HOW TO HEDGE FOREIGN CURRENCY
A Canadian owning U.S. assets is exposed not only to the performance
risk of the asset, but also to exchange rate risk. You must make two
separate decisions - to buy the stock or not, and to hedge the FX or
not. The tax treatment of exchange rate gains and losses (for Canadian
tax) is covered by Interpretation Bulletin IT-95. You can find the exchange rates at the Bank of Canada site.
For some context, this page was written in 2004, after the Canadian Loonie bottomed at $0.65 and was in a clear rising trend along with all other currencies. At that time retail investors were asking repeatedly in the media "How can I hedge my US stocks?" Hedged ETFs had not been invented. With one uniform message the industry told them "Don't hedge". It was only after year after year of huge currency losses that hedged ETFs were invented, and (surprise, surprise) the industry changed their advice. In the intervening years they have never told retail investors 'how to hedge', other than with hedged ETFs.
Now in 2013, after the Loonie has created a top trading range, and then fallen some, the issue has left people's minds. The long list of excuses rehashed here are no longer heard in the media. The industry advice now is more like "Hedge when you think the Loonie will rise (with ETFs) and don't hedge when you think the Loonie will fall." In other words - play currencies at the same time you play stocks.
Does FX Exposure Reduce Risks ?
The financial industry would certainly like you to think that exposing yourself to currency exchange risk REDUCES overall risk. They tell you repeatedly that you have no need to hedge. But their reasons are garbage. Below are the arguments they use, with their counterarguments.
..............  1  .................
"Over the long term exchange rates have reverted to the mean." (or) "It all evens out over time." Garbage. Look at this 80-year chart of the Loonie. There is no cyclical repeating pattern.
When 15 big currencies are charted over 1900-2000 (ref:Optimists page 92) it is clear that currencies don't cycle. They
* make large one-time movements that never 'correct',
* move randomly in a range, or
* trend steadily in one direction for a generation or two.
Tell this line of bull to the Japanese or the Swiss. Any US dollar investment made in 1970, and held to 2009, by their citizens would have lost 74% and 76% of its value from exchange losses.
..............  2  .................
"Different countries' currencies move in different directions. Losses from one currency will be offset by gains in another." "Exchange losses over time are minimal if diversified."
Just the opposite. The 101 year chart of 16 currencies (ref:"Optimists" p.93) shows that only the Swiss and Netherlands gained against the US dollar and Loonie. Canadians and Americans would have lost from holding all the others. Their subsequent paper provides data for the 10 years to 2010, that show every single major currency gaining against the dollar.
|Annualized % Gain against US Dollar |
| ||1970-2009||2001-2010 ||1900-2000 |
|Swiss franc ||3.6 ||5.7 ||1.2 |
|Japanese yen ||3.4 ||3.5||-3.9 |
|Deutsche mark / Euro ||2.5 ||3.6||-3.0 |
|Danish krone ||0.9 ||3.6||-0.7 |
|Norwegian krone ||0.5 ||4.3|| |
|Canadian dollar ||0.1 ||4.2 ||-0.4 |
|Australian dollar ||-0.5 ||6.3||-1.5 |
|Swedish krona ||-0.8 ||3.4 ||-0.9 |
|British pound ||-1.0 ||0.5||-1.2 |
..............  3  .................
"If a currency depreciation or appreciation is purely due to a change in the relative purchasing power, the local price of equities should adjust in inverse proportion." Notice the big qualifying "if". The authors (MSCI paper) use this qualifying assumption to discount the effects of a FX loss. They assuming you get an offsetting higher equity return (and vice versa). Before you can evaluate this idea you must understand the relationships between inflation, interest rates and exchange rates.
The FX market is very large, but its players are short-term traders. The cash flows from their trades, and from foreign investments, determine FX rates in the short term. In the short term any investor should be prepared for 10% yearly losses or gains. The long-run annualized returns in the table above hide the yearly volatility. In the short-term a booming economy
* causes a high demand for capital, which
* causes inflation, which
* causes central banks to raise interest rates to combat inflation, which
* causes foreign investment inflows to buy that high-interest debt, which
* causes the currency to appreciate.
In the longer term the flows of money dry up. People start comparing purchasing-power-parity between countries. Inflation that originally drove up the currency has degraded the Purchasing Power Parity (PPP). Foreigners figure they can get more bang for their buck in another country. So the currency falls.
