Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Counterpoint
Posts: 45
Joined: Sun Jun 05, 2016 1:42 pm

Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by Counterpoint » Tue Sep 11, 2018 8:03 pm

This is one of the findings from an Aug 2018 research piece from Vanguard: https://personal.vanguard.com/pdf/ISGPCH.pdf They break down their analysis by home currency of the investor, but for US investors the key is in Figure 7b. It shows that over the last few decades, a US investor would have reduced volatility in most 3-year rolling periods by hedging currency risk on an international equity portfolio. (Hedging currency risk on international bonds is a no-brainer).

This is not new: Vanguard’s research echoes findings by (for example) JP Morgan’s institutional research in 2010, where JPM recommended that US investors hedge their currency exposure from international equities: see page 7 and Figure 4G in https://careers.jpmorganchase.com/jpmpd ... 198939.pdf

So while there are lots of good reasons to invest in international stocks, IMO diversification benefit of currency risk is not one of them.

JPM is more clear about their recommendations; Vanguard waffles around despite the clear nature of the historical data, perhaps because it would require too big a shift on their part to start hedging currency risk on international stocks (for example in their target date index funds), and the research folks don't want to rock the boat too much.

Hedging currency risk on developed market currencies is very straightforward and inexpensive (very low bid-offers), for instance using 1-month rolling forwards. There are 2 ETFs I know of that do this already - DBEF and HEFA, but they cost about 35bp more than Vanguard’s unhedged developed market index fund VTMGX, and I’m not willing to pay that much extra (the additional cost is not due to the cost of hedging, which is just a couple of basis points in developed markets). I have been waiting and hoping for Vanguard to come out with a currency-hedged developed market ETF/index fund at a reasonable ER - that would allow people like me to maintain exposure to other countries’ markets without taking on currency risk (in the institutional finance world, this currency risk is often called the “return-free risk”).


[As an aside, the general wisdom in institutional finance is to NOT hedge currency risk on emerging market stocks because (1) the cost of hedging (in terms of both bid-offer and carry) is often too high, and (2) over the very long term, EM currencies would be expected to appreciate in real terms as they converge towards their PPP values - see for instance bullet point #3 in the summary on pg 2 of Goldman’s research piece that came up with the term “BRICs”: http://www.goldmansachs.com/our-thinkin ... -dream.pdf ]
Last edited by Counterpoint on Wed Sep 12, 2018 10:22 am, edited 1 time in total.

columbia
Posts: 842
Joined: Tue Aug 27, 2013 5:30 am

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by columbia » Tue Sep 11, 2018 8:45 pm

Hedged EAFE ETF from iShares costs 0.35%. I feel certain that Vanguard could beat that price.

HEDGEFUNDIE
Posts: 646
Joined: Sun Oct 22, 2017 2:06 pm

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by HEDGEFUNDIE » Tue Sep 11, 2018 8:56 pm

I am following with interest WisdomTree’s “dynamic currency hedging” ETFs:

https://www.wisdomtree.com/strategies/c ... 96B4600300

Early results are promising:

https://www.portfoliovisualizer.com/bac ... ion3_3=100

Counterpoint
Posts: 45
Joined: Sun Jun 05, 2016 1:42 pm

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by Counterpoint » Tue Sep 11, 2018 9:40 pm

columbia wrote:
Tue Sep 11, 2018 8:45 pm
Hedged EAFE ETF from iShares costs 0.35%. I feel certain that Vanguard could beat that price.
Yup, this is the HEFA I was referring to. I mistakenly referred to it (and DBEF - Deutsche Bank's X-Trackers MSCI EAFE Hedged Equity ETF) as index funds rather than the ETFs that they are.
And I very much agree that Vanguard could do it for a much lower ER if they decided to do such an ETF/fund.
HEDGEFUNDIE wrote:
Tue Sep 11, 2018 8:56 pm
I am following with interest WisdomTree’s “dynamic currency hedging” ETFs:

https://www.wisdomtree.com/strategies/c ... 96B4600300

Early results are promising:

https://www.portfoliovisualizer.com/bac ... ion3_3=100
Yes, I also looked at WisdomTree's dynamic currency hedging, which is based on equal weightage to momentum, carry and value parameters. But evaluating the effectiveness of this is equivalent to evaluating active management with all the attendant complexity - my objective is to keep things simple with index funds.

