The Press Release Problem

Every six months, a new announcement lands: a pan-African payment system is live, a cross-border settlement corridor has launched, real-time intra-African payments are finally here. The press coverage is enthusiastic. The reality, if you work in institutional payments and actually trace payment flows across the continent, is considerably more qualified.

I want to be precise about this. The infrastructure is improving. PAPSS is real, it processes transactions, and it is a genuine structural development. Mobile money interoperability between certain markets is functional and growing. Progress is happening.

What is not happening is the transformation the announcements describe. The gap between what the press releases say and what the correspondent banking data shows is the story of African cross-border payments in 2026. Understanding that gap is the prerequisite for building anything in this space that actually works.

What PAPSS Actually Is — And What It Is Not

The Pan-African Payment and Settlement System was developed under AFREXIMBANK and formally launched for commercial operations in 2022. The architecture is sound: a multilateral settlement system that enables banks in member countries to settle intra-African transactions in local currencies, bypassing the need to route through New York or London correspondents.

In theory, a Ghanaian bank paying a Kenyan supplier should be able to settle directly in cedis-to-shilling, with PAPSS clearing the cross-currency leg. No US dollar intermediary. No correspondent in a third country. Lower cost, faster settlement, more transparent.

In practice, PAPSS adoption as of early 2026 is concentrated in West Africa, where the WAMZ monetary zone provides a pre-existing regulatory framework for multi-currency settlement. Nigeria, Ghana, The Gambia, Sierra Leone, Guinea, and Liberia have functional connectivity. East Africa — a region with arguably more active SME cross-border trade volumes — has limited connectivity. The East African Community's integration agenda has not yet produced the central bank bilateral agreements that PAPSS requires to process transactions.

The adoption gap matters because cross-border payment flows don't respect the map of which countries have joined which system. Nigerian exporters trade with Kenyan importers. Ghanaian fintechs process payments for Tanzanian consumers. When the settlement infrastructure doesn't cover the actual trade corridor, the "pan-African payment revolution" is a West African payment evolution — meaningful, but not the continental transformation the announcements imply.

The practical implication for banks: PAPSS is a usable rail for specific corridors now. It is not a replacement for correspondent banking relationships for the majority of intra-African payment flows. A payment strategy built on PAPSS alone, in 2026, will fail to cover most of the corridors it needs to serve.

The SME Corridor Problem: Where Correspondent Banking Is Actually Broken

The correspondent banking crisis in Africa is real, but its impact is not evenly distributed. For major institutional flows — sovereign bond transactions, multinational corporate treasury operations, large trade finance — correspondent chains remain functional, if expensive. The structural breakdown is most acute in SME corridors.

Consider the Nigeria-Kenya corridor. An SME exporter in Lagos needs to pay a supplier in Nairobi. The transaction size is $15,000 — too small for bespoke treasury operations, but significant for the businesses involved. The typical payment chain in 2026 looks like this:

  1. Lagos commercial bank → Nigerian correspondent in a Gulf bank (hop 1)
  2. Gulf bank → regional hub bank in Nairobi or Johannesburg (hop 2)
  3. Hub bank → Kenyan receiving bank (hop 3)

Three hops, three FX conversion events or dollar intermediation points, three sets of correspondent fees, and three points where the transaction can fail, delay, or sit in a compliance queue. End-to-end settlement time: two to five business days. Total cost: 3–5% of transaction value in a corridor that should cost 0.5–1%.

The reason this hasn't been fixed is structural. SWIFT's correspondent banking network deprioritised direct bilateral relationships between African institutions throughout the de-risking period of the 2010s. Building those relationships back requires capital commitment from both sides — Nigerian and Kenyan banks establishing nostro accounts with each other, negotiating bilateral FX agreements, and building joint compliance frameworks. That investment is not happening at the pace the trade volumes justify.

PAPSS partially addresses this for WAMZ-connected countries. For Nigeria-Kenya specifically, there is no short-term PAPSS solution. The practical bridge is the fintech layer — established payment aggregators have built proprietary correspondent networks that effectively pre-fund positions across corridors and aggregate liquidity. This is functional, but it introduces a new dependency on non-bank settlement infrastructure that carries its own counterparty and regulatory risk.

Why Instant Payment Schemes Stop at Borders

Domestic instant payment infrastructure in Africa is genuinely impressive. Nigeria's NIP (NIBSS Instant Payments), Kenya's RTGS and PesaLink network, Ghana's GhIPSS Instant Pay — these are functional, high-throughput, low-latency systems. Nigerian NIP processed over 9 billion transactions in 2023. These systems work.

The problem is architectural: domestic instant payment systems are designed for intrabank and interbank settlement within a single central bank's jurisdiction. Cross-border settlement requires:

  1. Foreign exchange conversion — which requires a market mechanism, pricing, and settlement in a shared currency or through bilateral currency agreements
  2. Dual regulatory compliance — the payment must satisfy AML/CFT requirements in both the originating and receiving jurisdiction
  3. Multilateral settlement — the net positions between all participating banks need to clear through a mechanism that all parties trust

Solving all three for a cross-border instant payment requires either a multilateral clearing house (PAPSS's model), pre-funded bilateral liquidity (the fintech aggregator model), or a shared settlement asset (the stablecoin or CBDC model). None of these solutions are trivially fast.

The honest answer is that "real-time" cross-border payments in Africa, for the majority of corridors, means same-day or next-day — not sub-second. The infrastructure simply doesn't support universal instant settlement across 54 currencies and regulatory jurisdictions. The "real-time" framing in press releases often refers to the UI experience (the sender sees a "payment sent" confirmation instantly) while the actual settlement takes hours or a day.

