Swedroe On Int'l Allocation

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
pascalwager
Posts: 906
Joined: Mon Oct 31, 2011 8:36 pm

Swedroe On Int'l Allocation

Post by pascalwager » Mon Aug 19, 2013 1:53 pm

Larry Swedroe recommended a 50% int'l allocation when the world market was about 55% int'l. I think he intended the 5% reduction to cover costs and currency risk. Now that the world market is about 50% int'l, should we now reduce our int'l allocation down to 45 or 46 percent? Or, more generally, should we regularly adjust our int'l allocation to stay below the current market int'l percentage?

User avatar
InvestorNewb
Posts: 1570
Joined: Mon Sep 03, 2012 11:27 am

Re: Swedroe On Int'l Allocation

Post by InvestorNewb » Mon Aug 19, 2013 2:03 pm

Why not hold 20-30% International instead? Most of the experts seem to suggest this as the "going allocation" for international.
My Portfolio: VTI [US], VXUS [Int'l], VNQ [REIT], VCN [Canada] (largest to smallest)

User avatar
cflannagan
Posts: 948
Joined: Sun Oct 21, 2007 11:44 am
Location: Palm Harbor, Florida

Re: Swedroe On Int'l Allocation

Post by cflannagan » Mon Aug 19, 2013 2:05 pm

InvestorNewb wrote:Why not hold 20-30% International instead? Most of the experts seem to suggest this as the "going allocation" for international.


"Most experts"? Citation?

User avatar
Ketawa
Posts: 1884
Joined: Mon Aug 22, 2011 1:11 am
Location: DC

Re: Swedroe On Int'l Allocation

Post by Ketawa » Mon Aug 19, 2013 2:11 pm

I regularly adjust my domestic/intl split (and my developed/emerging split) to be exactly market cap. It's one entry into my spreadsheet and 50/50 wouldn't be worth the "simplification" to me, since what happens if U.S. declines to 40% sometime in the next 50 years of my investing lifetime? What if it declines to 25%? Using market cap for the split makes this decision easier and has a sound theoretical basis.

I suspect Swedroe chose 50/50 since it was "close enough". If anything, he has recommended tilting towards intl in some cases since everything else in your net worth is denominated in or can be exchanged for only dollars - home, SS, pension, etc.

User avatar
G-Money
Posts: 2852
Joined: Sun Dec 09, 2007 7:12 am

Re: Swedroe On Int'l Allocation

Post by G-Money » Mon Aug 19, 2013 2:12 pm

pascalwager wrote:I think he intended the 5% reduction to cover costs and currency risk.


I'm not sure this is right, but it's been a while since I read his books, and I don't have one handy.

Honestly, I don't think it matters much.

Say you're 60% stocks. How much do you think it will matter if your US/ex-US split is 30%/30%, or 33%/27%? A 50/50 split has the advantage of simplicity when figuring out your allocations.
Don't assume I know what I'm talking about.

larryswedroe
Posts: 15598
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Swedroe On Int'l Allocation

Post by larryswedroe » Mon Aug 19, 2013 2:19 pm

Last I had checked the US was below 45%.

tilting slightly to US may make sense but NOT for currency risk reasons--which is a two way and not one way street

The reasons for slight tilt to domestic would be lower costs and possible tax inefficiency

On other hand if your labor capital is large percent of portfolio, it's likely US tied, and thus that would argue for more international.

And given current valuations you certainly could make the case for more of tilt to international, especially EM which is only place things look perhaps cheap.


Bottom line 50/50 is fine and IMO only reason for less is if you cannot take the tracking error
Larry

Dave76
Posts: 564
Joined: Sat Mar 03, 2012 9:05 pm

Re: Swedroe On Int'l Allocation

Post by Dave76 » Mon Aug 19, 2013 2:42 pm

I read somewhere that BAM recommends 40% int. for most of its clients. (?)

User avatar
InvestorNewb
Posts: 1570
Joined: Mon Sep 03, 2012 11:27 am

Re: Swedroe On Int'l Allocation

Post by InvestorNewb » Mon Aug 19, 2013 2:45 pm

cflannagan wrote:
InvestorNewb wrote:Why not hold 20-30% International instead? Most of the experts seem to suggest this as the "going allocation" for international.


"Most experts"? Citation?


I used the word "seem" ... all of the portfolios below are from authorities in the space; most do not pass the 30% range for international:
http://www.bogleheads.org/wiki/Lazy_Portfolios

I don't want to derail the thread, but I think there are more authorities in the space recommending 20-40% international. (Not 20-30% like I originally said - my bad.) I am not discrediting Larry in any way, as he could very well be right - just stating my own observations.

