International stock allocation: currency risk

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
Topic Author
Kevin M
Posts: 15789
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

International stock allocation: currency risk

Post by Kevin M »

Currency risk is often mentioned as a reason to limit one's allocation to international stocks. For example, in his most recent investing book, "Rational Expectations", Bill Bernstein says that a market-cap allocation of around 50% is too much because "your retirement liabilities will be in dollars, so a 52% foreign allocation is inappropriately high".

On the other side of the argument, Larry Swedroe says currency risk is not a factor that should affect one's allocation to international stocks, since the risk has upside as well as downside. See this article by Larry for more details: Swedroe: Int’l Diversification Is Free.

I just read a Vanguard research paper, Global equities: Balancing home bias and diversification that indicates that currency risk is a positive, not a negative:
Although currency movements tend to be unpredictable and can be large, they have historically been uncorrelated to movements in stock prices. As a result, over time, currency movements have helped to reduce the correlation between non-U.S. equities and U.S. equities, thus contributing the diversification benefits of foreign holdings
I didn't see any discussion of the point about liabilities being in dollars.

I am undecided on this point. How about you?

Everyone agrees that cost is a factor, and that the higher costs of foreign stock trading justifies a lower-than-market-cap allocation.

The research paper recommends a foreign stock allocation range of 20% to 40% of equities, Bill recommends a range of 30%-45% based on current higher expected returns, and Larry recommends a starting point of market-cap weight (about 50%), with an adjustment down for costs, and adjustments up or down depending on one's tolerance for tracking-error and and one's exposure to specific risk factors, with a minimum 30% allocation.

My foreign stock allocation has been 40% of equities since I worked out this part of my AA some years ago, and I'm sticking with that, but I tend to recommend 30% of equities as a default position, primarily because that's what Vanguard's Target Retirement and LifeStrategy funds hold. It also happens to be the minimum recommended by both Larry and Bill, and not surprisingly in the middle of the range recommended by the Vanguard research paper.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
nisiprius
Advisory Board
Posts: 52216
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: International stock allocation: currency risk

Post by nisiprius »

I think it's a small but meaningful factor.

The point that currency risk is a positive, not a negative because "Although currency movements tend to be unpredictable and can be large, they have historically been uncorrelated to movements in stock prices. As a result, over time, currency movements have helped to reduce the correlation between non-U.S. equities and U.S. equities, thus contributing the diversification benefits of foreign holdings" is, I'm convinced, bogus. I'm going against Vanguard and Larry Swedroe when I say that, though.

As far as I know, it's generally agreed that the expected long-term return of a foreign currency is zero.

The big win in a portfolio is two assets that have low correlation and at least roughly comparable return and roughly comparable volatility. If you don't have all three, you don't get any substantial diversification advantage. A clear case where low correlation does nothing would be cash as a diversifier for stocks. Cash has zero correlation with stocks, but it lacks the other two factors--roughly comparable return and roughly comparable volatility. It just dilutes both stock risk and stock return, without changing the risk-adjusted return at all.

Now, currency risk gives a different "two out of three" situation. Adding currency risk to a stock portfolio increases the risk of that component, and it increases it, not enormously, but enough to matter. Vanguard Total International is in a higher risk category than Total Stock. The dollar index went from 80 to 160 to 80 1981-1987. That is, a dollar can buy twice as much of a global stock portfolio, measured in shares, at one time than another. This increased risk is probably not compensated by increased return. Why should it be? There's no need to take that risk, and it's not experienced by home-country investors.

So, does it improve a portfolio to invest part of in in an asset that is uncorrelated but has zero expected return? Interestingly enough, the question is debated, and some people actually believe it does. I don't want to go over the ground again, but will just say that I don't. I just don't believe that adding pure zero-return noise to a portfolio can help it. There aren't any fair gambling games out there, of course, but if adding zero-return noise to a portfolio improved it, then in principle you could improve your portfolio by taking 1/10th of it to the casino every month and betting it on red at the roulette wheel. That would give you a completely uncorrelated movement with respect to stocks.

