Opinion on International or Global Funds

Exiting after 20 years of underperforming is probably not market timing. I hold about 5-7% international and this has been an interesting thread. I too have seen the underperformance. But I've only been without a financial advisor for two months so have not acted on it. Since we are living on our taxable portfolio, slowly dumping the international equities in exchange for food, gas, utilities, and travel might be a good idea

If one takes portfolio theory seriously I do not think you want to exit any full class of assets since you theoretically never know which one may be the leader next year. And aren't all classes underperforming compared to the SP500?

And in the parlance of many isn't selling all your international stocks and instead titling to US megacaps "too risky"?

Not aimed at you specifically EastWest Gal, just a discussion point. :)
 
Last edited:
I would be so happy to see it go! The foreign tax credit increases your taxable income to ~match. For example for 2024 I had something like $1225 foreign tax credit, but $1230 in additional income due to the foreign tax paid by the funds. I didn’t receive that $1230, but I am still taxed on it and it still adds to my AGI. Also impossible to know or predict until after year end, so it interferes with accurate tax/AGI planning.

The whole thing is a royal pain in the butt. Not only the Form 1116, but also dealing with carryovers some years.

In other words, there is no benefit to receiving a foreign tax credit in this case of passive international income. Just a lot of extra paperwork, and a higher AGI so having to leave space when trying to stay below a certain IRMAA threshold and hoping you guessed well.
You know I think it depends on what vehicles you use for international investments (other than the S&P500). Last year, I had a total of @9% between Schwab International (SCHF) and Vanguard FTSE Emerging (VWO), with 75% of my international equities invested in Large Cap (SCHF). Dividends amounted to $6300 with $487 going to taxes.
 
I hold 10% International (Schwab International Equity SCHF). I try and follow the mantra of being diversified. But it certainly hurts when you compare the S&P total return vs International as shown in this pic.
View attachment 54162

But, yet I still want to hold some of these big world companies so I keep it in my portfolio (. C'mon... Nestle ...the world needs its chocolate. ;) . Plus I own 2 Toyota Tacomas; great trucks!

Schwab International Equity ETF (SCHF)
Top 15 Holdings
SymbolHolding% Assets
SAPSAP SE
1.47%​
ASMLASML Holding NV
1.34%​
NOVO.BNovo Nordisk A/S Class B
1.26%​
NESNNestle S.A.
1.07%​
7203
Toyota Motor Corp.
1.05%​
ROGRoche Holding Ltd Dividend Right Cert.
1.03%​
AZNAstraZeneca PLC
1.00%​
NOVNNovartis AG
0.99%​
SHELShell Plc
0.96%​
HSBAHSBC Holdings Plc
0.91%​
MCLVMH Moet Hennessy Louis Vuitton SE
0.88%​
5930
Samsung Electronics Co., Ltd.
0.86%​
RYRoyal Bank of Canada
0.83%​
CBACommonwealth Bank of Australia
0.80%​
SIESiemens Aktiengesellschaft
0.75%​
The problem with this comparison is that none of the foreign stocks in the Schwab International (SCHF) ETF are in the S&P 500. So their returns cannot match the S&P500. But it is also important to know that a very tiny percentage of SPY is owned by American companies that are part of the S&P500 and their income is reported by the parent companies. Here's the breakdown of where in the world S&P 500 companies earn income in the SPY ETF. Source is Charles Schwab, and you can see you do have earnings from foreign countries. Digging deeper in the SPY Prospectus may tell you which companies in the countries below are generating income in the SPY basket of investments.


1738539430790.png
 
The Jack Bogle recommendation was that ex US is not necessary. Somehow the BogleHEADS have morphed that into keeping global cap weight in ex US stocks, which is about 40%.
Bogle's recommendation has been right for the last 15 or so years, as ex US (VXUS at Vanguard) has returned very little beyond dividends since its inception. The 3 Fund Portfolio, imo, has been a smelly dog for years but the geniuses over at Bogleheads won't admit it.
The Boglehead idea, to the extent that a monolithic one exists, is that a truly agnostic approach would hold everything at market-weight. This means a US vs. ex-US allocation split also at market weight. The rationale is related to Efficient Markets Theory, which in some form implies that barring some insider information or lucky guesses, there is nothing to justify trading beyond market-cap. The additional argument is that P/E is high in the US and lower ex-US, so that for $1 of future earnings, a US investment is more expensive.