Here is Vanguard's paper on the subject:
https://institutional.vanguard.com/iam/ ... lbrief.pdf
My Portfolio: VTI [US], VXUS [Int'l], VNQ [REIT], VCN [Canada] (largest to smallest)

larryswedroe
Posts: 15598
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Swedroe On Int'l Allocation

Post by larryswedroe » Mon Aug 19, 2013 4:50 pm

Dave
We don't recommend any specific allocation to clients. We use however 40% as starting point for discussions because it is our experience that most investors, even with hand holding and education, have a hard time with Tracking error. If you can get over the TE problem IMO 50% should really be the starting point. My book The Only Guide You'll Ever Need for the Right Financial Plan asks the questions that lead to the answers as to who should tilt more or less

Best wishes
Larry

Investing is boring
Posts: 237
Joined: Tue Apr 09, 2013 11:55 am

Re: Swedroe On Int'l Allocation

Post by Investing is boring » Mon Aug 19, 2013 4:51 pm

InvestorNewb wrote:Why not hold 20-30% International instead? Most of the experts seem to suggest this as the "going allocation" for international.


Some experts think that 20-30% is a good amount for a typical investor. You are over-playing your hand.

Calm Man
Posts: 2917
Joined: Wed Sep 19, 2012 9:35 am

Re: Swedroe On Int'l Allocation

Post by Calm Man » Mon Aug 19, 2013 5:48 pm

larryswedroe wrote:Dave
We don't recommend any specific allocation to clients. We use however 40% as starting point for discussions because it is our experience that most investors, even with hand holding and education, have a hard time with Tracking error. If you can get over the TE problem IMO 50% should really be the starting point. My book The Only Guide You'll Ever Need for the Right Financial Plan asks the questions that lead to the answers as to who should tilt more or less

Best wishes
Larry


Larry, I like your 50/50 recommendation. It is simple and gives plenty of wiggle room to be "about right" as long as the US is above 40% or so. After that one could still stick to it and have some home bias. I like how you cite the 2 main reasons: fees and tax inefficiencies. I have researched Tax managed International (Vanguard). It only comes in Admiral Shares. The ER is 0.1% which is very low although admittedly double the insanely low 0.05% of US Total stock market. And 100% of it's dividends are qualified as well as providing the foreign tax credit. To my calculations, with the foreign tax credit and 100% QDI, tax managed international is actually MORE tax efficient that US TSM (if yields were the same, which they aren't). Although the yield is well over 1% more so in that sense it costs more in taxes.

Tax managed international does not have parts that Total International does such as small caps and emerging markets to any meaningful extent. I plotted however the 10 year performance chart of tax managed international and total international at the Vanguard site and guess what: they are IDENTICAL. Sort of like index 500 vs TSM but even closer in tracking.

When you indicate an issue of tracking error, what do you mean? Failure to track the index?

User avatar
Noobvestor
Posts: 4502
Joined: Mon Aug 23, 2010 1:09 am
Contact:

Re: Swedroe On Int'l Allocation

Post by Noobvestor » Mon Aug 19, 2013 8:06 pm

Bogle used to say 'up to 20%' but now says 'decide for yourself'. Vanguard, the brainchild of Bogle, says 'we'll make it 20% of our Target Retirement equities' originally then changes its mind in 2010 and makes it 30%. Aside from that, I don't know of any experts that recommend 30% or less in international, and I wouldn't be surprised to see the percent in Vanguard's TR funds creep up, since if you actually read that paper you linked (instead of just its conclusions) you'll note that the optimal range was anywhere from 30% to 70% depending on the time period, so 50% makes a lot of sense. So we have Bogle/Vanguard, a related pair, and ... who else? I'd like to see some other sources please.

I do find it fascinating though that somehow these few data points (I suppose because this is a Bogle/Vanguard forum) manage to translate into people thinking a majority or even a plurality of experts recommend such low international allocations. 20-30% is a severe tilt toward a single country and should be understood as such.
"In the absence of clarity, diversification is the only logical strategy" -= Larry Swedroe

User avatar
nisiprius
Advisory Board
Posts: 34335
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Swedroe On Int'l Allocation

Post by nisiprius » Mon Aug 19, 2013 8:19 pm

Noobvestor wrote:Bogle used to say 'up to 20%' but now says 'decide for yourself'. Vanguard, the brainchild of Bogle, says 'we'll make it 20% of our Target Retirement equities' originally then changes its mind in 2010 and makes it 30%. Aside from that, I don't know of any experts that recommend 30% or less in international, and I wouldn't be surprised to see the percent in Vanguard's TR funds creep up, since if you actually read that paper you linked (instead of just its conclusions) you'll note that the optimal range was anywhere from 30% to 70% depending on the time period, so 50% makes a lot of sense.

I do find it fascinating though that somehow these few data points (I suppose because this is a Bogle/Vanguard forum) manage to translate into people thinking a majority or even a plurality of experts recommend such low international allocations....

Fidelity Freedom 2040 fund, FFFFX: Domestic equity funds 60.87%, International equity funds 21.74% = 26% international.
T. Rowe Price Retirement 2040, TRRDX: 60.09% U. S. stock, 28.66% Non U.S. Stock = 32% international.
Schwab Target 2040, SWERX: 62.80% U.S. stock, 23.96% Non U.S. stock = 28% international.
MFS Lifetime 2040 A, MLFAX: 53.52% U.S. stock, 34.01% Non U.S. stock = 39% international.
iShares Target Date 2040, TZV: 55.60% U.S. stock, 25.88% Non U.S. stock = 32% international.
JHancock2 Retirement Choices at 2040, JRROX: 50.29% U.S. stock, 20.55% Non U.S. stock = 29% international.
American Century One Choice 2040, ARDVX: 58.10% U.S. stock, 16.35% Non U.S. stock = 22% international.
MassMutual RetireSmart 2040, MRFAX: 55.76% U.S. stock, 27.82% Non U.S. stock = 33% international.

