Read Emerging Market Currency Swings as a Regime Variable, Not a Fixed Property

A durable piece of market shorthand says that emerging market currencies are structurally more volatile than the dollar, and that this is a permanent feature of how the system is built. The structural half of that claim is defensible. The permanent half is not, and the period from early 2025 through the first quarter of 2026 is an unusually clean test of the difference.

Consider the sequence. The dollar index fell 10.8% in the first half of 2025, its worst first-half performance since 1973. Over an overlapping window, JPMorgan's emerging market volatility gauge sat below the comparable G7 gauge for 209 consecutive trading days. That streak, the longest in records going back to 2000, ended on 3 March 2026, when Brent crude reached its highest level since July 2024 amid Middle East tensions. For most of a year, the currencies supposed to swing harder swung less, and the reserve currency was the one repricing.

The depth argument is not wrong. It answers a different question than the one it is usually deployed to answer. Depth governs how much a given order size moves a price; it says nothing about how large the orders will be, or in which direction. Treating a transmission coefficient as a directional forecast is the error. What follows works through the data in three layers, then reads them against each other.

Layer One: The Depth Table

The Bank for International Settlements runs its Triennial Central Bank Survey every three years; the April 2025 round is the current reference. Global over-the-counter FX turnover averaged USD 9.6 trillion per day, up 28% from USD 7.5 trillion in April 2022. The instrument split matters as much as the headline: FX swaps were roughly 42% of turnover, spot 31%, outright forwards 19%, with options and currency swaps making up the remainder. Spot's share rose from 28% in 2022.

Currency	Share of one side of all trades, April 2025	Direction vs 2022
US dollar	89.2%	Up from 88.4%
Euro	28.9%	Down from 30.6%
Japanese yen	16.8%	Broadly unchanged
Pound sterling	10.2%	Down from 12.9%
Chinese renminbi	8.8%	Up from 7.0%
Swiss franc	6.4%	Up; now sixth most traded
Hong Kong dollar	3.8%	Up from 2.5%
Singapore dollar	2.4%	Up
Indian rupee	1.9%	Up

Two technical points. The shares sum to 200%, not 100%, because every transaction involves two currencies and each is counted once. And the survey measures turnover at sales desks: 75% of global trading was booked in four jurisdictions, the United Kingdom at roughly 38%, the United States at roughly 19%, Singapore at 11.8% and Hong Kong SAR at 7.0%. Where a currency trades and where its economy sits are different questions.

The aggregate is the number that undercuts the "permanently thin" framing. Emerging market economy currencies collectively reached 29% of global turnover in April 2025, against 26% in April 2022 and under 10% through the 2000s. Renminbi turnover grew 56% over the three years, and USD/CNY overtook USD/GBP to become the third most traded pair. Brazilian real turnover grew 37% on BIS figures. Beyond the renminbi and the Hong Kong dollar, however, gains were modest, and turnover as a share of GDP plateaued for the median emerging market economy between 2022 and 2025. Depth improved, but it improved unevenly, and concentration inside the emerging market bucket is itself a structural fact.

Layer Two: The Balance Sheet Table

Turnover describes the plumbing. It does not describe the pressure in the pipes. That comes from the stock of foreign currency obligations, and the BIS global liquidity indicators track it. At end-December 2025:

Measure	Level	Annual change
US dollar credit to non-bank borrowers outside the US	USD 14.3 trillion	+8.5%, fastest since Q3 2014
Of which, to emerging market and developing economies	USD 4.3 trillion	Up from USD 3.2 trillion at end-2015
Euro credit to non-bank borrowers outside the euro area	EUR 4.9 trillion	+11%
Of which, to emerging market and developing economies	EUR 858 billion	Up from EUR 437 billion at end-2015
Yen credit	Not comparable in scale	Contracted 4.9% during 2025

Dollar credit to emerging markets grew by roughly a third over a decade, which is real but not explosive. Euro credit to the same borrowers roughly doubled from a much smaller base, so the funding mix is diversifying at the margin. Yen credit shrank in 2025, which is what a funding currency does when its rate differential compresses.

None of these figures capture the largest exposure, which sits off balance sheet. BIS work on FX swaps and forwards put total outstanding dollar payment obligations in those instruments at roughly USD 85 trillion as of end-June 2022, including about USD 26 trillion owed by non-banks outside the United States and about USD 39 trillion by banks headquartered outside the United States. These are contractual obligations that standard debt statistics do not show. Any account of currency stress that works only from published external debt is working from a fraction of the exposure, and the fraction is not stable across countries.

