A 22% decline in active correspondent banks between 2011 and 2019 would be unremarkable if it were about consolidation. It is more interesting because volumes rose over the same period. The network is not shrinking because there is less to carry; it is thinning because fewer institutions are willing or able to carry it. That distinction turns a market statistic into a systemic story.

Concentration in a few hubs

When relationships decline while traffic grows, activity concentrates. A smaller set of large institutions and hubs handles a larger share of the flow. That is efficient up to a point, and then it is not. Concentration reduces the number of available paths, so a single disruption or a single decision to withdraw has a wider effect.

The periphery feels this first. In Latin America, active relationships have fallen by around 30% since 2012. Thin corridors lose their last few routes, and the cost of reaching them rises precisely where access matters most.

Why relationships disappear

Correspondent banking is built on bilateral, personal trust. Each relationship is maintained separately, reviewed separately and priced separately. As compliance expectations rise, the cost of keeping a low-volume relationship open can exceed its value to the institution. Withdrawal is rational for the bank and expensive for the corridor.

What a systemic answer looks like

The architectural response is to reverse the direction of relationships. Instead of every institution opening accounts everywhere, a hub can hold correspondent accounts for institutions. Trust can then be codified in a multilateral reliance agreement, the basis for which already exists in FATF Recommendation 17, rather than negotiated bilaterally and made non-transferable.

This is not a claim that the problem disappears. Cross-border reliance is not permitted everywhere, and local-law limits must be checked jurisdiction by jurisdiction. But it reframes the question from "which relationship survives?" to "which rule set keeps the corridor open?"

In short

  • Declining relationships are a structural story, not simple consolidation.
  • Concentration shrinks the number of routes and widens the impact of withdrawal.
  • Latin America has seen roughly a 30% fall since 2012.
  • Bilateral trust is expensive to maintain at low volume.
  • A hub model and multilateral reliance offer a systemic alternative.

See how the protocol reframes cross-border relationships at /protocol/.