There is, however, intrinsic diversification because national economies and businesses are imperfectly correlated.

So, I see it as a balancing act. If there were a single world currency, then the optimum portfolio would be globally cap-weighted. Since there isn't, you are getting some intrinsic diversification from the fundamentals of international investments themselves. You are also getting some unrewarded risk from currency fluctuations. Even though that risk is bidirectional--it was a win from 2002-2008--it is still extra uncertainty, extra fluctuation, extra volatility. I don't put too much faith in those efficient frontier charts, but I believe the reason why the optimum always shows up at some proportion that's considerably lower than cap-weighting is because it is a balancing act, and currency risk goes on the scale as one of the factors.

Really, though, the most salient characteristic of international equities is that all the data seems to show that it doesn't actually matter very much. I'm not sure what's behind what I see as the overpromotion of international stock investing. I don't really know what's supposed to be different between 1990, when Burton Malkiel was recommending that 1/6th of equities be international, and 2011, when he was recommending 1/2. The proportion of U.S. as a percentage of global actually increased between those endpoints.
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.
steve_14
Posts: 1507
Joined: Wed Jun 20, 2012 12:05 am

Re: International stock allocation: currency risk

Post by steve_14 »

I invest internationally to mitigate country risk (see: Japan). The fact that I use an unhedged foreign stock fund tells you that I don't worry too much about currency risk, though I'd never view it as a positive thing (for the reasons nisiprius points out). Do keep in mind that when USD gets weaker, our products become more competitive abroad, theoretically mitigating the effects. I maintain a 50/50 US/foreign ratio.
Longtimelurker
Posts: 471
Joined: Fri Dec 13, 2013 7:23 am

Re: International stock allocation: currency risk

Post by Longtimelurker »

What makes you think you aren't taking currency risk holding the S&P 500? Aren't 30-40% of earnings foreign? Aren't a good portion of total international earnings in dollars? Seems to me a very overblown concern...

Add to that that debasing currency is typically additive to GDP through exports and what is the real effect of this? Tiny in my mind...
Stay the course. If you can't resist greed, and fear is proven to be 2x as strong, you are doomed as an investor.
bnes
Posts: 404
Joined: Thu May 22, 2014 12:29 am
Contact:

Re: International stock allocation: currency risk

Post by bnes »

Note: where's no Vanguard "non-globalized companies" fund. Thus you're getting a lot of currency exposure anyway, even with an all USA portfolio.
User avatar
Topic Author
Kevin M
Posts: 15789
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: International stock allocation: currency risk

Post by Kevin M »

The Vanguard research paper discusses the questions related to whether or not US multinational corporations provide US investors with enough exposure to foreign markets. Apparently they don't think so, for these reasons:
  1. No stake in leading, global companies domiciled elsewhere (e.g., Samsung, Nestle, Toyota).
  2. Many firms hedge the currency risks of their foreign operations.
  3. A US-only portfolio is underweighted in "old world" industries; e.g., autos, electrical equipment and durable household goods.
Note that hedging foreign currency risks eliminates or at least greatly reduces the currency diversification effect.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
letsgobobby
Posts: 12073
Joined: Fri Sep 18, 2009 1:10 am

Re: International stock allocation: currency risk

Post by letsgobobby »

I like currency risk, and feel between TSM and TISM the goal is to own as many securities in as many currencies as possible.
Karamatsu
Posts: 1447
Joined: Mon Oct 27, 2008 2:42 am

Re: International stock allocation: currency risk

Post by Karamatsu »

I suppose I don't worry about currency issues too much when allocating equities. It's just another ingredient in the diversification blender, so I just use a simplified (whole-number) allocation that was roughly tuned to global market cap (as expressed in dollars) when I worked out my AA. Since I don't live in the US that means 40% US and 60% elsewhere. If I lived in the US... I think I might bias things the other way for the reason Bernstein gave. My "financial security" (both perceived and real) would be measured in dollars, so I'd want that to be primary.
User avatar
Topic Author
Kevin M
Posts: 15789
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: International stock allocation: currency risk

Post by Kevin M »

I'm not as convinced as Nisiprius that the currency-diversification-is-good is a bogus argument, but I've had thoughts along similar lines. Even though the risk is two-sided, why introduce the additional uncertainty unless the risk is compensated by a risk premium?