So inflation has the exact opposite effect in the short run as in the long run. Investors care about the short-run. They gravitate to countries with smoking-hot economies and highly profitable companies. Their money flows strengthen a currency. The relationship between exchange rates and equity returns is direct, not inverse.
..............  4  .................
"The local price of equities should adjust in inverse proportion to inflation. The local price of a stock will appreciate in a country with high inflation. So FX losses due to inflation will be offset by higher nominal returns." Maybe it 'should' theoretically but it doesn't in fact. Inflation causes investors to discount prices more. They demand a lower earnings' multiple because market yields are high. There is no correlation between inflation and equity returns. This was discussed on the Risks page.
..............  5  .................
"Diversification is good therefore diversification of currencies is good." This statement is heard from retail investors, not really from the industry. The argument misrepresents what diversification of investment securities accomplishes. Even while you presume all your investment securities will increase in value over time, not all will go up at the same time. Diversification of securities smoothes the volatility of your portfolio's value and makes sure you do not mis-pick only losers.
But no one expects all currencies' exchange rates to increase over time. Exchange rates vary in both directions. Exposing yourself to a foreign currency increases your risk, not decreases it. Increasing the number of currencies you are exposed to increases the point-sources of your FX risk, but it does not reduce your risk. You get rid of FX risk by staying invested in your home country or hedging. Even the MSCI paper referenced above had to conclude that hedging FX reduced the portfolio's volatility.
..............  6  .................
"You can protect yourself from home-country inflation by diversifying in other currencies. Inflation at home will devalue the purchasing power of your own currency, so holding stronger foreign currencies will protect you." This won't work because PPP works in the long term, but not the short term, as discussed above. Even if it were working, as an investor you will gravitate to countries with smoking-hot economies - with their own inflation.
"Inflation at home may be caused by the increasing cost of imported good resulting from the importing country's weakening currency. Holding the foreign currency will hedge that inflation." This second argument concerns inflation that is imported, as opposed to being caused by home-grown demand for capital. But not all (probably little) inflation is caused by FX rates. For Americans in particular, their buying power is so great they can dictate the currency in which deals are done. In mid 2000's foreign suppliers found that purchase orders received from the US were changed to be denominated in US dollars - because the US dollar was falling and Americans wanted suppliers to accept the FX risk. Suppliers were told to suck it up, and they did.
..............  7  .................
Hedging does not systematically improve (or lower) equity returns. Well duh! What the authors mean by 'systematic' is 'averaged across all the different currencies and across time. But of course for every currency gaining strength there is the opposite losing value. On average they cancel each other out. You improve your returns by hedging when the currency of your investment weakens. And by NOT hedging when the currency of your investment strengthens.
..............  8  .................
"What should matter most for international investors are the real (net of the foreign country's inflation) currency returns and not nominal returns." This is the excuse used in most academic papers, whether stated so blatantly or not. Common sense should tell you only the actual return and the FX change are relevant (the top row below). Inflation in the foreign country is irrelevant for your purchasing power, or for deciding whether to hedge currency. Even home-country inflation is irrelevant because all assets owned (and wealth generally) are equally hit.
Here is how they fudge the math to get the conclusion they want (that FX changes are irrelevant but foreign inflation is). Each row adds across and each column adds down.
|Foreign Stock Return ||+/- ||FX Δ ||= Canadian Currency Return |
|less Foreign Inflation ||+|| Foreign Inflation |
less Canadian Inflation
|= less Canadian Inflation |
|Foreign Stock REAL Return ||+/-|| REAL FX Δ ||= Canadian Currency REAL Return |
Their top line starts with the correct, useful, and intuitive calculation of returns for a Canadian investor measured in Loonies. Then inflation is factored in to both returns and FX changes. The bottom line shows the foreign return (adjusted for the foreign inflation) , plus/minus the change in FX (adjusted for the difference in inflation between countries), equals a Canadian's real return. Nobody disputes any of that math.
Nobody disputes the measurement of the bottom-right box, derived by the right column ... if there is some need to measure real returns. But they say a Canadian should look at the bottom-left box instead ... because the bottom-middle box (REAL FX Δ) is essentially = 0.
