Getting banks ready for an El Niño-driven NPL wave

DAR ES SALAAM: THE Tanzania Meteorological Authority (TMA) has not minced its words.

Dr Ladislaus Chang’a, the agency’s Director General, has described the El Niño now forming over the Indian Ocean as “very strong”, warning that this year’s Vuli rains, expected to begin in late September or early October, intensify through November 2026 and persist into January 2027, will be shaped by the weather system.

The northern coastal belt, Lake Victoria basin, northeastern highlands and parts of Kigoma have been identified as among the areas most exposed.

President Samia Suluhu Hassan has already ordered regional and district authorities to prepare, warning that delays “could result in significant damage to lives, property and infrastructure.”

For disaster-management officials, that warning triggers a familiar checklist: Drainage, evacuation routes, health supplies and hydropower safeguards. For commercial banks and Development Finance Institutions (DFIs), it should trigger an equally urgent one.

Heavy, sustained rainfall does not only flood fields and wash away bridges. It can also flood loan books. Tanzania’s banking sector enters the rainy season in reasonable shape on paper.

The Non-Performing Loan (NPL) ratio stood at 4.7 per cent in the first half of 2026, just below the Bank of Tanzania’s regulatory ceiling of five per cent. Beyond that threshold, lenders face dividend restrictions, mandatory action plans and closer supervisory scrutiny.

That leaves the sector with a relatively thin buffer, while exposure to climate-related disruption is far from evenly distributed.

Agriculture, still the largest employer in the economy, sits directly in the path of the forecast rains, along with the value chains built around it, input suppliers, produce traders, transporters and small agro-processors.

Construction and real estate loans in flood-prone urban areas, SME financing tied to markets near rivers and low-lying coastal areas and transport loans along vulnerable road networks also carry heightened risks.

This is precisely the territory Tanzania’s DFIs, with the Tanzania Agricultural Development Bank (TADB) chief among them, were established to serve. Their portfolios could therefore carry a disproportionate share of the coming season’s climate risk.

Tanzania does not need to imagine what a strong rainy season can do to borrowers; it has experienced the consequences before.

El Niño-linked floods during the 2023/24 season affected an estimated 200,000 people across several regions and caused fatalities. The following season brought further disruption.

In Kilimanjaro alone, about 300 homes were submerged and more than 2,000 acres of farmland were destroyed, while the collapse of bridges at Somanga and Matandu disrupted transport links along the coast and affected trade with neighbouring countries.

Behind every affected farm, home or transport route may be a borrower whose ability to repay depends on those assets remaining productive. When a road to market disappears, a trader’s ability to sell can quickly disappear with it.

When crops are destroyed, a farmer’s repayment capacity can deteriorate just as quickly. The lesson is not simply that floods cause defaults.

It is that lenders that treat climate shocks only as a collections problem after the event risk recovering less, more slowly and at a higher cost than those that adjust their approach before the rains arrive.

The first move should be internal: stress-test loan books against the specific geography identified by TMA rather than relying on national averages. A branch in Same or Moshi faces materially different risks from one in Mtwara.

Head-officewide provisioning and risk assessments can therefore miss important concentrations.

Portfolios heavily exposed to the flagged regions deserve sector- and district-level reviews now, while there is still time to adjust exposure limits, moderate new lending in the riskiest areas and identify accounts that may require early intervention.

ALSO READ: Banks urged to digitise complaint handling

The second step is a shift in posture towards borrowers likely to experience temporary difficulties through no fault of their own.

Restructuring a loan for a farmer whose crop is destroyed in November may be cheaper in both financial and reputational terms than waiting until February, when the account has already fallen deeply into arrears.

Banks can proactively consider grace periods, revised repayment schedules or seasonal moratoriums for genuinely affected borrowers.

Early engagement with the regulator will also be important, particularly where distress is clearly linked to an external climate shock.

The third issue is logistical but easy to overlook: Loan officers cannot collect from borrowers on flooded roads.

Banks with significant rural exposure should therefore shift collections and customer engagement towards mobile money, agency banking and other digital channels before physical access becomes difficult.

Alternative routes and staffing arrangements should also be prepared for districts likely to be cut off. For TADB, TIB Development Bank and similar institutions, the calculus is even sharper because their mandates place them closer to the agricultural and rural borrowers most exposed to the forecast.

A DFI cannot simply de-risk by retreating from agriculture; doing so would undermine its purpose.

Risk management therefore has to happen within the portfolio. That means identifying financed farms and value chains located in high-risk districts, coordinating early with agricultural extension officers and insurers on mitigation measures and building repayment flexibility into loan structures rather than treating it as an emergency response.

Where crop or weather-index insurance is available, lenders should also encourage greater uptake.

A successful insurance claim can provide a more reliable recovery mechanism than relying on a borrower whose harvest has been wiped out. None of this is an argument for going soft on collections.

It is an argument for better sequencing. Institutions that identify their exposure, engage borrowers early, restructure viable loans and keep collection channels open during the rains are likely to recover more of what they are owed than those that wait for arrears to accumulate and then chase borrowers through disrupted communities.

Tanzania’s regulators, banks and DFIs have several months of warning this time, courtesy of TMA’s forecast and the President’s call for early preparedness.

The loan book deserves the same urgency as the drainage channels, evacuation routes and other disaster-preparedness measures.