No, the Swiss broker or bank will completely wash their hands of it. At the end of the day, you’re the one responsible (or your heirs), not them. Whether you actually pay those US taxes or not doesn’t really matter to anyone - except the IRS, obviously.
I honestly can’t imagine that a big chunk of non-US shareholders are properly reporting and paying US estate tax on significant holdings. How many people even hold meaningful amounts of US stocks outside the US?
I’m not telling anyone to evade taxes, but let’s be real - I have my doubts about how much of this is actually being declared and paid.
I’ve already told my girlfriend to move everything over to a Swiss broker if something happens to me. And if I get old and don’t die in some random accident, I’ll probably do the transfer myself while I’m still around. Who knows what the tax situation will look like by then - maybe Australian ETFs will suddenly make the most sense.
It’s one of those topics where opinions are pretty divided. My personal view is this: if someone really wants to minimize risk today, the cleanest setup is an Ireland-domiciled ETF with a Swiss broker. But then, to be consistent, you’d also have to accept not investing in the US market at all.
I am absolutely certain that US-domiciled funds such as VT qualify as US-situs assets.
The estate administrator is legally required by US law to file Form 706-NA (US Estate Tax Return for NRAs) within 9 months of death if US-situs assets exceed $60’000.
When the estate tries to transfer or liquidate the US fund shares, even at a foreign broker, the transfer agent or US custodian in the chain has to require a transfer certificate (IRS Form 5173) confirming estate tax has been addressed.
Why on earth would you inflict such risk and legal uncertainty on your heirs? What do you get in exchange of taking such risks? I am genuinely curious to hear why.
There is an abundance of funds and a plethora of providers in Europe. Why do you think VT is vastly better, and why do you think Vanguard is vastly better?
Even basic Physical IE UCITS funds tracking the same FTSE Global All Cap Index can be at least as good as VT, as you can see below.
Could you please explain what does VT bring more, compared to the tax-optimized, Swap-based portfolio I posted above?
This reminds me of what was being said (and written!) at the Oil&Gas trading desk of a renowned bank in Geneva some time ago.
That bank was a fiercely independent entity, and the pride and joy of Geneva’s financial industry, from its prestigious offices overseeing the river and the lake.
But the powers of the IRS and the DOJ were more far-reaching than expected. The group ended up paying a 9 billion fine. Many smart people were not smart enough, and experienced life-changing events. The entity is now extinct, replaced by a simple branch.
I strongly advise against washing hands of US law. I am dead serious.
You should not compare a FTSE All-World vehicle with a FTSE Global All Cap vehicle, they track different benchmarks. Invesco’s fund does not include small caps, which are included in VT. Small caps have underperformed significantly over the past five years. I also would not use Invesco as a benchmark implementation, as it is relatively new and heavily sampled, so its recent performance is less informative due to higher tracking error, whereas VWCE has a low tracking error (6 bps over the last three years).
It comes down to the fact that VT underperforms its gross benchmark by 15–20 bps annually, while VWCE underperforms by 55–60 bps. In other words, VWCE costs you an additional 35–40 bps per year. That is the cost of insuring your heirs against having to deal with the IRS.
I tend to agree with you intuitively that 35–40 bps is a worthwhile insurance premium to pay. Over a period of 25–30 years, terminal wealth may be around 10% lower with VWCE than with VT, but your estate will be spared the burden of dealing with the IRS.
Yes in theory, no in practice. Vanguard owns the exclusive rights to this specific index, therefore no other third-party fund providers offer direct trackers for it.
Nevertheless, we must compare VT with comparable products. And clearly, they are at least as good.
If you want a better comparison, compare VT to SPYI (MSCI ACWI IMI, which is practically identical to FTSE Global all cap).
EDIT: not to SPYI, which is accumulating, but to SPSA GY, which is distributing. But SPSA is new, there is not much history. There are no otehr all-in-one vehicles in UCITS that are comparable to VT index-wise.
It might be a reasonable cost, but also might not based on the exact circumstances. 30 bps more in TER and 15% lost of the dividends (let’s say 1%) means that you are paying 0.45% of X yearly for an insurance of “if I die, make sure my heirs receive X CHF a couple of years earlier than they would’ve”. The absolute numbers might be too small for you to care (and in that case it’s fine) but that’s a very high premium for such an insurance. For comparison , an actual life insurance (“pay X CHF to my heirs if I die”) could cost you ~0.1% of X yearly for a healthy 30-40yo (according to my LLM of choice, citation needed).