Don't make this out to be some Vanguard thing. MFS is the only one that goes higher than 1/3 international and it's still under 40%.

Ah, found one. But I must say this a fairly... unconventional fund. It is only 25% stock, it is leveraged, it is 77% bonds, it is 36.68% "other." But for the stocks:
Pimco RealRetirement 2040, POFAX: 10.21% U.S. stock, 14.91% Non U.S. stock = 59% international.
Last edited by nisiprius on Mon Aug 19, 2013 8:39 pm, edited 7 times in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
InvestorNewb
Posts: 1570
Joined: Mon Sep 03, 2012 11:27 am

Re: Swedroe On Int'l Allocation

Post by InvestorNewb » Mon Aug 19, 2013 8:23 pm

nisiprius wrote:Fidelity Freedom 2040 fund, FFFFX: Domestic equity funds 60.87%, International equity funds 21.74% = 26% international.
T. Rowe Price Retirement 2040, TRRDX: 60.09% U. S. stock, 28.66% Non U.S. Stock = 32% international.
Schwab Target 2040, SWERX: 62.80% U.S. stock, 23.96% Non U.S. stock = 28% international.
MFS Lifetime 2040 A, MLFAX: 53.52% U.S. stock, 34.01% Non U.S. stock = 39% international.
iShares Target Date 2040, TZV: 55.60% U.S. stock, 25.88% Non U.S. stock = 32% international.
JHancock2 Retirement Choices at 2040, JRROX: 50.29% U.S. stock, 20.55% Non U.S. stock = 29% international.
American Century One Choice 2040, ARDVX: 58.10% U.S. stock, 16.35% Non U.S. stock = 22% international.
MassMutual RetireSmart 2040, MRFAX: 55.76% U.S. stock, 27.82% Non U.S. stock = 33% international.
Don't make this out to be some Vanguard thing. MFS is the only one that goes higher than 1/3 international and it's still under 40%.


+1 :happy
My Portfolio: VTI [US], VXUS [Int'l], VNQ [REIT], VCN [Canada] (largest to smallest)

User avatar
Noobvestor
Posts: 4502
Joined: Mon Aug 23, 2010 1:09 am
Contact:

Re: Swedroe On Int'l Allocation

Post by Noobvestor » Mon Aug 19, 2013 8:48 pm

nisiprius wrote:Don't make this out to be some Vanguard thing. MFS is the only one that goes higher than 1/3 international and it's still under 40%.

Ah, found one. But I must say this a fairly... unconventional fund. It is only 25% stock, it is leveraged, it is 77% bonds, it is 36.68% "other." But for the stocks:
Pimco RealRetirement 2040, POFAX: 10.21% U.S. stock, 14.91% Non U.S. stock = 59% international.


Point conceded. You're right. It's not just Vanguard. It's most (if not all) major fund companies targeting less than 50%. I really should have known better and done what you did for me - appreciate you calling me out and providing the data to boot. In my defense, I was not thinking about fund companies so much as professionals making recommendations in the space, but nonetheless, I did (mistakenly) think Vanguard's was lower than the industry average and I was wrong.

I want to get that out of the way with an apology (sorry!) before I ask about the other half of my point: does anyone know what professionals/experts (I'm thinking authors and academics aligned with passive indexing or otherwise) suggest less than 1/3 international based on real data or sound theory? My best guess is that Target Retirement funds have figured out what Larry alluded to - namely that people will freak out about too much international when it doesn't do so well (Tracking Error) and are selling low-international funds accordingly.

However, I'm open to being wrong if someone can answer my question - i.e. provide examples of neutral parties with sound analysis who provide compelling arguments for being less than 30-40% international. I can't recall reading any studies (including Vanguards, FWIW) that points to an actually optimal allocation of around 30%. Even Vanguard just says you get 'most of the benefit' in that range (from their study: "mean-variance analysis suggests that an investor would have been best off in terms of lowest average volatility by allocating 40%").
"In the absence of clarity, diversification is the only logical strategy" -= Larry Swedroe

larryswedroe
Posts: 15598
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Swedroe On Int'l Allocation

Post by larryswedroe » Mon Aug 19, 2013 9:02 pm

Tracking error refers to not performing like the S&P 500, which is considered a proxy for the market.
Larry

pascalwager
Posts: 906
Joined: Mon Oct 31, 2011 8:36 pm

Re: Swedroe On Int'l Allocation

Post by pascalwager » Mon Aug 19, 2013 9:29 pm

Postby larryswedroe » Mon Aug 19, 2013 3:19 pm

Last I had checked the US was below 45%.


According to VG TWS fund, int'l is now 50.4% (7/31/13).