User avatar
Noobvestor
Posts: 4750
Joined: Mon Aug 23, 2010 1:09 am
Contact:

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by Noobvestor » Tue Sep 11, 2018 10:32 pm

Counterpoint wrote:
Tue Sep 11, 2018 8:03 pm
It shows that over the last few decades, a US investor would have reduced volatility in most 3-year rolling periods by hedging currency risk on an international equity portfolio.
That's a very narrow way to look at things. Sample size: one country. Sample period: a few decades. Sample type: 3-year rolling periods.
Counterpoint wrote:
Tue Sep 11, 2018 8:03 pm
So while there are lots of good reasons to invest in international stocks, diversification benefit of currency risk is not one of them.
Are we reading the same whitepaper? Because in the one I'm reading, Vanguard puts forward some sound arguments to the contrary:
vanguard wrote:Although currency exposure can be hedged away ... portfolio risk goes beyond simple “volatility” measures. Tail risk, or unexpected portfolio impacts, can stem from events such as an economic crisis, geopolitical developments, and execution risk; the list is long and often unknowable. So it may be advisable to consider risks beyond volatility, and holding some nonlocal currency exposure, which can reduce overall portfolio risk.
In all regions, the hedging effect is relatively marginal and sometimes removes volatility. Hedging at times may actually lead to higher volatility
[Currency can also protect against] tail risks—a currency, economic, or military crisis—not captured in the historical data.
[A] hedging strategy does have an associated cost and risks, so a reduction in short-term volatility should be weighed against diminished return.
Putting the potential portfolio result in terms of that magnitude is an important exercise. Under both a 90/10 and a 60/40 portfolio, we see the most benefit between a 100% hedged and 100% unhedged portfolio. Given the high equity allocation, however, the overall benefit of volatility reduction from hedging appears to be marginal. For instance, a 60/40 portfolio with a 30% hedge has 18 basis points less volatility than an unhedged portfolio at the Vanguard home-bias position. (A basis point is one one-hundredth of a percentage point.)
A U.S. investor has a unique circumstance as well. The U.S. dollar represents approximately 50% of all equity and fixed income issuance, and depending on the strategic asset allocation, a majority of a U.S. investor’s exposure may already be held in one currency. A U.S. investor therefore must weigh the potential volatility reduction with structural and behavioral variables (including U.S. dollar market representation).
In the end, this all comes down to the question: will the future look like the past? And, related: is the US persistently special? In other words: do you want to actively extrapolate that the next few decades will look just like the last few, or concede that we just can't know?

Personally, I'd rather 'hedge' my US dollar risk (70% of my total portfolio) with some unhedged international exposure.
"In the absence of clarity, diversification is the only logical strategy" -= Larry Swedroe

Counterpoint
Posts: 45
Joined: Sun Jun 05, 2016 1:42 pm

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by Counterpoint » Wed Sep 12, 2018 10:19 am

Noobvestor wrote:
Tue Sep 11, 2018 10:32 pm
Thanks for the response, noobvestor.

First of all, I should have separated out the facts/analytical results (that the historical data shows that US investors would have reduced volatility by hedging) from my opinion (that currency diversification is not a good reason for engaging in international equity investments) - so I’ll add an “IMO” to that section of my original post.

I also want to mention that my opinion actually makes life complicated for me, so I welcome contrary views. It would be much simpler for me if I could convince myself that currency risk in my international equity investments was good for my portfolio - and I’m a big believer in simplicity. So I’ll lay out my thought process below, and look forward to feedback.


Overall Context

Vanguard’s paper (and hence overall language) looks at investors with home bases in various currencies (US, UK, Australia, Canada) for whom the overall currency hedging picture can be quite different - I was only referring to the portion relevant to US-based investors. In fact, of the 4 countries studied, the US is the only one where (1) both the equity-currency correlation and the relative historical performance consistently suggest the benefit of hedging, and (2) where this relationship is stable over time.

Also, Vanguard has its own opinion/spin which in my view is not unbiased (possibly for the reasons I mentioned in my original post) - I’ve noticed this in their previous currency hedging pieces over the years. This is not to discount all their qualifiers (of which more below).

But it’s worth reading JP Morgan’s piece as well, which is much clearer in recommending that most US investors fully hedge their international equity positions. From their piece: “For investors based in Switzerland, Germany, Japan, the United Kingdom, and the United States, full hedging is appropriate for most investors (see 4C, 4D, 4E, 4F and 4G).” It’s also interesting to read about the structural reasons why currencies of countries like Australia and Canada behave very differently from the others.


Vanguard’s section for US investors

In Vanguard’s piece (with their language in quotes followed by my commentary):

This analytical approach suggests that, all else being equal, the U.S. investor should hedge the international equity position.
Ok, that’s the big (initial) takeaway.

“As we discussed in the “Variables affecting the currency-hedging decision” section on page 8, a number of other factors that can alter that view are not included in the hard numbers. For instance, a long- term investor may be indifferent to short-term volatility, and the magnitude of the gain may be limited.”
True, but the traditional main argument for any diversification benefit is that it lowers portfolio volatility (the free lunch). If taking currency risk actually has increased portfolio volatility historically, then we need to move on to other arguments why taking currency risk may make sense.

