be paid (and thus subject to withholding tax) for swapping ETFs.

Just looking at dividend withholding taxes, investing in U.S. equity seems worse in most cases, due to the 30% or 15% non-refundable part. And it remains to be seen whether we will still have relatively easy means of avoiding this through U.S.-domiciled funds, come 2020.

Last but not least, one should put into perspective the effect of withholding taxes on total returns. Many companies will retain (as least parts of) their earnings to reinvest anyway.

[Grog]

@San_Francisco

…which generally wouldn’t matter anyway, as elevant ETFs or other funds aren’t domiciled in Switzerland. Rather in the U.S., Ireland or Luxembourg, maybe the U.K. So in effect, there would be no withholding tax when distributing to Swiss investors.

it depends on the treaties those country (where the ETf is domiciled) have with the frontier countries. The fund itself may receive a netto dividend without the possibility to claim back L1 dividend. Again, it depends on the tax treaties between Ireland/ Luxemburg and Vietnam or whatever.
What I was hinting to is the process today for which you may purchase US etf and claim 100% of the taxes back, thus having an effective TER of 0.04% or whatever.

Another poin is that those frontier ETF are traded more rarely and have higher spread, that leads to additional costs.

If you are set on the strategy, that’s fine, I’m not attacking you directly. I want to provide more point of view for people who are investigating this approach.
4 years ago I was leaning towards a similar strategy. But then I simply decided to simplify my portfolio and in taxable accounts I only keep VT (usa stock only) for which right now I have the most ability to reduce TER, spread and hidden costs. Of course then I’m heavily invested in the US, but I’m fine with it since it more or less correlate with the world stock.

[San_Francisco]

@Grog

What I was hinting to is the process today for which you may purchase US etf and claim 100% of the taxes back, thus having an effective TER of 0.04% or whatever.

Oh, I was fully aware of what you meant there

For investors like me and probably many others here (retail investors domiciled in EFTA), there seem to be so few options of achieving that in a cost-conscious fashion that you could essentially call that it a small loophole - which might close in a few months.

Because I am not aware of (m)any brokers that still allow EFTA retail investors to buy U.S. funds. The exception being IBKR (maybe Schwab/who else?). And IBKR they have at least indicated to a member of this forum that they might drop such access to U.S. funds as of next year.

@Grog

Again, it depends on the tax treaties between Ireland/ Luxemburg and Vietnam or whatever.

Well, not in this particular case, as Vietnam (see above) doesn’t levy withholding tax on distributed dividends. There is no tax in the first place to be reduced by double taxation treaty.

@Grog

it depends on the treaties those country (where the ETf is domiciled) have with the frontier countries. The fund itself may receive a netto dividend without the possibility to claim back L1 dividend

Using the U.S. and its equity market as a “benchmark” (which seems sensible, as it constitutes about half of world market capitalisation), investing in emerging markets is on average more tax-efficient than U.S. equity. In emerging markets dividend tax burden will be lower, whereas U.S. non-treaty withholding tax rate of 30% is one of the highest in the world. And the non-reclaimable portion of 15% (in Irish ETF) or 30% is higher or at least not lower than for the great majority of emerging market countries.

So only considering dividend taxation from the perspective of a European retail investor investing through ETFs, U.S. market is among the worst to invest in, when compared to emerging economies.

…unless of course you happen to

1. live in that one European country (CH), where you can…
2. be customer at that one broker (IBKR) that still offers to…
3. buy ETFs domiciled in that one country (the U.S.)

…that will allow you to effectively lower your tax burden on U.S. dividends to 0%. That would be the small remaining loophole.

@Grog

Another poin is that those frontier ETF are traded more rarely and have higher spread, that leads to additional costs.

True, but for me it’s more buy&hold, so spreads are negligible in the long run.

@Grog

I’m heavily invested in the US, but I’m fine with it since it more or less correlate with the world stock.

I would rather put it the other way round: World stock market is highly correlated to the U.S. - alone by the fact that the U.S. constitutes about half of it.

I’m consciously willing to reduce my correlation to the world market (or U.S. market, for that matter) by overweighting non-U.S. (and especially EM) equity.

[paul]

I would suggest you to take a look at this:
https://twitter.com/mebfaber/status/1120390153162371074?lang=en
and in particular to the Bridgewater study linked:
https://www.bridgewater.com/research-library/daily-observations/geographic-diversification-can-be-a-lifesaver/Geographic_Diversification_Can_Be_a_Lifesaver.pdf

This decade, the US has been the best performer so far, but it was one of the weaker performers in the previous decade following the dotcom bust; it was one of the best performers in the 1990s, but before that you have to look back to the 1920s to find a decade in which US equity performance was better than middling.

Apparently the strenght of equal weight is exactly in the rebalancing. The assumption is that no country will always outperform the others. If a country outperforms in a period, rebalance by putting your money on the countries that underperformed. This has brought more gains and less volatility in the past.

