Currency Hedging Foreign Investments

In Thinking About Your Portfolio we discussed framing your financial view in a way independent of a currency anchor.  A currency is a security like any other, except we use it as our default unit of value.  But as we discussed, this introduces distortion into our thinking because a currency is not inherently any more superior than other assets as a standard of value.  Our choice may be one of habit, or even government fiat, but it’s not one forced by the laws of economics.

We also cited some ways in which our investment decisions might be improved as a result of removing these distortions.  Let’s take a look at another one.

Investors are sometimes urged to hedge investments for currency risk. But in most cases it amounts to adding a currency risk, not cancelling out one, as might be implied by the term “hedge”. Looked at in our clearer frame of reference, we can see flaws in the line of thinking the strategy is based on.  It assumes that the local currency performance of that market is the “real” performance, and that if we invest on an unhedged basis we incur an additional risk due to currency translation.  But this isn’t true … if we own those stocks or bonds the only investment performance we will experience is that of the stocks or bonds.  If we hedge for “currency risk”, in fact we’re layering on an additional speculation on the relative performance of the currencies involved.

Why?  No currency is absolute.  The performance of foreign securities as measured in their local currency is no more real than the performance of those as measured in yours.  In the first case, the return being measured is the ratio of return of the foreign securities to that of the foreign currency. It’s a relative return, not an absolute one. In the second case, we’re looking at the return of the foreign securities relative your domestic currency. 

The return of those foreign securities is the same no matter unit what you use to measure it. If the numbers come out different measured in different currencies, it’s solely due to changes in the value of the currencies. Changing from one currency to another is a simple change of units, but the underlying quantity remains the same. Like the temperature outside being 10°C or 50°F … you have a different reading but the actual temperature is the same.

As an example let’s assume, say, a US investor buying Japanese stocks. You buy the Nikkei index. You find that the return you get is different than what’s reported for the Nikkei. But that’s because of the convention of reporting the Nikkei itself in Japanese yen. If you were to calculate the Nikkei in dollars, the Nikkei would be reported differently. You may be told you must “hedge” the currency to “protect” your returns. That’s a fallacy.

The “hedging” argument is based on the assumption that the yen-based calculation is the “real” performance of the Japanese stock market. And that your performance in another currency is somehow deficient. It’s not … it’s just measured using a different unit. The real performance of your collection of Japanese stocks is the same either way.

So suppose you “hedge” by an offsetting position in the dollar/yen. You can now match the yen-based performance of Japanese stocks in dollars. But this is not the same actual, real, return … what you’ve actually done is changed the return on your “Japanese stocks” investment by adding a currency trade.

In reality, you haven’t “hedged” anything … you’ve added a currency position to your stock position in order to make your dollar-based return look the same as the one reported in yen!

It’s as if you heated the temperature up to 50°C to make it look like the same as 50°F.

Now the result may (or may not) be better than if you hadn’t “hedged”. But if so, it’s because you added a profitable currency trade, not because your hedge has protected you against currency losses and allowed you to capture the “real” performance of Japanese stocks.

Although we’ve used stocks in this example, none of this is unique to stocks. Or to US investors buying overseas. The same logic would apply for example to a non-US investor buying US bonds.

12 thoughts on “Currency Hedging Foreign Investments

  1. Jk says:

    I’m sure “hedged”is a lot easier to sell, which is all to often the only thing that counts.

    assuming you don’t have a strong feeling about the currencies involved, the argument in favor of the hedged holding is that you live your life in your native currency. if you live in the u.s. you pay your rent or mortgage in dollars, you buy your groceries in dollars, and so on. if you decide to sell your investment in e.g. the nikkei you will want the proceeds in dollars.

    Otoh should “diversification”include holding a variety of currencies? it’s not usually defined that way.

    is the global equity fund you use currency hedged? or does it recalculate the conversions daily to publish it’s nav?

    1. Bill Terrell says:

      The native currency argument doesn’t hold up under scrutiny either. All a currency “hedge” does is give you the return of the foreign market relative to the foreign currency, denominated in your local currency. Why would that be inherently less risky than the return of the foreign market by itself?

