Timor-Leste × Bangladesh Stablecoin Settlement Bridge
A corridor case study: USD-pegged wholesale stablecoin bridge between Timor-Leste (USD-dollarised, UTC+9, thin correspondent banking) and Bangladesh (BDT, bKash/BEFTN, UTC+6). This page analyses three settlement architectures (wholesale CBDC bridge, stablecoin escrow, nostro pre-funding), scored against a 3-hour UTC overlap constraint, with collateral pool sizing, repo sweep mechanics, and an ISO 20022-aligned double-entry ledger schema. Grounded in BIS Project Agorá, IMF AREAER 2024, and World Bank Remittance Prices data. To run live corridor inputs, use the AINumbers emerging-market-fx-corridor chain ↗.
- Frontier Markets · Settlement
- T75 · T58 · T218 · T217 · T221 · T220 · T157 · T325
- USD-Stablecoin Bridge
- ISO 20022
- Zero PII
- UTC Overlap
- 3.0 h
- Business-hours window
- TLS GDP 2024
- ~$2.0B
- USD-dollarised economy
- BGD Remittance
- $22.4B
- Annual inflow, 2024
- Corridor Hops
- 3–5
- Correspondent intermediaries
- Architectures
- 3
- CBDC · Stablecoin · Nostro
- Schema
- ISO 20022
- pacs / camt aligned
Three Options Scored Against Corridor Constraints
Structural constraints: USD dollarization (TLS), BDT capital controls (BGD), thin correspondent network, 3-hour UTC overlap
USD-Pegged Stablecoin Escrow
Recommended architectureUSD-pegged wholesale stablecoin on a permissioned ledger (Corda or Fabric). TLS correspondent locks USD collateral into an escrow smart contract; Bangladeshi bank releases BDT via bKash/BEFTN on confirmed settlement signal. Eliminates the 3–5 hop SWIFT chain; collapses settlement to near-real-time within the UTC overlap window.
- Settlement latency
- 45–90 s
- All-in cost
- 0.45–0.85%
- Herstatt risk
- Eliminated (PvP)
- Capital controls fit
- High
Wholesale CBDC Bridge
Conditional: requires BDT wholesale CBDCA bilateral CBDC bridge (BIS Project Agorá pattern) requires Bangladesh Bank to issue a wholesale BDT CBDC, which is not in active deployment as of mid-2026. Viable at best cost/latency if the programme advances; requires Treaty-level interoperability and shared settlement infrastructure.
- Settlement latency
- 8–15 s
- All-in cost
- 0.30–0.55%
- BDT CBDC prerequisite
- Not live
- Timeline to viability
- 36–48 months
Nostro Pre-Funding (Baseline)
Current state: legacy comparisonTLS correspondent pre-funds a Dhaka nostro in USD. Familiar to both banking systems but ties up capital, exposes both legs to Herstatt risk across the 48–120 h window, and produces 4.2–6.5% all-in cost through a 3–5 correspondent hop chain.
- Settlement finality
- 2–5 days
- All-in cost
- 4.2–6.5%
- Herstatt window
- 48–120 h
- Trapped capital
- High
3-Hour Business-Hours Window
Timor-Leste UTC+9 · Bangladesh UTC+6 · simultaneous business hours 04:00–07:00 UTC
Timor-Leste (UTC+9, 09:00–17:00 local) and Bangladesh (UTC+6, 10:00–18:00 local) share only a ~3-hour window of simultaneous business hours. This constraint is critical for any architecture requiring human confirmation steps or same-day nostro reconciliation. SWIFT MT103 cutoff in Dhaka falls outside the Dili close window. Any instruction initiated after 14:00 Dili time cannot receive same-day confirmation in Dhaka.
- UTC Overlap Window
- 3.0 h
- 04:00–07:00 UTC · Mon–Fri
- Intraday Liquidity Buffer
- 140%
- Of peak daily volume, minimum recommendation
- SWIFT MT Cutoff Conflict
- Yes
- Dili close falls outside Dhaka MT processing window
- D+0 Feasibility
- Stablecoin only
- CBDC and nostro architectures cannot achieve D+0
The stablecoin escrow is the only architecture achieving D+0 finality within the overlap window. Smart contract settlement is asynchronous: the BDT release signal fires on confirmed contract execution even outside Dhaka banking hours, provided the bKash/BEFTN integration handles the queued credit. This is the key structural advantage over both CBDC and nostro architectures for this corridor pair.