“Furthermore, the impacts of several variables, often difficult to quantify, influence the investor decision. These include the fact that the period studied may not be prologue,”
Of course. But the evidence for a US investor also appears very stable over the many 3-year rolling periods from 1970 through 2016 which covers a wide range of economic scenarios and US dollar appreciation/depreciation.

“the risk of large concentrations in a single currency,”
This seems to be a possible substantive argument (together with tail risk, mentioned elsewhere in the piece), so more on these in the next section

“the costs and risks of executing a hedging program,”
These are trivial for short-term hedging in developed markets (which is what I was referring to) - Vanguard’s piece itself earlier says that “Today, most hedging costs amount to a handful of basis points” other than for EM currencies

“the investor behavioral risks of another investment choice,”
Hmmm… So we’re being patronized that we cannot be trusted to make our own decisions? Enough said for a BH audience.

“and the performance volatility that hedging can bring compared with local practice and benchmarking.”
More of an issue for institutional investors when they are compared to their peers, but less so for retail investors (and in any case there does seem to be a healthy disagreement on the currency hedging issue at least as judged by this forum)


Tail Risk

Yes, there will be tail risk events. But given the reserve currency status of the US dollar (this is a reality for quite a few years to come, not a starry-eyed view of American exceptionalism), an unexpected tail risk event will often result in a flight to quality into US Treasuries and hence an appreciation of the dollar. If this direction of flow is more likely in a tail risk event, it would argue that on average having unhedged non-US equities would be worse during a tail risk event than hedged non-US equities.


Concentration Risk in Dollar-Denominated Assets

I’m trying to understand this one better. The argument seems to be that a US-based investor is too concentrated in dollars if they have just dollar-denominated assets. If this is the case, shouldn’t it apply also to a very conservative portfolio that is 100% FI (fixed-income) that is both US + non-US? The conventional wisdom including from Vanguard is to hedge all non-US FI back to dollars because the unhedged currency risk is too high relative to FI risk. But this is a pragmatic reason, and does not address my conceptual question of why US dollar concentration risk is not mentioned in the context of a 100% FI portfolio, but is brought up if the portfolio contains equities. In fact, if concentration risk were that significant, it would probably be a higher portion of the risk of a (lower-risk) FI portfolio than of a (higher-risk) equity portfolio, and hence more important to address in the FI portfolio.

User avatar
Noobvestor
Posts: 4750
Joined: Mon Aug 23, 2010 1:09 am
Contact:

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by Noobvestor » Wed Sep 12, 2018 12:37 pm

Counterpoint wrote:
Wed Sep 12, 2018 10:19 am
In fact, of the 4 countries studied, the US is the only one where (1) both the equity-currency correlation and the relative historical performance consistently suggest the benefit of hedging, and (2) where this relationship is stable over time.
I think we're looking at the same data and drawing opposite conclusions. I see this and I think 'OK, so if you happened to be a US investor over this period, you happened to get marginally higher volatility, but you shouldn't bank on being in the right country at the right time.'
Also, Vanguard has its own opinion/spin which in my view is not unbiased (possibly for the reasons I mentioned in my original post) - I’ve noticed this in their previous currency hedging pieces over the years. This is not to discount all their qualifiers (of which more below).
I see your point, but find it a little hard to believe that Vanguard is engaged with circular reasoning. I mean, I hope they're not.
“This analytical approach suggests that, all else being equal, the U.S. investor should hedge the international equity position.”
But then they go on to explain that the actual magnitude of volatility benefit of hedging (strictly analytically) has been marginal.
“the costs and risks of executing a hedging program,” These are trivial for short-term hedging in developed markets (which is what I was referring to) - Vanguard’s piece itself earlier says that “Today, most hedging costs amount to a handful of basis points” other than for EM currencies
But again, per the above, the benefits are also marginal (and at times and in some places: nonexistent).
“the investor behavioral risks of another investment choice,”
Hmmm… So we’re being patronized that we cannot be trusted to make our own decisions? Enough said for a BH audience.
I don't see it that way. If I were 100% USD-hedged and the dollar suffered for an extended period, I might second-guess my choices. I would hope that I wouldn't cave at the wrong time, but better to simply hold a portfolio that has less risk of that in the first place.
Tail Risk: Yes, there will be tail risk events. But given the reserve currency status of the US dollar (this is a reality for quite a few years to come, not a starry-eyed view of American exceptionalism), an unexpected tail risk event will often result in a flight to quality into US Treasuries and hence an appreciation of the dollar. If this direction of flow is more likely in a tail risk event, it would argue that on average having unhedged non-US equities would be worse during a tail risk event than hedged non-US equities.
Historically this has been true, and it's a reason I hold a sizable chunk of US Treasuries. But it's not just about short-term flights to safety - we're talking about long-term investing. The US dollar is the reserve currency now, but wasn't always, and won't always be. I have no idea if its status will change in my lifetime, but I'm not going to place all of my bets on the status quo. Disasters can unfold quickly, like the sudden crash of the Japanese market decades ago, or slowly, like its torpid pace in the decades since. And who knows, maybe the crisis will be dollar-related.
Concentration Risk in Dollar-Denominated Assets: I’m trying to understand this one better. The argument seems to be that a US-based investor is too concentrated in dollars if they have just dollar-denominated assets. If this is the case, shouldn’t it apply also to a very conservative portfolio that is 100% FI (fixed-income) that is both US + non-US? The conventional wisdom including from Vanguard is to hedge all non-US FI back to dollars because the unhedged currency risk is too high relative to FI risk. But this is a pragmatic reason, and does not address my conceptual question of why US dollar concentration risk is not mentioned in the context of a 100% FI portfolio, but is brought up if the portfolio contains equities. In fact, if concentration risk were that significant, it would probably be a higher portion of the risk of a (lower-risk) FI portfolio than of a (higher-risk) equity portfolio, and hence more important to address in the FI portfolio.
I don't see anyone recommending 100% FI for even ultra-conservative investors. Even Vanguard's target date funds bottom out at around 30% equities. But sure, as you approach spending, having more local currency may make sense for volatility reasons. In most cases, though, this is handled by gliding toward bonds. If you wanted to go 100% TIPS (I'd advise against it but) I'd say at least then you're hedging your own currency's inflation - but even then, I'd be nervous about the government monkeying with inflation metrics down the line or some other low-probability, high-impact problem. If I were personally going for an ultra-conservative portfolio, I'd maybe do 70% TIPS, 30% global stocks (US + international). Beyond a certain point (Benjamin Graham's rule of thumb was 75/25 to 25/75 stocks/bonds) you take on risks best avoided through diversification.

