VTI/VXUS split

Have a question about your personal investments? No matter how simple or complex, you can ask it here.
Topic Author
ilyaraz
Posts: 5
Joined: Thu Aug 27, 2020 2:52 pm

VTI/VXUS split

Post by ilyaraz »

What is a good framework to think how to split the stock index fund portfolio between US and international stocks? Ideally, if it's at least somewhat quantitative.

My current allocation is close to 80% VTI, 10% VXUS and 10% VNQ (not counting some individual stocks and other assets not relevant to the discussion), and I'm thinking of bumping up the VXUS share, but I'd like to do it in a principled way rather than just blind guessing.

How do you choose the allocation domestic/international beyond the gut feeling?
livesoft
Posts: 85971
Joined: Thu Mar 01, 2007 7:00 pm

Re: VTI/VXUS split

Post by livesoft »

Vanguard had / has a white paper, but you know, there is not much quantitative about this i think. Do you see why after reading the white paper?
Wiki This signature message sponsored by sscritic: Learn to fish.
Silk McCue
Posts: 8912
Joined: Thu Feb 25, 2016 6:11 pm

Re: VTI/VXUS split

Post by Silk McCue »

There are an almost unending and contentious threads regarding allocation between US and International. There will be nothing new added in this thread. It’s certainly fine to ask but I learn the most by searching topics I have questions on.

I hold Vanguard Total World which is roughly 60/40 US/ Intl. some subscribe to no more / no less than 20. Others say 0% is the only answer. All Vanguard Target Date and Lifestrategy funds hold around 60/40 equities US/Intl.

There is no definitive answer as to what is right for you.

Cheers
stan1
Posts: 14235
Joined: Mon Oct 08, 2007 4:35 pm

Re: VTI/VXUS split

Post by stan1 »

There's no quantitative answer since the future can't be predicted. If there was hedge funds would always get it right and they don't.

So you are down to what assumptions you want to personally make and there are three groups:

1) Total World Market Cap weight (roughly 60 US/40 ex US) - the future can't be predicted so trust market cap weighting and efficient markets to take all the risks and rewards into account and give you a portfolio that is likely to perform better than most other choices

2) 100% US - believe US is the best investment environment available and will be for some time to come (inclusive of securities regulation, government support for business and investing, innovation, reserve currency, etc.)

3) Both sides have some merit so split the difference somewhere around 80/20 US/exUS to 67/33 US/exUS. Basically if you don't know the answer pick something in the middle.

A slice of international less than 20% of equities is highly unlikely to make a difference.

That's about it. Pick one and stay with it for the rest of your life (or at least until there is new information that is actionable).
Last edited by stan1 on Sun Sep 19, 2021 12:58 pm, edited 1 time in total.
MotoTrojan
Posts: 11259
Joined: Wed Feb 01, 2017 7:39 pm

Re: VTI/VXUS split

Post by MotoTrojan »

stan1 wrote: Sun Sep 19, 2021 12:55 pm There's no quantitative answer since the future can't be predicted.

So you are down to what assumptions you want to personally make and there are three groups:

1) Total World Market Cap weight (roughly 60 US/40 ex US) - the future can't be predicted so trust market cap weighting and efficient markets to take all the risks and rewards into account and give you a portfolio that is likely to perform better than most other choices

2) 100% US - believe US is the best investment environment available and will be for some time to come (inclusive of securities regulation, government support for business and investing, innovation, reserve currency, etc.)

3) Both sides have some merit so split the difference somewhere around 80/20 US/exUS to 67/33 US/exUS. Basically if you don't know the answer pick something in the middle.

A slice of international less than 20% is highly unlikely to make a difference.

That's about it.
You forgot those of us that weight more closely to the fundamental weighting, which is closer to 45% US (I am 40% US, 40% DM, 20% EM myself, all in value).

https://www.rafi.com/content/dam/rafi/d ... et-usd.pdf
stan1
Posts: 14235
Joined: Mon Oct 08, 2007 4:35 pm

Re: VTI/VXUS split

Post by stan1 »

MotoTrojan wrote: Sun Sep 19, 2021 12:57 pm
stan1 wrote: Sun Sep 19, 2021 12:55 pm There's no quantitative answer since the future can't be predicted.

So you are down to what assumptions you want to personally make and there are three groups:

1) Total World Market Cap weight (roughly 60 US/40 ex US) - the future can't be predicted so trust market cap weighting and efficient markets to take all the risks and rewards into account and give you a portfolio that is likely to perform better than most other choices

2) 100% US - believe US is the best investment environment available and will be for some time to come (inclusive of securities regulation, government support for business and investing, innovation, reserve currency, etc.)

3) Both sides have some merit so split the difference somewhere around 80/20 US/exUS to 67/33 US/exUS. Basically if you don't know the answer pick something in the middle.

A slice of international less than 20% is highly unlikely to make a difference.

That's about it.
You forgot those of us that weight more closely to the fundamental weighting, which is closer to 45% US (I am 40% US, 40% DM, 20% EM myself, all in value).

https://www.rafi.com/content/dam/rafi/d ... et-usd.pdf
Yeah, I was trying not to over complexify but you got me :-)
MotoTrojan
Posts: 11259
Joined: Wed Feb 01, 2017 7:39 pm

Re: VTI/VXUS split

Post by MotoTrojan »

stan1 wrote: Sun Sep 19, 2021 12:59 pm

Yeah, I was trying not to over complexify but you got me :-)
Just teasing. But the OP asked for a quantitative approach, of which the fundamental weighting is a great one as it is immune to relative over/under-valuation (bubble) scenarios where market-cap weighting isn't a great metric.
User avatar
TxFrog
Posts: 92
Joined: Mon Mar 01, 2021 7:45 pm

Re: VTI/VXUS split

Post by TxFrog »

The equities in my taxable account are split 60% VTI and 40% VXUS.