Perhaps the currency risk shouldn't be thought of as an independent risk factor. If the Vanguard paper is correct, currency movements have reduced the correlation between US stocks and foreign stocks. So although the expected return of the currency risk itself may be zero, the expected return of foreign stocks is not zero, so the lower correlation of US stocks and foreign stocks due to currency effects would be a benefit.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
steve_14
Posts: 1507
Joined: Wed Jun 20, 2012 12:05 am

Re: International stock allocation: currency risk

Post by steve_14 »

Kevin M wrote:So although the expected return of the currency risk itself may be zero, the expected return of foreign stocks is not zero, so the lower correlation of US stocks and foreign stocks due to currency effects would be a benefit.
Well that all depends. In the 50% of cases where foreign stocks outperform, I'd like as high a correlation as possible, thank you very much :) . If we expect exchange rates to float randomly, I wouldn't expect them to affect correlations (can bounce toward you as easily as away from you).
rnitz
Posts: 282
Joined: Thu May 10, 2007 12:47 pm
Location: San Diego

Re: International stock allocation: currency risk

Post by rnitz »

I also have less concern about currency volatility risk of Int'l equities. Why? Because you also have currency risk even if you only hold dollar denominated assets (S&P500 or TSM) and dollar denominated expenses (US retiree). To paraphrase Bastiat, it's the difference of what's seen versus what's not seen.

With TISM, when the dollar drops (or rises) you see it directly and linearly in your returns. But try this thought experiment: if you only hold TSM (and you're a US retiree with US only expenses), if the dollar dropped by 50% would you see any effect of this currency volatility? Any US company that imported sub-components (or outsourced labor, i.e. iPhones) would have cost increases and you'd feel them. The price of oil (itself an effective world currency) would nearly double, and you'd feel that. What effect would it have on US inflation? To be fair, there are a lot of compensating factors (exporters would have better sales/profits, etc.) but a US only investor/retiree is still exposed to this volatility.

Being exposed to international currency volatility also compensates for (hidden) exposure to US dollar volatility. I don't know the exact counterbalancing factors (which is why I still show some home-country US bias) but I think the idea should be considered.
User avatar
Rob5TCP
Posts: 3812
Joined: Tue Jun 05, 2007 7:34 pm
Location: New York, NY

Re: International stock allocation: currency risk

Post by Rob5TCP »

Are there any Vanguard ex-US that are NOT hedged?
User avatar
Topic Author
Kevin M
Posts: 15789
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: International stock allocation: currency risk

Post by Kevin M »

Rob5TCP wrote:Are there any Vanguard ex-US that are NOT hedged?
Vanguard international stock funds are not hedged, as far as I know. The international bond funds hedge currency risk, but here we're discussing stock funds.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
Rob5TCP
Posts: 3812
Joined: Tue Jun 05, 2007 7:34 pm
Location: New York, NY

Re: International stock allocation: currency risk

Post by Rob5TCP »

Kevin M wrote:
Rob5TCP wrote:Are there any Vanguard ex-US that are NOT hedged?
Vanguard international stock funds are not hedged, as far as I know. The international bond funds hedge currency risk, but here we're discussing stock funds.

Kevin

Thank you - I thought both stock and bond funds were hedged.
I wish the bond was unhedged as well.
User avatar
kramer
Posts: 1953
Joined: Wed Feb 21, 2007 1:28 am
Location: World Traveler

Re: International stock allocation: currency risk

Post by kramer »

Also, for the typical investor who has, say, a 30% allocation to foreign equities in a 60/40 portfolio:

* He probably has all or mostly dollar-denominated bonds
* Much of his wealth is in future dollar-denominated Social Security payments and Medicare

So his foreign currency exposure is much less than it initially seems.

