Skip to the content
HomePUBLIC-EVENT RECONSTRUCTION

Public Event Reconstruction

Bab-el-Mandeb: How a Chokepoint Shock Propagates

A reconstruction of how disruption at a strategic maritime chokepoint propagates through routing, transit time, fleet utilisation, freight capacity, inventory requirements, working capital, and downstream availability.

Reconstruction Brief0:00 / 1:11

Methodological disclaimer

This reconstruction is an analytical demonstration using publicly available information. It is not a client engagement and not a record of contemporaneous advice. The event chronology is drawn from public reporting; the analytical outputs are a reconstruction of method, not a historical observation. Where a value is illustrative it is labelled as such.

Decision support

The decision this reconstruction bears on

The decision this serves

Big betComplex
  • Big bet · Argue the case against before committing: alternatives, what must be true, and what would make this wrong.
  • Complex · Probe, sense, respond: define the probe, its stopping condition, and what would change the reading.
Who faces it
Importers, manufacturers and the lenders exposed to the Europe–Asia container string.
The choice
Whether to hold additional inventory at destination, diversify the routing mix, or re-sequence production before the next disruption.
Standing
The reading is available on public evidence. The exposure it implies is not quantified here, and no figure is asserted that the public record does not carry.

The case against this reading

If carriers restore effective capacity faster than this reconstruction assumes — through reallocation, faster steaming or new tonnage entering the string — the schedule shock is shorter than modelled, and buffer held against it is cost without return. If destination cover is deeper than the public record shows, these propagation paths describe a system the reader is not in. Both cases are stated rather than argued away.

The question

What happens to downstream inventory, transit time, freight economics, and exposed industries when traffic through the Bab-el-Mandeb is materially disrupted?

Period

The reconstruction window runs from the onset of sustained disruption in late 2023 through the following twelve months.

Geography

The Bab-el-Mandeb and southern Red Sea, the Cape of Good Hope routing, European and Mediterranean destination nodes, and Asian origin ports.

Industries considered

  • ·Container shipping and freight forwarding
  • ·Retail and consumer goods importers
  • ·Automotive and industrial manufacturing
  • ·Energy and refined products
  • ·Port and terminal operations

Inside the model

  • Vessel routing and effective capacity
  • Transit time and schedule reliability
  • Freight pricing and war-risk insurance
  • Inventory in transit and destination cover
  • Working-capital timing

Outside the model

  • Air freight as a full substitution path (treated as an indicator only)
  • Domestic road and rail distribution inside destination markets
  • Contractual and legal remedies between counterparties
  • Second-order political consequences beyond trade flow

A boundary that is never written down cannot be challenged. These exclusions are a design decision, not an oversight.

The system

How the shock travels

Select any node to inspect its definition, its upstream and downstream relationships, the evidence behind it, and its limitations.

Modelled shock originBab-el-MandebCape of Good Hope diversionAsian origin portsGulf of AdenRed SeaSuezMediterranean & European destinations

The modelled sequence

  1. Bab-el-Mandeb
  2. Red Sea Routing
  3. Transit Time
  4. Vessel Utilisation
  5. Available Freight Capacity
  6. Freight Economics
  7. Inventory Requirements
  8. Working Capital
  9. Regional Availability
  10. Downstream Business Effects

The chain is the structure of the model, not an observation of traffic. Each node carries its own evidence, confidence and limitations, shown in full below.

Schematic · Indicative geometry · Not a map · Not live

No vessel positions, current volumes or present conditions are shown. The corridor is drawn to make the model's structure legible, not to represent geography at scale.

The strait between the Horn of Africa and the Arabian Peninsula, carrying a large share of Europe–Asia container and energy traffic.

Upstream

— none —

Downstream

routing

Indicators

Transit volume against trailing baseline · Vessel targeting reports

Confidence

High

Evidence

Public maritime traffic reporting and published trade route data.

Limitation

Volume shares vary by source and by quarter; the model uses a range, not a point value.

propagates to
propagates to
propagates to
propagates to
propagates to
propagates to
propagates to
propagates to
propagates to

The shock

What was observed, inferred, assumed, and unknown

These categories are never blurred. An inference presented as an observation is the most common failure in this kind of analysis.

Observed

  • Sustained targeting of commercial shipping in the southern Red Sea from late 2023.
  • Major container lines suspending Red Sea transits and routing via the Cape of Good Hope.
  • Naval coalition escorts announced, with transits continuing for some operators.

Inferred

  • That diversion would become the default routing for a sustained period rather than a short interruption.
  • That effective capacity, not just transit time, would be the binding constraint.

Assumed

  • That freight economics would transmit to inventory policy within one planning cycle.
  • That destination markets would absorb variance through buffer rather than through service reduction.