I've probably spent way too many hours researching S&P500 vs TSM and what's the "right" amount of international. I finally came to the realization Vanguard and Fidelity have trillions of dollars in AUM and their target date fund equities are 60% US and 40% international. I'm not smarter than them and might as well use their "free" research.
User avatar
burritoLover
Posts: 4097
Joined: Sun Jul 05, 2020 12:13 pm

Re: VTI/VXUS split

Post by burritoLover »

ilyaraz wrote: Sun Sep 19, 2021 12:27 pm What is a good framework to think how to split the stock index fund portfolio between US and international stocks? Ideally, if it's at least somewhat quantitative.

My current allocation is close to 80% VTI, 10% VXUS and 10% VNQ (not counting some individual stocks and other assets not relevant to the discussion), and I'm thinking of bumping up the VXUS share, but I'd like to do it in a principled way rather than just blind guessing.

How do you choose the allocation domestic/international beyond the gut feeling?
How would you answer the following:
Over a 10 year period, your U.S. allocation returns 16%, while your international allocation makes 8%. What do you do?
A. Nothing - you continue contributing to and rebalancing into international as before, maintaining the same allocation.
B. You stop contributing and rebalancing to international because obviously it is not worth it.
C. You sell off some or all international to put more in the U.S.
Da5id
Posts: 5058
Joined: Fri Feb 26, 2016 7:20 am

Re: VTI/VXUS split

Post by Da5id »

stan1 wrote: Sun Sep 19, 2021 12:55 pm There's no quantitative answer since the future can't be predicted. If there was hedge funds would always get it right and they don't.

So you are down to what assumptions you want to personally make and there are three groups:

1) Total World Market Cap weight (roughly 60 US/40 ex US) - the future can't be predicted so trust market cap weighting and efficient markets to take all the risks and rewards into account and give you a portfolio that is likely to perform better than most other choices

2) 100% US - believe US is the best investment environment available and will be for some time to come (inclusive of securities regulation, government support for business and investing, innovation, reserve currency, etc.)

3) Both sides have some merit so split the difference somewhere around 80/20 US/exUS to 67/33 US/exUS. Basically if you don't know the answer pick something in the middle.

A slice of international less than 20% of equities is highly unlikely to make a difference.

That's about it. Pick one and stay with it for the rest of your life (or at least until there is new information that is actionable).
Yeah, I agree with the above.

IMO worst case is to pick something, not be confident in it, then chase performance. Buy international, dump it when US outperforms for a while. Then flop back and buy ex-US somewhere into the next cycle where international outperforms.

But OP your desire for a quantitative answer based on the imperfect data and very limited visibility of the future, well, not going to happen. I have 60% US/40% International, but don't think that number is anything special. I choose to own international, and having chosen that I want to own enough to matter. My specific number is ~what Vanguard Life Strategy funds do, which is roughly market cap these days. But those aren't quantitative arguments, nor do I have any to provide.
Last edited by Da5id on Sun Sep 19, 2021 2:13 pm, edited 1 time in total.
MotoTrojan
Posts: 11259
Joined: Wed Feb 01, 2017 7:39 pm

Re: VTI/VXUS split

Post by MotoTrojan »

burritoLover wrote: Sun Sep 19, 2021 1:56 pm
How would you answer the following:
Over a 10 year period, your U.S. allocation returns 16%, while your international allocation makes 8%. What do you do?
A. Nothing - you continue contributing to and rebalancing into international as before, maintaining the same allocation.
B. You stop contributing and rebalancing to international because obviously it is not worth it.
C. You sell off some or all international to put more in the U.S.
If you answer B or C, then it is important to follow-up with how you'd respond if just after doing that, the following decade is 0% for US, and 8% again for international (no change)?
lostdog
Posts: 5361
Joined: Thu Feb 04, 2016 1:15 pm

Re: VTI/VXUS split

Post by lostdog »

burritoLover wrote: Sun Sep 19, 2021 1:56 pm
ilyaraz wrote: Sun Sep 19, 2021 12:27 pm What is a good framework to think how to split the stock index fund portfolio between US and international stocks? Ideally, if it's at least somewhat quantitative.

My current allocation is close to 80% VTI, 10% VXUS and 10% VNQ (not counting some individual stocks and other assets not relevant to the discussion), and I'm thinking of bumping up the VXUS share, but I'd like to do it in a principled way rather than just blind guessing.

How do you choose the allocation domestic/international beyond the gut feeling?
How would you answer the following:
Over a 10 year period, your U.S. allocation returns 16%, while your international allocation makes 8%. What do you do?
A. Nothing - you continue contributing to and rebalancing into international as before, maintaining the same allocation.
B. You stop contributing and rebalancing to international because obviously it is not worth it.
C. You sell off some or all international to put more in the U.S.
D. You buy equity at world market cap is at the time. No need to rebalance. Just buy and sell at world market cap weight.

E. Just buy VT.
Stocks-80% || Bonds-20% || VTI/VXUS/AOR
Topic Author
ilyaraz
Posts: 5
Joined: Thu Aug 27, 2020 2:52 pm

Re: VTI/VXUS split

Post by ilyaraz »

Thanks for your replies!