As a US citizen investor who lives abroad and spends in non-dollar currencies, for the most part, I have struggled to get enough non-dollar exposure. It's not easy given the above factors.
User avatar
Chan_va
Posts: 870
Joined: Wed Dec 05, 2012 6:15 pm

Re: International stock allocation: currency risk

Post by Chan_va »

Currency risk gives you some inflation diversification since long term expected real return in forex is 0. That to me is worth the premium.
steve_14
Posts: 1507
Joined: Wed Jun 20, 2012 12:05 am

Re: International stock allocation: currency risk

Post by steve_14 »

kramer wrote:Also, for the typical investor who has, say, a 30% allocation to foreign equities in a 60/40 portfolio:

* He probably has all or mostly dollar-denominated bonds
* Much of his wealth is in future dollar-denominated Social Security payments and Medicare

So his foreign currency exposure is much less than it initially seems.

As a US citizen investor who lives abroad and spends in non-dollar currencies, for the most part, I have struggled to get enough non-dollar exposure. It's not easy given the above factors.
Right, with a 100% stock, 50% foreign portfolio, I manage to get enough (I think), but everyone should be mindful of their total financial picture.
User avatar
nisiprius
Advisory Board
Posts: 52216
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: International stock allocation: currency risk

Post by nisiprius »

Chan_va wrote:Currency risk gives you some inflation diversification since long term expected real return in forex is 0. That to me is worth the premium.
I once took a look at how effective a basket of foreign currencies was as inflation protection. I'm not completely sure I did it right, and the thread is here, but this is what I came up with.

Image

The blue line is the purchasing power of a dollar (measured by CPI, that is to say the purchasing power when used to purchase consumer goods within the United States). The red line is the purchasing power of the basket of foreign currencies used to calculate the "trade-weighted dollar index." It is not obvious that the foreign currencies provide much long-term inflation protection, and with regard to the short term, all they seem to do is make things worse with fairly meaningful random short-term fluctuations.

It seems to me that if inflation is the concern, TIPS are far more effective and reliable than foreign currencies.

It makes sense--I don't think the U.S. is any paragon of monetary integrity, but I don't think the rest of the world is any better--it's not as if everyone in the world had sound currency but us.
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
ogd
Posts: 4876
Joined: Thu Jun 14, 2012 11:43 pm

Re: International stock allocation: currency risk

Post by ogd »

With all due respect to Larry, I don't buy this argument as stated here or in a paragraph of the article.
Kevin M wrote:On the other side of the argument, Larry Swedroe says currency risk is not a factor that should affect one's allocation to international stocks, since the risk has upside as well as downside. See this article by Larry for more details: Swedroe: Int’l Diversification Is Free.
Even if a risk averages to zero, it's still a risk and I need to get paid for it. For example, leveraging a stock portfolio creates risk with both an upside and a downside, but I get paid for assuming it.

Now what the article actually argues is that currency risk has been worth it, because of the gains from diversification make up for it (in a stock/bond portfolio, but). I can accept that. But the right way to think about this is not that currency risk is free, but that it offset part or all of the diversification benefits. In other words, diversification is a free lunch that got eaten into by currency risk, in a proportion that depends on the overall shape of the portfolio.
User avatar
Topic Author
Kevin M
Posts: 15789
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: International stock allocation: currency risk

Post by Kevin M »

ogd wrote: Even if a risk averages to zero, it's still a risk and I need to get paid for it.
Yes, this is the doubt already expressed by Nisiprius and me.
ogd wrote:But the right way to think about this is not that currency risk is free, but that it offset part or all of the diversification benefits. In other words, diversification is a free lunch that got eaten into by currency risk, in a proportion that depends on the overall shape of the portfolio.
But the interesting point made by the Vanguard research paper is that it hasn't worked out this way since 1970. Currency fluctuations didn't eat into diversification, bit actually improved it by lowering the correlation between US stocks and foreign stocks. The correlation between foreign stocks denominated in US dollars and US stocks has been lower than the correlation between foreign stocks denominated in their local currencies and US stocks.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
ogd
Posts: 4876
Joined: Thu Jun 14, 2012 11:43 pm

