The setup

Emerging-market local-currency debt has returned roughly 14% in the trailing twelve months, beating both US investment-grade credit and global aggregate by a wide margin. That outperformance has been quiet — there are no investor letters celebrating it, no fund-of-the-year nominations, and almost no flow chasing it.

That quiet is the point. EM debt outperforms most reliably when nobody is positioned for it.

Key takeaway — Local-currency EM debt is offering a 4.2 percentage-point real-yield premium over US Treasuries — the widest in a decade. Historically, that premium has compressed to 1.5 points within 24 months. Today's spreads do not require optimism to work.

What changed in 2024–2025

Three things shifted the regime:

  • EM central banks moved first. Brazil, Mexico, Indonesia, and Czechia all began cutting policy rates before the Fed. By the time the Fed pivoted in late 2024, EM real yields were structurally higher than DM real yields — an unusual configuration historically associated with strong forward returns in EM debt.
  • Fiscal trajectories diverged. US debt-to-GDP ticks up toward 130%. Brazil's stabilises near 75%. Mexico's holds at 53%. India's debt-to-GDP, though high in nominal terms, is funded almost entirely in domestic currency. The conventional "EM is fragile" narrative reads worse than the actual numbers in 2026.
  • Currency overshoot in 2022–23 reversed. The 2022 dollar squeeze pushed EM currencies to two-standard-deviation cheapness against PPP. Most have only recovered halfway. The carry-plus-spot setup is therefore not asymmetric on the downside the way it was in 2013 or 2018.

What the numbers look like today

A composite local-currency EM debt portfolio yields 7.4% as of March 2026 against US Treasuries at 4.2%. Inflation in the same EM basket runs at 3.8% versus US CPI at 3.1%. That puts EM real yields at 3.6% versus US real yields of 1.1%.

The forward-return relationship between starting real yields and subsequent 5-year returns in EM debt is one of the most stable in fixed-income markets. The R-squared of the regression is over 0.7. At today's starting point, the model implies a base-case 5-year nominal return in EM local debt of 9–11% annualised.

What can go wrong

Three things, in order of importance:

  1. A dollar squeeze. If the Fed is forced to re-tighten by a US inflation surprise, EM currencies sell off. The 2022 episode showed this is the dominant risk factor. But starting valuations are far less stretched.
  2. A China growth shock. A faster-than-expected China slowdown hits commodity exporters (Chile, Indonesia, South Africa) hardest. Asia equity exposure is the buffer most multi-asset portfolios use, but the bond side of an EM allocation is not immune.
  3. A specific-country accident. Idiosyncratic risk — a default, a coup, a populist fiscal reversal — is permanent in EM. Diversification across countries and managers is the only defence; concentrated single-country bets are not the trade.

How to think about manager selection

Beta to the EM debt index is easy and cheap. Where active managers earn their fees is in country selection, duration positioning, and FX overlays. The names worth paying for in 2026, by track record and capacity discipline:

  • Hard-currency specialists who can rotate out of sovereigns into corporates when sovereign spreads compress;
  • Local-currency managers with genuine FX research, not just bond traders bolting on a carry signal;
  • Blended-strategy funds that can flex between the two without strict mandate constraints.

The portfolio context

Most multi-asset portfolios are structurally under-weight EM debt — 2–4% versus an index-implied 6–9%. A move to neutral is not a tactical call. It is a recognition that the asset class has matured, that local-currency markets are deeper than they were in 2013, and that real yields above 3.5% should not sit in benchmark-relative under-weight positions in any income-oriented portfolio.

Bottom line

The trade is not "EM debt is going to rip". The trade is "EM debt offers a real-yield premium that no longer requires luck to harvest". Starting valuations, central-bank positioning, and fiscal arithmetic all point in the same direction. The risk is being paid; the position size should be sized to the conviction, and the conviction should be sized to the data.