A pre-funding-free remittance rail between the UAE and India — settled in e‑Rupee and a regulated AED token, bound by atomic payment-versus-payment. Opposing flows fund each other locally; only the net imbalance ever needs to settle.
Discussion draft — counterparties, volumes and regulatory characterisations are indicative and subject to confirmation.
Cross-border retail payments are slow and expensive not because messages travel slowly, but because providers must lock working capital in destination accounts before value has arrived. Faster messaging — or swapping fiat for stablecoins — changes the instrument, not the requirement.
Each matched pair settles through an asymmetric atomic PvP: a native hashlock on the public-chain AEDZ leg, an emulated conditional hold on the e-Rupee leg at a regulated settlement-node bank. The regulated party always moves first. Click through the lifecycle:
A sender’s order enters as a marketable limit order, priced within a volatility-elastic band anchored to the interbank reference rate. A multilateral netting cycle (~5 minutes, dynamic) pairs opposing AED→INR and INR→AED flows, with partial and divisible fills to maximise the match ratio.
Rate certainty for the consumer; no AMM, no slippage, no front-running.
Move the sliders. The model is illustrative, but the structure is the point: under correspondent banking, working capital is a multiple of daily gross flow; under netted PvP it collapses to the unmatched residual plus a marginal timing buffer.
Every gross leg rides a framework RBI already supervises — RDA or MTSS for inbound personal remittances, LRS via an AD-II for outbound. Money already does not cross the border under these schemes. The rail changes how the legs are funded, not how they comply.
Permission to net an independently-compliant inbound e-Rupee leg against an independently-compliant outbound e-Rupee leg. Each gross leg stays scheme-compliant; only liquidity is netted — and programmability gives the regulator a per-leg audit trail that no correspondent chain can offer.
| Dimension | Correspondent / SWIFT today | This rail (v1 sandbox) |
|---|---|---|
| Working capital | Pre-funded nostro, scales with gross volume | Net imbalance + marginal timing buffer |
| Settlement risk | Managed by delay and exposure limits | Eliminated by construction (atomic PvP) |
| Speed | T+1–3, batch cut-offs, business hours | Near-real-time, 24/7 netting cycles |
| Auditability | Fragmented across intermediaries | Per-leg, on-ledger, purpose-coded |
| Consumer rate | Margins stack across the chain | One firm quote in an IBR-anchored band |
| Compliance perimeter | Established, well understood | Unchanged — same schemes, same originator obligations |
The building blocks are individually proven in BIS Innovation Hub work. What has never been done is joining them on a live retail corridor — with a regulator-grade liquidity answer. India’s e-Rupee pilot makes the AED–INR corridor the first place this can exist in production.
The sandbox application is deliberately bounded to what must be proven. The network properties the infrastructure unlocks are claimed openly — and committed as post-sandbox extensions, not smuggled into v1.
Saber builds cross-border payments infrastructure and operates regulated money movement at scale today. The rail proposed here is not a whitepaper exercise — it is the next architecture for flows we already run.
The full concept note, technical annex and legal comparison note are available on request. We’d welcome a working session with your team.
edul@mudrex.com