This means that, for example, you have 1M of US stocks, you can either:
Move them to UCITS, and if you die your wife gets 1M very fast
Pay 4k of life insurance per year and keep them in US, and if you die your wife gets 4M very fast, and after a while gets also the extra 1M
I’ve read a bit about unfunded swap etf and I still don’t understand all the worst case scenarios. the 10% loss might be OK to accept, but I don’t get what and when it could happen that the ETF provider need to sell the 90% part. A normal ETF would not have such issues, everything is just worth less until it goes up (when). But the Swap based must sell the 90%. I’m confused.
That’s may be valid reason not to invest in US assets.
I understood your original post and the discussion in this thread to be about the risk of sudden, punitive changes to US law regarding these securities.
Estate tax and concerns about mental capacity are not that - they’re something different. Which is why I feel that shifts a case. Let’s not move goalposts. You have to think about estate taxes and sour risk tolerance today.
„It is also important to dispel the misconception that U.S. stocks owned in accounts outside the U.S. are not subject to the U.S. estate tax. As discussed above, this analysis is based on the type of investment within an account, and not where the account itself is physically located. An individually-owned account in Switzerland with shares of individual U.S. stocks will also potentially be subject to the U.S. tax upon the death of the individual foreign owner."
Let’s be real: the U.S. IRS is one of the last administrations in the world that I’d risk evading taxes from. Or risk, let alone rely on non-reporting.
That goes for banks and brokers and girl- and boyfriends alike.
That’s not to say that moving assets to a Swiss intermediary is wrong. But it shouldn’t be with the motive of knowingly or unwittingly „getting around" estate taxes. And if any „Western" bank’s afraid of any foreign administration, it’s the U.S.
Many ETFs are covered by collateral worth more than 90% of NAV in practice. A Federal Reserve analysis found that the synthetic ETFs it examined were overcollateralized, on average, by about 2%.
Another loss I know of was a small (less than 70M) ETF, Xtrackers S&P Select Frontier Swap UCITS. Nigerian stocks made up roughly 4% of the fund until 2023, when S&P removed Nigeria from the S&P Select Frontier index at a “zero-price”, causing that portion of the fund’s portfolio to be written down to zero.
But that’s pretty much it. Synthetic ETFs behaved as expected, and incurred losses only on two occasions (Russia and Nigeria) as far as I know.
I know that the mustachian way is to reduce costs to the absolute minimum, but:
US dividend yield is barely above 1%
there are one-fund UCITS world ETFs with a TER of 0.07% like XALL
Swiss brokers like Saxo are not that expensive anymore nowaydays
A while back in another thread, I compared buying VT and FWRA on IBKR, Swissqote, neon etc. I just updated my tables with Saxo and XALL. Here’s the gist of it:
Investing for 20 years
IBKR + VT
Saxo + XALL
CHF 100’000 lump-sum
CHF 347’543
CHF 340’968 (-1.9%)
CHF 1’000 per month
CHF 475’158
CHF 468’804 (-1.3%)
CHF 100’000 lump-sum + CHF 1’000 per month
CHF 822’700
CHF 809’771 (-1.6%)
CHF 250’000 lump-sum + CHF 2’500 per month
CHF 2’057’464
CHF 2’026’545 (-1.5%)
Assumptions
7% return p.a. before TER and taxes
FTSE Global All-Cap and FTSE All-World perform identically
share and dividend yield of US vs. ex-US: 62% vs. 38%, 1.2% vs. 2.5%
marginal tax rate: 25%
the entire amount of US withholding taxes are credited each year
IBKR Fixed Pricing, automatic currency conversion
Saxo Classic (0.08% fee, CHF 3 minimum)
So if you:
can’t get the entire US withholding tax credited every year, or
don’t want your heirs to have to deal with the IRS after your death, or
feel safer with a Swiss broker, or
don’t want to accept the risk of swap-based funds,
…you’re looking at ~1.5% underperformance after 20 years.
Because XALL was just launched, I would consider buying it on its most liquid market (probably XETRA in EUR) and in a very liquid day (the first or last business days of the month are safe bets.)
But if we are talking about a monthly investment (as opposed to a big lump sum) honestly I wouldn’t bother optimizing the spread too much, and would stick to a limit order on SIX.
I would avoid poor liquidity days though, for example around Christmas, where the spread could be bigger than normally imaginable.
For a one-fund portfolio, you might consider the unfunded BNP Paribas Easy MSCI ACWI UCITS (LU3086265710) as it has no L1 drag, a 0.06% TER (that’s less than VT) and a higher AUM.
Mit dem Lesen und der Teilnahme an diesem Forum bestätigst du, dass du die Forum-Richtlinien gelesen hast und damit einverstanden bist sowie den Haftungsausschluss auf http://www.mustachianpost.com/de/ akzeptierst.