For me to be convinced that hedging was worthwhile, I'd want to see (a) that it had relatively consistent benefits across many/most countries over long periods, and (b) that those benefits were worth paying for in terms of cost and risk. Right now I don't see either of those. I really don't think this is a case of Vanguard going against its own data - I think they're looking at the data, seeing a small marginal historical benefit that is dependent on a particular time window and country, and concluding that pursuing it isn't worth the associated costs and risks.
"In the absence of clarity, diversification is the only logical strategy" -= Larry Swedroe

columbia
Posts: 842
Joined: Tue Aug 27, 2013 5:30 am

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by columbia » Wed Sep 12, 2018 1:11 pm

The VG Global Minimum Volatility fund is currency hedged and I presume that’s their test bed for hedging international equities.

Counterpoint
Posts: 45
Joined: Sun Jun 05, 2016 1:42 pm

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by Counterpoint » Sun Sep 16, 2018 9:50 am

Concentration Risk: Vanguard’s FI hedging implies this risk is very small

Thanks for your thoughtful response, Noobvestor. Even if you and I have different perspectives on the same information, it was instructive to see your reasoning. I’ll address a couple of specific issues in this post, and follow up with a more general response.

In responding to my earlier post on concentration risk, you seem to have been somewhat diverted into what kind of AA is appropriate for a conservative investor, which was not my point at all. My point is to say that there’s inconsistency between an investor being concerned about concentration risk in dollar-denominated securities for a hedged equity portfolio, while not expressing concern about this same risk for a (lower-risk) hedged fixed-income (FI) portfolio. Vanguard and others suggest hedging 100% of non-US FI back into USD for US investors, because they view the currency risk as too high relative to FI risk. But since this results in moving to a 100% “concentration” in USD, it also implies that they see the resulting USD concentration risk as being much less of an issue than taking on currency risk - otherwise they would not have hedged the full 100% back to USD. With that as the backdrop, if you then believe that currency risk is marginal in the context of an equity portfolio (as you quoted from the Vanguard paper), that implies that concentration risk must therefore be even less than marginal for an equity portfolio.

This is one illustration of why I think the Vanguard piece is slanted : While they acknowledge that the hard numbers mean that US investors should hedge their international equity positions, they add qualifiers like concentration risk without analytical justification.

Another illustration of Vanguard’s slant: You correctly pointed out the following quote from Vanguard - “Given the high equity allocation, however, the overall benefit of volatility reduction from hedging appears to be marginal. For instance, a 60/40 portfolio with a 30% hedge has 18 basis points less volatility than an unhedged portfolio at the Vanguard home-bias position.” However, there were 8 different cases reported in Table 7 of the paper, depending on whether you are looking at 90/10 or 60/40 portfolios, 100% or 30% hedging, 40% or market cap allocation to international. The reduction in volatility from hedging ranges from 18 bp to 83 bp. Vanguard chooses to only highlight the lowest case of benefiting from hedging: 18 bp.