As I suspected, there is no good answer so it ultimately just boils down to "reasonably guessing" initially and then staying the course.
User avatar
Noobvestor
Posts: 5944
Joined: Mon Aug 23, 2010 1:09 am

Re: VTI/VXUS split

Post by Noobvestor »

ilyaraz wrote: Sun Sep 19, 2021 12:27 pmMy current allocation is close to 80% VTI, 10% VXUS and 10% VNQ (not counting some individual stocks and other assets not relevant to the discussion), and I'm thinking of bumping up the VXUS share, but I'd like to do it in a principled way rather than just blind guessing. How do you choose the allocation domestic/international beyond the gut feeling?
You can go market weights, which is fully neutral. I'm 50/50, because when I started that was close to market weights (and still isn't that far off) and it's simple - round/even numbers keep me from wanting to tinker and make rebalancing easier.

Vanguard's whitepapers show volatility reduction anywhere from 1% to 60%+ (with max benefits in the 30-50 range IIRC). If you want a principled approach, though, again: market weights is neutral - hard to get more robust than that.
"In the absence of clarity, diversification is the only logical strategy" -= Larry Swedroe
User avatar
Noobvestor
Posts: 5944
Joined: Mon Aug 23, 2010 1:09 am

Re: VTI/VXUS split

Post by Noobvestor »

ilyaraz wrote: Sun Sep 19, 2021 5:15 pm As I suspected, there is no good answer so it ultimately just boils down to "reasonably guessing" initially and then staying the course.
Yes, I think that's true to an extent. But with 80%+ US, staying the course may be hard if US underperforms for a long time. One of the reasons to go somewhere closer to market weights is to reduce potential tracking error -- regrets that might lead to bad future changes to your AA.
"In the absence of clarity, diversification is the only logical strategy" -= Larry Swedroe
User avatar
happyisland
Posts: 907
Joined: Thu Oct 03, 2013 1:36 pm
Location: nos baranca tan stima

Re: VTI/VXUS split

Post by happyisland »

Why not just go by market capitalization?
Da5id
Posts: 5058
Joined: Fri Feb 26, 2016 7:20 am

Re: VTI/VXUS split

Post by Da5id »

Noobvestor wrote: Sun Sep 19, 2021 7:52 pm Vanguard's whitepapers show volatility reduction anywhere from 1% to 60%+ (with max benefits in the 30-50 range IIRC). If you want a principled approach, though, again: market weights is neutral - hard to get more robust than that.
I'm in on international (40%, which is near market weight these days), but I think the volatility reduction argument is not particularly strong. Figure 3 panel a in the white paper https://personal.vanguard.com/pdf/ISGGE ... Online.pdf panel shows a very modest decrease in volatility. Peaking at maybe 25-40% international, starting to decrease at 50% actually. But the magnitude of the effect just isn't very impressive. I think the curve has shifted a bit since last I looked, as they updated the white paper in April this year.
UpperNwGuy
Posts: 9446
Joined: Sun Oct 08, 2017 7:16 pm

Re: VTI/VXUS split

Post by UpperNwGuy »

ilyaraz wrote: Sun Sep 19, 2021 12:27 pm What is a good framework to think how to split the stock index fund portfolio between US and international stocks? Ideally, if it's at least somewhat quantitative.

My current allocation is close to 80% VTI, 10% VXUS and 10% VNQ (not counting some individual stocks and other assets not relevant to the discussion), and I'm thinking of bumping up the VXUS share, but I'd like to do it in a principled way rather than just blind guessing.

How do you choose the allocation domestic/international beyond the gut feeling?
I would sell all your VNQ and buy VXUS in its place. That would bring the VXUS up to 20% of your all-equities portfolio, and that's the percentage of international that many of us hold. It's not quantitative, but it seems right.
roth evangelist
Posts: 253
Joined: Wed Jul 01, 2020 9:56 pm

Re: VTI/VXUS split

Post by roth evangelist »

I'm uncomfortable with the international-heavy market weight of 40%, so I've arbitrarily decided on 25%.
40 Years' Gatherin's
Posts: 203
Joined: Sun Jul 26, 2020 7:39 am

Re: VTI/VXUS split

Post by 40 Years' Gatherin's »

Forget ex-US. You don't need it. VTSAX all the way!
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: VTI/VXUS split

Post by Marseille07 »

I think there are 3 reasonable choices:
a) just hold VT and take world market weight
b) up to 20% ex-US because Mr. Bogle said so
c) 100/0, because Mr. Bogle actually didn't believe in ex-US
40 Years' Gatherin's
Posts: 203
Joined: Sun Jul 26, 2020 7:39 am

Re: VTI/VXUS split

Post by 40 Years' Gatherin's »

Marseille07 wrote: Sun Sep 19, 2021 9:54 pm I think there are 3 reasonable choices:
a) just hold VT and take world market weight
b) up to 20% ex-US because Mr. Bogle said so
c) 100/0, because Mr. Bogle actually didn't believe in ex-US
I'll take "Mr. Bogle actually didn't believe in ex-US" for $800,000, Alex.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: VTI/VXUS split

Post by Marseille07 »

