"Currency adds volatility - and that can be costly." - Sales Pitch
One of the main selling points for hedging currency exposure in foreign equities is a reduction in volatility.
The argument typically goes as follows: as a U.S. investor, why take on additional currency volatility if you don't have to? Simply hedge the foreign currency exposure, leaving you with just the equities, a lower volatility exposure.
On the face of it, this seems to make perfect sense. Currencies have volatility, and so adding them to the risk equation would appear to make an investment more volatile (equity volatility + currency volatility = more volatility). But is this actually the case? Let's take a look…
The oldest and largest currency-hedged ETF is the WisdomTree Japan Hedged Equity ETF (NYSEARCA:DXJ), with assets under management of over $5 billion. Since its inception in June 2006, DXJ has an annualized volatility of 22.9% versus 21.6% for EWJ (the unhedged iShares MSCI Japan ETF).
Note: calculated using daily total returns. Data source for all charts/tables herein: YCharts.
Looking at rolling 1-year periods, we find that DXJ had higher volatility than EWJ 59% of the time.
A second Japan hedged equity ETF (NYSEARCA:DBJP) was launched in June 2011. Since then, it has an annualized volatility of 21.4% versus 17.1% for EWJ, with higher volatility in 94% of rolling 1-year periods.
What gives? How can currency-hedged ETFs exhibit more volatility than their unhedged counterparts?
The volatility of an instrument depends on its variation (deviation from the mean). Including currencies in a foreign equity holding could increase variation and make the path more volatile but it could also decrease variation and make the path less volatile. The latter scenario is precisely what we've seen since the start of these two currency-hedged ETFs.
I realize this is not intuitive, so let's run through an example. You are bullish on Japanese equities and bearish on the Japanese yen, the perfect combination for a hedged Japan exposure. Over the next five days, Japanese equities are up every day (+0.5%, +2%, +1.25%, +0.75%, +1%) while the yen is down every day (-0.2%, -0.3%, -0.4%, -0.25%, -0.5%). The hedged exposure is a success, returning 7.35% versus 5.62% for the unhedged exposure.
But what about volatility? Was it lower?
No. As you can see in the table below, hedging the currency, in this case, increased the volatility from 9.14% to 9.81%. As the variation of the daily gains increased by implementing the hedge, the volatility increased as well.
Now let's say the yen rallied every day instead of declined. This time around the unhedged exposure outperforms the hedge (5.62% to 3.90%). It was better not to hedge but was it also less volatile? No. In this case, the variation of returns is lower after implementing the hedge.
These are just 2 examples. There are an endless array of possibilities when it comes to foreign equities and currencies, and no simple rule that says a currency-hedged exposure has to be less volatile than the alternative.
So, does this mean one should not hedge foreign currency risk?
No, just that one should not do so with the expectation that it will necessarily reduce volatility. As we have seen, you could very well experience just the opposite, and for long periods of time. This can be true even if the hedge appears to be working (hedged exposure has a higher return).
Are there other reasons besides volatility to hedge?
Yes, if you do not believe in the benefits of currency diversification and/or expect the foreign currencies to significantly underperform the U.S. dollar over time. But the direction of currencies can be quite hard to predict, which is why proponents of currency hedging are often saying that by not hedging you are essentially "making a bet" on their direction. There is truth in that statement, but judging by the timing of asset flows in the currency-hedged ETFs, buyers of these products are also making a bet on direction.
In the chart below, you'll note that the bet they are making tends to follow the recent direction of the currency. When the Japanese yen was in a freefall in early 2013, assets in DXJ went from $1 billion to over $12 billion. Investors were betting that the yen would continue lower. During the next leg, down for the yen from summer of 2014 to its low a year later assets grew to a peak of over $18 billion. Again, the bet was a lower yen. From there, the yen rallied sharply over the next year and assets in DXJ plummeted to less than $7 billion. This time, sellers were betting on a stronger yen.
Clearly, the purchasers of this product were "making a bet," whether they acknowledged it or not.
To hedge or not to hedge?
Perhaps the decision to hedge or not to hedge should rest on understanding your own behavior:
- If you are more likely to stick with a currency-hedged exposure because you do not believe in currency diversification or you believe foreign currencies are in a secular decline, that may be preferable.
- If you are more likely to stick with an unhedged exposure because you value currency diversification or believe the dollar will ultimately weaken, that may be preferable.
- If you are more likely to stick with a partially hedged exposure (yes, there's an ETF for that) because that will prevent you from FOMO-induced chasing, that may be preferable.
The best strategy is the one you can stick with long enough to reap the benefits of compounding.