How to Trade EMFX: Carry, Volatility, and What Actually Works

Chapter 4 is where this book gets really practical. The authors stop talking about what drives EM and start talking about how to trade it. And they begin with the most famous strategy in FX: the carry trade.

But before we get there, they ask a question that sounds simple but actually isn’t. What even counts as an emerging market currency?

The Line Between EM and G10 Is Blurry

You’d think the distinction between EM and G10 currencies would be obvious. Rich countries on one side, developing countries on the other. But when you look at actual trading behavior, the line almost disappears.

The authors run correlations between a bunch of currencies and the usual macro drivers: US rates, the S&P 500, US high yield credit, the VIX, and commodities. What they find is that most G10 currencies and most EM currencies react to these drivers in pretty similar ways. The Norwegian krone doesn’t behave all that differently from the South African rand when it comes to responding to global risk. Both sell off when things get scary. Both rally when things calm down.

The one currency that truly stands out? The Japanese yen. It’s the only currency that consistently goes up when risk goes up. When stocks fall, when the VIX spikes, when credit spreads blow out, the yen rallies. No other currency does this reliably. The euro comes kind of close, but the effect is much weaker.

On the EM side, some currencies act more like G10 than you’d expect. The Czech koruna barely moves with global risk. The Chinese yuan is so managed it basically ignores everything. And if you compare a basket of G10 commodity currencies (like AUD, NOK, NZD, CAD) against a basket of EM commodity currencies (like ZAR, BRL, RUB, COP), their betas to global macro drivers are shockingly similar. Sometimes the EM basket actually has lower betas.

The authors’ conclusion: it makes more sense to sort currencies by commodity producer vs. commodity consumer than by EM vs. G10.

So What Actually Makes EMFX Different?

If the macro correlations are similar, what’s the real difference? Two things.

First, institutions are weaker. On most trading days, this doesn’t matter. But during elections or political crises, it matters a lot. A market-unfriendly candidate winning in Brazil can cause way more damage to the currency than something similar happening in, say, Canada. In emerging markets, individuals can overpower institutions more easily. That means capital controls, surprise policy changes, and other things that just don’t happen in developed markets.

Second, interest rates work differently. This is the big one. In G10 FX, when a central bank hikes rates, the currency usually strengthens. Higher rates attract capital, currency goes up. Simple.

In EMFX, this relationship breaks down completely. When EM central banks hike rates, it’s often because the currency is already falling. Rates go up because things are bad, not because the economy is strong. The authors test this with a thought experiment: imagine you have perfect foresight on interest rate changes. In G10, if you know rates are going to rise, you can reliably predict the currency will strengthen. The P&L of this perfect-foresight strategy goes up in a straight line for G10 currencies.

For high-volatility EMFX? The P&L goes straight down. Even for low-volatility EMFX, it loses money. Knowing which way rates are moving doesn’t help you trade EM currencies. That’s a fundamentally different world.

This is actually the authors’ preferred way to define what counts as “true” EM. If rate changes help predict FX moves, it quacks like G10. If they don’t, it’s EM. Using this definition, most Latin American and CEEMEA currencies are true EM. But most Asian currencies actually behave like G10. The ones that don’t follow the pattern are Taiwan (TWD) and Singapore (SGD), which is surprising since they’re not the usual high-yield suspects.

The Carry Trade: It Works If You’re Japanese

OK, now we get to carry. The idea is dead simple. Borrow in a low-interest-rate currency. Buy a high-interest-rate currency. Collect the difference. In good times, this is basically free money because high-yield currencies tend to stay stable or even appreciate.

The authors look at what happens when you buy the top four highest-yielding EM currencies. Against the USD, against the EUR, and against the JPY.

The results are sobering if you’re a dollar-based investor. The USD-funded carry trade peaked in 2011 and basically went nowhere for years after that. The 2011 peak happened right when the broad USD bull market began, which makes sense. When the dollar is getting stronger, buying EM currencies against the dollar is swimming against the current.

