A head of a microfinance institution preparing for an inspection generally prepares their ratios. That is what they were taught, that is what their returns produce, and that is indeed what the regulator looked at for a long time.
It is no longer the heart of what is examined.
The 2015 reform of the prudential framework extended supervision beyond ratios alone. It introduced requirements on operational control, compliance, risk management and internal audit — that is, on arrangements, procedures and evidence, not on numbers. A transition period ran until 2020. And COBAC intensified its inspections from 2021.
Five years later, many institutions still present a body of procedures designed for the old regime: documents that explain how one calculates, not how one controls. The gap does not show in the returns. It shows in an inspection.
What "beyond ratios" means concretely
A ratio is an outcome. An arrangement is a process that produces that outcome, and one that must be describable, testable and provable.
The difference is easy to illustrate on a case every institution knows. Take risk division. Under the old regime, the question was: does your exposure to a single beneficiary exceed the limit? You answered with a figure, drawn from your return.
Under the current regime, the question becomes a chain:
- How do you identify that a set of beneficiaries constitutes a single risk — same directors, same collateral, same cash flow?
- Is that grouping automatic in your system, or does it depend on the memory of a relationship manager?
- Who checks that grouping, how often, and where is the evidence?
- What happens when the limit is approached: does the system block, or does it alert someone who can override?
- If someone can override, who, under what written delegation, and is that override logged?
An institution can be perfectly within the limit on the ratio and perfectly deficient on the chain. That is in fact the most common situation, because the ratio is a by-product of an information system, whereas the chain is an organisational choice nobody formally made.
The four recurring breaches, read through this lens
Four norms recur with striking regularity in the sector's files: the solidarity fund, fixed-asset coverage, risk division and the liquidity ratio. They are usually presented as a list of ratios to watch. That reading is incomplete, and it explains why they recur.
None of these four deteriorates because of poor commercial activity. All four deteriorate through failure to steer the balance-sheet structure or failure of arrangement.
- The solidarity fund does not build itself: it presupposes an allocation decision, repeated, evidenced, and a control that the allocation actually occurred.
- Fixed-asset coverage — COBAC regulation EMF 2002/09 requires net capital funds, increased by borrowings of more than five years allocated, to cover at least 100 % of net fixed assets — tips at once on the signature of an acquisition, not gradually. It ranks, moreover, among the least respected norms in the sector, and particularly among microfinance institutions.
- Risk division depends entirely on the quality of the grouping of linked beneficiaries, that is, on an arrangement.
- The liquidity ratio depends on the reliability of the maturity ladder, therefore on data-entry quality and update discipline.
In other words: these four are governance indicators disguised as ratios. That is why they reappear in the same institutions, corrected one quarter and breached the next.
The silent killer, and why the regime changed
What the reform targeted must be named. The sector's crises in the CEMAC zone had as their principal cause regulatory shortcomings and poor governance practices within the structures themselves. Institutions placed under provisional administration were placed there because of management irregularities — irregularities orchestrated by their directors, not suffered by them.
This is an uncomfortable point and it is central. The risk that brings down a microfinance institution is not, statistically, credit risk. It is internal risk: misappropriation, irregular commitment, untraced transaction, circumvented delegation. And internal risk shows in no ratio — by construction, whoever commits it also controls what the ratios display.
Supervision therefore stopped asking "are your figures good?" and started asking "who can do what, and who verifies it?". An internal control arrangement that exists on paper but whose control activities nobody follows produces exactly the gap inspections now report.
The other transformation: the data leaves your building
Two developments under way change the very nature of the exercise.
Reporting has changed tool. The 44 SESAME returns a microfinance institution produces quarterly — resources, commitments, risk division, internal control arrangements, among others — are now filed through the SPECTRA platform, which incorporates the SESAME 4.0 module dedicated to microfinance regulatory reporting. COBAC is actively pushing institutions onto it. A change of platform is never neutral: it makes consistency checks more systematic, and it makes visible what a manual filing let through.
Borrower information is becoming shared. The CEMAC Credit Information Bureau entered service in January 2026, operated by Creditinfo Central Africa under the aegis of the BEAC and the International Finance Corporation. It collects from banks, financial institutions and microfinance institutions, but also from telecommunications operators and water and electricity suppliers. The BEAC targets the integration of at least 60 % of credit and microfinance institutions in the sub-region within three years, and full coverage by 2028.
The common meaning of these two movements: what you declare becomes verifiable elsewhere. A gap between what your returns say and what other sources show ceases to be invisible.
What to re-read in your procedures
The fastest test is not to re-read your ratios. It is to take your body of procedures and look for traces of the current regime. Five questions are enough to know where you stand:
- Do your procedures describe controls, or only operations? A procedure that says how to grant a loan without saying who verifies the grant, when, and on what sample, belongs to the old regime.
- Are delegations written, quantified and current? Who commits what amount, alone, and above what threshold a second pair of eyes becomes mandatory. An implicit delegation cannot be defended in an inspection.
- Are overrides logged? An arrangement that permits no exception is an arrangement that gets circumvented. An arrangement that permits evidenced exceptions is one that holds.
- Does internal audit have a plan, and has that plan been executed? The existence of a function is worth nothing without a completed work programme and dated reports.
- Does segregation of duties survive the absence of one person? In smaller structures, control often rests on a single trusted individual. That is precisely the configuration internal fraud files describe.
The real stake, and it is not only regulatory
Bringing yourself up to date with this regime costs management time and little money. Not doing so costs in two ways.
The first is known: observation, injunction, then heavier measures.
The second is less so, and it is financial. Since the BEAC suspended, on 2 April 2026, refinancing for medium-term credit, long resource has become scarce across the zone and banks are rationing, starting with counterparties that are expensive to assess. An institution whose internal control arrangement is documented, whose delegations are written and whose audit reports exist is, literally, cheaper to analyse. International refinancing vehicles apply the same reasoning, more strictly still.
Compliance with the 2015 regime was long treated as an administrative burden. In a market where resource is becoming scarce, it has become a financing argument.
Published 14 August 2026. The regulatory references and infrastructure developments cited are verified as at that date.