This distinction matters enormously for banks building treasury products, cash management solutions, or trade finance infrastructure. If your product design assumes instant cross-border settlement and the underlying rail delivers same-day, your product has a reliability problem you need to design around.

Stablecoins as Rails: Separating Signal from Noise

The stablecoin narrative in African cross-border payments has gone through several phases. The current phase — where stablecoins are presented as the solution to the correspondent banking gap — contains real signal alongside significant noise.

The signal: dollar-pegged stablecoins have demonstrably reduced the cost of certain remittance corridors by removing the correspondent banking intermediary entirely. A sender in the diaspora converting to a stablecoin and a recipient in Lagos or Accra converting out of it — if both sides have functional on-ramps and off-ramps — can execute a payment for 0.5–1% rather than the 5–8% that traditional wire transfers charge. This is real, it is growing in volume, and it is already affecting remittance competitive dynamics.

The noise: stablecoin rails for institutional B2B and financial institution payments remain constrained by three unsolved problems. First, regulatory status: FSCA in South Africa, CBN in Nigeria, and most African central banks have not issued clear frameworks for stablecoin settlement in institutional contexts. Banks cannot build core payment infrastructure on rails that may be regulated out of existence. Second, liquidity depth: the African stablecoin on-ramp/off-ramp infrastructure lacks depth for large transactions — a $2 million trade finance payment cannot currently be executed on stablecoin rails in most corridors without moving the market. Third, FX basis risk: a stablecoin denominated in dollars solves the correspondent banking problem but does not solve the local currency FX problem — the cedi-to-stablecoin and stablecoin-to-shilling legs still require FX conversion events.

The pragmatic view: stablecoins are a useful tool for specific use cases (diaspora remittance, small B2B payments in corridors with functional on-ramps) and a growing factor in competitive positioning for retail-adjacent payment products. They are not yet viable rails for institutional payment infrastructure at scale. Banks that have already allocated meaningful engineering resources to stablecoin integration for their core cross-border offering are likely two to three years ahead of when the regulatory and liquidity infrastructure will support that bet.

What Mid-Tier Banks Should Actually Build

Given the infrastructure reality — PAPSS functional in specific corridors, correspondent banking fragmented for SME flows, instant settlement not universally achievable, stablecoins useful but not institutionally viable — what is the right investment thesis for a mid-tier African bank trying to improve its cross-border payments capability?

The wrong answer, which I see proposed repeatedly: build another mobile wallet. Africa does not lack mobile wallets. The continent has dozens of functional mobile money schemes. The bottleneck is not the consumer-facing interface — it is the institutional plumbing underneath.

The right answer is a payment orchestration layer.

A payment orchestration layer sits between your customer-facing products (corporate banking portal, treasury management interface, trade finance platform) and the available settlement rails (PAPSS for eligible corridors, direct bilateral correspondents where you have them, fintech aggregators for SME flows, SWIFT for institutional volumes). It makes routing decisions in real time based on cost, speed, and reliability data for each corridor.

The architecture requires four components:

1. Rail inventory and monitoring. A live, queryable map of which settlement rails are available for which corridors, with real-time status on uptime, current processing times, and cost. This sounds basic and is genuinely absent from most mid-tier bank infrastructure.

2. Routing logic. Given a payment with specific corridor, amount, speed requirement, and cost tolerance, the orchestration layer selects the optimal rail. Nigeria-Ghana flows route over PAPSS. Nigeria-Kenya flows route over a fintech aggregator. A $500,000 institutional transfer routes over SWIFT. This logic is rule-based at the start and model-based as data accumulates.

3. FX transparency. The orchestration layer captures the all-in FX cost for each route, including correspondent spreads that are often not disclosed in the payment chain. This data is surfaced to the customer in the quote before payment authorisation — not hidden in a settlement advice three days later.

4. Compliance integration. Dual-jurisdiction AML screening integrated into the routing layer, not bolted on at the end. The compliance check runs in parallel with rail selection, and the routing logic accounts for compliance processing time in the speed estimate.

This is not a transformative product vision. It is infrastructure engineering. It is exactly the kind of infrastructure that global transaction banks have spent fifteen years building and that African mid-tier banks systematically lack.

The banks that build this in the next two to three years will have a durable competitive advantage in corporate and SME cross-border payments. The banks that wait for a "pan-African payment revolution" to deliver the infrastructure for them will still be routing payments through a Gulf correspondent in 2030.

The Path Forward Is Unglamorous

I have sat through enough payments industry conferences to know that the gap between announcement and reality in African payments infrastructure is structural. Every new system — PAPSS, CBDC pilots, stablecoin initiatives, bilateral swap lines — is announced at maximum ambition and deployed at minimum viable adoption.

That is not a reason for pessimism. It is a reason for realism about timelines and a reason to build for the infrastructure that exists now, with a migration path to the infrastructure that will exist.

For cross-border payments Africa, the practical priorities for a mid-tier bank in 2026 are clear: map your actual rail topology (not the theoretical SWIFT topology), build direct bilateral relationships with regional hubs in Mauritius, Kenya, and South Africa, connect to PAPSS for West African corridors now, and build the orchestration layer that lets you route intelligently across all of it.

The press release will come after the plumbing is in place. Build the plumbing first.

Ready to map your cross-border payment infrastructure and identify the corridors where you are losing margin unnecessarily? Talk to Amara.

Amara leads Cross-Border Payments research at Aicura Consulting, with a focus on sub-Saharan African financial infrastructure and institutional payment flows.