Moving from VT to VWRL (or VWRD, or VEVE/VFEM)

[Dr.PI]

@Twain

I decided to invest in VWRD as my core ETF going forward, and not only for the sake of simplicity and “laziness”.

That surprises me, because as an investor, fees are basically the only thing that you can control. And a TER advantage of 0.09% is not that small as you have shown yourself.

@Twain

in a 90%/10% allocation

If you want to do it strictly according to the market state, that should be proportional to the capitalization of underlying indices. Check FTSE index factsheets. Now it is around 90/10, if emerging markets do better than developed it can go to 85/15, if not to 95/5, and so on. And once 2 funds are brought into the market capitalization proportion they will stay like this for quite some time, might be years even, before the proportion is going to drift away because of dividends, index changes etc.

@Twain

a quick look at the factsheets will tell us that the VWRD contains 3784 number of stocks, and VEVE+VFEM 2254 + 1893 = 4147 number of stocks

Yes, and for me it was an argument for 2 funds portfolio, as it is actually invested in all securities tracked.

@Twain

As a result slight deviations of the allocation would occur constantly and influence the short- and consequently the long-term performance of the portfolio,

With such an inequal ratio between two subindices, you should really deviate a lot to have a substantially different investment returns. Like 15% or 5% EM instead of 10%.

@Twain

Comparative Overview of IB Commissions and Exchange/Transfer Fees,

I think you have “fixed” commission type. With my “tiered” I see lower numbers in “order preview”. And all in all it looks like I guessed before: buy in the currency (USD/CHF/EUR) that you have at hands now.

But I also see your point about a simplicity of buying one fund only. I can suggest a hybrid approach:

• Split your current VWRwhatever holdings into VEVE and VFEM.
• Buy VWRD/L only until you get to 20/30/50/something k,
• then split what you have accumulated.

Or, even better, you can do a “pragmatic” investment in your 2 funds portfolio. When you add money to your portfolio, calculate how much should go into the each fund according to the current market capitalization, but only buy smaller position if the amount to be invested is more than for example 1k. Like this you can avoid small purchases, and the more your portfolio is getting, the less significant is your “tracking error”.

[yetanothername]

Minor issue, but why bother with reinvesting dividends manually when you can buy VWCE which is accumulating?

[Twain]

Both @jay and @Dr.PI made some very good points, which might just make me reconsider my choice.

I realise now I’ve been a victim of a “simplicity” bias after having invested in VWRL and especially lately in VT on IB, with just one ETF to buy every month at fees of 0.35 USD per trade (@Dr.PI I’m on a tiered acount as well; regarding the calculations above I just took them from the IB fee page for tiered accounts) and had an inner resistance to the “two-fund portfolio with higher fees and periodic rebalancing” issue.

And, considering that we are already better off than the many that do not invest their savings at all, or invest them with active managers with 2% annual management fees, this is so true:

@"jay

That said, it’s perfectly fine to simply invest in VWRD, of course. The overall fees are still very low. And compared to lots of other investment decisions, this won’t have a big impact. It’s much more important to stay the course. I.e. choose a solution that you’re comfortable with.

so from a psychological point of view I was reluctant to chase a “potentially great” portfolio instead of a good one, and keep it as simple as possible.

But because I totally agree with this:

@Dr.PI

…because an investor, fees are basically the only thing that you can control.

I was willing to research how I could optimize the fees with VEVE + VFEM.

---

Now I do have some comments on the above:

@jay

When VFEM performs better than VEVE (or the other way round), the ratio between the two in your portfolio will indeed change a bit. However, the ratio of developed markets vs. emerging markets within e.g. VT will also approximately change the same way. I.e. such performance differences wouldn’t move you away from a global market cap weighted portfolio.

Does the current weighting of DM et EM (currently 89.9 / 10.8% per VWRL factsheet) really change short-term if one pertforms better than the other? I imagine it’s more like @Dr.PI said:

@Dr.PI

And once 2 funds are brought into the market capitalization proportion they will stay like this for quite some time, might be years even …

Like this, as long as I check the weighting and keep as close to it as possible every month there shouldn’t be major deviations.

@"jay

I check which ETF is further below the target allocation and then buy shares of that ETF.

@Dr.PI

[…] calculate how much should go into the each fund according to the current market capitalization, but only buy smaller position if the amount to be invested is more than for example 1k

Although this is what I had been doing with my VT/BRK combo because the short term allocation swings didn’t matter to me, I don’t know if I would do this with VEVE/VFEM, because I want to fully invest the same amount monthly, putting for example 2k in VEVE for 9 months, and then 2k in VFEM for 1 month would not be practical.