In the very long-run, it is inflation (PPP) that determines changes to FX. So the change in FX in the top line will equal and offset the difference in inflation of the middle line, resulting in net 0, over the very long run.
Dimson, et al (page 6 and Figure 3) conclude from this that "The gap between the two return measures (bottom-right and left) is for every country under 1% per annum." and "Currency risk has generally added only modestly to the US dollar risk of foreign investments." But those conclusions are only valid when measuring century long returns. FX is not determined by differential inflation in the short or medium time-frames of an investor. Heck, the most common FX trading strategy depends on the disconnect between the two. The bottom-middle box will not = 0.
What matters to the investor is the foreign return, the change in FX rates, and (if a real return measure is required) the home country's inflation.
Excuses for Not Hedging
- "Retail Investors cannot hedge." This is presented
by the industry that WILL not let you hedge. There is a difference
between 'cannot' and 'will not'. The industry denies everyone (except
'qualified investors' who are worth $1,000,000 with income of $100,000)
access to the futures market, where a year's hedge can be bought for
$3. Your response should be to lie to the brokerages. Claim you are
'qualified' and use these restricted products yourself. (See definition)
- "I don't know how"
has been offered up on national TV by portfolio managers with seven-digit salaries. Bald-faced lies.
- "Hedging involves speculating on exchange rates."
This qualifies as being 'unclear on the concept'. When your position is
hedged, by definition, all exchange movements will have NO impact
whatsoever. Derivative products can be used to EITHER increase risk
(leverage) OR to decrease risk (hedge). You are speculating on FX when
you are NOT hedged.
- "Futures contracts are highly leveraged, so using them to hedge is very risky."
First consider whether your existing exposure to FX is leveraged. When
a Canadian spends $100,000 buying US stocks, how much money is invested
in the stocks, and how much in the US currency? In fact ALL your
investment went to buy the stock. Your resulting exposure to the US$ is
100% leveraged. Putting a hedge in place offsets your leverage; it neutralizes your leverage.
- "I made money on the falling loonie so I don't mind losing money as it strengthens."
These people don't understand that investing is about making money AND KEEPING IT.
- "I buy US positions in my Cdn $ account"; or "I buy Cdn interlisted companies"; or "I use ETFs trading on the Cdn exchange."
The error here is mistaking the appearance for reality. It is the
business' operations (purchases/sales and assets/liabilities) that
create the exchange risk. It does not go away when the company reports
in another currency, or lists on another exchange, or when the investor
uses another currency to purchase the shares.
- "It is too expensive to hedge."
The cost to buy a $100,000 futures contract good for a year, is about $5. Pittance. Yes you must provide collateral but that is not a cost. You will also find that buried somewhere in whatever method you chose, is a cost/benefit equal to the difference in interest rates between the countries. Since Canada and US rates are so close you can cancel out any cost by using limit orders that let normal market volatility make up the difference.
The fund managers who say hedging is expensive are referring to using options. This method IS expensive. Options are a one-sided bet on the direction of FX, expiring within a specific time span. They are not hedges.
The cash funding of open futures contracts come from the daily settlements. If your position gains $1,000 in value one day, there will be $1,000 put into your account. More importantly, if your position loses $1,000, it will be taken from your account. If there is not sufficient cash your broker will consider it borrowed and charge you interest. You must realize that because this is a hedge, you are not 'losing' that cash. For every dollar you might lose in the futures account, the offsetting investment in the foreign security will have gained the same amount. Agreed, you cannot liquidate that cash daily, but the value is there.
For Canadians choosing between passive ETFs of the S&P500, hedged vs not (SPY), the hedged products underperform yearly, often by many percentage points. This is quite a high cost, and correctly causes people to pause. There really is no excuse for this underperformance because their hedging with futures contracts simply does not cost that much. The reason for the underperformance is their pre-set rules for action. No doubt they are ignoring FX trends, instead reacting after the fact to every change ... that promptly reverses.
- "But what if the Loonie goes back down?"
This is the same mind space of investors who refused to sell as
Nortel's stock fell from its highs. They had already suffered so much
pain that they could not abide the thought of selling, only to see the
stock go back up. Investors
must realize that past losses do NOT make future gains more likely. The
past is done and gone. Only the future should be considered.
Why Not Hedge ?