Uncertain

  • Duration of the disruption.
  • Whether a durable escort regime would restore carrier confidence.
  • The speed at which shippers would substitute sourcing rather than absorb cost.

Observation timeline

Late 2023OBSERVED

Targeting of commercial shipping in the southern Red Sea begins and is publicly reported.

Source
Public reporting
Relation to prior assessment
Shock onset — the initial condition of the reconstruction.
December 2023SUPPORTED

Major container lines suspend Red Sea transits; Cape of Good Hope routing is adopted on affected strings.

Source
Carrier service advisories
Relation to prior assessment
Confirms the routing layer of the model.
Early 2024PARTIALLY SUPPORTED

Sustained diversion persists; effective capacity absorption becomes the dominant commercial effect.

Source
Published freight indices and schedule data
Relation to prior assessment
Confirms the capacity layer; the model had ranked this second to transit time.
Through 2024REQUIRES REVISION

Air freight substitution is more limited than the model's substitution path implied.

Source
Published air cargo data
Relation to prior assessment
Challenges the substitution pathway.
Through 2024UNRESOLVED

Working-capital effects are visible in some sectors but not traceable at the resolution the model implied.

Source
Sector reporting
Relation to prior assessment
Unresolved — the layer could not be tested with public data.

Propagation paths

Four routes through the system

Each pathway states its evidence, its confidence, the indicator that would confirm it, and the condition under which it fails.

Path 01

Capacity absorption

  1. Routing disruption
  2. Longer transit
  3. Reduced effective capacity
  4. Higher freight pressure
Evidence
Routing distance, service rotations, published freight indices.
Confidence
High
Observable indicator
Capacity absorbed as a share of deployed fleet.
Possible failure mode
New-build deliveries offset absorption faster than modelled.
Path 02

Inventory in transit

  1. Longer transit
  2. More inventory in transit
  3. Higher working-capital requirement
  4. Inventory planning change
Evidence
Transit extension, stated payment terms (inferred).
Confidence
Low to moderate
Observable indicator
Days of goods in transit; destination cover by category.
Possible failure mode
Firms absorb the extension inside existing buffer without changing policy.
Path 03

Lead-time variance

  1. Freight disruption
  2. Supplier lead-time variance
  3. Production planning uncertainty
  4. Downstream availability pressure
Evidence
Sector reporting on lead times.
Confidence
Low to moderate
Observable indicator
Lead-time variance on affected supplier lanes.
Possible failure mode
Substitution at the sourcing layer removes the pressure before it reaches production.
Path 04

Sourcing substitution

  1. Regional exposure
  2. Alternative sourcing
  3. Supplier substitution
  4. Cost and lead-time change
Evidence
Trade flow data; published sourcing announcements.
Confidence
Low
Observable indicator
Share of affected volume re-sourced.
Possible failure mode
Substitution is slower and narrower than assumed — this is what occurred.

Indicators to watch

What would confirm or challenge the assessment

No live values are shown. Where an indicator cannot be observed with public data, it is marked as such rather than estimated.

Vessel routing

Rising
Current
Elevated diversion
Baseline
Trailing 12-month norm
Confidence
High
Timestamp
Reconstruction window
PUBLIC SOURCEPublic maritime traffic reporting

Primary observable for the routing layer.

Transit time

Rising
Current
Extended
Baseline
Pre-disruption schedule
Confidence
High
Timestamp
Reconstruction window
PUBLIC SOURCECarrier schedule data

Drives capacity absorption.

Freight capacity

Tightening
Current
Absorbed
Baseline
Deployed fleet norm
Confidence
Moderate
Timestamp
Reconstruction window
ILLUSTRATIVEModel output — ILLUSTRATIVE

Binding constraint in the model.

Freight pricing

Rising
Current
Elevated
Baseline
Trailing index
Confidence
Moderate
Timestamp
Reconstruction window
PUBLIC SOURCEPublished freight indices

Transmits cost into landed cost.

War-risk insurance

Rising
Current
Elevated
Baseline
Pre-disruption premium
Confidence
Moderate
Timestamp
Reconstruction window
PUBLIC SOURCEInsurance market reporting

A cost layer the model under-weighted.

Inventory cover

Unresolved
Current
Not observed at firm level
Baseline
Seasonal norm
Confidence
Low
Timestamp
—
DATA NOT AVAILABLEDATA NOT AVAILABLE

Cannot be tested with public data.

Supplier lead times

Unresolved
Current
Variable
Baseline
Contracted lead time
Confidence
Low
Timestamp
Reconstruction window
PUBLIC SOURCESector reporting

Partially observable through sector reporting.