EUR-funded carry does better. The cumulative returns from 2003 to 2018 are about the same as USD-funded carry, but the Sharpe ratio is much higher. Less pain for similar gain.

JPY-funded carry is the clear winner. Significantly higher cumulative returns than either USD or EUR funding. If you’re a Japanese investor doing the carry trade, the last two decades have been pretty good to you. This explains why the carry trade is so popular in Japan.

The authors’ suggestion: even if you’re not based in Japan, think about diversifying your funding currency. Don’t just automatically fund carry trades in USD.

Volatility Adjustments: A Small But Real Improvement

One tweak that consistently helps: adjusting for volatility. Instead of just ranking currencies by how much carry they offer, rank them by carry divided by volatility. Then size your positions based on how volatile each currency is.

The idea is straightforward. A currency offering 8% carry with 20% volatility is less attractive than one offering 6% carry with 8% volatility. The second one gives you more carry per unit of risk.

The numbers back this up. A volatility-adjusted long carry strategy had an information ratio (IR) of 1.16, compared to 0.83 for the unadjusted version. That’s a meaningful improvement. The returns are about the same (8.6% vs. 8.7%), but the volatility drops from 10.6% to 7.4%.

This is a theme throughout the chapter: volatility adjustment helps almost everything. Carry, momentum, other strategies. If you’re trading EMFX and not adjusting for vol, you’re leaving money on the table.

Why Simple Carry Broke Down

The obvious explanation for why carry stopped working would be that carry itself fell. Lower global rates mean less carry to collect. And yes, carry was higher in the early 2000s when the strategy was printing money. But the authors show it’s not the full story.

Since 2011, there have been periods where carry was actually at similar levels to 2007 or 2010, times when the strategy worked fine. The carry was there. The returns weren’t.

Their conclusion is that we probably need carry to rise back to the levels of 2004 for the simple strategy to work again. But that’s hard to imagine in a world of low global interest rates. So the simple carry trade, at least funded in USD, is not what it used to be.

Long-Short Carry: Expensive But Improvable

One attempt to fix the carry trade: go long the top yielders and short the bottom yielders. This strips out the USD directionality. You’re not betting on the dollar anymore, just on the spread between high and low carry within EM.

The bad news: long-short carry has lower returns than long-only carry. The short leg costs you. You’re shorting currencies with low carry, which means you’re paying carry on those shorts. The cumulative performance is worse than just buying the high yielders outright.

The better news: volatility-adjusted long-short carry actually made new highs after 2011, peaking in 2016. So it’s not dead. It’s just not as good as you’d want.

Here are the numbers that tell the whole story:

All strategies beat the EMFX benchmark. Vol-adjusted versions are consistently better. But none of these are home runs anymore.

Finding the Best Hedges

Even though the short leg is expensive, there’s a useful concept buried in it. Some currencies are volatile, highly correlated with global risk, and offer low or even negative carry. These are your best hedges.

At the end of 2018, the three most interesting hedges were the Hungarian forint (HUF) and Swedish krona (SEK), both with betas greater than one and negative carry. The euro and Czech koruna could round out a hedge basket to reduce single-country risk.

How did this play out? In 2019, those four currencies weakened by an average of 3.4% in spot terms. The EMFX index only lost 0.4% on a vol-adjusted basis. So the hedges did their job beautifully, and you got paid carry to hold them.

The authors show that systematically finding the cheapest hedge, picking the three currencies with the lowest cost after adjusting for their beta to EMFX, and using them to hedge a long EMFX position works decently over time. It won’t make you rich, but it stops you from losing during the bad times.

EMFX Is Afraid of Its Own Shadow

Here’s one of the more colorful findings. The authors build a risk index using the maximum z-score of implied volatilities across five asset classes: EMFX, G10 FX, US rates, the S&P 500, and oil. When any one of these volatilities spikes above two standard deviations, you cut your EMFX exposure. When it falls back below two, you get back in.

