When the Money Stops Moving

The notification never comes. You have been sending money home on the same schedule for two years, same operator, same Tuesday morning, and then one week the transfer sits in processing for four days before a hold notice appears in small print. You call the number. The queue runs forty minutes. Nobody explains. By the third time this happens, you stop trying, and you find a man who carries cash, and you pay more, and you trust him less. The service, for its part, has already made its calculation in a governance meeting you were never invited to attend.

Which corridors survive and which get abandoned is not, in any serious analysis, primarily a question of demand. Demand persists for years after a corridor has become operationally dead. It is a question of internal governance: who sits on the risk committee, how compliance costs are allocated across the network, and whether the operators running the books in a given country have the institutional standing to argue for their corridor's survival when the numbers turn uncomfortable.

The Ledger That Decides Everything

A remittance network is, at its core, a set of bilateral agreements wrapped around a shared liquidity pool. An operator in Toronto takes in Canadian dollars from a Somali diaspora sender. In Mogadishu or Hargeisa, an affiliated agent pays out in Somali shillings or US dollars from a local float. The network reconciles the difference on a schedule, weekly or monthly, moving value through a correspondent banking relationship.

That correspondent relationship is the choke point. A single large correspondent bank, headquartered in New York or London, may service dozens of corridors simultaneously through one master account held by the remittance operator. When the bank's compliance team decides that a particular country's transactions carry too much regulatory risk, it does not necessarily close the master account. It imposes enhanced due diligence: transaction-level documentation, beneficial ownership verification, source-of-funds declarations above a certain threshold. The corridor does not die immediately. It starts bleeding.

The compliance cost of meeting those requirements gets distributed internally, and this is where governance structure bites hardest.

How Compliance Costs Get Passed Down the Chain

Larger networks with genuine corporate headquarters tend to allocate compliance costs using a formula tied to transaction volume and margin. A high-volume corridor to the Philippines or Mexico generates enough throughput that a per-transaction compliance overhead of a dollar or two gets absorbed without drama. A low-volume corridor to South Sudan or Haiti may process only a few hundred transactions a month. The same overhead, at the same absolute dollar figure, can eliminate the corridor's entire margin. Not reduce it. Eliminate it.

The internal governance question then becomes: who has the authority to subsidize that corridor, and for how long?

In networks structured as franchises or loose agent arrangements rather than vertically integrated companies, the answer is almost always nobody. The Toronto franchise owner running the Somalia corridor cannot accept a cross-subsidy from the Manila corridor owner in Vancouver. There is no central treasury to redistribute margin. The governance architecture produces, by its own logic, a structural bias toward corridor abandonment whenever a low-volume route hits a compliance cost spike. It is less a decision than a default.

Vertically integrated operators face a different but related problem. Governance concentrates decision-making at headquarters, typically in the UK, US, or Gulf states. The executives on the risk committee have career incentives tied to the company's regulatory standing with large Western financial institutions. Preserving access to a major correspondent bank is a legible, career-relevant win. Preserving a corridor to a country that accounts for 0.3% of transaction volume is not. The votes in that room go a predictable way, and have gone that way, in various institutional forms, since the colonial-era clearing houses decided which trade routes were worth maintaining.

A Tale of Two Operators

Consider two operators who built their businesses on the same East African corridor, same decade, same London-based correspondent bank, same diaspora population concentrated in three British cities.

The first, call her Amara, structured her operation as a fully licensed money service business: a compliance officer on staff, a dedicated ledger for the corridor, a formal integration with a local mobile money provider at the receiving end. When the correspondent bank raised its documentation requirements, she had the systems to meet them. The cost per transaction rose, but the mobile money integration had already reduced her payout costs at the other end by roughly the same margin. She absorbed the shock because she had built the absorbers in advance.

The second operator, call him David, ran a leaner arrangement, effectively a sub-agent relationship under a larger network's licence. When the compliance requirements changed, the decision about whether to continue the corridor was not his. It sat with the network's regional director, who was simultaneously managing pressure from seven other sub-agents across three countries. The corridor was suspended pending a systems upgrade that was never funded. David's customers moved to informal channels: slower, more expensive, less traceable. The regional director filed a brief note and moved on.

The difference was not demand, not regulatory environment, not even cost structure in any simple sense. It was governance, specifically who owned the decision and what their incentives were when they made it.

The Quiet Exit, and What It Costs

Operators rarely announce corridor closures. The more common pattern is gradual degradation: slower settlement times, lower transaction limits, more frequent compliance holds, a customer service queue that quietly stops being answered. The corridor does not close. It becomes unusable, which is a different thing and leaves a different paper trail.

This matters because the regulatory cost of formally exiting a corridor can itself be substantial. Depending on the licensing regime, an operator may need to notify regulators, wind down open transactions, and demonstrate that outstanding balances have been settled. Gradual degradation sidesteps that process. It is, from a governance perspective, the path of least resistance, and it is, from any other perspective, a form of abandonment dressed as maintenance.

The people paying for that institutional convenience are the senders and receivers. A transfer arrangement that becomes unreliable does not just cost the family waiting on the money. It drives volume toward informal hawala networks or cash-carrying intermediaries, less traceable, less regulated, and often more expensive for small transactions. The regulatory logic that imposed the compliance costs in the first place, intended to reduce financial crime risk, ends up concentrating that risk in the very informal channels it was designed to displace. This is not an unintended consequence at this point. It is a documented pattern, and continuing to treat it as a surprise is a choice the industry makes.

Remittances to low-income countries collectively exceed official development assistance by a substantial multiple, and a significant share of those flows move through corridors that are, by the internal governance logic of their operators, perpetually at risk of exactly this kind of quiet exit.

What Determines Survival

The corridors that tend to survive compliance shocks share a few structural features. Receiving-end infrastructure matters enormously. A corridor connected to a robust mobile money ecosystem, where the payout agent is large and well-capitalised, can absorb compliance overhead because the total transaction cost is lower everywhere else in the chain. The mobile money provider has a genuine incentive to invest in the relationship, which means the corridor has an advocate on the other side of the ledger.

Governance structures that give corridor-level operators a formal voice in risk allocation decisions produce more durable corridors. This sounds obvious. It requires deliberate design. A network whose governance documents specify that corridor managers sit on the compliance committee, and that corridor-level cost impacts must be quantified before any new compliance requirement is adopted, is structurally different from one where those decisions happen at headquarters without granular input. The difference shows up not in mission statements but in which corridors are still running five years later.

Regulatory engagement helps too, perhaps counterintuitively. Operators who maintain an active dialogue with financial regulators in both the sending and receiving country tend to have more warning when compliance requirements are about to change. That warning time allows them to adapt systems rather than absorb a sudden cost spike that forces an emergency corridor review. Treating regulatory bodies as adversaries to be minimised, rather than counterparties to be managed, is a governance philosophy that tends to end badly, and expensively, for the communities the operator nominally serves.

The diaspora communities at the end of these corridors have no seat in any of these governance structures. No vote on the risk committee, no voice in the cost allocation formula, no standing to challenge a suspension. Their only recourse is exit: a competitor, or the informal channel. By the time they exercise it, the internal decision has usually already been made, filed, and forgotten.

To call this a market outcome is to mistake the container for the contents. The failure is a governance failure, specific, structural, and in principle correctable. The more interesting question is who benefits from leaving it uncorrected.