EMFX Event Guide: Interventions, Emergency Hikes, IMF Packages, and Elections
EM policymakers really, really care about their exchange rates. Way more than developed market policymakers do. And for good reason. FX matters more for inflation in emerging markets. There’s way more USD-denominated debt floating around. And politically, a collapsing currency is basically a death sentence for the sitting government. The FX rate is the most visible report card for whether the government is doing a good job.
So when things start falling apart, policymakers fight back. Chapter 5 lays out the playbook for how they fight, and more importantly, how you can trade around each step.
The pattern is pretty consistent. First they try FX intervention. If that doesn’t work, emergency rate hikes. If those don’t work, it’s either capital controls or calling the IMF. Each stage gets more painful and more desperate. And like generals fighting a war, EM policymakers fight to the bitter end, burning through reserves long after everyone can see they’re going to run out.
Intervention: Do It Rarely, or Do It Big
FX intervention is the first weapon because it’s cheap. Carrying reserves actually costs money (local rates are usually higher than USD rates), so spending them has zero or even negative direct costs. The problem is that the impact is usually short-lived.
The authors looked at intervention data across Latin America and found that the details are surprisingly country-specific. But one rule cuts across all of them: size matters, and rarity helps.
In Brazil, large interventions (above $2.5 billion per day) work. When done for the first time or when firepower gets meaningfully increased, the BRL outperforms the EMFX index by about 2.5% over 10 days. Small interventions? No significant impact. In Mexico, first-time interventions also work, generating about 1.5% outperformance over 10 days, with follow-up interventions being less effective. Chile’s rare interventions also tend to work for about 10 days. Colombia? Not much of a pattern, possibly because they use an options-based structure that dampens volatility but doesn’t clearly move the spot rate.
The practical lesson for traders: for countries that intervene all the time, only the big ones matter. Go with the intervention if the central bank brings serious firepower. For countries that rarely intervene, any intervention is surprising by definition, so trade with it. And here’s a nice bonus: when a large EM country gets an intervention that stabilizes its currency, that positive vibe often spreads to the rest of EM. In those cases, shorting USD broadly is better than just going long the specific currency against an EMFX index.
Interestingly, interventions on the strong side (buying USD to weaken the local currency) tend to be more effective and longer-lasting than interventions defending a weak currency. It’s easier to buy dollars than to sell scarce ones.
Emergency Rate Hikes: They Often Work
Emergency rate hikes are the second line of defense. They only come after intervention has failed, because the costs are much higher. Higher rates slow the economy, can trigger recessions, and politicians hate them. Central banks usually need political cover to pull the trigger, even formally independent ones.
The authors found 20 emergency hikes across 8 countries in 21 years. That’s roughly one every 8 years per country. Turkey is the serial offender, averaging one every three years. Mexico is at the other extreme with very few.
Here’s the good news: they usually work. For the majority of cases where only one hike was needed, the currency showed a strong positive response for about 10 days. After that, the initial pop fades, but the currency stays stable around pre-hike levels. Investors can at least earn the now-very-high carry.
But there’s a really important caveat.
Don’t Make the Emergency Call Too Early
The authors split emergency hikes into two groups: ones where the currency was already cheap (below its 8-year moving average real effective exchange rate) and ones where it was still expensive.
For cheap currencies, emergency hikes are a slam dunk. The currency stabilizes, the USD peak is in, and you probably won’t see those weak levels again for a long time. Go long.
For expensive currencies? The hike barely works for a day before the selloff continues with the same force.
The Argentina 2018 episode is the perfect example of what not to do. In April and May 2018, Argentina implemented three emergency rate hikes in seven days. The problem? The ARS had only depreciated 2% in the two months before the hikes. Two percent. That’s nothing for an EM currency. It wasn’t actually an emergency. The currency hadn’t cheapened enough for longs to have been flushed out and shorts to have been established. Without that positioning reset, emergency hikes can’t work their magic.