      Especially considering that the whole point of buying assets outside of your own cash and equivalents is to have alternatives to your own currency. If for instance you wanted those returns, why would you buy stocks at all?

      And if as a US investor you sell your Japanese stock fund, you will receive dollars for it … “hedging” has no effect on that.

      VT is not currency hedged, nor are most international and global stock ETFs … really just those that are explicitly marketed that way. And you’re bang on about the marketing … adds a selling point and justifies higher fees.

      The whole thing is predicated on nothing more than a historical convention. The Nikkei you see quoted in the media is the return of Japanese stocks relative to the yen. It could just as easily have been defined as the return of Japanese stocks relative to dollars … and very well might have been, if it had been originally intended for a US audience. And that’s just what you get if you buy it unhedged.

      Which is the “real” return of Japanese stocks? Neither!

      So there’s no reason to “hedge” the trade unless you believe one currency will outperform the other. In which case you could just trade the currency pair. Why bother with the stocks … they have nothing to do with it.

      In sum, if you buy Japanese stocks unhedged you get the return of Japanese stocks. If you buy them hedged you get the return of Japanese stocks plus or minus your currency trade. Regardless of what they actually do, investors should not be under the impression it’s the other way around … there is no “currency risk” whatsoever in owning foreign stocks.

  2. Milton Kuo says:

    While I understand the reasoning why currency hedging of foreign stocks is not particularly necessary (especially if the stock is in a foreign company that does a lot of international business), I do not see hedging as meaningless or unimportant when buying a bond in a foreign currency.

    In the “Deja Vu” article, I had mentioned something about currency exchange risks for an Australian who is interested in buying U.S. Treasuries to participate in what Systems and various other indicators suggest is an impending bull market in Treasuries. In this scenario, the Australian is buying Treasuries to participate in the upward movement in price of the Treasuries as denominated in U.S. dollars and is not seeking to own a piece of a productive business. It is a trade, not an investment.

    Thus, there is a possibility, however small, that the Treasury price could increase as anticipated but the U.S. dollar falls noticeably relative to the Australian dollar. Such a scenario could ruin the trade. Generally speaking, the cost of hedging currencies is very small (single digit percentage) as compared against the anticipated price increase in Treasuries (double digit percentage).

    If a foreign investor is bullish on the U.S. dollar and bullish on Treasuries, then it probably makes sense to buy Treasuries without any currency hedge. However, being that the U.S. dollar index is any multi-year highs, it’s possible that a foreign investor may be quite bullish on Treasuries but is far less sanguine on the U.S. dollar. In such case, I really think hedging the U.S. dollar exposure makes a lot of sense and is not inconsistent with what the intent of the trade really is: capturing the movement in price of the Treasury but independent of the purchasing power of the U.S. dollar.

    1. Bill Terrell says:

      Well taken, Milton, but your case still boils down to making a currency bet. My point is simply that it’s a separate issue. I’m not saying making a currency trade is never appropriate, rather that it’s not made appropriate simply by virtue of trading foreign securities. Hedging is often sold on the premise of avoiding currency risk when trading foreign stocks or bonds. If you unpack it though, you’re not taking a position in a currency by in owning securities in a different asset class. Of course it’s true that you may benefit by adding a currency trade, but only if you’re right about the currencies.

      1. Milton Kuo says:

        >Well taken, Milton, but your case still boils down to making a currency bet.

        For an investor buying into a bond denominated in a foreign currency, isn’t he making a currency bet even if unknowingly? Unlike buying foreign equities which gives him a piece of a business that, hopefully, is increasing its profits and revenue and should at least keep up with the rate of inflation, a foreign bond is foreign money on foreign money. Its purchasing power is not directly backed by the productivity and profitability of a business.

        I am thinking of various hyperinflations in the twentieth century where bondholders were ruined but holders of equities in companies that survived managed to keep much of their purchasing power. That is the most extreme example I can think of that illustrates why it may be worthwhile to put a currency hedge on a foreign bond. An investment in a foreign bond is a very much a long term investment in a foreign currency.