40 Years' Gatherin's wrote: Sun Sep 19, 2021 10:13 pm
Marseille07 wrote: Sun Sep 19, 2021 9:54 pm I think there are 3 reasonable choices:
a) just hold VT and take world market weight
b) up to 20% ex-US because Mr. Bogle said so
c) 100/0, because Mr. Bogle actually didn't believe in ex-US
I'll take "Mr. Bogle actually didn't believe in ex-US" for $800,000, Alex.
We're on the same page. My choice is c) as well :sharebeer
automatic
Posts: 21
Joined: Thu Jan 07, 2016 7:48 pm

Re: VTI/VXUS split

Post by automatic »

I went 50:50 when I started and have stuck with it. I felt that it was simple, similar to the global market cap (at the time) and something I could stick with. I liked that it was so simple - half and half - and that you could bounce back and forth between different 10% increments in either direction based on the current sentiment of the day. The only change I’ve ever considered is just buying VT and forgetting about maintaining the exact via 50:50 split that I rebalance to. I feel it’s WAY MORE IMPORTANT that you just define a split and stick with it in perpetuity. The consistency of strategy is what’s key, not the exact split. Simplicity is sophistication, as they say.
stefan_lec
Posts: 88
Joined: Fri Mar 08, 2019 10:43 pm

Re: VTI/VXUS split

Post by stefan_lec »

+1 that ability to stay the course is more important than quantitative stuff. I used to have a 20% international split, but I kept having to fight the urge to tinker with it, because the percentage was completely arbitrary.

World market cap weight (VT) seems least arbitrary to me personally, so I switched and haven't felt the urge to tinker since. Owning everything and letting the ratio float gives me peace, since I no longer care how well one sector or country is doing.

Others seem very confident that the US is an intrinsically better investment than everywhere else, and will remain so for decades. I don't share that certainty, but if you feel confident enough in that belief to not feel the urge to tinker, 100% US could also be a good choice.

I say go with what feels most peaceful for you personally. Avoiding shooting yourself in the foot with behavioral errors is way more important than optimizing the statistical properties of your portfolio. :sharebeer
Portfolio: 75% VT, 25% BNDW/I-bonds/HYSA
DonFifer
Posts: 85
Joined: Wed May 05, 2021 1:59 pm

Re: VTI/VXUS split

Post by DonFifer »

When I first encountered the BH philosophy, I searched for a reasonable (recommended?) AA primarily using ETFs, and ended up with a whole litany of possibilities. I eventually found the "Vanguard ETF Strategic Model Portfolio" brochure on line, and ended up selecting the Core Series 60/40 AA model. Like the Vanguard white paper and the Total World Stock ETF (VT), the Core Series has equities split 60 U.S. / 40 ex U.S.
While there may be 100s of possibilities, all with valid points, I'm staying where I am.
Stick5vw
Posts: 241
Joined: Sun Dec 18, 2016 5:46 am

Re: VTI/VXUS split

Post by Stick5vw »

I tend to follow the weights for the MSCI ACWI benchmark. That is roughly 59% US, 32% International, and 9% EM.

I have found that VXUS actually has a higher weighting to EM than I expected (Roughly 25%) which made it hard to marry up to ACWI with any precision.

What this means in practice is that I use VEA (developed international) alongside VXUS to ensure the EM exposure doesn’t get too out of kilter. I use ITOT for most of my US exposure.

In hindsight I would’ve done a building block approach via ITOT, VEA, and EEM, or simply just VT exposure. But I’ve put a decent chunk of change into VXUS and don’t want to sell it and incur a tax hit.
User avatar
ruralavalon
Posts: 26297
Joined: Sat Feb 02, 2008 9:29 am
Location: Illinois

Re: VTI/VXUS split

Post by ruralavalon »

ilyaraz wrote: Sun Sep 19, 2021 12:27 pm What is a good framework to think how to split the stock index fund portfolio between US and international stocks? Ideally, if it's at least somewhat quantitative.

My current allocation is close to 80% VTI, 10% VXUS and 10% VNQ (not counting some individual stocks and other assets not relevant to the discussion), and I'm thinking of bumping up the VXUS share, but I'd like to do it in a principled way rather than just blind guessing.

How do you choose the allocation domestic/international beyond the gut feeling?
That's 10% of stocks in international stocks. That's lowish in my opinion.

I suggest around 20 - 30% of stocks in international stocks. Vanguard paper (March 2012), "Considerations for investing in non-U.S. equities", available as an archived pdf. Historically, allocating 20% of an equity portfolio to non-U.S. stocks would have captured about 84% of the maximum possible diversification benefit, and allocating 30% of an equity portfolio to non-U.S. stocks would have captured about 99% of the maximum possible diversification benefit (p. 6). The diversification benefit has varied over time.

You can find lots of debate here on international allocation, opinions ranging all the way from 00% to 50% of stocks in international stocks. If you want more viewpoints on international stocks please try the Google search box, upper right, this page.

Asset allocation is a very personal decision. You must decide on an allocation that is comfortable for you based on your own ability, willingness and need to take risk.

What is your age? About how long to expected retirement? Are you eligible for a substantial pension? Do you have a fixed income allocation? A fixed income allocation will be much more important to diversification than an international stock allocation.
"Everything should be as simple as it is, but not simpler." - Albert Einstein | Wiki article link: Bogleheads® investment philosophy
User avatar
ruralavalon
Posts: 26297
Joined: Sat Feb 02, 2008 9:29 am
Location: Illinois

Re: VTI/VXUS split

Post by ruralavalon »

MotoTrojan wrote: Sun Sep 19, 2021 1:07 pm
stan1 wrote: Sun Sep 19, 2021 12:59 pm

Yeah, I was trying not to over complexify but you got me :-)
Just teasing. But the OP asked for a quantitative approach, of which the fundamental weighting is a great one as it is immune to relative over/under-valuation (bubble) scenarios where market-cap weighting isn't a great metric.
"Immune to" overweight/underweight seems an odd way to phrase this. Market weight is a right qualitative choice if, and only if, you religiously believe in the Efficient Market Hypothesis (EMH) as a doctrine or creed. But the EMH is unproven.