- Many Canadians plan on retiring to the US - spending their wealth in US dollars. These folks should not hedge their US exposure. But then they should hedge their Loonie exposure, and track their wealth in USD. Snowbirds going south for only half each year would want exposures to both currencies.
- If you are pretty sure the exchange rate will not change more
than a percent or two, it is reasonable to save yourself the hassle.
- If you think the rate change will work to your benefit, AND
YOU HONESTLY ACKNOWLEDGE YOU ARE SPECULATING, you would feel that
hedging is an unnecessary restriction on your profits.
- If the only methods of hedging available to you were more costly than your expected FX loss, you would not hedge.
- Before making this decision use this FX charting service (Pacific Exchange Rate Service) to get a clear idea of short and long term exchange movements.
HOW? ... To Hedge a U.S.$ Asset Create a U.S.$ Liability
- Borrow US dollars from your broker. Canadian banks will not lend you US
dollars. You should complain. The Government borrows US dollars. The
bank themselves borrow US dollars. Our businesses can borrow US
dollars. But retail Canadians can't.
In 2003 US margin debt from your broker was dirt-cheap, and again in 2009 some brokers charge less than 1%. This method is preferable to buying Futures contracts when there is risk that the Loonie may fall in value for any extended period in which you would have to fund the loss from a futures contract. But when US borrowing rates rise, this strategy ends.
The proceeds from creating the debt must be invested in Canadian assets. You can buy the Loonie currency ETF (e.g. N-FXC) when you are less sure of a leveraged play. Or you can create a 'carry trade' by
- buying Canadian companies dual listed on US exchanges.
- buying a Canadian market index ETF from a US exchange that Americans would normally buy.
- repatriating the cash into Loonies in order to buy a wider variety of Canadian assets. (You will be charged FX exchange fees (about 1.5%) each way.)
- If you own real estate in the US, take out a US dollar
mortgage there. Move the money into Loonies to pay
off Canadian debt or invest.
- If you own US index ETFs (e.g. Spyder (SPY) consider instead buying the Future Contract of that index instead. At the end of
the contract you would have to pay for it in US dollars. In other words
you have created a US liability. Futures contract cost next to nothing to purchase. If they gain in value, the profit $$ will be deposited in your account. No one objects to that. But the reverse is also true. If the contract loses value, the $$ loss will be withdrawn, daily, from your account. This is much the same process as most brokers use for short selling stocks.
The future's market price starts with the current index value, reduces it for the dividends you will not receive, and increases it for the bank interest you will earn by keeping your cash until the contract's end. A Canadian keeping his $$ in Loonies will earn a different interest rate, so the cost of the hedge includes the interest rate differential between the countries. It has been only a very small percentage for a long time.
- It is more simple for many people to buy the hedged ETF products
(S&P, EAFE) available on the Toronto Exchange. The hedging may not work perfectly though because they use a rules-based approach.
- If you own individual US stocks, buy a Futures
contract on the Canadian Dollar. As with Index Futures above, this creates a US dollar
liability. These are valued at CAN$100,000 so you may have to share
with a friend. The gains/losses on these contracts are treated as
'capital' for Canadian tax purposes - only 1/2 taxable (IT-346R). As discussed above, the contracts are essentially free, and the cash value of gain/losses with be credited/debited to/from your account daily.
There are also mini-contracts worth only US$10,000 that make this strategy more accessible to small investors. There may be much less trading on these contracts, so watch the pricing. Notice that these contracts are reversed from the large contracts. These are "to buy US dollars and to be paid for in Loonies at the end of the contract". So to create a US dollar liability you would SHORT this contract.
- Contracts-for-Difference (CFD) have a very bad reputation. However you should know that there is a brokerage, CMC Markets,
that allows you to create FX positions which will hedge the currency
exposure of your stocks. They charge no commission on FX trades, and
collect their income from the bid/ask spreads, and the interest earned
on long overnight positions and the collateral. The spreads are quite
small and you can avoid the interest charged on overnight positions by
picking the contracts that allow you to 'short'.
While the collateral
required for normal FX futures contracts can be invested in productive
assets, here it is cash. Your 'cost' is the income forgone on that
collateral. Even though their stated requirement is only 1% collateral,
you must also fund the changes in value. So in order to maintain the
hedge through price volatility, your collateral must be more like 5%. This is not really a good option.
- There are now ETF products that invest in specific
currencies (list). Not all uses of these create good currency hedges.