User avatar
ofcmetz
Posts: 2259
Joined: Tue Feb 08, 2011 8:09 pm
Location: Louisiana

Re: Swedroe On Int'l Allocation

Post by ofcmetz » Mon Aug 19, 2013 11:38 pm

pascalwager wrote:Larry Swedroe recommended a 50% int'l allocation when the world market was about 55% int'l. I think he intended the 5% reduction to cover costs and currency risk. Now that the world market is about 50% int'l, should we now reduce our int'l allocation down to 45 or 46 percent? Or, more generally, should we regularly adjust our int'l allocation to stay below the current market int'l percentage?


Being close seems good enough when it comes to asset allocations. I would maintain my desired international allocation at whatever percentage that I set it at while maybe revisiting things every few years. Would having 45% vs 55% of equities in international even make that much of a difference?
Never underestimate the power of the force of low cost index funds.

User avatar
ofcmetz
Posts: 2259
Joined: Tue Feb 08, 2011 8:09 pm
Location: Louisiana

Re: Swedroe On Int'l Allocation

Post by ofcmetz » Mon Aug 19, 2013 11:40 pm

larryswedroe wrote:Tracking error refers to not performing like the S&P 500, which is considered a proxy for the market.
Larry


I can definitely see how this could be taken as meaning that an index fund was not closely following its benchmark.
Never underestimate the power of the force of low cost index funds.

User avatar
JoMoney
Posts: 4472
Joined: Tue Jul 23, 2013 5:31 am

Re: Swedroe On Int'l Allocation

Post by JoMoney » Tue Aug 20, 2013 5:40 am

Noobvestor wrote:.. I can't recall reading any studies (including Vanguards, FWIW) that points to an actually optimal allocation of around 30%. Even Vanguard just says you get 'most of the benefit' in that range (from their study: "mean-variance analysis suggests that an investor would have been best off in terms of lowest average volatility by allocating 40%").


FWIW, I'm not trying to open up a debate on this, just trying to contribute. If I crossed a line in our prior discussion, I'm sorry.

Here is a Forbes article that comes to the same conclusion that you're seeing with the other portfolios / Vanguard study. It may provide a better explanation:
Using historical data, and essentially a "Return to the Mean" expectation (they don't use the RTM term).
They explain that some may find historical evidence to show optimal compound returns in the 40-60% range.
Then acknowledge a return-to-the-mean anecdotal expectation .
Using that expectation, explain that the only expected gain from international exposure is reduced volatility and diversification.
Explains that you get different results depending on different time periods, but once you exceed 20% the volatility goes up, and since there is no long-term expected gain (the only goal is to reduce volatility), it's non-optimal to exceed that amount.
http://www.forbes.com/sites/advisor/201 ... -too-much/
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

Call_Me_Op
Posts: 6579
Joined: Mon Sep 07, 2009 2:57 pm
Location: Milky Way

Re: Swedroe On Int'l Allocation

Post by Call_Me_Op » Tue Aug 20, 2013 7:01 am

I have been 50-50 for many years. Don't know about the rest of you, but I like a little tracking error. After all, tracking error is likely to be positive about as often as it is negative. And if you understand diversification, you shouldn't want your portfolio's value strongly coupled to the value of a single market or investment.
Best regards, -Op | | "In the middle of difficulty lies opportunity." Einstein

Tom_T
Posts: 1461
Joined: Wed Aug 29, 2007 2:33 pm

Re: Swedroe On Int'l Allocation

Post by Tom_T » Tue Aug 20, 2013 7:09 am

Just because the 'experts' recommend 30% or under doesn't mean they are right. It could very well be that all of the major investment firms, including Vanguard, recommend that percentage because that is what their clients are comfortable with.

User avatar
InvestorNewb
Posts: 1570
Joined: Mon Sep 03, 2012 11:27 am

Re: Swedroe On Int'l Allocation

Post by InvestorNewb » Tue Aug 20, 2013 7:44 am

Noobvestor wrote:
nisiprius wrote:Don't make this out to be some Vanguard thing. MFS is the only one that goes higher than 1/3 international and it's still under 40%.

Ah, found one. But I must say this a fairly... unconventional fund. It is only 25% stock, it is leveraged, it is 77% bonds, it is 36.68% "other." But for the stocks:
Pimco RealRetirement 2040, POFAX: 10.21% U.S. stock, 14.91% Non U.S. stock = 59% international.


Point conceded. You're right. It's not just Vanguard. It's most (if not all) major fund companies targeting less than 50%. I really should have known better and done what you did for me - appreciate you calling me out and providing the data to boot. In my defense, I was not thinking about fund companies so much as professionals making recommendations in the space, but nonetheless, I did (mistakenly) think Vanguard's was lower than the industry average and I was wrong.

I want to get that out of the way with an apology (sorry!) before I ask about the other half of my point: does anyone know what professionals/experts (I'm thinking authors and academics aligned with passive indexing or otherwise) suggest less than 1/3 international based on real data or sound theory? My best guess is that Target Retirement funds have figured out what Larry alluded to - namely that people will freak out about too much international when it doesn't do so well (Tracking Error) and are selling low-international funds accordingly.