The lesson: if you see an emergency hike on a currency that hasn’t sold off much yet, fade the rally. If the currency has been absolutely crushed and is now genuinely cheap, buy it.
Capital Controls: Even Severe Ones Come in Many Shapes
When intervention and rate hikes both fail, countries face a choice: capital controls or the IMF. Which path they pick usually comes down to politics. If a country has goodwill with Washington and the IMF, an IMF package is the better option because it spreads the adjustment over time. If the country has been on a collision course with the US, capital controls might be the only politically feasible move.
The authors are pretty blunt: severe capital controls are basically untradable. Both shorts and longs can lose money because the FX fixing gets set arbitrarily. Shorts might not be able to transfer local currency out. Longs might get stuck buying back at artificially strong rates. A black market always pops up, but foreign investors usually can’t access it at the scale they need.
Soft capital controls, on the other hand, don’t really do much. Brazil’s IOF tax is a good example. The authorities used this financial transaction tax to fight off excessive capital inflows, hiking it several times between 2008 and 2010. It barely moved the BRL. In the end, it took a turn in the broader dollar index to actually change the trend.
El Cepo: Argentina’s Capital Trap
Argentina’s “el cepo” (literally “the trap”) from 2011 to 2015 is the textbook case for severe capital controls. In late 2011, the government required all foreign exchange transactions to be pre-approved by the tax agency. They deployed 4,000 inspectors to enforce it. They slapped a 15% tax on credit card use abroad. A black market predictably appeared, with the dollar trading almost 20% above the official rate.
But here’s what’s wild: the authorities never closed one big loophole. Investors could swap onshore equities or bonds (priced in pesos) for their offshore equivalents (priced in dollars), using things like American Depositary Receipts. This was perfectly legal. The “blue chip swap rate” that emerged from this mechanism was insanely volatile, at times 80% weaker than the official exchange rate. So it was very cheap to get dollars into Argentina and very expensive to get them out.
Despite all this chaos, the total returns on ARS during el cepo were actually strong. NDF implied yields peaked at 65% in late 2013, and rolling one-month NDFs turned out to be profitable overall. The drawdowns were severe, but the carry was enormous.
Fun with Onshore-Offshore Spreads
The 2015 Argentine election created one of those beautiful arbitrage opportunities that capital controls sometimes produce.
Everyone expected a big peso devaluation after the election. So the offshore one-month NDF priced in a huge move. But the outgoing government didn’t let the onshore forward market (ROFEX) reflect any of that weakness. They were intervening heavily in onshore futures to project strength going into the election.
A massive gap opened up between onshore and offshore one-month ARS forwards. Onshore investors could buy dollars cheap onshore and sell them expensive offshore. It was basically a free subsidy from the Argentine government to anyone who could access both markets. Even offshore investors could participate because ROFEX only required margins to be sent into Argentina, not full principal.
IMF Packages: They Stabilize FX, Eventually
When a country goes to the IMF, what happens to the currency?
The authors looked at all IMF packages since 2001. On average, the currency sells off hard going into the announcement. And here’s the thing that surprises people: the announcement itself usually doesn’t stabilize the currency right away.
Only the best-case outcomes (top 10th percentile) see an immediate turnaround. For the median case, it takes about 25 trading days for the currency to start strengthening. For the average, including some ugly outliers, it takes about three months.
So IMF programs do work. They stabilize currencies in the big picture. But not on day one. You need patience.
One reason for the delayed reaction is that some IMF programs actually require a weaker currency as part of the adjustment. The IMF knows that the country’s external balance is out of whack, and a weaker FX is part of the medicine.
Elections: Don’t Be Afraid of Your Average Left-Wing Populist
Political risk in EM is higher than in developed markets because institutions are weaker and less credible. But that’s a constant background factor. What makes it tradable is elections.
The authors studied 23 presidential elections in Latin America where a market-unfriendly leftist candidate got at least 25% of the vote. The pattern they found is remarkably consistent.
