At end-March 2026, net claims of the monetary system on CEMAC states reached 11,397.6 billion CFA francs, equivalent to 83.5 % of the 13,655.4 billion of credit extended to the economy.
That ratio deserves a slow reading. It does not say banks lend little. It says that for every franc going to the real economy, 0.84 goes to the sovereign signature — a counterparty that does not default, asks for no collateral, costs almost nothing to appraise and can be refinanced.
No rational credit committee behaves otherwise. That is precisely what makes the subject difficult.
The paradox, in two more figures
The zone is not short of liquidity: bank over-liquidity there exceeds one thousand billion francs. And yet credit to the private sector represents around 17.5 % of Cameroonian gross domestic product, against 28.7 % on average in sub-Saharan Africa.
Abundant liquidity, weak private credit: this is not a contradiction, it is a trade-off. Capital is not lacking; what is lacking is a private counterparty whose appraisal costs less than its marginal return.
Why the trade-off tilts mechanically
Cost of appraisal. A sovereign security is analysed in minutes. A corporate credit file in the CEMAC zone takes tens of days — often sixty to ninety — and most of that is consumed by documentary back-and-forth, because four companies out of five there operate outside official records, without certified accounts or auditable statements.
Capital treatment. At comparable return, local-currency sovereign exposure structurally consumes less regulatory capital than unrated corporate exposure. In an exercise where capital becomes the scarce resource, that gap weighs heavily.
Exit liquidity. A security can be sold or refinanced. A non-performing corporate receivable stays on the balance sheet until recovery or write-off: there is no organised secondary market for non-performing loans in the OHADA zone — unlike what is being built in Morocco, where a draft law governing their direct transfer, developed with the support of the International Finance Corporation, will open acquisition beyond regulated institutions alone.
What has just changed, and hardens the trade-off further
Two facts from 2026 push the same way.
On 2 April 2026, the BEAC's Monetary Policy Committee suspended refinancing operations for medium-term credit intended for productive investment. The Governor confirmed on 29 June that the suspension answered a request from the International Monetary Fund, whose position is that the mechanism should be phased out. Banks lose one of the few arrangements matching medium-term lending with a corresponding resource — precisely the tool that made investment credit sustainable.
And COBAC regulation R-2025/02, in force since 1 January 2026, raises minimum share capital from 10 to 25 billion francs, with a first step at 14 billion on 31 December 2026, then 18, 22 and 25 billion through 2029. A bank that must raise capital favours, in the meantime, the uses that consume least of it.
In short: regulatory constraint and resource constraint both push towards the sovereign, at the moment the economy most needs the opposite.
The blind spot in this trade-off
Sovereign exposure is treated as a risk-free asset. It is free of default risk. It is not free of concentration risk.
A banking system whose claims on states represent 83.5 % of credit to the economy carries a massive correlation: every institution is exposed to the same signature, at the same time, at the same maturities. A regional budgetary strain does not diversify — it adds up.
That concentration is perfectly visible in the returns. It is rarely treated as an exposure, because regulatory weighting does not require it. It is exactly the kind of risk that does not show in a ratio and is discovered in a cycle.
What actually moves the trade-off
The usual discourse asks banks to "do more SME lending". That is not a request, it is a wish: as long as the cost of appraisal remains what it is, the trade-off will not move.
What moves it is measurable, and rests on three levers.
Reduce the cost of appraisal, not the risk. A file that arrives complete — restated accounts, clear ownership, registered and enforceable security, a cash plan tied to the real cycle — does not ask for less rigour: it asks for fewer hours. The trade-off is settled on hours, not on appetite.
Make exit possible. As long as an impaired receivable can neither be sold nor outsourced for recovery, every disbursement commits the balance sheet to term. Preparing a sellable portfolio today — described in cohorts, with verified security and a documented recovery history — requires no new legislation.
Use the instruments that exist. Portfolio guarantees from development institutions cover a substantial share of the risk — up to 50 % for loans to small and medium-sized enterprises, more for microfinance institutions. They are under-used not out of mistrust, but because the bottleneck is not risk: it is, once again, the file.
What to take away
83.5 % is not a figure for indignation. It is the measure of a rational trade-off, made every day, by committees doing their work correctly with the files they receive.
It will not be corrected by willingness. It will be corrected the day appraising a company costs a bank something comparable to what a security costs it — and that day is prepared upstream of the credit committee, not inside it.
Published 14 August 2026. The figures cited are verified as at that date.