Pool Sizing, 3× Spike Stress & Repo Sweep
BIS CPMI guidance: frontier-market corridors require 3× baseline stress sizing
The stablecoin escrow requires a collateral pool held at a TLS-side custodian in USD. Pool sizing must cover normal daily flow plus a 3× spike (remittance surge or corporate FX event). A repo sweep facility converts excess collateral into overnight yield. Under GENIUS Act compliance, eligible reserve assets must have ≤93-day maturity; overnight repo or T+1 MMF shares qualify.
- Example daily volume
- $5M
- Trade-flow scenario
- Normal collateral pool
- $7M
- 140% coverage ratio
- 3× spike requirement
- $21M
- Remittance surge or FX event
- Repo sweep yield (ann.)
- 4.8–5.2%
- Overnight UST / MMF · GENIUS Act eligible
Net cost of maintaining the collateral pool after repo sweep: ~0.3–0.5% annually on the stressed pool size. Total corridor economics: 0.45–0.85% all-in vs 4.2–6.5% on legacy correspondent rails, a 70–80% cost reduction.
Double-Entry Ledger Schema: pacs.008 / camt.053 Aligned
Five account types · IFRS / BIS Triennial conventions · ISO 20022 message identifiers
Settlement generates five account types in a double-entry ledger structured to align with ISO 20022 pacs.008 (credit transfers) and camt.053 (bank-to-customer statement). The schema supports real-time gross settlement at the wholesale layer with netting at the retail bKash disbursement layer.
ACCOUNT STRUCTURE // pacs.008 / camt.053 aligned nostro USD nostro at TLS correspondent Dr on send vostro BDT vostro at Dhaka agent bank Cr on receive confirmation collateral Stablecoin escrow pool (USD) Dr on lock; Cr on release stable_float Issued stablecoin in circulation Cr on mint; Dr on burn fx_revaluation USD/BDT mark-to-market (P&L) Daily sweep // Posting sequence — $500K trade settlement Step 1 — Lock collateral Dr collateral 500,000 USD // Escrow contract funded Cr nostro 500,000 USD Step 2 — Mint stablecoin Dr stable_float 500,000 USDTL // Issued 1:1 against collateral Cr collateral 500,000 USD Step 3 — PvP atomic swap (pacs.008) Dr vostro_bdt [BDT equiv] // Atomic — simultaneous with Step 2 burn Cr stable_float 500,000 USDTL Step 4 — FX revaluation (pacs.009 / camt.054) Dr/Cr fx_revaluation [gain/loss] // Daily BDT/USD MTM
Why this corridor is structurally hard
Timor-Leste runs on USD without a correspondent network of scale. Bangladesh runs BDT through BEFTN / bKash with capital controls and weak USD liquidity in-country. Together that is a 3-hop SWIFT route, 5–7 day finality, 4–6% all-in cost, and a 3-hour UTC overlap.
A USD-pegged stablecoin on a permissioned wholesale rail collapses the hops to one, provided collateral reserves, intraday liquidity, and the repo sweep are sized for the spike.
Refs: BIS Project Agorá, BIS Project Mariana, IMF AREAER 2024, World Bank Remittance Prices Q3 2024, IOSCO/CPMI principles for stablecoin arrangements (2022).
Run the live emerging-market FX corridor model
The AINumbers emerging-market-fx-corridor chain sequences all eight tools referenced in this case study (T58, T75, T157, T217, T218, T220, T221, T325) and exports a composite Policy Mandate. Use it to run your own corridor inputs against the same analytical framework.
Designing issuance or settlement architecture?
MiCA / GENIUS readiness, FATF Travel Rule, tokenized RWA and wholesale settlement corridors. We’ve advised on tokenized rail implementations across frontier markets. Let’s talk.
Post Oak Labs · advisory on production tokenized payment deployments · emerging and frontier markets worldwide