Air freight substitution

Weaker than modelled
Current
Limited
Baseline
Pre-disruption share
Confidence
Moderate
Timestamp
Reconstruction window
PUBLIC SOURCEPublished air cargo data

Challenged pathway 04.

Consequence map

What the model identified

Consequences are not presented as equally certain. Confidence degrades down the chain, and the lowest tier is stated plainly.

Primary consequence

  • ·Effective container capacity absorbed by longer voyage rotations.
  • ·Elevated freight and insurance cost on affected strings.

Secondary consequence

  • ·Schedule variance at destination nodes, absorbed initially by buffer.
  • ·Working-capital extension where goods remain in transit longer.

Conditional consequence

  • ·Category-level availability pressure — conditional on buffer depth and service-level policy.
  • ·Production planning change — conditional on supplier concentration and substitution options.

Low-confidence / uncertain

  • ·Attribution of margin effects specifically to corridor cost versus other concurrent inputs.
  • ·Duration of the disruption and the point at which routing normalises.

Scenario framing

Scenario framing

Short interruption

Assumption

Diversion reverses within one planning cycle.

Exposed systems

Freight and insurance cost only.

Sustained disruption

Assumption

Diversion persists across multiple planning cycles.

Exposed systems

Capacity absorption, inventory policy, working capital.

Persistent disruption

Assumption

Diversion becomes structural, with network reconfiguration.

Exposed systems

Sourcing structure, network design, capital allocation.

Scenario outputs are distributions and conditional pathways, not predictions. Confidence is conditional on the stated boundary, the available evidence, and the model assumptions.

Assessment record

The full record

Assessment date
Reconstruction window — analytical demonstration
System boundary
See system boundary above.
Initial conditions
Established Europe–Asia container routing through the Bab-el-Mandeb, with destination buffer held at seasonal norms.
Shock / scenario
Sustained disruption to commercial transit through the strait.
Propagation pathways
Four propagation pathways — capacity absorption, inventory in transit, lead-time variance, sourcing substitution.
Expected observable indicators
Diversion share, transit days, capacity absorption, freight pricing, air-freight substitution share.
Scenario distribution
Three scenario bands rather than a point forecast: short, sustained, persistent.
Confidence
High at the routing and transit layers; low at the inventory, capital, and availability layers.
Known limitations
Firm-level inventory and payment terms are not public. Substitution behaviour was under-modelled.
Subsequent observations
See the observation timeline above.
Model revision
Capacity layer promoted; insurance layer added. See the revision sequence below.
What was missed
Duration, insurance transmission speed, and the weakness of substitution.

Where the assessment failed

Publishing the failure modes is the point, not an admission against interest. An assessment that cannot show where it broke cannot be trusted where it held.

Duration

The model treated the disruption as bounded by a short escort-and-normalise cycle. It persisted far longer, which moved the binding constraint from transit time to capacity absorption.

Insurance as a cost layer

War-risk premiums were modelled as a minor input. They proved material and moved faster than freight indices, so the cost layer was under-weighted.

Substitution speed

Pathway 04 assumed sourcing and modal substitution would relieve pressure. Substitution was slower and narrower than assumed, so the relief did not arrive.

Resolution mismatch

Working-capital effects were asserted at a resolution the available data cannot support. The layer should have been marked unresolved from the outset rather than modelled.

What changed

Initial assessment

Assessed as a transit-time shock, with capacity and inventory effects following at one remove.

New observation

Sustained diversion made capacity absorption the dominant commercial effect, and insurance moved ahead of freight pricing.

Model revision

The capacity layer was promoted above transit time in the causal ordering; insurance was promoted from a cost line to a distinct layer.

Updated consequence map

The revised ordering is reflected in the node sequence on this page: routing → transit → utilisation → capacity → freight economics.

Limitations

What this reconstruction cannot establish

  • ·Firm-level inventory, payment terms, and margin data are not public, so the lower half of the model is inferred rather than observed.
  • ·Attribution is contested: corridor cost is one input among several concurrent macro factors.
  • ·The model is calibrated to public sources only; no proprietary data was used.
  • ·Scenario outputs describe conditional consequences under stated assumptions — they are not predictions.
  • ·Confidence is materially lower for the inventory, working-capital, and availability layers.

Model output is not a single-point forecast

The Institute models relationships and conditional consequences. Outputs describe distributions, pathways, indicators, and uncertainty rather than asserting that one predetermined future outcome will occur.

Model

≠

Certainty

Scenario

≠

Prediction

Indicator

≠

Outcome

Bring us a system.

The same method applies to a defined decision, a shock, an exposure, or an operating question. We begin by writing down the system boundary and stating what cannot be established.

Start with the question