Layer Three: The Reserve Composition Table

The slowest of the three series is the IMF's Currency Composition of Official Foreign Exchange Reserves. For 2026 Q1, total allocated and unallocated reserves stood at USD 13.10 trillion, with allocated shares as follows.

Currency	Share of allocated reserves, 2026 Q1	Prior quarter
US dollar	57.13%	56.42%
Euro	20.03%	Modestly higher
Japanese yen	5.44%	5.84%
Renminbi	1.99%	1.95%

The dollar share rose in the quarter, after a year in which the dollar had fallen sharply and the diversification narrative was at its loudest. Reserve shares are partly a valuation artefact and partly an allocation decision, and a single quarter separates neither. The narrower point: the renminbi holds under 2% of allocated reserves while holding 8.8% of daily turnover. Trading depth and reserve status do not move together.

Reading the Three Layers Against Each Other

Depth is a coefficient, not a forecast

A market clearing USD 9.6 trillion a day absorbs a USD 500 million order differently than one clearing a few tens of billions. That is the whole of the depth argument, and it holds. It produces a larger price move per unit of flow, and says nothing about whether flow arrives. During the 209-day stretch, the coefficient had not changed. Flow had.

The funding leg usually moves first

Carry positions are the mechanism that converts a rate expectation into a currency order. When the borrowed currency appreciates, the position must be closed, and closing it means buying the funding currency and selling the destination currency, regardless of what the destination economy is doing. Yen credit contracting 4.9% during 2025 is the visible trace of positions being unwound in an orderly way. The disorderly version is what makes headlines.

The dollar is not the low-volatility asset by construction

The 10.8% first-half decline in 2025 is the cleanest available counterexample. The dollar's 89.2% turnover share and 57.13% reserve share did not prevent it. Depth changes the shape of a repricing, spreading it over more sessions and more counterparties, but it does not cap the size.

The August 2024 Precedent and What It Does Not Cover

The most cited recent template for an emerging market currency shock is the yen carry unwind of early August 2024. BIS analysis records a VIX that briefly exceeded 60 in pre-market trading before closing under 40, a jump far exceeding what the historical relationship with equity returns implied. Speculative short yen futures positions unwound visibly. The Mexican peso and Brazilian real, both destination currencies, saw bouts of depreciation. The offshore renminbi appreciated alongside the yen and Swiss franc, behaving as a funding currency rather than a destination one.

That episode is a positioning shock: it began in the funding leg, propagated through margin calls, and resolved over days once positions were flat. The template it supplies is to identify who is short the funding currency, size the position, and expect a fast, largely mean-reverting reversal.

The shock that ended the 209-day volatility streak in March 2026 is a different object. The IMF's April 2026 outlook attributes the downgrade to a Middle East conflict that began in late February 2026, disrupting maritime and air traffic. It cut projected global growth to 3.1% for 2026 from a pre-conflict 3.4%, with 3.2% penciled for 2027, advanced economies at 1.8% then 1.7%, and emerging market and developing economies at 3.9% then 4.2%. Oil is assumed to rise 21.4% in 2026 to an average of USD 82 per barrel, with the reference forecast conditioned on the conflict resolving by mid-2026. Global headline inflation is projected at 4.4% for 2026 before easing to 3.7%, with emerging markets at 5.5% then 4.6%.

A terms-of-trade shock does not mean-revert on the same schedule as a positioning shock. The IMF is explicit that commodity-importing emerging market and developing economies are at risk of being hit harder, with currency depreciation compounding higher energy and food prices. That is a persistent adjustment, not a squeeze. Applying the August 2024 playbook to it is a category error, and it is the specific error the 2026 setup invites.

The Version of This That Is Wrong

Several objections to the framing above are serious enough to state plainly.

209 days is one observation. A single streak, however long in the record, is not a distribution. It is entirely consistent with the structural view that emerging market volatility is usually higher and that 2025 was an outlier driven by an idiosyncratic dollar repricing. Anyone reading the streak as proof that the structural gap has closed is over-fitting to one window.

Volatility indices are not the underlying. The gauges compare option-implied volatility across baskets whose composition is set by the provider. Basket weights, the currencies included, and the tenor all affect the comparison. A crossover between two constructed indices is a fact about those indices before it is a fact about currencies.