It's called a "hypothesis" for a reason. A hypothesis is "a supposition or proposed explanation made on the basis of limited evidence as a starting point for further investigation" or "a proposition made as a basis for reasoning, without any assumption of its truth" bing search link.
"Everything should be as simple as it is, but not simpler." - Albert Einstein | Wiki article link: Bogleheads® investment philosophy
MotoTrojan
Posts: 11259
Joined: Wed Feb 01, 2017 7:39 pm

Re: VTI/VXUS split

Post by MotoTrojan »

ruralavalon wrote: Mon Sep 20, 2021 9:40 am
MotoTrojan wrote: Sun Sep 19, 2021 1:07 pm
stan1 wrote: Sun Sep 19, 2021 12:59 pm

Yeah, I was trying not to over complexify but you got me :-)
Just teasing. But the OP asked for a quantitative approach, of which the fundamental weighting is a great one as it is immune to relative over/under-valuation (bubble) scenarios where market-cap weighting isn't a great metric.
"Immune to" overweight/underweight seems an odd way to phrase this. Market weight is a right qualitative choice if, and only if, you religiously believe in the Efficient Market Hypothesis (EMH) as a doctrine or creed. But the EMH is unproven.

It's called a "hypothesis" for a reason. A hypothesis is "a supposition or proposed explanation made on the basis of limited evidence as a starting point for further investigation" or "a proposition made as a basis for reasoning, without any assumption of its truth" bing search link.
If you believe in bubbles or over/under-valuation you by definition don't subscribe to EMH. In a bubble, market weight will overweight the bubble. If you use a fundamental weight you'll maintain your weighting to economic scale, no matter what happens with valuations. In that sense it is immune.
User avatar
ruralavalon
Posts: 26297
Joined: Sat Feb 02, 2008 9:29 am
Location: Illinois

Re: VTI/VXUS split

Post by ruralavalon »

MotoTrojan wrote: Mon Sep 20, 2021 7:54 pm
ruralavalon wrote: Mon Sep 20, 2021 9:40 am
MotoTrojan wrote: Sun Sep 19, 2021 1:07 pm
stan1 wrote: Sun Sep 19, 2021 12:59 pm

Yeah, I was trying not to over complexify but you got me :-)
Just teasing. But the OP asked for a quantitative approach, of which the fundamental weighting is a great one as it is immune to relative over/under-valuation (bubble) scenarios where market-cap weighting isn't a great metric.
"Immune to" overweight/underweight seems an odd way to phrase this. Market weight is a right qualitative choice if, and only if, you religiously believe in the Efficient Market Hypothesis (EMH) as a doctrine or creed. But the EMH is unproven.

It's called a "hypothesis" for a reason. A hypothesis is "a supposition or proposed explanation made on the basis of limited evidence as a starting point for further investigation" or "a proposition made as a basis for reasoning, without any assumption of its truth" bing search link.
If you believe in bubbles or over/under-valuation you by definition don't subscribe to EMH. In a bubble, market weight will overweight the bubble. If you use a fundamental weight you'll maintain your weighting to economic scale, no matter what happens with valuations. In that sense it is immune.
Sorry, I don't understand what you are saying. Probably my fault.

I just do not see proof that market weighting in international stocks is the objectively the best weight for diversification. That is an unproven idea.

I see proof that historically less than market weight in international stocks has given very good diversification benefits.
"Everything should be as simple as it is, but not simpler." - Albert Einstein | Wiki article link: Bogleheads® investment philosophy
User avatar
jakehefty17
Posts: 345
Joined: Fri May 11, 2018 6:48 pm
Location: New York

Re: VTI/VXUS split

Post by jakehefty17 »

ilyaraz wrote: Sun Sep 19, 2021 12:27 pm What is a good framework to think how to split the stock index fund portfolio between US and international stocks? Ideally, if it's at least somewhat quantitative.

My current allocation is close to 80% VTI, 10% VXUS and 10% VNQ (not counting some individual stocks and other assets not relevant to the discussion), and I'm thinking of bumping up the VXUS share, but I'd like to do it in a principled way rather than just blind guessing.

How do you choose the allocation domestic/international beyond the gut feeling?
Personally, I look at VT (Vanguard Total World Stock Index Fund ETF) and set my asset allocation near that. That led me to a 60/40 split US/ex-US.

That's good enough for me. Good luck!
"The problem with the world is that the intelligent people are full of doubts, while the stupid ones are full of confidence." -Charles Bukowski
MotoTrojan
Posts: 11259
Joined: Wed Feb 01, 2017 7:39 pm

Re: VTI/VXUS split

Post by MotoTrojan »

ruralavalon wrote: Tue Sep 21, 2021 9:56 am
MotoTrojan wrote: Mon Sep 20, 2021 7:54 pm
ruralavalon wrote: Mon Sep 20, 2021 9:40 am
MotoTrojan wrote: Sun Sep 19, 2021 1:07 pm
stan1 wrote: Sun Sep 19, 2021 12:59 pm

Yeah, I was trying not to over complexify but you got me :-)
Just teasing. But the OP asked for a quantitative approach, of which the fundamental weighting is a great one as it is immune to relative over/under-valuation (bubble) scenarios where market-cap weighting isn't a great metric.
"Immune to" overweight/underweight seems an odd way to phrase this. Market weight is a right qualitative choice if, and only if, you religiously believe in the Efficient Market Hypothesis (EMH) as a doctrine or creed. But the EMH is unproven.

