How to Trade EM Credit: The Sweet Spot and What Buffett Would Do

Here’s the thing about EM credit that nobody tells you upfront: the structural trade is basically dead. You’d think that because emerging markets grow faster than developed ones, their credit spreads would keep compressing over time. More growth, less risk, tighter spreads. Makes sense, right?

Except it hasn’t worked that way. The all-time low spread for EM high yield was hit back in 2007. Since then, despite EM averaging 5.3% GDP growth versus only 1.5% for developed markets, EM high yield spreads have just bounced around in a 250-to-500 basis point range. No trend. No compression. Nothing.

So if the structural trade is dead, what’s left? The cycle. And this chapter is all about how to trade it.

EM Credit vs US High Yield: Not What You’d Expect

Before we get into strategy, let’s clear up one big misconception. People often treat EM sovereign credit and US corporate high yield as interchangeable. They’re not.

EM sovereign defaults are rarer. The default probability for sub-investment-grade sovereigns was 2.5% from 1983 to 2016. For US high yield corporates? 4.2%. Almost double.

And when EM sovereigns do default, you get more of your money back. The average recovery value for sovereigns was 65% (45% if value-weighted). For US high yield? Only 37%.

Lower default rates. Higher recovery. That alone gives EM sovereign high yield a structural edge over US high yield on a total return basis. Even without spread compression.

Carry Still Works in Credit

Remember how carry strategies stopped working in EMFX after the 2013 taper tantrum? Well, good news: carry is still alive and well in EM credit.

The simplest version of this trade is just overweighting high yield over investment grade. The HY index has consistently beaten the IG index on a cash-neutral basis. Drawdowns were contained too, with 2008 being the one big exception.

Why does carry still work in credit but not in FX? One theory: global credit has been in a structural bull market. EMFX has been in a bear market. Carry just doesn’t work well in bear markets. Simple as that.

At the country level, carry works even better. The authors built a strategy that goes long the top eight highest-yielding credits and short the bottom eight, with volatility adjustments. It performed in line with the index before 2008, had a much smaller drawdown during 2008, and then crushed it afterward.

You can improve the carry strategy further by using a risk indicator. The authors borrowed the volatility-based risk indicator they built for EMFX and applied it to credit. It worked beautifully. The information ratio for plain spread returns was 0.53. With the risk indicator, it jumped to 0.85. That’s a big deal.

Momentum also works. A 12-month look-back momentum strategy, where you go long the top eight credits and short the bottom eight (all volatility-adjusted), generated an information ratio of 0.7 from 2004 to 2018. The index only managed 0.4.

The Sweet Spot: 3-to-5 Year BB-Rated Bonds

OK, so you want to pick up extra carry. You have two levers. You can go down in credit quality (riskier names, higher spreads). Or you can increase duration (longer bonds, steeper curve, more carry).

The authors looked at every combination and found the sweet spot.

On the ratings side, BB-rated bonds had the best information ratio at 0.89. B-rated bonds were next at 0.72. A-rated were solid at 0.70. And BBBs? A sad 0.46. We’ll get to why BBBs are so bad in a minute.

On the duration side, the front end crushed everything:

The 3-to-5 year bucket had the highest IR of any duration segment. And the 10+ year bucket had the worst.

So the sweet spot is BB-rated bonds in the 3-to-5 year range. Not the sexiest trade. Not the one that’ll make you look brilliant at a dinner party. But it’s the one that actually makes money consistently.

What Warren Buffett Would Do

Warren Buffett’s whole thing is buying safe, boring companies and leveraging them up. The idea is that levering up a portfolio of safe assets gives you a higher risk-adjusted return than just owning risky assets outright.

This works in EM credit too. If you take the short-duration part of the EMBI index and lever it up (using six months of rolling volatility to calculate the right ratio), you get higher information ratios than simply owning the riskier long end. Since 2012, the excess performance has been impressive. Investors do not get paid for venturing out to the long end of the credit curve.

But here’s a catch. If there’s an actual default, all bonds on the curve start trading at roughly the same dollar price. So the short-duration strategy is basically selling jump-to-default risk. That means it works best for investment grade or strong BB credits. Nobody is defaulting overnight there.

Now, within each ratings bucket, the Buffett principle holds up perfectly. A-rated beats BBB. BB beats B. The safer credit in each category outperforms the riskier one.