Re: International stock allocation: currency risk

Post by ogd »

Kevin M wrote: But the interesting point made by the Vanguard research paper is that it hasn't worked out this way since 1970. Currency fluctuations didn't eat into diversification, bit actually improved it by lowering the correlation between US stocks and foreign stocks. The correlation between foreign stocks denominated in US dollars and US stocks has been lower than the correlation between foreign stocks denominated in their local currencies and US stocks.

Kevin
I don't read it the way you seem to, which is that currency risk is an absolute positive, but rather a net positive at certain allocations. If currency risk was good on its own, they would have recommended market weights, but as it is it seems that it stops working for you at about 40% and above that your efficiency is heading back down, as the volatility from currencies exceeds the diversification benefits.

I am also onboard with the proposition that this all changes over time, btw.
countmein
Posts: 653
Joined: Fri Dec 06, 2013 8:10 pm

Re: International stock allocation: currency risk

Post by countmein »

Here is the quote from Larry's article:

---
Finally, we need to discuss one reason for having a low allocation to international stocks: the currency risk it entails—a risk for which there is no risk premium. However, this isn’t a good reason, because currency risk is actually a two-way, not a one-way street. There’s just as much risk of the dollar falling in value as there is that it will rise in value.
---

If you have the opportunity to play a simple coin flip betting game (win/loss odds 50/50, even payout, expected value = 0); is it better or worse to play or abstain? Or is it that either choice is equally good?
letsgobobby
Posts: 12073
Joined: Fri Sep 18, 2009 1:10 am

Re: International stock allocation: currency risk

Post by letsgobobby »

I would approach this from the other angle. Start with a cap-weighted, unhedged global stock index like VT. If you want to deviate from that, be specific as to why.

I decided I didn't have a great reason to deviate from that at current valuations, so I'm 40/60 US/international. If relative valuations were equal, I'd be closer to 50/50 or a bit more US, due to dollar-based future obligations, etc. Also if I were closer to decumulation, I would be closer to 70/30 due to dollar-based present obligations.

As of today, no one can predict whether the dollar will be up or down relative to a basket of foreign currencies when I will need my money in 20 or 30 years. I don't want to put all my eggs in one basket so some semblance of splitting the difference 50/50 seems best to me.
User avatar
ogd
Posts: 4876
Joined: Thu Jun 14, 2012 11:43 pm

Re: International stock allocation: currency risk

Post by ogd »

countmein wrote:Here is the quote from Larry's article:

---
Finally, we need to discuss one reason for having a low allocation to international stocks: the currency risk it entails—a risk for which there is no risk premium. However, this isn’t a good reason, because currency risk is actually a two-way, not a one-way street. There’s just as much risk of the dollar falling in value as there is that it will rise in value.
---

If you have the opportunity to play a simple coin flip betting game (win/loss odds 50/50, even payout, expected value = 0); is it better or worse to play or abstain? Or is it that either choice is equally good?
Yes, it's the paragraph I was referring to, as in not agreeing.

In your game it's far better to abstain because you should demand to be paid for risk. It's easier to see why if you compare rewarded games: a game with a return expectation of 5% and one with double the risk but the same expectation. It's far better to play the first game, because you can put up double the money (e.g. leverage) and get 10% with the same risk as the second game. So the second game should pay 10%, not 5%.
User avatar
Topic Author
Kevin M
Posts: 15789
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: International stock allocation: currency risk

Post by Kevin M »

ogd wrote: If currency risk was good on its own, they would have recommended market weights, but as it is it seems that it stops working for you at about 40% and above that your efficiency is heading back down, as the volatility from currencies exceeds the diversification benefits.
The paper doesn't argue that currency risk is good on its own, but that it's a positive in that it reduces correlation between US and non-US stocks. So they're not recommending less than market weight in foreign stocks because of currency effects, but rather because it optimizes the overall correlation and diversification benefit.