        1. Bill Terrell says:

          Yes it is. But we may have left some original context behind. We’re assuming that all things considered, you’ve already decided the asset class is attractive. Then the question arises, do I need to separately hedge the currency? The answer is no.

          The issue arose in connection with the broad asset class outlook, including Synthetic Systems forecasts. The latter for instance show long term UST outperforming over the next few quarters. The question arose how it would be impacted from the point a foreign investor using a different coin of the realm.

          Not at all. Past or future, the relative performance of the asset classes is completely unaffected by what your unit of account is. If Treasuries outperform stocks in one currency, they will outperform by exactly the same amount in any other.

          Of course we could debate the investment merits of Treasuries just like any other asset class, but that’s a separate issue.

          1. Peter Fife says:

            “But we may have left some original context behind. We’re assuming that all things considered, you’ve already decided the asset class is attractive. Then the question arises, do I need to separately hedge the currency? The answer is no.”

            At this point, I have some serious concerns with the above statement. And, for me, it is not just an academic exercise, I have actually experienced the situation where I made a profit in USD, but if I had transferred the total USD, including the profit, back to AUD, I would have made a loss in AUD. Of course, it is possible I’m taking the above statement out of context and/or not understanding the whole argument correctly.

            However, I have created a spreadsheet which gives an example of a USD investment of 100k. It nominally returns 20%, USD 20k. If there is sufficient movement in the underlying exchange rate in the meantime, the profit of 20k plus the original 100k, may become a negative return when converted back to AUD.

            I will email the spreadsheet, showing what could happen. But, if the 100k is hedged, at the time as the purchase of 100k with AUD, and assuming a zero cost hedge for purposes of the argument, I believe the 20k, whatever the exchange rate is when I wish to convert it back to AUD, will be exactly the same, independent of the exchange rate movement.

            What I’m trying to say, is that with a hedge in place, if the 20k converts to AUD 10k, or AUD 30k, the hedge will ‘protect’ it and the original AUD amount required to purchase USD. Without the hedge, the cost of the initial amount of AUD to buy USD plus the 20k in profit, when converted back to AUD, may amount to a loss in AUD.

            I’ll email the spreadsheet separately. First, check that the numbers are correct, as I’ve never actually done a hedge before.

            And, please add spreadsheet, if correct.

            1. Bill Terrell says:

              Yes Peter I think you’re not getting the fundamental argument. In the example you cite, the AUD apparently appreciated against the USD. So obviously going long the AUD:USD would be profitable.

              But that’s not intrinsic to stocks or bonds or gold or any other asset.

              My point isn’t whether the currency trade is profitable or not – rather it’s that an investment in any of these other assets classes doesn’t intrinsically involve a currency position. The notion of “hedging” imagines that it does, and that adding a currency trade neutralizes this intrinsic currency position. That is false. Such a “hedge” adds a currency position to where there was none.

              You see my point? It’s not about whether the currency trade is profitable, it’s about whether it neutralizes an implicit currency position (it doesn’t) or adds a currency position to where there was none (it does).

  3. Peter Fife says:

    Hi Bill, Just out of interest, did you look at my spreadsheet? If yes, is there an error in the spreadsheet? BTW, just so you know, I have no great desire to put a “hedge” in place, but I know I would have lost money if I had transferred all my USD back to AUD, since the AUD had appreciated, even though I had made a profit in USD. Peter

    1. Bill Terrell says:

      Yes, Peter, thanks. My answer remains the same, because my point is not whether or not a currency trade is profitable … that depends on the circumstances … sometimes it will be and sometimes it won’t. Rather it’s about whether it neutralizes an implicit currency position (it doesn’t) or adds a currency position to where there was none (it does).

      Whether to make a currency trade at any given time is an investment decision, not a principle or fact. The gist of my argument is that making one on the premise of neutralizing the currency effect of a position in something that is not a currency is a fallacy; it’s not a “hedge”, it’s an active position.

  4. Peter Fife says:

    It appears on this point, we are going to have to agree to disagree. I’ll give it one last shot.