But it does NOT work across buckets. You can’t lever up A-rated bonds and expect to beat high yield. There’s clearly an excess premium for taking on HY risk. Buffett’s rule works within your weight class, not across weight classes.

One big reason BBBs do so poorly: the downgrade cycle. When a country loses its investment grade rating, going from BBB to BB, it’s the most painful ratings move in all of EM credit. These downgrades crushed BBB performance while actually helping the BB bucket (since the newly downgraded credits join the BB universe at cheap levels). Avoid BBBs when the growth cycle looks weak.

Improving on Buffett with the Business Cycle

You can make the Buffett trade even better by timing it with the US business cycle.

When the ISM manufacturing index is in the high 50s (basically, the economy is running hot and about to slow), that’s when you want to rotate from high yield into investment grade. High yield spreads widen more than IG spreads during downturns. And the carry isn’t enough to offset the spread widening.

The opposite is true on the way back up. Early in the recovery, the most beaten-up HY credits snap back the hardest.

The rule of thumb for timing: watch the US yield curve. Be long EM credit most of the time. But when the US 2s/5s curve inverts and then disinverts, cut your positions. The disinversion is often more dangerous than the inversion itself. In 2001 and 2006, waiting for the disinversion before getting negative on EM credit would have been the right call.

The one exception: if the inversion is driven by an EM crisis (like 1998), don’t wait for the disinversion. Get out on the inversion itself.

Even on duration, there’s a cycle play. When the ISM is falling, front-end spreads actually do worse than back-end spreads on an index level. But the authors looked at individual countries (like Poland, one of the safest EMs during 2008) and found that even for very safe credits, front-end bonds outperformed back-end bonds during severe downturns. The lesson: when the ISM peaks, reduce duration even on the safest curves. If you want to benefit from falling US rates, just buy Treasuries directly.

What Ray Dalio Would Do

The classic way to pick EM credit winners is to look at current account balances, fiscal deficits, debt levels, and inflation. The typical red flags: current account deficit above 3-4%, fiscal deficit above 7-8%, external debt above 45%, inflation above 20%.

But the authors argue these widely-watched indicators are already priced in. If a credit looks cheap compared to its fundamentals, the market is probably pricing in future deterioration that you haven’t seen yet.

A better approach borrows from Ray Dalio’s framework on debt crises. Instead of looking at traditional fundamentals, look for signs of exuberance:

• Equity rally of more than 25% over three years? Warning.
• Capital inflows above 12% of GDP over three years? Warning.
• Real exchange rate appreciation of more than 10% over three years? Warning.
• Positive output gap above 2.5%? Severe overheating.

The problem is timing. Exuberance can last a long time. Argentina in 2017 ticked every single box on Dalio’s checklist. Weak fundamentals, bubbly equity markets (up 83% in USD over three years), massive current account deficit (5.8% of GDP), rapidly rising external debt. But almost nobody saw the bust coming because the narrative was all about President Macri and market-friendly reforms.

So the authors suggest two approaches to the timing problem. First, keep a list of countries showing signs of exuberance, but only short them when a negative shock changes the consensus narrative. A commodity price drop, a bout of USD strength, a political surprise. Second, use technicals. A break below the 200-day moving average for a country on your exuberance watchlist is a good entry point for a short.

For commodity producers specifically, the credit story often boils down to one commodity price. Oil producers with pegged exchange rates are the easiest shorts during oil downturns. Bolivia, Iraq, and Oman all underperformed the broader EM commodity credit basket during falling oil prices. Pegged currencies can’t adjust, which makes everything worse.

IMF Packages: They Work, But Not Right Away

When a country blows up and the IMF comes in, what happens next?

The authors studied all 32 IMF packages since 2001. Here’s the pattern:

Spreads widen by an average of 130 basis points in the two months before the announcement. Then they stabilize about 15 days before (when rumors start leaking). The announcement itself doesn’t immediately help. Spreads actually widen another 50 basis points on average in the first 20-30 days after the announcement, as some investors use the stability to cut their exposure.

But then things get good. 150 trading days after the announcement, the median spread is more than 100 basis points tighter than before the deal. Interestingly, the size of the IMF program doesn’t seem to matter for bond performance. Maybe because the IMF is pretty good at sizing its packages to fit the problem.

The trade: go long after the IMF announcement, but don’t rush. Wait for the first pullback. That’s your entry.