However, I'm open to being wrong if someone can answer my question - i.e. provide examples of neutral parties with sound analysis who provide compelling arguments for being less than 30-40% international. I can't recall reading any studies (including Vanguards, FWIW) that points to an actually optimal allocation of around 30%. Even Vanguard just says you get 'most of the benefit' in that range (from their study: "mean-variance analysis suggests that an investor would have been best off in terms of lowest average volatility by allocating 40%").


BTW my "+1" comment wasn't directed at knocking down your post or anything. I only did that because nisiprius' findings seems to support what I said earlier about the 20-40% international range.
My Portfolio: VTI [US], VXUS [Int'l], VNQ [REIT], VCN [Canada] (largest to smallest)

User avatar
nisiprius
Advisory Board
Posts: 34335
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Swedroe On Int'l Allocation

Post by nisiprius » Tue Aug 20, 2013 7:45 am

Noob, first, a little gift for you:
In 'Common Sense on Mutual Funds,' John Bogle, in 2000 wrote:The global investing strategy is a favorite of academic theorists, some of whom recommend that the ideal equity portfolio should consist of holdings of each nation's stocks at their world-market weight.
A pity he doesn't name the theorists or cite sources, but I assume they are out there for someone to discover.

Noob, again not trying to re-open the stale debate, because I think we agree on the essentials. Here are some things I think which I believe you probably agree with. Investing is not a science. There's no way to prove what will be the optimum going forward. What mutual fund companies do when choosing percent international in their target funds is mostly about "fashion trends," with an eye to what the competition is doing, and what the big guns in the company personally think, customers want, etc. The actual evidence is all over the map and not definitive and you can find data and studies to prove anything you like.

It's a plain objective fact that the only international index that goes back more than a couple of decades is EAFE. I think it's a plain objective fact that the big thing is that actual results of an EAFE allocation over the full time the index has existed have amounted to almost no effect at all. The much smaller thing is that the small effect that is there has nevertheless had a very broad, very shallow peak and the peak is somewhere around 30-40%, not 50-55%. And that 20% has been about as good as 50-55%. AND IT'S ALL THE PAST, and it's all MPT efficient frontier charts which are extremely unstable depending on the endpoints you pick. These are a few years old; I don't know how much they've wriggled around since then.

The point is, I think the various numbers that have been chosen, that are less than 50-55%, actually are minimums or maximums on some curve someone can point to. Yeah, I think that numbers like Vanguard's former 20% are someone's arbitrary weighting in which they put a little home bias on one side of the scale and say, Aaaaah, I can enjoy my home bias and still get most of the goodness.

But if you take your thumb off the scale, it does read less than 50-55%.

Fidelity put this out around 2009:
Image

From my 9th edition of "A Random Walk Down Wall Street," dunno if it's changed in the latest edition, anyone have a copy? I am not sure why he chooses to show and describe the "portfolio with the least risk" rather than the tangent line drawn from the riskless asset, but he does, and what he says in the text is "It turns out that the portfolio with the least risk had 24 percent foreign securities and 76 percent U.S. securities." Unsaid, of course, is how small the risk reduction is from 100% U.S.

Image
Last edited by nisiprius on Tue Aug 20, 2013 8:45 am, edited 1 time in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
Taylor Larimore
Advisory Board
Posts: 26240
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

When experts

Post by Taylor Larimore » Tue Aug 20, 2013 8:39 am

Bogleheads:

When experts disagree--it is often because it doesn't make much difference.

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

User avatar
Ketawa
Posts: 1884
Joined: Mon Aug 22, 2011 1:11 am
Location: DC

Re: Swedroe On Int'l Allocation

Post by Ketawa » Tue Aug 20, 2013 8:54 am

nisiprius wrote:It's a plain objective fact that the only international index that goes back more than a couple of decades is EAFE. I think it's a plain objective fact that the big thing is that actual results of an EAFE allocation over the full time the index has existed have amounted to almost no effect at all. The much smaller thing is that the small effect that is there has nevertheless had a very broad, very shallow peak and the peak is somewhere around 30-40%, not 50-55%. And that 20% has been about as good as 50-55%. AND IT'S ALL THE PAST, and it's all MPT efficient frontier charts which are extremely unstable depending on the endpoints you pick. These are a few years old; I don't know how much they've wriggled around since then.


Something else that throws a wrench in all these numbers is that although we often use EAFE since it has the longest history, it is substantially smaller than the US (at least in the present day). If you were limited to US/EAFE today, market cap weighting would be about 57/43 since the EAFE leaves out emerging markets and Canada.

User avatar
G-Money
Posts: 2852
Joined: Sun Dec 09, 2007 7:12 am

Re: Swedroe On Int'l Allocation

Post by G-Money » Tue Aug 20, 2013 9:50 am

Noobvestor wrote:I want to get that out of the way with an apology (sorry!) before I ask about the other half of my point: does anyone know what professionals/experts (I'm thinking authors and academics aligned with passive indexing or otherwise) suggest less than 1/3 international based on real data or sound theory? My best guess is that Target Retirement funds have figured out what Larry alluded to - namely that people will freak out about too much international when it doesn't do so well (Tracking Error) and are selling low-international funds accordingly.