It's called a "hypothesis" for a reason. A hypothesis is "a supposition or proposed explanation made on the basis of limited evidence as a starting point for further investigation" or "a proposition made as a basis for reasoning, without any assumption of its truth" bing search link.
If you believe in bubbles or over/under-valuation you by definition don't subscribe to EMH. In a bubble, market weight will overweight the bubble. If you use a fundamental weight you'll maintain your weighting to economic scale, no matter what happens with valuations. In that sense it is immune.
Sorry, I don't understand what you are saying. Probably my fault.

I just do not see proof that market weighting in international stocks is the objectively the best weight for diversification. That is an unproven idea.

I see proof that historically less than market weight in international stocks has given very good diversification benefits.
I think your confusion is in that I didn't say market weight was objectively the best, I said fundamental weight was as it is immune to mispricings (whether they overvalue the US or ex-US) since it's weighting is not in any way linked to price, only fundamentals.

The "proof" you speak of is just past returns, no? The US has been less volatile, and had higher returns, so it isn't a surprise that you got the best diversification with more of it. Does that tell you about the future though? In particular, does it tell you anything valid if the US outperformed primarily due to valuations growing? That could suggest the least optimal portfolio if we expect that to revert, and if we expect that valuation spread to persist, it still would suggest optimal had more US than the future would be expected to have.

https://www.aqr.com/Insights/Perspectiv ... ing-to-You
User avatar
drumboy256
Posts: 673
Joined: Sat Jun 06, 2020 2:21 pm

Re: VTI/VXUS split

Post by drumboy256 »

60/30 is the best split, objectively, you’re as diversified as you can be.
Promise is one thing. Fulfilling that promise is quite another. - Sir Alex Ferguson | 20% IVV / 40% IBIT / 20% IXUS / 20% VGLT + chill
Northern Flicker
Posts: 15289
Joined: Fri Apr 10, 2015 12:29 am

Re: VTI/VXUS split

Post by Northern Flicker »

Da5id wrote: Sun Sep 19, 2021 8:35 pm
Noobvestor wrote: Sun Sep 19, 2021 7:52 pm Vanguard's whitepapers show volatility reduction anywhere from 1% to 60%+ (with max benefits in the 30-50 range IIRC). If you want a principled approach, though, again: market weights is neutral - hard to get more robust than that.
I'm in on international (40%, which is near market weight these days), but I think the volatility reduction argument is not particularly strong. Figure 3 panel a in the white paper https://personal.vanguard.com/pdf/ISGGE ... Online.pdf panel shows a very modest decrease in volatility. Peaking at maybe 25-40% international, starting to decrease at 50% actually. But the magnitude of the effect just isn't very impressive. I think the curve has shifted a bit since last I looked, as they updated the white paper in April this year.
Moreover, samples from other time periods, often do not reproduce Vanguard's result.
Northern Flicker
Posts: 15289
Joined: Fri Apr 10, 2015 12:29 am

Re: VTI/VXUS split

Post by Northern Flicker »

MotoTrojan wrote: Mon Sep 20, 2021 7:54 pm
ruralavalon wrote: Mon Sep 20, 2021 9:40 am
MotoTrojan wrote: Sun Sep 19, 2021 1:07 pm
stan1 wrote: Sun Sep 19, 2021 12:59 pm

Yeah, I was trying not to over complexify but you got me :-)
Just teasing. But the OP asked for a quantitative approach, of which the fundamental weighting is a great one as it is immune to relative over/under-valuation (bubble) scenarios where market-cap weighting isn't a great metric.
"Immune to" overweight/underweight seems an odd way to phrase this. Market weight is a right qualitative choice if, and only if, you religiously believe in the Efficient Market Hypothesis (EMH) as a doctrine or creed. But the EMH is unproven.

It's called a "hypothesis" for a reason. A hypothesis is "a supposition or proposed explanation made on the basis of limited evidence as a starting point for further investigation" or "a proposition made as a basis for reasoning, without any assumption of its truth" bing search link.
If you believe in bubbles or over/under-valuation you by definition don't subscribe to EMH. In a bubble, market weight will overweight the bubble. If you use a fundamental weight you'll maintain your weighting to economic scale, no matter what happens with valuations. In that sense it is immune.
If small caps are overvalued, equal weighting will overweight them.
User avatar
FoundingFather
Posts: 407
Joined: Fri Dec 18, 2020 9:20 pm

Re: VTI/VXUS split

Post by FoundingFather »

stefan_lec wrote: Mon Sep 20, 2021 1:38 am +1 that ability to stay the course is more important than quantitative stuff. I used to have a 20% international split, but I kept having to fight the urge to tinker with it, because the percentage was completely arbitrary.

World market cap weight (VT) seems least arbitrary to me personally, so I switched and haven't felt the urge to tinker since. Owning everything and letting the ratio float gives me peace, since I no longer care how well one sector or country is doing.