One of the most interesting things about the paper, to me, is that they don't make the argument about decreasing international allocation because most of one's liabilities are in dollars, which seems to be the most commonly stated reason to do so (the one that Bill Bernstein states).

Kevin
If I make a calculation error, #Cruncher probably will let me know.
steve_14
Posts: 1507
Joined: Wed Jun 20, 2012 12:05 am

Re: International stock allocation: currency risk

Post by steve_14 »

Kevin M wrote:The paper doesn't argue that currency risk is good on its own, but that it's a positive in that it reduces correlation between US and non-US stocks.
I'd watch the part of this TED talk about "Confusing the past with the possible" http://www.youtube.com/watch?v=c-4flnuxNV4 . Random price movements are as apt to increase correlations as decrease them.
User avatar
Topic Author
Kevin M
Posts: 15789
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: International stock allocation: currency risk

Post by Kevin M »

letsgobobby wrote:Also if I were closer to decumulation, I would be closer to 70/30 due to dollar-based present obligations.
If you are typical, you will have a much higher fixed income allocation in decumulation (aka retirement), in which case the currency risk of the foreign stocks in your portfolio is even less of an issue.

More generally, the higher one's fixed income allocation, the less difference the US/International stock split makes from a diversification perspective (at least in terms of volatility reduction based on historical results). This is evident from the graph in Figure 3 on page 5 of the paper, which shows less variation in portfolio volatility as one varies the US/foreign ratio for a 60/40 stock/bond portfolio.

I am in "decumulation" and am at 60/40 US/foreign, but I'm only 30/70 equity/debt. My Mom is in her mid-80s, and we also have her at 60/40 US/foreign, but with stocks/fixed-income at 80/20, any possible currency risk in the foreign stocks is not even on the radar.

Kevin
Last edited by Kevin M on Thu Jul 03, 2014 3:24 pm, edited 1 time in total.
If I make a calculation error, #Cruncher probably will let me know.
letsgobobby
Posts: 12073
Joined: Fri Sep 18, 2009 1:10 am

Re: International stock allocation: currency risk

Post by letsgobobby »

Kevin M wrote:
letsgobobby wrote:If relative valuations were equal, I'd be closer to 50/50 or a bit more US, due to dollar-based future obligations, etc. Also if I were closer to decumulation, I would be closer to 70/30 due to dollar-based present obligations.
But this really is the crux of the discussion. The reason you state as a justification for more home bias is the "liabilities in dollars" argument, yet the research paper does not mention this as a factor, nor does Larry mention this as a justification for home bias for US investors.

On the other hand, Larry does say that a US citizen living in a foreign country should try to get more exposure to that country's local currency, perhaps by investing in local-currency fixed income. This seems to me to be somewhat inconsistent.

Kevin
I see that 'liabilities in dollars' is not their justification/argument for holding more dollar denominated securities in a portfolio. However for someone in the decumulation stage, I do not think they are saying that is not a factor at all, are they? As you say, Larry's point about expats can't be true if the reverse is not also true.
User avatar
ogd
Posts: 4876
Joined: Thu Jun 14, 2012 11:43 pm

Re: International stock allocation: currency risk

Post by ogd »

Kevin M wrote:
ogd wrote: If currency risk was good on its own, they would have recommended market weights, but as it is it seems that it stops working for you at about 40% and above that your efficiency is heading back down, as the volatility from currencies exceeds the diversification benefits.
The paper doesn't argue that currency risk is good on its own, but that it's a positive in that it reduces correlation between US and non-US stocks. So they're not recommending less than market weight in foreign stocks because of currency effects, but rather because it optimizes the overall correlation and diversification benefit.
The paper indeed doesn't but it seemed to me like you did. Again, they conclude that after 40% it becomes a negative. If it was nothing to worry about, there'd be nothing wrong to going all the way to 52% or beyond.
Kevin M wrote:One of the most interesting things about the paper, to me, is that they don't make the argument about decreasing international allocation because most of one's liabilities are in dollars, which seems to be the most commonly stated reason to do so (the one that Bill Bernstein states).
Not explicitly, but it's implied in this paragraph and others:
Vanguard wrote:Investments in foreign markets are exposed to fluctuations in
foreign exchange rates. Figure 9 illustrates that currency
fluctuations have periodically added to or subtracted from the
return for U.S. investors of international investments. For
example, currency movements subtracted 17% from the 12-month
returns of international stocks in 1984 and then added 35% in
1986.
They don't talk about "liabilities" per se, but I'm quite sure it's what "exposure" means.
User avatar
Topic Author
Kevin M
Posts: 15789
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: International stock allocation: currency risk