However, I'm open to being wrong if someone can answer my question - i.e. provide examples of neutral parties with sound analysis who provide compelling arguments for being less than 30-40% international. I can't recall reading any studies (including Vanguards, FWIW) that points to an actually optimal allocation of around 30%. Even Vanguard just says you get 'most of the benefit' in that range (from their study: "mean-variance analysis suggests that an investor would have been best off in terms of lowest average volatility by allocating 40%").

I believe Rick Ferri generally recommends 30% in equities. Older versions of Larry Swedroe's books had model portfolios recommending about 30% in equities. I believe Bill Bernstein's older model portfolios were also significantly below 50% international. In his book Unconventional Success, David Swensen's model portfolio recommends 28.5% equities (20% of a portfolio with 70% in equities); in a lecture, he later recommended 35% international (25% of portfolio, 70% equities).

I can't speak to what the latest versions of their books say. I know Larry in particular recommends global weight as a default now. Don't recall if they cited particular studies. But, certainly, card-carrying Boglehead experts have and sometimes still do recommend portfolios with international allocations in the 20-40% range.
Don't assume I know what I'm talking about.

Calm Man
Posts: 2917
Joined: Wed Sep 19, 2012 9:35 am

Re: Swedroe On Int'l Allocation

Post by Calm Man » Tue Aug 20, 2013 10:11 am

larryswedroe wrote:Tracking error refers to not performing like the S&P 500, which is considered a proxy for the market.
Larry


Hmmm. I thought for international, it would be tracking error vs an international index. We would WANT tracking error of an international fund compared to the S&P 500, otherwise why bother with international, right?

Calm Man
Posts: 2917
Joined: Wed Sep 19, 2012 9:35 am

Re: Swedroe On Int'l Allocation

Post by Calm Man » Tue Aug 20, 2013 10:19 am

Looking at nisi's 2 graphs a few posts above......
When presented with the top one by the Vanguard CFP when doing a review for me and recommending 30% international, it was pointed out that between about 30 and 70% the volatility was reduced by 1%. But extending it would way further reduces the volatility by 0.5%. This seems so trivial. And going to 0% international or 80%, there is no change in volatility. Who says volatility translates to long term returns at all? So I am still confused by what one should really do.

User avatar
nisiprius
Advisory Board
Posts: 34335
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Swedroe On Int'l Allocation

Post by nisiprius » Tue Aug 20, 2013 10:42 am

Calm Man wrote:Looking at nisi's 2 graphs a few posts above......
When presented with the top one by the Vanguard CFP when doing a review for me and recommending 30% international, it was pointed out that between about 30 and 70% the volatility was reduced by 1%. But extending it would way further reduces the volatility by 0.5%. This seems so trivial. And going to 0% international or 80%, there is no change in volatility. Who says volatility translates to long term returns at all? So I am still confused by what one should really do.
In theory, lower volatility should translate to higher long-term returns if you have what I call a "risk budget" for your portfolio, because if you can lower the volatility of your stocks you can nudge up your stock allocation and not exceed your risk budget.

It's really interesting how rarely anyone displays the calculation that way or puts numbers on it. Mostly people just say "Look! Less risk! Woo hoo!" I think--just making these numbers up--that the reason is that if they did, it would turn out that you're lowering the risk within your stock portfolio by, you know, 5-10%, so if you were at 60% stocks you can now increase your stock allocation up to, you know, maybe 64%, so if stocks return 3% more than bonds you've just upped your returns by, you know, 3% x 4% = 0.12%. That is, the risk reduction is unimpressive to begin with, but when you translate it into increased return it becomes even less impressive.

As for being confused, you should be confused. The reality of the situation is that it really didn't make much difference in the past, although if you are allowed to choose your favorite flavor of international and your own pair of endpoints you can find cases where it did. So you are left with the need to make some decision based on something other than past data, because the past data is ambiguous and the past differences are small.

One of the reasons I like Vanguard is that their own paper, Considerations for Investing in Non-U.S. Equities IS so honestly vague, so straightforwardly mealy-mouthed. They present quite a lot of very interesting data and the conclusion is "we think probably you sorta-kinda ought to maybe be somewhere in the range from 20% up to cap-weighted, but we're not sure even about that, but as a rough starting point maybe, and beyond that, gosh, don't ask us, we have no clue, do whatever you think best." They state it in a much more dignified way, of course: "The exact allocation to non-U.S. equities will depend on the investor’s view regarding the short- and long-term trade-offs." Yeah.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

vesalius
Posts: 777
Joined: Tue Jul 13, 2010 7:00 pm
Location: Texas

Re: Swedroe On Int'l Allocation

Post by vesalius » Tue Aug 20, 2013 11:37 am

Calm Man wrote:
larryswedroe wrote:Tracking error refers to not performing like the S&P 500, which is considered a proxy for the market.
Larry


Hmmm. I thought for international, it would be tracking error vs an international index. We would WANT tracking error of an international fund compared to the S&P 500, otherwise why bother with international, right?