I am not persuaded that the BH approach is correct, nor am I trying to justify it. But might as well at least state their position....
 
I am put in mind of the observation that "in theory there is no difference between theory and practice, while in practice there is."

1882 February, The Yale Literary Magazine, Conducted by the Students of Yale College, Volume 47, Number 5, Portfolio: Theory and Practice by Benjamin Brewster, Quote Page 202, New Haven, Connecticut.
 
I posted a similar question 6 years ago and got a lot of different opinions:

I got rid of some ex-US funds but my portfolio is still about 5% ex-US. I need to think about this again too, thanks for the reminder.
 
This morning, I traded all Global and International funds to other funds. I think and hope I did the right thing.
 
ss
If one takes portfolio theory seriously I do not think you want to exit any full class of assets since you theoretically never know which one may be the leader next year. And aren't all classes underperforming compared to the SP500?
I tend to think of market timing as a short term activity, about three years or less, most often one year or less.

I compared my total US stock market index fund to my total international index fund. It’s not even a close race. Sp500 is a nice indicator to watch on a day to day basis, but my money and my financial heart are in the total US market. That’s the gal I take to the dance and go home with.
 
I can hear the Beatles record, "I've Got a Feeling."

Each year the global outlook reports do reinforce the int'l component many indexers have. This is of course repeating for 2025.

My int'l AA since 2007 has dropped each year. Retired in 2020, and during the 401(k) moves to IRA, I halved the int'l AA. I have no regrets about that. Maybe I will at the end of next year. We'll see.

200722%
200815%
200921%
201020%
201119%
201221%
201320%
201418%
201517%
201618%
201718%
201816%
201913%
202011%
20215%
20229%
20239%
20249%
 
I'm not sure why indexers are so negative on international stocks, enough to give up on their entire investment thesis.

After all your bond indices have woefully underperformed. Is that the next thing you jettison?
 
I'm not sure why indexers are so negative on international stocks, enough to give up on their entire investment thesis.

After all your bond indices have woefully underperformed. Is that the next thing you jettison?
Well, I have given those up too back in 2018 and replaced them with CD/TSY/MYGA's.
 
If one takes portfolio theory seriously I do not think you want to exit any full class of assets since you theoretically never know which one may be the leader next year. And aren't all classes underperforming compared to the SP500?
There are a lot of commentaries on MPT. But one of the biggest assumption of MPT is there is a negative correlation between instruments. IMHO we have lost negative correlation (among many investment classes: bonds, international, RE) for over two decades and I am not sure it is coming back with the amount of financial engineering we have been doing.
 
International funds have underperformed my US stock exposure by a significant amount. I bought them for diversification but would have been much, much better off buying a diversified fund such as VG Total Stock Index.
Same and it is annoying that Total international stock from vanguard has a higher expense ratio (.08 vs .04) and generates a lower percentage of qualified dividends. I plan to cut back on international when I pay taxes on my roth conversions this year.
 
International developed makes up 12% of my equities with Emerging Markets another 4%. I figure all companies work to make money no matter where on the globe they are.
 
It's funny because on the various invest sub-reddits I follow, people always preach diversity in their holdings, and love to point out the time periods where Intl outperformed US. But if you point out that US has outperformed Intl for a decade and a half, they'll echo "past performance doesn't guarantee future returns" (which flies in the face of their example where they cite past performance).

The reality is that today, companies in the US have vast international exposure already, as they do business globally. Sure, you could argue that some domestic policy may affect US holdings that wouldn't affect Intl holdings, but have we really seen that yet? Maybe with the current administration we could see some of that, but it's unknown at this point.

YTD, Intl seems to be about on par (slightly ahead) of US. I hold about 20% of my investment in Intl currently, but depending on how the next couple of years that may end up changing.
 
I have had holding with global funds for over 40 years. I finally decided to exchange them for the same reason as many have expressed. In my other funds I hold there are global funds within those funds, and I was tired of the taxes on those foreign holding.
 