Others seem very confident that the US is an intrinsically better investment than everywhere else, and will remain so for decades. I don't share that certainty, but if you feel confident enough in that belief to not feel the urge to tinker, 100% US could also be a good choice.

I say go with what feels most peaceful for you personally. Avoiding shooting yourself in the foot with behavioral errors is way more important than optimizing the statistical properties of your portfolio. :sharebeer
+1

This is my exact same philosophy and approach.

Founding Father
"I do not think myself equal to the Command I am honored with." -George Washington (excerpt from Journals of the Continental Congress, 16 June 1775)
User avatar
ruralavalon
Posts: 26297
Joined: Sat Feb 02, 2008 9:29 am
Location: Illinois

Re: VTI/VXUS split

Post by ruralavalon »

FoundingFather wrote: Thu Sep 23, 2021 12:46 am
stefan_lec wrote: Mon Sep 20, 2021 1:38 am +1 that ability to stay the course is more important than quantitative stuff. I used to have a 20% international split, but I kept having to fight the urge to tinker with it, because the percentage was completely arbitrary.

World market cap weight (VT) seems least arbitrary to me personally, so I switched and haven't felt the urge to tinker since. Owning everything and letting the ratio float gives me peace, since I no longer care how well one sector or country is doing.

Others seem very confident that the US is an intrinsically better investment than everywhere else, and will remain so for decades. I don't share that certainty, but if you feel confident enough in that belief to not feel the urge to tinker, 100% US could also be a good choice.

I say go with what feels most peaceful for you personally. Avoiding shooting yourself in the foot with behavioral errors is way more important than optimizing the statistical properties of your portfolio. :sharebeer
+1

This is my exact same philosophy and approach.

Founding Father

+2

I still think I that in the area of 20-30% of stocks in international stocks is probably about right for good diversification, but don't feel that there is any clear right answer objectively.

When there is no clear right answer that necessarily means that there is no clear wrong answer either.

I have become pretty lackadaisical about adhering to that allocation in practice, I don't pay much attention to what it is at any given time or even rebalance U.S./international.

Our portfolio is a tinker-free zone :D .

Any factual evidence that exists for any viewpoint on this is necessarily found in past returns. For the future we only have opinions, there are no available facts yet.
"Everything should be as simple as it is, but not simpler." - Albert Einstein | Wiki article link: Bogleheads® investment philosophy
MotoTrojan
Posts: 11259
Joined: Wed Feb 01, 2017 7:39 pm

Re: VTI/VXUS split

Post by MotoTrojan »

Northern Flicker wrote: Wed Sep 22, 2021 11:38 pm
MotoTrojan wrote: Mon Sep 20, 2021 7:54 pm

If you believe in bubbles or over/under-valuation you by definition don't subscribe to EMH. In a bubble, market weight will overweight the bubble. If you use a fundamental weight you'll maintain your weighting to economic scale, no matter what happens with valuations. In that sense it is immune.
If small caps are overvalued, equal weighting will overweight them.
I suppose it is a good thing that I said fundamental weight, not equal weight :D.

While I think a fundamental weight is more efficient than equal-weight, an EW will still not overweight the individual names that are overvalued as it by definition is not tied to price. But yes if small-caps in aggregate are overvalued, equal-weight will likely underperform, but I wouldn't say it is overweighting them since it is a static weighting scheme.

Market-cap weight however will increase it's weighting to overvalued names as they become overvalued. If we assume some companies are randomly over and some under valued, then we have to also acknowledge that market-cap is by definition holding more of those (unknown) over-valued ones, and less of the under-valued ones, than a dart-board would (or equal-weight, or fundamental weight). This isn't an opinion, it is the only possible outcome assuming the over/under-valuation is random.
lostdog
Posts: 5361
Joined: Thu Feb 04, 2016 1:15 pm

Re: VTI/VXUS split

Post by lostdog »

FoundingFather wrote: Thu Sep 23, 2021 12:46 am
stefan_lec wrote: Mon Sep 20, 2021 1:38 am +1 that ability to stay the course is more important than quantitative stuff. I used to have a 20% international split, but I kept having to fight the urge to tinker with it, because the percentage was completely arbitrary.

World market cap weight (VT) seems least arbitrary to me personally, so I switched and haven't felt the urge to tinker since. Owning everything and letting the ratio float gives me peace, since I no longer care how well one sector or country is doing.

Others seem very confident that the US is an intrinsically better investment than everywhere else, and will remain so for decades. I don't share that certainty, but if you feel confident enough in that belief to not feel the urge to tinker, 100% US could also be a good choice.

I say go with what feels most peaceful for you personally. Avoiding shooting yourself in the foot with behavioral errors is way more important than optimizing the statistical properties of your portfolio. :sharebeer
+1

This is my exact same philosophy and approach.

Founding Father
+2 Well said. You're at peace. No confirmation needed.
Stocks-80% || Bonds-20% || VTI/VXUS/AOR
terran
Posts: 2966
Joined: Sat Jan 10, 2015 9:50 pm

Re: VTI/VXUS split

Post by terran »

My framework is to match the US/International split used by major target date index funds. It used to be that Fidelity was 70/30 and Vanguard was 60/40, so I went with 65/35. Now Both Fidelity and Vanguard are 60/40, so I do the same. I figure they have more time and expertise to do the research, so if that's where they're putting their customers money then it's good enough for me.
Northern Flicker
Posts: 15289
Joined: Fri Apr 10, 2015 12:29 am

Re: VTI/VXUS split

Post by Northern Flicker »

MotoTrojan wrote: While I think a fundamental weight is more efficient than equal-weight, an EW will still not overweight the individual names that are overvalued as it by definition is not tied to price. But yes if small-caps in aggregate are overvalued, equal-weight will likely underperform, but I wouldn't say it is overweighting them since it is a static weighting scheme.