Research arms of financial firms always have an eye out for their firms’ business needs (having worked on Wall Street, I’m quite familiar with this bias, whether explicit or implicit). And Vanguard has tens if not hundreds of billions of dollars invested for US investors in unhedged international stocks for their target date funds or indexed portfolios. Imagine if their research piece was more unequivocal (as is the JP Morgan piece) about the recommendation to hedge currency risk for US investors, without any qualifiers - it could create quite the kerfuffle among their investors. I actually applaud them for at least presenting the numbers straight up, but one does have to read between the lines based on the realpolitik.

Counterpoint
Posts: 45
Joined: Sun Jun 05, 2016 1:42 pm

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by Counterpoint » Sun Sep 16, 2018 10:00 am

Rationale for International Investing and Undesired Currency Overlay

More broadly, part of the relative important of currency risk depends on an investor’s portfolio. My portfolio is a 100% allocation to equities since we are in a comfortable financial position and don't need to draw down the portfolio, and so are the portfolios of a couple of young family members who are just starting out their careers given their long time horizons and risk tolerance. Hence the issue of whether our international stock allocations are hedged or not has a greater impact on our overall portfolio risk than for those with more balanced portfolios.

My objective in international stock investments is to diversify my stock portfolio globally, while staying within the equities asset class. I am looking for pure stock market risk given the expected real returns over time on stocks, but to diversify away from the US. To me, a currency overlay acts like a separate asset class from my desired equity exposure. So if I have a 40% allocation to international equities (as an example) and that’s unhedged, I not only have exposure to international equities for 40% of my portfolio, but I inherit a currency overlay position of 40% without asking for it.

(Noobvestor: On the other hand, your perspective seems to be that you take the combined stock market risk plus currency risk as your starting point, and you need a high bar to be convinced to peel away the currency risk. That’s a different perspective.)

For me, I’d like to be given the choice of either investing in unhedged international stocks or hedged international stocks. A good chunk of my concern about my exposure to the US economy may already be satisfied by buying hedged international stocks. Then if I’m further concerned about US dollar risk, I can decide whether to add some protection against that by undertaking a partially unhedged strategy. In the example above, I don’t want to be told that I have to take a 40% currency overlay - 10% or 20% may suffice. (In fact both the Vanguard and JP Morgan papers look at the scenario of partially hedged international stock investments.) Or perhaps I’d prefer to hedge this risk with a small allocation to gold - or whatever. The point is that I want my portfolio to be constructed on my terms, given my risk concerns, without being handed a pre-determined risk allocation to currency risk which is highly unlikely to match my needs.
columbia wrote:
Wed Sep 12, 2018 1:11 pm

The VG Global Minimum Volatility fund is currency hedged and I presume that’s their test bed for hedging international equities.
That’s an interesting fund I was not aware of - thanks for pointing it out. While it does hedge currency risk to keep down volatility, it also does a lot more to reduce volatility, like significant deviations from the sectoral composition of the overall stock market (e.g. much lower weights for technology and energy). This would be inconsistent with my (more or less) 2-fund philosophy, but I hope you’re right that it may make Vanguard feel more comfortable with offering a currency hedged international stock fund in the future.

User avatar
Noobvestor
Posts: 4750
Joined: Mon Aug 23, 2010 1:09 am
Contact:

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by Noobvestor » Sun Sep 16, 2018 11:54 am

Counterpoint wrote:
Sun Sep 16, 2018 9:50 am
Concentration Risk: Vanguard’s FI hedging implies this risk is very small

Thanks for your thoughtful response, Noobvestor. Even if you and I have different perspectives on the same information, it was instructive to see your reasoning.
Same!
Counterpoint wrote:
Sun Sep 16, 2018 9:50 am
In responding to my earlier post on concentration risk, you seem to have been somewhat diverted into what kind of AA is appropriate for a conservative investor, which was not my point at all. My point is to say that there’s inconsistency between an investor being concerned about concentration risk in dollar-denominated securities for a hedged equity portfolio, while not expressing concern about this same risk for a (lower-risk) hedged fixed-income (FI) portfolio. Vanguard and others suggest hedging 100% of non-US FI back into USD for US investors, because they view the currency risk as too high relative to FI risk. But since this results in moving to a 100% “concentration” in USD, it also implies that they see the resulting USD concentration risk as being much less of an issue than taking on currency risk - otherwise they would not have hedged the full 100% back to USD. With that as the backdrop, if you then believe that currency risk is marginal in the context of an equity portfolio (as you quoted from the Vanguard paper), that implies that concentration risk must therefore be even less than marginal for an equity portfolio.
My bold above. Maybe I wasn't clear, but it's impossible to decouple this from an AA conversation. In pushing the question to a logically possible (but not actually implemented) extreme, I see your point entirely - if Vanguard advised 100% FI portfolios, and they stuck to their 100% hedged position, then there would be currency concentration risk. I agree. But the first of those premises is false - they don't - so it's purely academic.
"In the absence of clarity, diversification is the only logical strategy" -= Larry Swedroe