Yes I think most of us would, intellectually. I think Larry is more addressing the emotional lability of many investors when their $$ do not track the S&P 500, which so many of the talking heads on TV and in the press use as the barometer for the market. When the Large Cap US Stocks of the SP 500 are up and International relatively underperforms, will swimming against that tide and the resultant emotional regret rule the day? Either giving you insomnia and ulcers or potentially result in selling Intemational at a relative low to join the herd and buy the S&P 500 (or total stock market) at a relative high?

feh
Posts: 1123
Joined: Sat Dec 15, 2012 11:39 am

Re: Swedroe On Int'l Allocation

Post by feh » Tue Aug 20, 2013 11:47 am

nisiprius wrote:Image


Interesting. The effect of int'l on an equities portfolio is very similar to the effect tilting to SV has.

Code Commit
Posts: 137
Joined: Mon Mar 04, 2013 11:42 am

Re: Swedroe On Int'l Allocation

Post by Code Commit » Tue Aug 20, 2013 12:08 pm

nisiprius wrote:From my 9th edition of "A Random Walk Down Wall Street," dunno if it's changed in the latest edition, anyone have a copy? I am not sure why he chooses to show and describe the "portfolio with the least risk" rather than the tangent line drawn from the riskless asset, but he does, and what he says in the text is "It turns out that the portfolio with the least risk had 24 percent foreign securities and 76 percent U.S. securities." Unsaid, of course, is how small the risk reduction is from 100% U.S.

Image


Funny how the same chart can confirm anyone's current, personal opinion. You could also say, for the same small amount of risk increase, the returns increase to 9% (at ~ 40% EAFE?). So, for about the same risk, you could have 100% US with 8.3% return, or with 60/40 US/Intl, you could have 9% return (of course, in the past).

Since Malkiel is quoted here, I will quote from his Tenth edition book :

One of the biggest mistakes that investors make is to fail to obtain sufficient international diversification. The United States represents only slightly more than 40 percent of the world economy. To be sure, a U.S. Total Stock Market fund does provide some global diversification because many of the multinational U.S. companies such as General Electric and Coca-Cola do a great deal of their business abroad. But the emerging markets of the world (such as China, India, and Brazil) have been growing much faster than the developed economies. China, for example, is still considered an emerging market. But it is now the second-largest economy in the world and is expected by the International Monetary Fund to continue to be the fastest-growing large economy in the world. Hence, in the recommendations that follow, you will note that I suggest that a substantial part of every portfolio be invested in emerging markets.

followed by
A SPECIFIC INDEX-FUND PORTFOLIO FOR AGING BABY BOOMERS
Cash (5%)* Fidelity Money Market Fund (FORXX), or Vanguard Prime Money Market Fund (VMMXX)
Bonds (27½%)† Vanguard Total Bond Market Index Fund (VBMFX)
Real Estate Equities (12½%) Vanguard REIT Index Fund (VGSIX)
Stocks (55%)
U.S. Stocks (27%) Fidelity Spartan (FSTMX), T. Rowe Price (POMIX), or Vanguard (VTSMX) Total Stock Market Index Fund
Developed International Markets (14%) Fidelity Spartan (VSIIX), or Vanguard (VDMIX) International Index Fund
Emerging International Markets (14%) Vanguard Emerging Markets Index Fund

i.e. about 50/50 US/Intl. In the end, I agree with nisiprius, that this is a stale debate and we agree on the essentials, but differ on some of the execution (which is okay). The calls to papers, charts, stats, research, or experts usually ends up being just to confirm our own current bias.

Tigermoose
Posts: 549
Joined: Sat Jul 21, 2012 5:02 pm

Re: Swedroe On Int'l Allocation

Post by Tigermoose » Tue Aug 20, 2013 12:16 pm

Code Commit wrote:Since Malkiel is quoted here, I will quote from his Tenth edition book :

One of the biggest mistakes that investors make is to fail to obtain sufficient international diversification. The United States represents only slightly more than 40 percent of the world economy. To be sure, a U.S. Total Stock Market fund does provide some global diversification because many of the multinational U.S. companies such as General Electric and Coca-Cola do a great deal of their business abroad. But the emerging markets of the world (such as China, India, and Brazil) have been growing much faster than the developed economies. China, for example, is still considered an emerging market. But it is now the second-largest economy in the world and is expected by the International Monetary Fund to continue to be the fastest-growing large economy in the world. Hence, in the recommendations that follow, you will note that I suggest that a substantial part of every portfolio be invested in emerging markets.



I thought it was acknowledged here on Bogleheads that growth does not equal or correlate with increased equity returns?
Institutions matter

Code Commit
Posts: 137
Joined: Mon Mar 04, 2013 11:42 am

Re: Swedroe On Int'l Allocation

Post by Code Commit » Tue Aug 20, 2013 12:35 pm

Tigermoose wrote:I thought it was acknowledged here on Bogleheads that growth does not equal or correlate with increased equity returns?

Do you think that acknowledgement is universal among all Bogleheads, or in any way, conclusive? (Don't shoot me, I am just the messenger :wink: ).

Tigermoose
Posts: 549
Joined: Sat Jul 21, 2012 5:02 pm

Re: Swedroe On Int'l Allocation

Post by Tigermoose » Tue Aug 20, 2013 12:48 pm

Code Commit wrote:
Tigermoose wrote:I thought it was acknowledged here on Bogleheads that growth does not equal or correlate with increased equity returns?

