Emerging market bonds and geopolitical risk: what each crisis does to them
We model 7 crises. For each, the model gives a typical holding in every sector a move in three cases, from calm to severe. Here is em bonds in each, next to the odds of that case and how sure we are of the number.
What each crisis does to emerging market bonds
Strait of Hormuz / Iran Conflict EMPIRICAL
| Case | Odds | EM bondsmodelled move |
|---|---|---|
| Calm caseNormalisation | about 1 in 4 (unlikely) | ▼ −3.3% |
| Middle caseArmed Truce | about 1 in 2 (roughly even chance) | ▼ −8.1% |
| Severe caseRe-escalation / Second Closure | about 1 in 4 (unlikely) | ▼ −21.0% |
These sensitivities were measured against the real 2026 Hormuz war.
Taiwan Strait / Semiconductor Chokepoint UNPRICED
| Case | Odds | EM bondsmodelled move |
|---|---|---|
| Calm caseGray-Zone Pressure | about 6 in 10 (likely) | ▼ −1.7% |
| Middle caseQuarantine / Blockade | about 1 in 4 (unlikely) | ▼ −21.4% |
| Severe caseInvasion / Fab Denial | about 1 in 8 (very unlikely) | ▼ −44.9% |
No Taiwan blockade has ever happened, so these are reasoned estimates anchored to Bloomberg Economics — a map, not a forecast.
Red Sea / Bab el-Mandeb UNPRICED
| Case | Odds | EM bondsmodelled move |
|---|---|---|
| Calm caseSuez Return | about 1 in 7 (very unlikely) | 0.0% |
| Middle caseSaudi Blockade Enforced | about 7 in 10 (likely) | ▼ −2.8% |
| Severe caseStrait Closed | about 1 in 5 (unlikely) | ▼ −4.5% |
The Red Sea reroute happened, but listed shares did not price it, so these are reasoned sensitivities to world growth — a map, not a forecast. Shipping shares carry no figure.
Korean Peninsula conflict UNPRICED
| Case | Odds | EM bondsmodelled move |
|---|---|---|
| Calm caseTense Standoff | about 9 in 10 (very likely) | 0.0% |
| Middle caseRegime Collapse | about 1 in 14 (very unlikely) | ▼ −3.5% |
| Severe caseWar | about 1 in 33 (almost no chance) | ▼ −27.3% |
A Korean war has never happened, so these are estimates anchored to Bloomberg Economics, with share effects borrowed from the Taiwan model: a map, not a forecast.
US tariff shock EMPIRICAL · one real event, replayed
| Case | Odds | EM bondsmodelled move |
|---|---|---|
| Calm caseRelief | about 1 in 5 (unlikely) | — |
| Middle caseCurrent Law | about 6 in 10 (likely) | 0.0% |
| Severe caseRe-escalation | about 1 in 5 (unlikely) | ▼ −3.2% |
Relief: no figure. We do not size relief: the relief we measured (the May 2025 truce, the February 2026 ruling) moved shares in ways no rule we tested could predict.
Measured from one real event: the re-escalation case replays what shares did in the week after April 2025's tariffs, as they moved then. Not scaled to any other size.
Russia–Ukraine war EMPIRICAL · one real event, replayed
| Case | Odds | EM bondsmodelled move |
|---|---|---|
| Calm caseCeasefire | about 1 in 7 (very unlikely) | — |
| Middle caseGrinding War | about 8 in 10 (likely) | 0.0% |
| Severe caseWider War | about 1 in 10 (very unlikely) | ▼ −6.0% |
Ceasefire: no figure. We do not size a ceasefire: past peace news (the 2022 Istanbul talks, the grain deal, the 2025 Trump–Putin talks) moved shares in ways our test could not predict.
Measured from one real event: the wider-war case replays what shares did in the first two weeks after the 2022 invasion, as they moved then. Not scaled to any other size.
No figure for em bonds in: Panama Canal drought. We give the sector no number there. That is not the same as saying it does not move.
What actually happened
- In the 2026 Iran war: −0.8% against the S&P 500, measured from daily prices. How it was measured →
The names we score in this sector
Coverage, not picks: these are the shares and funds the Workspace scores as em bonds.
iShares J.P. Morgan USD Emerging Markets Bond ETF (EMB), VanEck J.P. Morgan EM Local Currency Bond ETF (EMLC)
Your own holdings
The figures are for a typical holding in the sector. To see how each crisis reaches the shares you hold, add them in the Workspace: free, no sign-in needed.
We measure the risk. You make the call.
DELAX describes exposure. It is not investment advice. The modelled moves are a map of exposure, not price forecasts.