Post by Kevin M »

letsgobobby wrote: I see that 'liabilities in dollars' is not their justification/argument for holding more dollar denominated securities in a portfolio. However for someone in the decumulation stage, I do not think they are saying that is not a factor at all, are they? As you say, Larry's point about expats can't be true if the reverse is not also true.
Yes, this is one of the motivations for the post. I find it puzzling that the research paper doesn't address the liabilities-in-dollars argument, since it seems to be a main argument used by many to justify home bias. No, they don't say it's not a factor, but they don't say it is a factor either; they simply don't address it--at least not directly.

I read another Vanguard research paper on this topic some time ago; I'll try to find it to see if they address it.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
joer1212
Posts: 655
Joined: Sun Feb 17, 2013 12:55 pm

Re: International stock allocation: currency risk

Post by joer1212 »

I don't see any benefit at all in assuming currency risk, unless currency risk had an expected long-term real return, which it does not. Currency risk only adds uncompensated volatility to one's portfolio. If Vanguard offered a low-cost, hedged Total International ex-US fund, I would sell my VTIAX position and purchase it immediately. I think John Bogle is right about this one.
User avatar
nisiprius
Advisory Board
Posts: 52216
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: International stock allocation: currency risk

Post by nisiprius »

Kevin M wrote:...The paper doesn't argue that currency risk is good on its own, but that it's a positive in that it reduces correlation between US and non-US stocks...
Yes, and in my opinion that that argument is bogus, because I've fiddled around with it quite a bit and I am totally convinced that an asset (the currency risk component in this case) with zero return cannot improve a portfolio unless it has a negative correlation with the rest if the portfolio--not low correlation, not zero correlation, but negative correlation, and not just negative correlation from a luck-of-the-draw short period of time, but persistent, reliable negative correlation. As far as I know nobody claims that currency movements behave that way.

You can "reduce correlation" in your portfolio by taking, let's say 50% of your stocks, calling them your "roulette stocks," and taking them to the casino every weekend and gambling them all on red with one spin of the roulette wheel. In fact let's pretend it's a fair roulette wheel, with 36 slots 1-36 and no 0 or 00. Since roulette wheels are not correlated with the stock market, there is no doubt at all that this will reduce the correlation between your roulette stocks and the total U.S. stock market, but it will not improve the portfolio.
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.
joer1212
Posts: 655
Joined: Sun Feb 17, 2013 12:55 pm

Re: International stock allocation: currency risk

Post by joer1212 »

countmein wrote:Here is the quote from Larry's article:

---
Finally, we need to discuss one reason for having a low allocation to international stocks: the currency risk it entails—a risk for which there is no risk premium. However, this isn’t a good reason, because currency risk is actually a two-way, not a one-way street. There’s just as much risk of the dollar falling in value as there is that it will rise in value. ---
Currency risk may be a two-way street, but that amounts to volatility in one's portfolio. So, what is Swedroe's rationale for assuming that uncompensated volatility? What is the benefit?
Unless currency volatility has the potential to consistently reduce the overall volatility of one's portfolio, I see no benefit. And, even then, this would only be beneficial in the decumulation stage (retirement).
Last edited by joer1212 on Mon Sep 21, 2015 3:16 pm, edited 1 time in total.
joer1212
Posts: 655
Joined: Sun Feb 17, 2013 12:55 pm