The Boglehead idea, to the extent that a monolithic one exists, is that a truly agnostic approach would hold everything at market-weight. This means a US vs. ex-US allocation split also at market weight. The rationale is related to Efficient Markets Theory, which in some form implies that barring some insider information or lucky guesses, there is nothing to justify trading beyond market-cap. The additional argument is that P/E is high in the US and lower ex-US, so that for $1 of future earnings, a US investment is more expensive.

I am not persuaded that the BH approach is correct, nor am I trying to justify it. But might as well at least state their position....
A good summary by Diogenes.

mrhfeh in post #21 has an excellent link to a Bogleheads thread where the first post lists arguments (some with cites) for and against US/International allocation.

My 2 cents: Each asset class has long (>decade) historical periods of outperforming the other. p.118 of Bernstein's The Four Pillars of Investing (2002) has a graph showing S&P500 5.5% / EAFE 10% from 1970-79, 17%/22% 1980-89, and 18%/7.5% 1990-99. (All numbers are my eyeball approximation of the graph). Presumably, in 1990 those who are now dropping international would have dropped US after 20 years of underperformance and missed out on US outperformance since then. Nevertheless, over a longer time period (50 years, 1970-2019) S&P500 total returns were 8.82% and EAFE 9.13% (p.201 graph Malkiel's A Random Walk Down Wall Street 2023). EAFE volatility was higher; 21.4% vs. 16.5%. So, unless something fundamental has changed (this time it's different), long-term returns should be roughly equal. Reversion to the mean suggests higher international returns over the next 20-30 years. Swedroe article link showing various valuation metrics US/international as of July 2024:
https://www.morningstar.com/markets/can-us-continue-outperform

Our AA has 34% Developed Markets and 6% Emerging Markets in the equity portion, so 60/40 US/Ex-US overall. We plan to continue with this. In my view, this is diversification and rebalancing at work; the better performing US stocks are being rebalanced into the poorer performing international stocks. Outperformance in one asset class covers for underperformance in another class. Expected returns and actual returns vary (volatility). It's just going on longer than many folks are comfortable with.
 
Fun fact - When I bought my first international fund several decades ago Europe and the USA had about the same GNP on a per-capita basis. Today, a recent article has claimed that the US per-capital GNP is 20% higher than Europe. This may be one reason why International funds have not done so well.
Not all economic activity is productive. The US spends about 10 percentage points more of its GDP (roughly double) on medical care than European countries, with inferior health results. Therefore, one needs to be cautious with that comparison.

There are other distortions of GDP in the world. For example, Ireland is not as rich as its GDP suggests because much of the world's aircraft leasing happens within its borders, inflating its GDP substantially. That activity has little real-world impact on the country's wealth.
 
International funds have underperformed my US stock exposure by a significant amount. I bought them for diversification but would have been much, much better off buying a diversified fund such as VG Total Stock Index.

I noticed this about a year ago. It's sad that when one of your mutual funds has been around for 20 years and your total gains is like 2% over that period. It's a small portion of my portfolio but this thread has been a kick in my rear to do something.

After two decades of underperformance, I sold all my international funds in 2019, right after I retired. Domestic US companies have plenty of international exposure.

This morning, I traded all Global and International funds to other funds. I think and hope I did the right thing.

I just executed my first set of trades this morning. I have international funds in 3 different accounts. They're a small portion of my portfolio (1-2%) but I'm kicking them out. Will dollar cost average over 6 months. Will reinvest the sales into my total US market funds.
 
I bought HLEMX (emerging markets which is probably than worse than regular international funds) at about $50 in 2007. I sold it for about $50 in 2014. It's now $37 plus 1.14% expense ratio!

Luckily it was just a tiny amount of my whole picture. Never again international stocks.
 
I noticed this about a year ago. It's sad that when one of your mutual funds has been around for 20 years and your total gains is like 2% over that period. It's a small portion of my portfolio but this thread has been a kick in my rear to do something.





I just executed my first set of trades this morning. I have international funds in 3 different accounts. They're a small portion of my portfolio (1-2%) but I'm kicking them out. Will dollar cost average over 6 months. Will reinvest the sales into my total US market funds.
Good for you!! I should have done it years ago.
 
I have ~10% international in my retirement portfolio (none elsewhere). I have an asset allocation for that portfolio that I follow and rebalance.
 
Back
Top Bottom