Market-cap weight however will increase it's weighting to overvalued names as they become overvalued.
Fundamental weighting is just adding some contrarian rebalancing at the individual security level.

I think this is inefficient. I would implement a strategy to achieve a similar result using factor tilts at the asset class level. An example would be 70% VTI & 30% VBR with rebalancing.
Horton
Posts: 1479
Joined: Mon Jan 21, 2008 2:53 pm
Contact:

Re: VTI/VXUS split

Post by Horton »

drumboy256 wrote: Wed Sep 22, 2021 11:23 pm 60/30 is the best split, objectively, you’re as diversified as you can be.
What do you with the other 10? :D
80% global equities (faith-based tilt) + 20% TIPS (LDI)
User avatar
ruralavalon
Posts: 26297
Joined: Sat Feb 02, 2008 9:29 am
Location: Illinois

Re: VTI/VXUS split

Post by ruralavalon »

Horton wrote: Fri Sep 24, 2021 9:03 am
drumboy256 wrote: Wed Sep 22, 2021 11:23 pm 60/30 is the best split, objectively, you’re as diversified as you can be.
What do you with the other 10? :D
Spend it on learning to do arithmetic :oops: , or doing something fun :D
"Everything should be as simple as it is, but not simpler." - Albert Einstein | Wiki article link: Bogleheads® investment philosophy
User avatar
UpsetRaptor
Posts: 1068
Joined: Tue Jan 19, 2016 4:15 pm

Re: VTI/VXUS split

Post by UpsetRaptor »

ruralavalon wrote: Fri Sep 24, 2021 9:59 am
Horton wrote: Fri Sep 24, 2021 9:03 am
drumboy256 wrote: Wed Sep 22, 2021 11:23 pm 60/30 is the best split, objectively, you’re as diversified as you can be.
What do you with the other 10? :D
Spend it on learning to do arithmetic :oops: , or doing something fun :D
There's 3 kinds of people in this world: Those who are good at math, and those who aren't.
User avatar
drumboy256
Posts: 673
Joined: Sat Jun 06, 2020 2:21 pm

Re: VTI/VXUS split

Post by drumboy256 »

Horton wrote: Fri Sep 24, 2021 9:03 am
drumboy256 wrote: Wed Sep 22, 2021 11:23 pm 60/30 is the best split, objectively, you’re as diversified as you can be.
What do you with the other 10? :D
VGLT or EDV of course :sharebeer
Promise is one thing. Fulfilling that promise is quite another. - Sir Alex Ferguson | 20% IVV / 40% IBIT / 20% IXUS / 20% VGLT + chill
User avatar
ruralavalon
Posts: 26297
Joined: Sat Feb 02, 2008 9:29 am
Location: Illinois

Re: VTI/VXUS split

Post by ruralavalon »

drumboy256 wrote: Fri Sep 24, 2021 10:19 am
Horton wrote: Fri Sep 24, 2021 9:03 am
drumboy256 wrote: Wed Sep 22, 2021 11:23 pm 60/30 is the best split, objectively, you’re as diversified as you can be.
What do you with the other 10? :D
VGLT or EDV of course :sharebeer
Use the other 10% learning to post fund names, not just just ticker symbols alone.
"Everything should be as simple as it is, but not simpler." - Albert Einstein | Wiki article link: Bogleheads® investment philosophy
User avatar
drumboy256
Posts: 673
Joined: Sat Jun 06, 2020 2:21 pm

Re: VTI/VXUS split

Post by drumboy256 »

ruralavalon wrote: Fri Sep 24, 2021 10:26 am
drumboy256 wrote: Fri Sep 24, 2021 10:19 am
Horton wrote: Fri Sep 24, 2021 9:03 am
drumboy256 wrote: Wed Sep 22, 2021 11:23 pm 60/30 is the best split, objectively, you’re as diversified as you can be.
What do you with the other 10? :D
VGLT or EDV of course :sharebeer
Use the other 10% learning to post fund names, not just just ticker symbols alone.
Digging the sarcasm! :mrgreen:
Promise is one thing. Fulfilling that promise is quite another. - Sir Alex Ferguson | 20% IVV / 40% IBIT / 20% IXUS / 20% VGLT + chill
Boglebiker
Posts: 29
Joined: Sun Jun 14, 2020 2:04 pm

Re: VTI/VXUS split

Post by Boglebiker »

I consulted the crystal ball, 75/25 is the absolute right answer. I hope this ends the all the debate around investing in international once and for all.
User avatar
ruralavalon
Posts: 26297
Joined: Sat Feb 02, 2008 9:29 am
Location: Illinois

Re: VTI/VXUS split

Post by ruralavalon »

Boglebiker wrote: Fri Sep 24, 2021 11:22 am I consulted the crystal ball, 75/25 is the absolute right answer. I hope this ends the all the debate around investing in international once and for all.
I used my time machine to learn about future performance, we are all wrong. You will all be surprised.
"Everything should be as simple as it is, but not simpler." - Albert Einstein | Wiki article link: Bogleheads® investment philosophy
Post Reply