Do you think that acknowledgement is universal among all Bogleheads, or in any way, conclusive? (Don't shoot me, I am just the messenger :wink: ).


Here is a decent article on this subject: http://www.investorschronicle.co.uk/2011/07/14/comment/chris-dillow/the-emerging-market-growth-myth-aRKx7RWKzYJtXSYEyNkYqI/article.html

from the article:

"All this raises a question. If economic growth doesn't translate into stock market performance, why have emerging markets, in aggregate, out-performed developed ones?

The answer lies not in growth, but in risk. I mean this in three senses.

One is volatility. ...

A second risk is that emerging markets are heavily exposed to the US yield curve; higher short-term interest rates and a flattening in the curve are usually bad for emerging market shares. ...

There is a third risk that justifies higher returns in emerging markets. It is that local investors in emerging markets are generally less able to bear equity risk than their counterparts in developed economies. One reason for this is that they are poorer. ...

And here's the thing. This reduced ability to bear losses arises precisely because emerging markets are - by definition - under-developed. In this sense, the case for investing in them is not rests not upon the fact that they are likely to grow, but rather upon the fact that they are poor."
Institutions matter

Code Commit
Posts: 137
Joined: Mon Mar 04, 2013 11:42 am

Re: Swedroe On Int'l Allocation

Post by Code Commit » Tue Aug 20, 2013 1:07 pm


http://www.hendershotinvestments.com/ne ... rid=331723 (Interview with Jack Bogle)

The long-term rate of earnings growth, on the other hand, while hardly as given to precision as the
current dividend yield, is relatively stable. The fact is that after-tax corporate earnings have historically
grown at about 5% in nominal terms, roughly the same rate as the growth of our economy
.
....
If we can but recognize that corporate earnings have, with remarkable consistency, grown at about the rate of the
GDP, we can understand that the fundamentals of our economy drive long-term stock returns.

You probably didn't read my first post to the end. We can always find articles/posts/papers that give contradictory opinions (from reasonable experts). But these are all opinions and not conclusive. Investing is, far from science. (as nisiprius said above).

Tigermoose
Posts: 549
Joined: Sat Jul 21, 2012 5:02 pm

Re: Swedroe On Int'l Allocation

Post by Tigermoose » Tue Aug 20, 2013 1:25 pm

Code Commit wrote:

http://www.hendershotinvestments.com/ne ... rid=331723 (Interview with Jack Bogle)

The long-term rate of earnings growth, on the other hand, while hardly as given to precision as the
current dividend yield, is relatively stable. The fact is that after-tax corporate earnings have historically
grown at about 5% in nominal terms, roughly the same rate as the growth of our economy
.
....
If we can but recognize that corporate earnings have, with remarkable consistency, grown at about the rate of the
GDP, we can understand that the fundamentals of our economy drive long-term stock returns.

You probably didn't read my first post to the end. We can always find articles/posts/papers that give contradictory opinions (from reasonable experts). But these are all opinions and not conclusive. Investing is, far from science. (as nisiprius said above).



They might both be correct. Perhaps in developed economies, there is some correlation between the stock market and economic growth. In emerging economies, perhaps their stock markets are more driven by foreign corporations and investors and less so by internal growth fundamentals.
Institutions matter

hafius500
Posts: 217
Joined: Sat Sep 10, 2011 4:31 pm

Re: Swedroe On Int'l Allocation

Post by hafius500 » Tue Aug 20, 2013 1:27 pm

nisiprius wrote:Noob, first, a little gift for you:
In 'Common Sense on Mutual Funds,' John Bogle, in 2000 wrote:The global investing strategy is a favorite of academic theorists, some of whom recommend that the ideal equity portfolio should consist of holdings of each nation's stocks at their world-market weight.
A pity he doesn't name the theorists or cite sources, but I assume they are out there for someone to discover.


I had quoted in this post Siegel and Dimson/Marsh/Staunton:
hafis50 wrote:
Finance theory dictates that investors should hold the broadest of all possible portfolios, each country weighted by its market value, to achieve maximal diversification

(Siegel, Future for Investors, p.232)

We concluded that an investor with no stock selection skills should hold as widely diversified portfolio as possible, thus avoiding exposure to diversifiable, and hence unrewarded, risk. This effectively means holding a stake in the overall market. The same principle holds when investing internationally. Risk can be reduced as the returns on different markets and currencies are less than perfectly correlated. Investors with no special insights about the prospects for different markets and currencies should, like their domestic counterpatrs, hold as widely diversified a portfolio as possible. If it were not for market imperfections and differences in tastes they should hold the "world market portfolio".


(Dimson, p.108. It's not entirely clear IMO if this portfolio is cap- or GDP-weighted)


Certainly there are lots of other "professional experts" (like, e.g., the Norwegian Sovereign Wealth Fund, or global fund managers) who reject restrained or fixed country weights.
If you wanted to analyze the benefits of international diversification, shouldn't you perform out-of-sample tests (i.e., consider the expected or realized benefits to Non-US investors)?
prior username: hafis50

Post Reply