[covfefe]

reviving an old thread here. with inclusion of China A shares, VWO is now ~ 60% (China + Taiwan) and another ~ 15% (India + Brazil). Not leaving much to other EM

I want to buy other EMs but without having to buy 10+ individual country EMs and the hassle of rebalancing. So I looked up if there is any alternative and came up combining 3 ETFs:

• Work-Horse: VWO or EMVL (Emerging Markets Value Factor)

• China-Dampner: EMXC (Emerging Markets ex China)

• Spice-Mix: FM or EMFM (Frontier Markets)

A 60 : 20 : 20 mix of VWO : EMXC : EMFM gives following country weights at TER of 0.22
Location Weight
China 26.3
Taiwan 13.4
India 8
Brazil 5
Saudi Arabia 4.7
Thailand 4.6
South Africa 4.5
South Korea 3.9
Malaysia 3.5
Russia 3
Mexico 2.6
Indonesia 2.1
Kuwait 2
Philippines 1.9
Vietnam 1.8
Qatar 1.7
United Arab Emirates 1.7
Hong Kong 1.1
Turkey 0.9
Poland 0.7
Chile 0.7

China + Taiwan is down from 60% to under 40 %
India + Brazil is down from 15% to 13%

The difference helps make a long tail of countries with > 1% allocation.

May not be perfect, but manageable buy and hold 3 ETF combination with TER or 0.22

Note 1: Choosing EMVL instead of VWO titls towards value
Note 2: FM is much larger (& possibly more liquid) than EMFM, but it massively overweights Kuwait (26+% in FM vs 7.72% in EMFM)

[glina]

There used to be an Emerging Market All-Country Equal Weight ETF (EWEM) but it got liquidated 2 months ago.

It didn’t outperform, it was actually very close to the bare EM index.

I don’t think you will be able to produce any significant outperformance with the mix above simply for the reason of lacking control over what is going on underneath.

During the recent crash, the equal weight strategy also significantly underperformed standard EM weight in general, China held up pretty strong while some countries had drawdowns of as much as 70% (Colombia, South Africa) or 50% (most others).

Recovery has not been synchronized. Now China is not moving that much, while some countries have already seen 30% recovery.

I would guess that if you are entirely passive, then in a year or two from now the equal weight EM will be back in sync with market cap EM, but being able to rebalance manually could generate some premium.

[covfefe]

@glina The structure certainly lacks the flexibility for manual rebalancing to make use of deep drawdowns and quick recoveries. For that, your approach of using most liquid single country ETFs would be the best.

To be honest, I am also not sure whether market cap weight / equal weight / inverse cap weight will outperform

As I see it, VWO is now a great China+Taiwan fund with competitive TER of 0.1
The country exposure in VWO (50% China + rest) now starts to look like VT (55% US + rest). While I am comfortable with 50% in US, not sure I’am as comfortable for half of EM money to be in China.

The result of this exercise would hopefully be to diversify away political risk, as well as invest in countries which show less correlation with the broad market. Or simply a way to create an more diversified EM fund

[glina]

On second thought, thumbs up on your allocation. I could live with that if I were to just leave my portfolio untouched for years.
I’d replace VWO with EEM to make sure all your EM indices follow the MSCI weighting. Otherwise you are leaving out South Korea which is a developed market according to FTSE.

[wapiti]

Equal weight tend to outperform due to the overweight of small companies.
So the easiest way would be to add a small cap ETF

[covfefe]

MSCI Emerging Markets Small Cap Index still has China (23%), South Korea (17.5%) and China (13.08%)

So still China + Taiwan heavy.

Imagine a political upheaval or military tussle or between China and Taiwan and your EM fund is in smoke!

[glina]

What I ment to say is to replace VWO with EEM as your base index, simply for the reason that VWO excludes South Korea and the other two (EMXC and EMFM) are also MSCI indexed.

[covfefe]

@glina I got your comment. Agree with it. (IEMG would be even better than EEM, TER of 0.13 vs 0.68).

My argument about MSCI EM small cap was a response to @wapiti comment

[wapiti]

yes EM markets are riskier. Each EM has his own risk and political issue. Thinking that China is worse is wrong. It is just more reported in the media. I would not try market timing based on the political context.

I don’t know how much is EM weight on your portfolio but I’m not worth it.

The only thing I’m doing is to add A-share because the current indexes include them partially.

[covfefe]

I don’t think China is worse than any average EM in terms of political or market risk. But the negative repercussions of any trouble with China is many multiple times any other EM!

China can single handedly make or break your regular EM allocation (VWO/EEM/IEMG) any given day in the next 20-30 year That high level of concentration of geographical/political/currency risk is worth consideration!

[leman]

Multifactor strategies generally include the size factor and can be a good proxy to get such tilt.

See the comparison between JPGL and a more classic V3AA.