    Okay, so what about the following? I have 2 brokerage accounts in which I can trade the AUD USD futures. For the purposes of this hypothetical example, although actual values are the closing values on the respective dates, I have estimated what the mid-price is between the bid ask spread and ignored commissions, for the purposes of the discussion.

    Opening date is 4th July 2022, contract is September 2022.

    1. Brokerage account 1 sells 1 contract at 0.68905; short USD $68905
    2. Simultaneously, account 2 buys 1 contract at 0.68905; long USD $68905

    Closing date is 15th July 2022

    1. Account 1 closes 1 contract at 0.67900; long USD $67,900
    2. Account 1 net loss $1,005
    3. Account 2 closes 1 contract at 0.67900; short USD $67,900
    4. Account 2 net gain $1,005

    Overall position is a net gain/loss of $0.

    Account 2 is a hedge offsetting the position in account 1, surely?

    I fail to see how this does not act as a hedge for an Australian investor; said investor must convert AUD to USD to invest in USD based securities, and hence has exposure to the AUD USD exchange rate, which is not fixed.

    1. Bill Terrell says:

      To disagree would be premature, Peter, because you haven’t yet understood my position. My post was about buying foreign stocks or bonds. These are just currency trades. Of course you can structure them to cancel out.

      The details are irrelevant anyway. The issue is conceptual. If you were to buy say VOO (S&P 500) or VGLT (10-30 year UST), you convert your AUD into USD to make the trade. At that point you are no longer invested in currency, you’re invested in stocks or bonds.

      Then after some time interval has passed, you convert your ETF back into USD, and then convert your USD back into AUD.

      Now suppose you look up the return of the corresponding asset class, say the S&P. It says it was up 10% in USD. But also suppose during that same time interval the USD:AUD moved so that your realized return in AUD was 5%. This is your ostensibly “unhedged” trade.

      In an alternate scenario, you do the same thing, except pair the stock trade with a AUD/USD trade, so that it gains 5% in AUD. In this scenario, your combined realized AUD return is the same 10% that was reported for the S&P in USD. You conclude that you have successfully “hedged” your stock trade, preserving the 10% gain you earned after converting back into AUD.

      But I object that you haven’t established one crucial point. You have assumed that the 10% reported return of the S&P was the “real” return on your stock investment, while the 5% realized in AUD was not. In order to convince me that you have truly “hedged” to protect against a 5% haircut due to currency fluctuation, you must first prove your assumption that there was in fact one and only one “real” absolute return to protect in the first place.

      This hasn’t been done. Instead you’ve merely assumed that an absolute 10% return existed to protect. Looked at objectively all we really see is a 10% relative return in USD and a 5% relative return in AUD, neither of which is any more real and absolute than the other.

      In fact in your “hedged” scenario, your 10% return in AUD comes out to 15% in USD. Will the real return please stand up!

      Why am I belaboring the point? It strikes at the heart of a much bigger issue … a core tenet of this site. That currencies are merely securities issued by central banks, and that while out of habit and convention we pretend they are reliable units like feet, pounds and kilograms, they are certainly not … and this deception distorts our thinking and stands in the way of our being better investors. Currencies aren’t fundamentally different than any other securities … we could just as well pretend GM or Apple stock or bonds are reliable standards of value and use them to price every other security. I submit that in fact stocks have not risen anywhere near the reported figures … that the overwhelming majority of their historical gains are figments of using a shrinking unit of measure. For proof, all you need to do is try an alternative pricing unit. Merely substituting dollars with ounces of gold reveals for example the S&P to have been in a bear market since the turn of the century. So indeed taking the return of anything as absolute is an error … it’s all relative.

      The stock indexes – denominated in dollars, or any other currency – in no way represent the return of stocks, only the relative differential between stocks and dollars. The price of oil is not an accurate reflection of the value of oil … it is merely the ratio between the value of oil and the value of dollars. Getting our arms around this basic concept clarifies our thinking, helps us understand the economy better, and especially and helps us understand the financial markets better.