The aggregate hides the dispersion. Emerging market currency turnover reaching 29% of the global total is driven substantially by the renminbi, the Hong Kong dollar and the Singapore dollar. Turnover as a share of GDP plateaued for the median emerging market economy. A currency at the thin end of the distribution has seen little of that improvement, and for that currency the classical depth argument holds close to unmodified.

Managed regimes suppress measured volatility without removing risk. Where a currency is heavily managed, low realised volatility reflects intervention capacity, not market equilibrium. The risk is displaced into reserve drawdown, capital account measures, or a step adjustment. Reading a flat quote as an absent risk is precisely the mistake such regimes are designed to encourage in the short run.

Reserve shares are noisy at quarterly frequency. The move from 56.42% to 57.13% is within the range that valuation effects alone can produce when the dollar strengthens against other reserve currencies. Building an argument on one quarter of COFER is not defensible in either direction.

Risk Factors Worth Watching

• A sustained oil move well beyond the USD 82 assumption. The IMF reference case conditions on de-escalation by mid-2026. If that condition fails, commodity importers face a widening current account gap and the depreciation channel reasserts itself with force. This is the highest-probability path to a sustained volatility gap reopening.
• Rapid dollar credit growth reversing. An 8.5% annual expansion is the fastest since Q3 2014. Credit growing quickly can stop quickly. A sharp deceleration in the USD 4.3 trillion stock owed by emerging market borrowers would tighten conditions without any policy rate changing.
• A yen rate differential shift. Yen credit contracted 4.9% in 2025. A further compression that forces the remaining carry positions to close would reproduce the August 2024 mechanism, with the same destination currencies on the wrong side.
• Stress in the off-balance-sheet leg. FX swap obligations are the least observable part of the system, and the part that failed to roll smoothly in prior dollar funding squeezes. Deterioration shows up in cross-currency basis before spot.
• Concentration in the trading location. With 75% of turnover booked in four jurisdictions, an operational or regulatory disruption in any one of them is a liquidity event independent of any macro shock.

What to Watch Next Week

• The EM-minus-G7 volatility spread. Not the level of either gauge, the spread. Whether the March 2026 crossover persists or reverts is the single cleanest read on whether the regime has actually changed.
• Cross-currency basis in the major emerging market pairs. Widening basis is the early tell for dollar funding pressure and typically leads spot by days.
• Brent relative to the USD 82 annual assumption. Sustained trading materially above that level puts the IMF reference case under strain and shifts the balance toward the persistent-adjustment scenario.
• Central bank reserve reporting from large commodity-importing emerging markets. Drawdowns are the substitute for visible depreciation under managed regimes.
• Any BIS or national central bank commentary on FX swap rollover conditions. This is the least-watched and most informative of the five.

Concrete Framework — Where to Look First

A monitoring sequence that runs from slow variables to fast ones, rather than the reverse.

1. Fix the depth baseline once, then leave it. The BIS Triennial updates every three years. Record the relevant turnover shares from the April 2025 round and treat them as constant until the next survey. Re-deriving depth from daily price action is circular.
2. Track the funding stock quarterly. BIS global liquidity indicators, released roughly four months after quarter-end. The two numbers that matter are total dollar credit to non-bank borrowers outside the United States and the emerging market portion of it, currently USD 14.3 trillion and USD 4.3 trillion.
3. Log COFER quarterly but assign it low weight. Shares move in tenths of a percentage point and are contaminated by valuation effects. Useful over eight quarters, close to meaningless over one.
4. Classify every shock before applying a template. Positioning shock or terms-of-trade shock. The first is fast and largely mean-reverting; the second is slow and persistent. The August 2024 playbook applies only to the first.
5. Separate managed from floating before comparing volatility. A currency under active management and one that floats freely are not comparable on realised volatility. Check reserve adequacy and intervention history first.
6. Write the opposite scenario down. For any expected move, state the conditions under which the reverse happens and what would have to be observed first. If those conditions cannot be specified, the view is not yet a view.
7. Review the spread monthly, not daily. The EM-minus-G7 volatility spread is noisy at daily frequency. Monthly closes over a rolling twelve months carry most of the regime information with a fraction of the false signals.

The honest summary is that the structural depth gap is real, measurable and slow-moving, while the volatility gap it is used to explain is fast-moving and spent most of the past year inverted. Both statements are true at once. A framework that cannot hold both is not describing the market.

This article is macroeconomic analysis for general information. It is not investment, financial or trading advice, and it does not constitute a recommendation regarding any currency, security or instrument.