The only way to do it and avoid too complex of calculations is to simply invest in both every month. And it’s not that expensive actually:

• The FX commision is unique, so really just the exchange/trade fees for the second trade (VFEM) are additional compared to buying VWRD.
• And if the 10% VFEM position to be added is anywhere between 1 and 2000 GBP, the fees on LSE are the same and equivalent to the minimum which is 1.16 GBP per trade, or about 1.35 CHF. So for about 15 CHF annually you assure a constant allocation without any adjustments.

And not only with the added benefit of having the lower TER work for you in the accumulation phase, but more importantly in the distribution phase and in retirement where on a 3M portfolio the extra 3k a year is not negligible (Naval Ravikant’s book mentions the idea that usually a scenario of short-term pain with associated long-term gain is preferable to the inverse, even though counter-intuitive, mostly due to compound interest; and not only when it comes to finances).

[Twain]

@yetanothername

Minor issue, but why bother with reinvesting dividends manually when you can buy VWCE which is accumulating?

For a Swiss investor, you pay taxes on dividends either way, and the consensus is that it simplifies the tax return having a more transparent portfolio report with the dividend amounts declared etc., and maybe it also creates issues with the capital tax when sold?

[Bojack]

@Twain

I decided to invest in VWRD as my core ETF going forward

I personally believe this is the best option. Especially if you’re using IB, which has very cheap currency conversion fees. USD is the base currency of that asset, and it pays dividend in USD, so I think it’s transparent to trade in USD as well.

Comparing TER (which can change in the future, so it is unknown, but we expect it to keep decreasing) on such small differences over a lifetime of investing seems like splitting hairs. You think “if 50 years from now I have 10m and I saved 1% more, that’s 100k”, but this number lies very far into the future and is riddled with uncertainties, and it’s not as simple as you giving your future self 100k.

With only having VWRD and not some manual mix, you lose the temptation to time the market and overweigh certain regions. This could have a far greater impact on return than the TER difference.

Minimalism is also a consideration.

[Twain]

@Bojack

…such small differences over a lifetime of investing seems like splitting hairs.
Minimalism is also a consideration.

Absolutely. As revealing as this theoretical exercise was, it doesn’t change the fact that in real life it just makes it easier and safer (psychologically) to invest in one fund.

And the splitting hairs can go deeper. In 20 years there might very well be a significant performance difference between the investors who chose VT and those who chose VWRD. It wouldn’t mean one was better than the other.

[Twain]

@CHRad

@Bojack

I have 10m and I saved 1% more, that’s 100k”

That’s not how it works… 1% more rolling fees make a huge difference over decades (it’s not -1% at the end of your portfolio, it’s the same with 5% or 6% interest, that makes a huge difference after longer time), I believe after 40 or 50 years you have half the amount.

Yes, the rolling fees after you already accumulated your (retirement) portfolio matter, but in this case would still be only 0.1% annually (the TERs might very well get lower with time but probably for all Vanguard funds proportionally).

To second what @Bojack said, if you have 10 million after 20 years, that means you’ve been investing 17k a month all this time (at 8% for 20 years) to get there. If you have 17k excess liquidity a month, the 100k is just 6 months additional contributions and even less significant - this “6 months rule” actually applies to any size portfolio.

[Bojack]

@CHRad

That’s not how it works… 1% more rolling fees make a huge difference over decades (it’s not -1% at the end of your portfolio

You have clearly misunderstood what I’m saying. I said that a small difference in TER might look huge when accumulated over many years and applied to the entire portfolio, but it’s a misleading perspective, as the 100k saved 50 years from now does not have a current value of 100k. I did not talk about an annual difference of 1%…

annual_perf_diff = 0.04%
year_count = 50

without compounding
0.04% * 50 = 2%

with compounding
1.0004 ^ 50 - 1 = 2.019726%

performance difference after 50 years => 2%

Btw, I checked, out of curiosity, what is the value of 100k using DCF method

future_value = 100'000
discount_rate = 10%
year_count = 50

present_value = future_value * (100% - discount_rate) ^ 50
100'000 * 90% ^ 50 = 515

This means, if you offered me a “ticket” that I could reimburse in 50 years and receive $100’00 for it, I would be willing to pay you $515 for it…

[Bojack]

@Twain

but in this case would still be only 0.1% annually (the TERs might very well get lower with time but probably for all Vanguard funds proportionally).

I can imagine that in 20 years VWRL/VWRD has a TER of 0.11% and VEVE+VFEM is 0.07%. But of course, we don’t know. Just wanted to point out that a “proportional decrease” will also cut the gap proportionally, and not keep it.