Counterpoint
Posts: 45
Joined: Sun Jun 05, 2016 1:42 pm

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by Counterpoint » Sun Sep 16, 2018 10:10 pm

Noobvestor wrote:
Sun Sep 16, 2018 11:54 am
In pushing the question to a logically possible (but not actually implemented) extreme, I see your point entirely - if Vanguard advised 100% FI portfolios, and they stuck to their 100% hedged position, then there would be currency concentration risk. I agree. But the first of those premises is false - they don't - so it's purely academic.
A quick search shows that Vanguard does include 100% FI portfolios as part of their recommended portfolios depending on individual situations/risk tolerance - for instance see their first portfolio in https://personal.vanguard.com/us/insigh ... ns?lang=en (they have all combinations from 100% bonds to 100% stocks). Or see a more detailed portfolio specification for a 100% FI portfolio in the first column in the pages titled Core Series and Income Series in https://advisors.vanguard.com/iwe/pdf/FASINVMP.pdf : these 100% FI portfolios have 29.4% in international FI all of which is targeted to be hedged back to USD. And even if Vanguard were not suggesting 100% FI portfolios, the use of a logical extreme or “boundary value condition” is often useful in clarifying principles.

However, your point about this potentially being an academic exercise is true in a different sense: Until there is a low-cost hedged international equity index portfolio (at least for the developed markets), for me this is still effectively an academic exercise, since I have little choice but to accept currency risk as the price to be paid (in my view) to have international stock diversification. As I mentioned earlier, I view the cost of the Deutsche Bank and iShares hedged index ETFs as too high - I’m hoping Vanguard (or perhaps Fidelity or Schwab in this new world of indexing frenzy) will come up with one soon. That way each of one us could truly pick the strategy - unhedged or hedged or partially hedged - that we prefer for our international equity investments.

HEDGEFUNDIE
Posts: 646
Joined: Sun Oct 22, 2017 2:06 pm

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by HEDGEFUNDIE » Sun Sep 16, 2018 10:13 pm

Counterpoint wrote:
Sun Sep 16, 2018 10:10 pm
However, your point about this potentially being an academic exercise is true in a different sense: Until there is a low-cost hedged international equity index portfolio (at least for the developed markets), for me this is still effectively an academic exercise, since I have little choice but to accept currency risk as the price to be paid (in my view) to have international stock diversification. As I mentioned earlier, I view the cost of the Deutsche Bank and iShares hedged index ETFs as too high - I’m hoping Vanguard (or perhaps Fidelity or Schwab in this new world of indexing frenzy) will come up with one soon. That way each of one us could truly pick the strategy - unhedged or hedged or partially hedged - that we prefer for our international equity investments.
Just out of curiosity, how much extra ER are you willing to pay for a hedged developed markets fund, and why that amount?

User avatar
Noobvestor
Posts: 4750
Joined: Mon Aug 23, 2010 1:09 am
Contact:

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by Noobvestor » Sun Sep 16, 2018 10:55 pm

Counterpoint wrote:
Sun Sep 16, 2018 10:10 pm
Noobvestor wrote:
Sun Sep 16, 2018 11:54 am
In pushing the question to a logically possible (but not actually implemented) extreme, I see your point entirely - if Vanguard advised 100% FI portfolios, and they stuck to their 100% hedged position, then there would be currency concentration risk. I agree. But the first of those premises is false - they don't - so it's purely academic.
A quick search shows that Vanguard does include 100% FI portfolios as part of their recommended portfolios depending on individual situations/risk tolerance - for instance see their first portfolio in https://personal.vanguard.com/us/insigh ... ns?lang=en (they have all combinations from 100% bonds to 100% stocks). Or see a more detailed portfolio specification for a 100% FI portfolio in the first column in the pages titled Core Series and Income Series in https://advisors.vanguard.com/iwe/pdf/FASINVMP.pdf : these 100% FI portfolios have 29.4% in international FI all of which is targeted to be hedged back to USD. And even if Vanguard were not suggesting 100% FI portfolios, the use of a logical extreme or “boundary value condition” is often useful in clarifying principles.