Election uncertainty starts building about six months out, with rising volatility and a weaker currency. It accelerates at the three-month mark. But about two weeks before the election, the trend stalls and the currency turns around. After the election, it typically rallies. Within their sample, every single election showed depreciation between three months and two weeks before the event. And in all but one case, the currency strengthened from two weeks before to three months after.
Why does this work so consistently? Because even market-unfriendly candidates have strong incentives to sound market-friendly once they win. The alternative is an immediate FX collapse, inflation spike, and growth hit. Most new leaders prefer to “boil the frog slowly” rather than pour boiling water on the market right away.
Wait, When Did Chavez Come to Power Again?
Even Hugo Chavez, possibly the most market-unfriendly EM leader in recent history, wasn’t immediately negative for markets. Most of the volatility around his election happened before the December 1998 vote. After he won, Venezuelan bonds actually outperformed Brazil for years. It took until his third term in 2006, when he pronounced “21st century socialism” and started nationalizing industries, for Venezuelan bonds to consistently underperform. So much for the market’s ability to forecast long-term political risk.
Nothing Ever Changes South of the Rio Grande
The AMLO election in Mexico in July 2018 followed the exact same playbook. AMLO was a genuine populist who had been trying to become president for three elections. His party Morena won such a strong majority that they could even change the constitution. No safety valves left. And the peso still rallied after the election.
The rally was partly driven by a market-friendly acceptance speech. But mostly it was driven by local investors who had built huge long-USD positions getting stopped out. The negative carry on those positions reduced their staying power. In a classic ironic twist, it was only after most dollar longs had been squeezed out that AMLO actually did something market-unfriendly: canceling the new Mexico City airport.
Forget About NFP: Follow the CPI and the FOMC
Since EMFX is mostly driven by global factors, which data releases matter most?
The old answer was nonfarm payrolls. Not anymore. As of late 2018, NFP has become unreliable for EMFX trading. Reversals happen too quickly, often the day after. The price action on NFP release day is mostly noise.
What still works? US CPI releases and FOMC meetings show meaningful follow-through. Hawkish surprises (defined by US two-year rates moving up by at least one standard deviation) lead to EMFX weakness of about 0.7-0.8% over the following 20 trading days. Dovish surprises generate the opposite effect, though slightly smaller.
The US manufacturing ISM is interesting but a bit tricky. It shows strong follow-through for hawkish surprises but not for dovish ones, which makes the authors cautious about relying on it.
For Chinese data, the results are less clear. The authors used AUD moves as a proxy for whether Chinese PMI or total social financing surprised the market (since Chinese rates don’t have a long enough trading history). There’s some mild follow-through, but it’s hard to separate the signal from the general bear market in EMFX that dominated much of their study period.
One important nuance: what the Fed is focused on matters. Sometimes they care most about inflation, sometimes about activity. The general patterns hold, but you need to adjust for where the Fed’s attention is at any given moment.
The Bottom Line
EM policymakers have a predictable escalation ladder when their currency is under pressure: intervention first, emergency hikes second, then either capital controls or the IMF.
For traders, the actionable rules are:
- Intervention: Trade with it if it’s large or rare. Fade it if it’s small and routine.
- Emergency hikes: Go long if the currency is already cheap. Fade the rally if it’s still expensive.
- Capital controls: Stop trading the asset. But watch for juicy onshore-offshore spreads.
- IMF packages: They work, but give it 3 months. Be patient.
- Elections: Buy the dip about two weeks before. The relief rally is one of the most consistent patterns in EM.
- Data: Forget NFP. Trade CPI and FOMC surprises.
This post is a retelling of Chapter 5 of “Trading Fixed Income and FX in Emerging Markets” by Dirk Willer, Ram Bala Chandran, and Kenneth Lam. Published by Wiley, 2020. ISBN: 978-1-119-59905-0.