Cross-Border Payments in Africa: What Banks Get Wrong

The Problem Nobody Admits Out Loud

Africa is not one market. It is fifty-four regulatory regimes, dozens of currencies, and three distinct correspondent banking ecosystems — West, East, and Southern. When global transaction banks design cross-border rails for the continent, they routinely collapse this complexity into a single "Africa strategy." That is the first, and most expensive, mistake.

I have reviewed hundreds of payment flows across the continent. The failures cluster around three systemic issues that banks refuse to acknowledge in public but admit privately.

Mistake 1: Treating Correspondent Banking as Solved Infrastructure

The de-risking trend of the 2010s gutted correspondent banking relationships across sub-Saharan Africa. US and European banks exited local correspondent relationships in Nigeria, Tanzania, and Zimbabwe citing AML compliance costs. The result: payment chains with four and five hops where two used to suffice.

Every additional hop adds cost, adds latency, and adds failure points. A payment from London to a Tanzanian SME can sit in limbo for five days because an intermediate correspondent in a Gulf state lacks a real-time SWIFT connection to the Tanzanian receiving bank.

What banks get wrong: They blame the local banking infrastructure. The real failure is the unwillingness to invest in direct bilateral relationships, or to use newer clearing mechanisms (PAPSS in West Africa, REPSS in COMESA zones) that bypass correspondent chains entirely.

The Pan-African Payment and Settlement System (PAPSS), launched commercially in 2022, enables intra-African payments in local currencies without routing through New York or London. Adoption is accelerating. Banks that still route Kenya-to-Ghana flows through a US correspondent in 2026 are paying a structural tax they do not need to pay.

Mistake 2: Ignoring FX Spread Opacity

In developed markets, FX spreads on wholesale transactions are visible, competitive, and regulatorily constrained. In African corridors, they are frequently opaque, wide, and compounding.

A corporate treasurer in Nairobi paying a supplier in Accra faces a multi-layer FX problem: the Kenyan shilling-to-dollar spread, the dollar-to-cedi spread, and the intermediary margin that no one discloses upfront. The all-in cost of a $10,000 B2B payment can reach 4–6% in some corridors — compared to sub-1% in SEPA.

What banks get wrong: They focus compliance and product efforts on the remittance corridor (consumer, high-volume, politically visible) and underinvest in the B2B and institutional corridors where the real margin destruction occurs.

SWIFT gpi has improved transparency considerably for SWIFT-connected transactions. But a large proportion of African financial institutions are not full SWIFT members — they access the network through sub-SWIFT arrangements with their correspondent, which means gpi tracking data never reaches the originating bank. The treasurer sees "payment processing" for three days with no visibility into where the funds actually are.

Fix: banks need to map their actual network topology, not just the theoretical SWIFT topology. Dual-currency accounts held in regional hubs (Mauritius, Kenya, South Africa) significantly reduce the number of FX conversions in a payment chain.

Mistake 3: Compliance as a Binary Gate, Not a Risk Spectrum

AML and sanctions compliance is non-negotiable. The problem is that many transaction banking operations in Africa have implemented compliance as a blunt filter rather than a calibrated risk model.

An agriculture export company in Zambia that has banked with the same institution for eleven years should not have every payment above $50,000 trigger a manual review. Yet that is what happens in many global banks operating in the region, because their compliance models were designed for high-risk onboarding scenarios and never recalibrated for long-standing, well-understood relationships.

The practical consequence: false-positive rates above 20% in some corridors. Every false positive costs the bank time, costs the client trust, and often causes them to route payments through alternative channels that actually carry higher risk.

What banks get wrong: They conflate risk management with risk avoidance. A 20% false positive rate is not strong compliance — it is a process failure that actively degrades the quality of the bank's risk signal.

The data to differentiate legitimate clients from high-risk ones exists. The challenge is integrating transaction history, behavioural analytics, and network analysis into a dynamic model rather than a static rule set.

Where the Opportunity Sits

The banks that will dominate African cross-border payments over the next decade are not the ones with the broadest compliance walls — they are the ones that have built deep bilateral relationships with regional clearing mechanisms, invested in FX transparency tooling, and implemented risk-based (not rule-based) compliance frameworks.

PAPSS, REPSS, and the growing network of central bank digital currency pilots across East Africa are restructuring the plumbing. The opportunity is to build payment infrastructure aligned with where the rails are going, not where they were.

The right strategy is not a single "Africa play." It is a corridor-by-corridor assessment: Kenya-Uganda, Nigeria-Ghana, South Africa-Mozambique. Each corridor has a different regulatory profile, FX regime, and clearing infrastructure. Generic strategies fail. Specific, well-researched corridor strategies win.

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