However, your point about this potentially being an academic exercise is true in a different sense: Until there is a low-cost hedged international equity index portfolio (at least for the developed markets), for me this is still effectively an academic exercise, since I have little choice but to accept currency risk as the price to be paid (in my view) to have international stock diversification. As I mentioned earlier, I view the cost of the Deutsche Bank and iShares hedged index ETFs as too high - I’m hoping Vanguard (or perhaps Fidelity or Schwab in this new world of indexing frenzy) will come up with one soon. That way each of one us could truly pick the strategy - unhedged or hedged or partially hedged - that we prefer for our international equity investments.
I see they list out all options from 0% to 100%, but I'm curious (and do not know the answer - this isn't rhetorical): do they ever actually recommend 0% stocks? I somehow can't imagine it - if so, I would assume it would include some TIPS (which would hedge USD inflation).

I also don't disagree that this is something Vanguard could productively offer (hedged international) assuming they could get the costs sorted.
"In the absence of clarity, diversification is the only logical strategy" -= Larry Swedroe

jalbert
Posts: 3585
Joined: Fri Apr 10, 2015 12:29 am

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by jalbert » Mon Sep 17, 2018 12:16 am

Hedging currency risk on developed market currencies is very straightforward and inexpensive (very low bid-offers), for instance using 1-month rolling forwards. There are 2 ETFs I know of that do this already - DBEF and HEFA, but they cost about 35bp more than Vanguard’s unhedged developed market index fund VTMGX, and I’m not willing to pay that much extra (the additional cost is not due to the cost of hedging, which is just a couple of basis points in developed markets)
Currently, forward currency contracts to hedge developed market currencies to USD should have a positive hedge return, ie the cost should be negative. With the recent substantial rise in 1-month rates in the US, I would guess that the hedge return is more than enough to cover the 35 bp of additional ER of the funds in question.
Index fund investor since 1987.

Counterpoint
Posts: 45
Joined: Sun Jun 05, 2016 1:42 pm

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by Counterpoint » Mon Sep 17, 2018 7:55 am

HEDGEFUNDIE wrote:
Sun Sep 16, 2018 10:13 pm
Just out of curiosity, how much extra ER are you willing to pay for a hedged developed markets fund, and why that amount?
Estimates of hedging cost due to bid-ask spreads are made in this Vanguard piece: https://personal.vanguard.com/pdf/icrifi.pdf See pages 16-17 and in particular Figure 11 and its footnote. This was estimated several years ago (costs may have come down a bit since), and at that time it was a weighted average of about 13 bps for those currencies (which appear to account for about 85-90% of the portfolio of VTMGX/VEA, the ex-US developed market unhedged index). This 13 bps assume rolling 1-month forwards, so paying 12 bid-ask spreads in a year. As the footnote mentions, in practice 3 or 6-month forwards could be used for at least part of the portfolio which would reduce total costs due to bid-ask spreads, but would presumably increase tracking error relative to a benchmark that uses 1-month rolling forwards.

So I hope that if offered at cost by Vanguard, it would come in for less than the 13 bps mentioned in the research piece. This would be over and above the cost of an unhedged fund/ETF such as VTMGX/VEA which has an ER of 7 bps. So total ER under 20 bps, hopefully closer to 15 bps.

There is another aspect to hedging which is the forward premium/discount between the 2 currencies - this is referred to in jalbert’s post. It is the result of short-term interest rate differentials between the two currencies involved in the hedging. It’s useful to separate this aspect from the bid-ask spreads which is the true transaction cost of hedging. To understand this, take a look at the fact sheet for a developed market index hedged back to USD (it’s an MSCI index and VTMGX uses a FTSE index, but the principles are the same): https://www.msci.com/documents/10199/a8 ... c24481b783

The Annual Performance columns at the right of the top page have 3 sets of numbers: The first is the performance of the MSCI EAFE index hedged back to USD using 1-month rolling forwards, the second is the performance of the MSCI EAFE in purely local terms, and the third is the performance of an unhedged investment in the MSCI EAFE when measured in USD. To quote the fact sheet, “There are two components to a MSCI Hedged Index return: 1) the performance of the unhedged index in the home currency; and, 2) the Hedge Impact (aimed to represent the gain or loss on the Forward contracts) in the home currency.” In other words, the impact of the forward premium/discount referred to by jalbert is the difference between Column 1 and Column 2. That is separate from the transaction costs i.e. bid-ask spreads, which would be the tracking error between an actual hedged fund versus the hedged index, and this is what would be included in an ER and not the impact of the forward premia/discounts. That's why you see the Vanguard piece only referring to the bid-ask spreads as the hedging cost, and whereas the forward premia/discount is considered a yield (or return) differential.