If you buy the ETFs you are using some of your principal. Hedges only work if they cost little or nothing - so they have no opportunity cost. E.g. a Canadian using $US to buy the Loonie ETF (N-FXC) with US dollars, when he would normally have bought US stocks with the cash, has created a hedge but with a cost equal to the profits lost by NOT investing in the US stocks. However, buying the Loonie ETF on US dollar margin - without foregoing any stock purchases - creates a hedge with a cost only equal to the interest cost of the margin debt.
Shorting these ETFs requires no principal. The cost to you equals the sum of (i) the normal transactions costs to buy/sell a stock, (ii) the stock's dividend payments that you must make good to the lender, and (iii) the cost to borrow. Canadians can short the USD by shorting the Canadian issued T-DLR (not the DLR-U). However one broker who does not charge to borrow always sais 'the stock is not available' and another broker charges 6% yearly to borrow from the only one person willing to lend. Americans can short the Loonie by shorting the American-issued ETF (N-FXC).
These ETF products work well to fine tune large futures contracts. Say a Canadian's US portfolio was worth US$100,000 when he put a US$100,000 hedge in place using a future's contract. If the stocks decline in value 10%, he will be over-hedged by $10,000. Since the futures contracts are for large denominations he cannot fine tune them. Instead, he short sells US$10,000 of the Loonie ETF. It has the effect of depositing US dollars into his account; thereby adjusting his US dollars assets back up to $100,000 - equal to the futures contract hedge.
However, if a Canadian owns Euro denominated assets, he may want to short the American-issued Euro ETF. The problem is that this deposits US dollars into your broker account. So while the Euro gets hedged, you now have US dollars NOT hedged. You could withdraw the proceeds from the short sale and repatriate them to Loonies, but then your broker would charge interest on the money withdrawn and the FX swap will have its own cost.
an Option on the Can$ will somewhat hedge your US exposure. Options
cost about 1% a month, and you have to correctly predict both the time
frame and rate movements to have it pay off. It will never be a perfect
hedge. It will not increase/decrease $1 for every $1 your investments
decrease/increase in value. Rather, it will be 'negatively correlated'
to the FX gains/losses you are trying to hedge. It will tend to move
(in value) in the opposite direction, but will never offset 1:1.
You can also buy options on individual stocks. If you buy the issue
far into the money, you will pay only the intrinsic value (say 15%).
The rest will be hedged if you keep the remaining money in your home
currency. See the discussion on Leverage in Options
Contracts are a better variant to the Future Contract, but are not
available to us. The company you work for may have access, so make nice to the
comptroller, and ask. Know what you are talking about first though.
There is a very good article on the difference somewhere on the CME
Where Do I Look For Currency Problems ?
- The registered or head office location of a company can be Canada, US or elsewhere. The currency of these locations will not cause FX gains/losses.
- The stock exchange used to list the shares will not
impact FX. Buying Alcan on a US exchange will not expose you to $US FX
risk. Buying Saputo on a Canadian exchange will not save you from FX
risk from its US operations.
- A company can report their financials in any currency they like without affecting the reality of its FX gains/losses experienced. But the currency used will determine the reported
measure of the gain/loss. It is no surprise that many Canadian
companies with US operations changed their reporting to US$ in
2003-2004 to hide the devastating effect of the 30% FX loss.
- Buying foreign (non US) index ETFs and ADRs listed in the US will not
expose the investor to $US-Loonie FX risk. The US dollar is just a
convenient intermediator. Think of it like a Turk and a Mexican talking
to each other in English. The investor is exposed to
the FX risk of the specific country vs Loonie, if that is where their
operations are. But most of these indexes and ADR companies will be
very large internationals with 50% of operations in the US. To that
extent, owning them will expose you to $US-Loonie FX risk.
Just to make things complicated: If what you want is to earn (as a
Canadian in Loonies) the same return as the posted return (in $US) for
the (e.g. European) index, then you must hedge the $US-Loonie FX. This
will still leave you exposed to the $US-Euro, but so is the index.
Similarly with an ADR. If you hedge the $US-Loonie then your return is
the same as the ADR's, but not the return of the stock on its home
- A company's operations and assets are where to look for FX exposure problems. The currencies of these countries will
give rise to FX effects. E.g. SinoForest (T-TRE) is listed in Canada
and reports in $US, but it is the Chinese currency of its operations
that will cause the currency gains/losses against the Loonie.