Why the maximum instead of the average? Because EMFX reacts badly even if only one thing starts going wrong. It doesn’t matter if equity vol is calm, rates vol is calm, and FX vol is calm. If oil vol goes crazy, EMFX suffers. The risk can come from anywhere.

This risk overlay significantly improves returns compared to just running a vol-adjusted carry strategy. The title of the chart in the book is “EMFX: Afraid of Its Own Shadow,” which is perfect. High-carry EM currencies are inherently risky. They know it. The market knows it. And when any kind of volatility shows up anywhere, these currencies run for the hills.

Playing It Safe Doesn’t Pay

After all this talk about carry being risky, you might think: why not just buy the safe EM currencies? The ones with current account surpluses. The investment-grade ones. The ones that won’t blow up on you.

The authors test this. They rank currencies by current account balance and go long the surplus countries against the deficit countries. The result: it loses money overall. The only times this strategy works are during actual crises (2008, the taper tantrum in 2013, the oil crash in 2014-2015). During normal times, it bleeds.

Why? Because current account surpluses and low interest rates go hand in hand. Countries with strong external balances don’t need to pay high rates to attract capital. So the “safe” strategy is basically the opposite of the carry strategy. It wins during crises and loses the rest of the time. Unless you can perfectly time when risk aversion is about to spike, it doesn’t work.

Only Risky Borrowers Pay High Rates

There’s a neat chart in the book (Figure 4.10) showing real interest rates versus current account balances at the end of 2018. The pattern is clear: every country with a current account deficit larger than 3% of GDP offers high real rates. The surplus countries get away with much less.

This makes total sense. If your country needs foreign capital to finance a deficit, you have to pay up for it. If your country is self-funding, why would you offer high rates?

The exception the authors note is Russia, which chose to keep high real rates despite having a current account surplus. But that’s more about political risk and central bank credibility than economics.

The practical takeaway: you can’t have it both ways. High carry currencies are high carry for a reason. They’re risky. Trying to find the safe, high-carry EM currency is like looking for a unicorn. They don’t really exist. And when they briefly appear (like Russia at times), there’s usually some other risk lurking underneath.

The better approach, the authors suggest, is to combine carry with improvements in trade balances rather than static current account levels. If a country’s trade balance is getting better and it still offers good carry, that’s a stronger signal than just looking at who has the best balance sheet right now. The combination of carry ranking plus trade improvement ranking produces a much better P&L than either factor alone.

What We’ve Learned So Far

This first half of Chapter 4 sets up a few key ideas:

1. The EM vs. G10 line is mostly artificial. Only the yen is truly special. Most currencies respond to global macro in similar ways. The real distinction is how they respond to their own interest rates.

2. Simple carry is struggling. If you’re funding in USD, the glory days are over. JPY funding helps. Vol adjustment helps. But the easy money from 2003-2011 isn’t coming back without much higher global yields.

3. Volatility adjustment improves everything. Carry, momentum, hedging. Whatever you’re doing in EMFX, adjusting for vol makes it better.

4. EMFX is inherently scared. High-carry currencies know they’re risky. When volatility shows up in any asset class, they sell off. A risk overlay that watches implied vols across markets can save you from the worst drawdowns.

5. You can’t buy only the safe stuff. Playing it safe means giving up carry, and the carry is the whole point of being in EM. The trick is combining carry with improving fundamentals, not avoiding risk entirely.

In Part 2, we’ll cover the rest of Chapter 4: momentum strategies, valuation (spoiler: it barely works), technicals, and how to put all these factors together into one trading model.

Book Details:

• Title: Trading Fixed Income and FX in Emerging Markets
• Authors: Dirk Willer, Ram Bala Chandran, Kenneth Lam
• Publisher: Wiley
• Year: 2020
• ISBN: 978-1-119-59905-0

Previous: China Part 2 Next: How to Trade EMFX Part 2