So why bother hedging if it results in some additional volatility anyway between the true local return and the hedged index local return (even if the expected value of this difference will wash out over time, as mentioned in the Vanguard paper)? Because this volatility is far less than the volatility of an unhedged portfolio relative to the true local return. Take a look at the summary stats on standard deviation at the bottom of the first page of the MSCI fact sheet: The standard deviations of the hedged index and the local index are almost the same, but the standard deviation of the unhedged index is quite different (a lot higher). You can also eyeball the relative annual performance numbers of the 3 index series on the fact sheet to see that the hedged index is far closer to the local index than is the unhedged index.

HEDGEFUNDIE
Posts: 646
Joined: Sun Oct 22, 2017 2:06 pm

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by HEDGEFUNDIE » Mon Sep 17, 2018 8:57 am

Counterpoint wrote:
Mon Sep 17, 2018 7:55 am
HEDGEFUNDIE wrote:
Sun Sep 16, 2018 10:13 pm
Just out of curiosity, how much extra ER are you willing to pay for a hedged developed markets fund, and why that amount?
Estimates of hedging cost due to bid-ask spreads are made in this Vanguard piece: https://personal.vanguard.com/pdf/icrifi.pdf See pages 16-17 and in particular Figure 11 and its footnote. This was estimated several years ago (costs may have come down a bit since), and at that time it was a weighted average of about 13 bps for those currencies (which appear to account for about 85-90% of the portfolio of VTMGX/VEA, the ex-US developed market unhedged index). This 13 bps assume rolling 1-month forwards, so paying 12 bid-ask spreads in a year. As the footnote mentions, in practice 3 or 6-month forwards could be used for at least part of the portfolio which would reduce total costs due to bid-ask spreads, but would presumably increase tracking error relative to a benchmark that uses 1-month rolling forwards.

So I hope that if offered at cost by Vanguard, it would come in for less than the 13 bps mentioned in the research piece. This would be over and above the cost of an unhedged fund/ETF such as VTMGX/VEA which has an ER of 7 bps. So total ER under 20 bps, hopefully closer to 15 bps.
A good rule of thumb, thanks.
There is another aspect to hedging which is the forward premium/discount between the 2 currencies - this is referred to in jalbert’s post. It is the result of short-term interest rate differentials between the two currencies involved in the hedging. It’s useful to separate this aspect from the bid-ask spreads which is the true transaction cost of hedging. To understand this, take a look at the fact sheet for a developed market index hedged back to USD (it’s an MSCI index and VTMGX uses a FTSE index, but the principles are the same): https://www.msci.com/documents/10199/a8 ... c24481b783

The Annual Performance columns at the right of the top page have 3 sets of numbers: The first is the performance of the MSCI EAFE index hedged back to USD using 1-month rolling forwards, the second is the performance of the MSCI EAFE in purely local terms, and the third is the performance of an unhedged investment in the MSCI EAFE when measured in USD. To quote the fact sheet, “There are two components to a MSCI Hedged Index return: 1) the performance of the unhedged index in the home currency; and, 2) the Hedge Impact (aimed to represent the gain or loss on the Forward contracts) in the home currency.” In other words, the impact of the forward premium/discount referred to by jalbert is the difference between Column 1 and Column 2. That is separate from the transaction costs i.e. bid-ask spreads, which would be the tracking error between an actual hedged fund versus the hedged index, and this is what would be included in an ER and not the impact of the forward premia/discounts. That's why you see the Vanguard piece only referring to the bid-ask spreads as the hedging cost, and whereas the forward premia/discount is considered a yield (or return) differential.

So why bother hedging if it results in some additional volatility anyway between the true local return and the hedged index local return (even if the expected value of this difference will wash out over time, as mentioned in the Vanguard paper)? Because this volatility is far less than the volatility of an unhedged portfolio relative to the true local return. Take a look at the summary stats on standard deviation at the bottom of the first page of the MSCI fact sheet: The standard deviations of the hedged index and the local index are almost the same, but the standard deviation of the unhedged index is quite different (a lot higher). You can also eyeball the relative annual performance numbers of the 3 index series on the fact sheet to see that the hedged index is far closer to the local index than is the unhedged index.
The WisdomTree ETFs I mentioned above appear to reduce standard deviation by 1/3 (from 9% to 6%). It’s offered at a 0.43 ER, so twice as much as the hypothetical at-cost hedged fund.

But taking a step back, would I pay an extra 0.3% to reduce my volatility by a third? I think I would...

User avatar
vineviz
Posts: 1291
Joined: Tue May 15, 2018 1:55 pm

Re: Vanguard: Hedging international equities’ currency risk would have reduced portfolio volatility for US investors

Post by vineviz » Mon Sep 17, 2018 2:51 pm

One advantage of leaving international equities unhedged is that they will probably be better protection against domestic inflation.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

Post Reply