Buy U.S. Stocks Now While The Dollar Is On Sale ... FALSE
By 2010, the Loonie had increased 50% in value against the dollar since 2003. So the Loonie buys more US shares than before. But that does not mean you get more value than before.
When you buy a consumable whose use is
measured in pleasure, the stronger Loonie allows you to buy "more"
enjoyment than in the past. E.g. you can buy a longer vacation in
Florida. It costs less now to get the same amount of pleasure. It is
correct: Now is the time to buy "with the dollar on sale".
But this is not the way to approach the purchase of investments.
With investments, you are not so interested in buying the 'asset', as you are
interested in buying 'income' or 'capital gains'. Those profits are denominated in the same
foreign currency. A stronger Loonie lets you buy more shares of US
companies, but the larger profit $$ shrink proportionally when
repatriated into Loonies. The shrinkage completely offsets the
increased number of shares.
At all exchange rates your 'rate of return' measure in Loonies will equal the rate of return earned by Americans measured in USdollars. No matter
how strong the Loonie was, the value of the profits in Loonies, will be
discounted by exactly the same rate that the cost of the assets was
The exchange rate CHANGES of the past created currency gains/losses - also in the
past. They have not made investments 'cheaper'. The concept of buying
stocks 'cheap' implies a higher rate of return. E.g. buying stocks with
a low P/E results in a larger earnings yield (E/P). But the
strengthening Loonie has impacted both the cost AND the profits of
investments equally. The 'rate of return' remains the same.
This argument is different from the argument, often made at the same
time, that now is the time to buy US dollars because "exchange rates
are cyclical and the Loonie is bound to retrace its gains".
How Do I Understand the FX Reported in Financial Statements ?
First consider revenue and expenses. These are not
a problem. The operating results from another country are translated
into the reporting currency at the average FX rate over the period.
Granted, the investor is forward looking, and would like them
translated at the most recent FX rate, but the average rate is disclosed in the notes, and he can make his own assumptions about the future. Most businesses consider it good management to match revenues from a
country with costs from that same country, so that the FX risk is
limited to the profit margin. It should be obvious to the investor when
this is NOT the case: such as mining or oil operations.
The problem lies in the translation of the value of foreign assets and debts. The FX translation effects are reported in three ways.
- Sometimes the change in FX value will be in the Income Statement, sometimes individually disclosed, sometimes not.
- Sometimes the change in FX bypasses the Income Statement and gets
posted directly to the Balance Sheet's Equity, in the account called
"Cumulative Translation". You can measure this by subtracting the
difference between the opening and closing values.
- Sometimes the change in FX is not reported or measured anywhere -
until the asset is eventually sold. There is nothing you can do about
this because you will not have the necessary information.
Management often say they have hedged their foreign assets with
offsetting foreign debt. Even where this is true, the reporting of the
FX effects may show up in different places. There is nothing an investor can do except measure what is available to be measured, and make sure it is all included in comprehensive earnings.
There are two arguments made for NOT including FX changes in the Income Statement.
- When the financials are used to evaluate management it is not appropriate to make management responsible for investors who live around the world. But for the investor the issue of evaluating operations managers is minor.
- When the foreign operations are considered permanent, with no expected repatriation. Why worry about a
write down that may never occur? But the presumption that
there will be no repatriation is not valid. Expansions into
specific countries often flop, with the assets withdrawn. Divisions are
often sold, with the proceeds redirected around the world. The US tax
amnesty of 2004 prompted millions of dollars to be repatriated from
abroad. Even that condo in Florida will probably be sold to fund
retirement home costs in Canada.
Their arguments are not consistent.
- Everyone agrees that all debts should be revalued to the current exchange rate, so why not all assets? Not revaluing assets reflects only the accountants' attachment to historical cost accounting.
- People that agrees with the revaluation of some Balance Sheet items still object to recognizing the change in value as Income. But Income is defined as 'the change in value'. You cannot both accept the Balance Sheet change AND reject its recognition as Income.
Investors should be able to see both the operating
results and the FX translation effects separately, so they can project the future,
separately, of both the operations and exchange rates. Unfortunately,
the reported information is not enough to make these two evaluations.