The Best Bonds of 2026 Bet That Inflation Never Left
The Global Agg has lost money this year. Broad EM local debt returned 1.5%. One $886 billion corner of the bond market returned 11.3% — by assuming inflation would stay. How the trade worked, and where the money is rotating next.
Global bond investors have had a miserable year. The Bloomberg Global Aggregate — the benchmark that defines "owning bonds" for most of the world's institutional money — is down 0.1% in 2026. Emerging-market local-currency debt, pitched at the start of the year as the cleanest way to get paid for a falling dollar, has returned just 1.5%.
And then there is one corner of the market that has returned 11.3%.
Emerging-market inflation-linked bonds — an $886 billion asset class most US investors have never owned and many have never heard of — are the best-performing major fixed-income trade of 2026, through mid-August. Not by a rounding error. By roughly eight times the broad EM local index, in a year when the global benchmark lost money.
The instrument itself is simple. A linker pays a real coupon on a principal that ratchets up with the consumer price index. When inflation runs hot, the bond's principal grows with it; the investor is made whole automatically, month after month, at whatever pace prices actually rise. The US version is TIPS. But the deep, high-coupon versions live in Latin America — Brazil and Mexico dominate an LatAm-focused segment that Bloomberg Intelligence puts at more than $80 billion — because those are the countries that spent the twentieth century learning, expensively, that inflation is the default state of the world. They built their bond markets to survive it.
In 2026, that institutional memory became the trade.
The year the world was built for linkers
Start with the macro. The Strait of Hormuz has been closed to commercial traffic since early March. Brent went from roughly $70 in January to above $100 by April. About a fifth of global oil supply was disrupted, and LNG flows through the strait collapsed by roughly 95% — close to one-fifth of the world's supply of the fuel. That shock has fully passed through into diesel, jet fuel, fertilizer, and power prices. Brazil's own statistical agency attributes its energy and fuel inflation — running at 7.71% as of June — directly to the Hormuz closure. Turkey's central bank just raised its year-end inflation forecast from 26% to 28%. Price pressure is showing up from India to South Africa to Romania to Colombia.
A nominal bond hates that world. Its coupon is fixed; every upside inflation surprise eats the real return, and every central bank that pauses its cutting cycle to fight prices pushes yields higher and prices lower. That is precisely why broad EM local debt — which is mostly nominal — went almost nowhere this year despite a dollar sitting near four-year lows.
A linker loves that world, three times over. First, the carry: Latin American real yields entered the year among the highest on earth, with Brazil's policy rate still at 14% even after four straight cuts. Second, the indexation: every month of hot CPI accrues straight into the bond's principal. Third, the currency: the same weak dollar that was supposed to power the broad EM trade did show up — but it compounded on top of the inflation accrual instead of substituting for it.
Briefings like this land in members' inboxes before the market prices them in. Join free →
The correction to the consensus trade
Here is the uncomfortable part for anyone who bought the standard 2026 playbook — including, in part, us. In July we made the case that EM local-currency bonds were the cleanest way to be paid for dollar weakness. The dollar did weaken. The broad trade still barely paid. The entire year's return in EM debt concentrated in the inflation-protected slice, because the variable that mattered was not the dollar — it was that inflation itself kept running above what nominal yields had priced.
That distinction is not academic. It tells you what actually drove 2026's best bond trade, and it tells you the conditions under which it stops working. Which is now the live question: after an 11.3% run, the managers who were early — BlackRock, Vontobel, Aberdeen, Aegon — are openly saying the easy phase is over. Brazilian and Mexican linkers look rich. The money is rotating, and it is rotating to specific places for specific reasons.
Where it is going — and the one trade inside Brazil that the rotation logic says to do instead — is below the line.
The rest of this briefing is for paid members: the four linker markets the rotation is moving into and the buyer-base mechanics behind each, the Brazil fork (why one major manager is selling the linkers and buying the nominals), the three conditions that break this trade — one of which is being tested this week — and how a US-based investor actually gets exposure.
AlphaBriefing Paid gets you every investment thesis, scenario framework, and catalyst brief we publish — the analysis private intel clients pay four figures for, at a fraction of that.