Btw, is it actually clear that the “actual” fee/cost of the ETF is the advertised TER? I tried once doing some comparisons, but it’s very tricky. I think that all the dividend-paying ETFs which track an index, eventually hug that index, so the difference is not visible in price, I think? So I assume that they deduct the fee from the dividend. I just want to say, I don’t have enough knowledge, but I wouldn’t just blindly assume that you can take the face value of TER and strictly apply it as cost.

[Bojack]

And is TER a value which is known a priori and the deductions are fixed for the future? As such, is it guaranteed by the fund that this value will not be exceeded? Or is it rather a summary of the funds total costs minus income (if they do securities lending, market making, or other shenanigans) from a past period? I guess, reading your comment, that it has to be the former. But in that case, the ETF would have to bear the risk and cover the difference between the income from the TER and the actual cost of running the fund.

[Bojack]

I just wonder why we don’t see this constant divergence between the S&P 500 and VOO or even VUSD. With a TER of 0.10%, you would expect a difference of 1% after 10 years. Or maybe we do see it, but it’s so small that it’s invisible on this chart without a proper zoom.

[Bojack]

That makes sense to me, but then we should see a steadily widening gap between the index and the ETF over the years.

[Dr.PI]

@Twain

In 20 years there might very well be a significant performance difference between the investors who chose VT and those who chose VWRD.

Which you can’t control or estimate a priori. Unlike fees.

@Bojack

I wouldn’t just blindly assume that you can take the face value of TER and strictly apply it as cost.

Splitting the world

I was thinking about updating this post, but decided not to. Too many variables that affect total investment costs. Instead I decided to outline what I think is a right way to calculate the total investment costs for an ETF. As a Swiss investor, one is only taxed on dividends, and namely up to 3 times. Let’s consider the whole process. All companies included in a fund are paying out dividends, which all together make gross dividend yield (in %) D_G. The gross dividend yield of various baskets …

@Bojack

As such, is it guaranteed by the fund that this value will not be exceeded?

There is a management fee which is fixed in advance and other costs, mostly trading fees. There are expected TER and actually realized TER, typically 1 year rolling. It should be possible to find both in reports etc.

@Bojack

I just wonder why we don’t see this constant divergence between the S&P 500 and VOO or even VUSD.

The fund management is using futures to “leverage” cash positions and bring the fund exposure into an agreement with the index tracked.

[Dr.PI]

Going back to the original question. I calculate actual costs saved based on the portfolio size and see if it is worth the effort. 0.09% with 200k portfolio size saves you 180 CHF per year. For me it is totally worth the effort and extra fees.

[Dr.PI]

@CHRad

The fund has a small cash buffer to cover that and I guess the buffer gets refilled when rebalancing (0.07% isn’t that much lol)

Considering that US taxation very strictly distinguishes between dividends and capital gains, selling securities to cover costs would be a big no-no, I guess.

[Bojack]

@Dr.PI

The fund management is using futures to “leverage” cash positions and bring the fund exposure into an agreement with the index tracked.

Huh? So they’re eventually making up for the TER?

[San_Francisco]

@Twain

For a Swiss investor, you pay taxes on dividends either way, and the consensus is that it simplifies the tax return having a more transparent portfolio report with the dividend amounts declared etc., and maybe it also creates issues with the capital tax when sold?

As long as your fund is listed in the federal tax administration‘s database (which VWCE is), there’s no simplification in tax matters. Software will calculate things for you, and you don’t have to reinvest dividends with an accumulating fund.

On the other hand, there have been reports of „refunds“ (or rather offsetting) of U.S. withholding tax having been denied with the U.S. ETFs - even though they did have a portfolio reports indicating withheld taxes.

[uji]

Thank you everyone for the great comments in this thread.
I currently own 100% VT (with IB), accumulation phase. I’m concerned of staying with VT for the reasons listed above. I am considering switching to VWRD.

Question on tax impacts: Is there a best time to sell the VT shares and buy VWRD?

• I’m currently negative on my VT investments. Would it be better to sell now and switch to VWRD to avoid capital gain taxes?

• Any benefits of keeping both VT and VWRD? And potentially selling VT later on.

[Burningstone]

@uji

I’m currently negative on my VT investments. Would it be better to sell now and switch to VWRD to avoid capital gain taxes?

There’s no capital gains tax in Switzerland, unless you are qualified as a professional investor.

How much VT are we talking about?

[Dr.PI]

@uji

Would it be better to sell now and switch to VWRD to avoid capital gain taxes?

There are no capital gain taxes for private investors in Switzerland.

@uji

Any benefits of keeping both VT and VWRD?

If you insist on distributing shares, there is no real difference between them.

Furthermore, you can buy VWRL with any European broker, it doesn’t have to be IB. (Or yes, margin loans).