Re: International stock allocation: currency risk

Post by joer1212 »

nisiprius wrote:an asset (the currency risk component in this case) with zero return cannot improve a portfolio unless it has a negative correlation with the rest if the portfolio--not low correlation, not zero correlation, but negative correlation, and not just negative correlation from a luck-of-the-draw short period of time, but persistent, reliable negative correlation.
+1
ji.isaacs
Posts: 103
Joined: Sat May 24, 2014 3:19 pm

Re: International stock allocation: currency risk

Post by ji.isaacs »

nisiprius wrote:
So, I see it as a balancing act. If there were a single world currency, then the optimum portfolio would be globally cap-weighted. Since there isn't, you are getting some intrinsic diversification from the fundamentals of international investments themselves.
Since there isn't, yet.

Really, though, the most salient characteristic of international equities is that all the data seems to show that it doesn't actually matter very much. I'm not sure what's behind what I see as the overpromotion of international stock investing. I don't really know what's supposed to be different between 1990, when Burton Malkiel was recommending that 1/6th of equities be international, and 2011, when he was recommending 1/2. The proportion of U.S. as a percentage of global actually increased between those endpoints.
I noticed the push, too. It was as if, on cue, global cap weighted portfolios were not just the rage, but there seemed to me a bit of salesmanship urgency to the new suggestion. Perhaps we are headed towards a global currency. Perhaps that is what's at the end of the flirt pole Rick Ferri is waving about. Waiting for some event that hasn't happened yet, but we're almost there.
rnitz
Posts: 282
Joined: Thu May 10, 2007 12:47 pm
Location: San Diego

Re: International stock allocation: currency risk

Post by rnitz »

I don't think it's reduction of volatility through negative or low correlation, it's the mitigation of a particular risk - the risk of the decline of the dollar. You may think that because you have US dollar only direct expenses and invest only in US dollar investments that you have absolutely no exposure to the risk of the value of the dollar, but I would posit that you may be wrong. International equity exposure (at least through Vanguard International equity index funds) is an incredibly cheap way to hedge/mitigate this risk.
User avatar
oneleaf
Posts: 2562
Joined: Mon Feb 19, 2007 4:48 pm

Re: International stock allocation: currency risk

Post by oneleaf »

rnitz wrote:I don't think it's reduction of volatility through negative or low correlation, it's the mitigation of a particular risk - the risk of the decline of the dollar. You may think that because you have US dollar only direct expenses and invest only in US dollar investments that you have absolutely no exposure to the risk of the value of the dollar, but I would posit that you may be wrong. International equity exposure (at least through Vanguard International equity index funds) is an incredibly cheap way to hedge/mitigate this risk.
I agree. There has even been some research (Crill and Davis) done over 53 years worth of data that shows that unhedged international stocks performed best during periods of high inflation. They found a slight positive correlation between unexpected inflation and international stocks. The currency movements can work just when you need them most. Of course, like commodities, this doesn't always happen since there are more than one type of bear market. Since our last major downturn was deflationary, and every diversifier you could think of went down (except for treasuries), many more than the US market, it is easy to shun diversifiers (like international and emerging stocks, commodities). We are even seeing references to Buffett's 90% S&P500 index recommendation nowadays. I think the current environment is challenging for those who like diversifiers, but it is times like these that we must stick to the strategy.
User avatar
patrick013
Posts: 3301
Joined: Mon Jul 13, 2015 7:49 pm

Re: International stock allocation: currency risk

Post by patrick013 »

I'm kind of a doubting Thomas about this, meaning when I see
it I'll believe it. I would think that an actively managed int'l
fund would be able to take advantage of currency fluctuations thru
research. I would think an int'l index fund would be advantaged
by diversification.

Some good returns are there but I haven't seen that they were
decidedly better than US funds. Disappointing is bond funds
where a little research should put the fund mostly where it
would be in favor. So I see int'l funds with potential.
age in bonds, buy-and-hold, 10 year business cycle
Post Reply