Monday, April 6, 2015

Using Leverage to Reduce Currency Risk?

Sounds weird that leverage could reduce any sort of risk. But in the context of currency risk when investing internationally, I believe you can.

At least that's my understanding. If I'm wrong, some one please correct me before I lose money.

First, an example of how currency risk hurts returns, then an example of how leverage mitigates that.

Hypothetical Assumptions:
Home currency is USD
begin EUR:USD:  1.10
end    EUR:USD:  0.88 (down 20%)
interest rate on local currency debt: 4%
simplistically, no taxes - should not matter in the example

Example 1 (Unhedged). How FX hurts returns. Say I invest in some European stock. I first convert USD to EUR, then I use the EUR to buy some stocks. Now say that stock then goes up 20% in local currency. Great, but EUR depreciated 20% against USD during that time. After converting back to USD, I end up with (1.2)*(0.8)= 0.96 of my original investment. I just lost 4% despite solid stock picking.

Example 2 (Hedged). But how if I borrow in local currency first? Say I want to buy 100 of this European stock. I'd borrow 100 first then use that to buy the stock. Again, stock goes up 20% in local currency, and EUR depreciates 20% against the USD.

My 100 of stock became 120. I use that to pay down 100 of local debt, along with 4 of interest on local currency debt for a net return of 16. Translate that back to USD, I'm net gain about ~$14 USD despite the 20% currency depreciation.  Net return is (20% - 4%) * 0.8 = 13%.

This makes sense because by borrowing in EUR first, I effectively hedged FX by shorting the euro against the dollar. Only the gain of 20 was unhedged. The interest expense here is my hedging cost.


Interactive Brokers let you do this and that's how I'm currently buying foreign stocks. I'm not sure how others are doing it. Is the hedging cost / interest expense worth it? I think so. After all when I buy stocks I'm looking for a lot more than 4%. On the other hand if local interest rate is 15% then this is probably not a good idea. The other consideration is my allocation for foreign stocks is limited by my own fear of leverage.

In my pre-crisis life of structuring asset backed securities, we had all sorts of ways to shift risk around. The total, absolute amount of risk does not change, but the idea is that each slice of risk went to their most appropriate owners - "better buyers", so that the collective utility for risk owners were increased. Here too, arguably I'm just shifting one form of risk into another (from FX to leverage risk). But some risk are more preferable than others depends on who you are and the situation. Here I prefer taking on leverage vs FX translation risk.

In hindsight - that whole MBS/ABS episode did not end up too well, and it's possible that this post is just one big brain fart. So again, if someone sees some false logic here please let me know.

*** updated 4/11/2015

Just got off the phone with a friend and I realized I have not made myself very clear. First, here's what we agree on. In example 2 in my original post, (for an USD based investor), the act of borrowing in EUR is equivalent to entering into an FX swap shorting EUR against the dollar. Like an FX swap, the idea is you locked into a particular exchange rate. But instead of paying whatever whatever hedging cost associated with an FX swap (bid/ask, transaction fees...etc), you pay an interest expense (conceptually equivalent to your hedge cost).

What this is NOT is hedging with an option, where you pay a cost and have asymmetric upside/downside. So in example 2 below, if EUR goes up the hedged positions will clearly underperform the unhedged approach. That would be true whether you use a currency swap or borrowing in a foreign currency.

The other thing is you have to be extra sure of your stock picks. The worst case is if stock pick AND currency both move against you. So if stock was down 20% and Euro was up 20%, then your net return would be (-20% return on stock - 4% interest expense) * 1.2 =  29%!.  The losses on your stock pick are magnified. That is because the principal amount of your loan (or think of FX swap notional) is hedged, but the subsequent gain/losses are not.

When attempted "hedges" are not hedges at all

That's why my friend argues this is not a hedge. I said even if you use a currency swap, you would have the same results. Then my friend says yes, but using a currency swap is not necessarily a hedge either because correlation complicates the matter.

So if you're investing in the stock of an European exporter, and euro was up 20% the stock would probably get killed. Then like the above paragraph, your losses would be magnified by a rising euro. In that scenario your "fundamental stock call" is really a macro bet shorting the euro, and by trying to "hedge the currency" (again via an FX swap or local currency loan), you're actually doing the opposite - you're shorting the euro more and doubling down on that directional FX bet.

In that situation he's definitely right. So always know what you're betting on exactly. If your single name stock idea has a macro component (and most companies have that to some extent), then be really careful about layering on "hedges".

No comments:

Post a Comment