Embrace Defaults

This might be the most counterintuitive section of the whole chapter. Defaults sound scary. They’re not. At least not if you’re buying after the restructuring.

The authors looked at every sovereign default over $1 billion since 1995. The median one-year return after the restructuring was 16%. Median excess return over the index was 11%. Greece returned 133% in the year after its 2012 restructuring. Ecuador returned 36% after its 2009 default.

Why are the returns so good? Because after a default, everything that was bad suddenly gets better. Leverage drops (because you just wrote off a chunk of the debt). The currency is cheap. The current account often swings to surplus because imports got crushed. A new government usually takes over, bringing hope for better management. And investors, still traumatized by the default, demand huge risk premiums to come back.

Markets have surprisingly short memories about defaults. Academic research shows that post-1991, the average time a sovereign remained in default was only about four years. And since 1998, only one well-known restructuring took longer than two years to resolve. After the restructuring, countries can borrow again pretty quickly. Spreads are about 400 basis points higher than normal in year one, 250 in year two, and then they converge to zero.

The trade going into a default is trickier. Most investors hide at the front end of the curve to clip coupons. This works for a while because defaults almost always take longer than expected. But when the default actually gets close, the curve flattens aggressively as front-end bonds sell off (since restructurings typically pay the same dollar price across the entire curve).

50 Cents Says: So What?

Here’s a fun stat. Between 1996 and 2018, there were 16 times when a country’s bond index fell below 50 cents on the dollar at month-end. Of those 16 episodes, 10 were actual defaults and 6 were false alarms. But only 3 of those defaults happened within six months. Four took longer than two years.

That’s a lot of coupons you can clip while waiting.

On average, bonds perform well after falling below 50 cents. The countries that don’t end up defaulting recover quickly. And even the ones that do default tend to see prices recover before the final leg lower, which happens on average three months before the actual default event. Prices bottom about one month after the default.

The Hemingway quote about bankruptcy applies perfectly: “Gradually. Then suddenly.” FX reserve losses can accelerate incredibly fast, making it nearly impossible to time the exact moment of default.

Rating Agencies: Late to the Party, But They Can Still Move Markets

Rating agencies are slow by design. They want to capture the long-term credit cycle, not short-term noise. But by the time they actually downgrade or upgrade someone, the market has usually already priced it in. Spreads widen into downgrades and tighten into upgrades well before the announcement.

So should you even care about rating changes? Mostly no. But there’s one huge exception: the investment grade threshold.

Losing IG status is different. The sell-off is 2.5 times larger than a normal downgrade. And here’s where it gets interesting: rating agencies herd. When one agency downgrades a country below IG, the second agency follows within six months 92% of the time. In 42% of cases, the second downgrade comes within a month.

The second downgrade is usually peak negativity. After that, especially for EM corporates, it’s time to buy. Give it two or three days after the second IG loss, then go long. Some investors wait until the end of the following month, when forced selling from IG-only mandates is fully done.

On the upgrade side, the first upgrade to IG has good follow-through. The second upgrade is peak short-term bullishness.

The lesson: rating changes are lagging indicators. Don’t react to most of them. But when a country crosses the IG/HY border in either direction, pay very close attention.

Wrapping Up Part 1

The big takeaways from this first half of the EM credit chapter:

The sweet spot exists. BB-rated bonds in the 3-to-5 year range give you the best risk-adjusted returns. Not the flashiest trade, but the most reliable.

Carry still works. Unlike EMFX, carry strategies in EM credit continue to deliver. Overweight HY over IG on a cash-neutral basis and you’ll do well most of the time.

Think like Buffett. Lever up safe, short-duration bonds rather than reaching for risky, long-dated ones. Within each ratings bucket, the safer credit wins.

Think like Dalio for credit selection. Watch for exuberance, not just bad fundamentals. When equity markets are bubbly, capital inflows are massive, and the currency is overvalued, trouble is coming.

Defaults aren’t the end of the world. Recovery rates for EM sovereigns are higher than US corporates. And the year after a restructuring is usually a great time to buy.

Rating agencies herd. The second downgrade below IG comes fast. Wait for it, then buy.

In Part 2, we’ll cover how EM credit events play out, external versus local bonds, and how to put it all together.

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: EM Rates Event Guide Next: EM Credit Part 2 - Events and External vs Local