Open in interactive viewer → charts, metric popovers & call review

Credicorp's El Niño Ladder: From 'Hiccup' to 'Potentially Extreme'

A 19.9% ROE quarter, a PEN 500 million weather provision, an extraordinary dividend that quietly evaporated — and a Yape that has run out of unbanked Peruvians to sign up.
BAP · Earnings Call · 2026-11-10

Credicorp's El Niño Ladder: From 'Hiccup' to 'Potentially Extreme'

One word, four rungs

The most informative thing about Credicorp's third quarter is not the 19.9% return on equity or the 14.8% year-over-year loan growth. It is the vocabulary. Six months ago the CEO was calling the weather a footnote: “El Nino is, I would say, a hiccup” — Gianfranco Piero Dario Ferrari de Las Casas, Chief Executive Officer · 2026-08-14. By August, the risk officer was “moving in the expectation of medium and strong El Nino” — César Ríos, Chief Risk Officer · 2026-08-14 and promising a mid-fourth-quarter reassessment. This quarter the base case moved again — “our base case shifted from a strong to a very strong El Niño scenario at quarter end” — Ignacio Belaúnde, Chief Financial Officer · 2026-11-10 — and then moved once more, with the outlook described as evolving from a very strong to potentially extreme scenario. Four rungs of one ladder in two quarters, and the impact of el niño now sits atop the company's keyword set with a bullet. What matters is not the escalation itself but that the guidance did not move with it. Full-year cost of risk is still pegged at 1.7%–2.1% on the whole book; underlying cost of risk was a healthy 1.6% this quarter against a headline 2.0%. The direct exposure is roughly 8% of the total loans — a PEN 160 billion book implies about PEN 13 billion — but the risk team is explicit that the gross number overstates the hit because geography, sector and client sensitivity have to be funnelled first: “with the figures that we have, we are not expecting to have 10% provision for El Niño for this amount” — César Ríos, Chief Risk Officer · 2026-11-10. The provision stock stands near PEN 500 million (PEN 100 million built in Q2, PEN 400 million in Q3), but about PEN 200 million of specific releases — a wholesale recovery, a mortgage recovery, insurance-covered loans, better-behaved pilots — meant only roughly PEN 200 million actually hit this quarter's P&L. That is the mechanical story. The more interesting one is the analogy management rejected. When Santander's Andres Soto asked why this event should be treated differently from the 2023 El Niño that was provisioned but never fully landed, the answer was a flat refusal of the comparison:

I would not take the 2023 as a reference point, because this situation is different. Our tools are different. The underlying situation of the country is significantly different. We were, in 2023, in a weakness situation. Now we are in a position of strength.

César Ríos, Chief Risk Officer · 2026-11-10
And then the honest caveat, from Gianfranco: “the temperature of the sea is higher today than it was in September 30th” — Gianfranco Ferrari, Chief Executive Officer · 2026-11-10. The models get re-run every fifteen days. Any figure management gives you is already stale.

The dividend that vanished without a question

Here is the tension worth carrying out of this call. In August, asked directly about capital returns, the CEO was categorical:

We believe that we are very well capitalized. So the potential provisions or additional provisions because of our very strong El Nino should not affect the extraordinary dividend that we may pay this year.

Gianfranco Piero Dario Ferrari de Las Casas, Chief Executive Officer · 2026-08-14
Eleven weeks later, buried in the prepared remarks rather than the Q&A: given the attractive opportunities we see to deploy capital to support profitable growth, we do not expect to declare an extraordinary dividend this year. The stated reason is redeployment, not weather — the CFO frames it as “consistent with our disciplined capital allocation framework and our focus on maximizing long-term shareholder value” — Ignacio Belaúnde, Chief Financial Officer · 2026-11-10. But note what happened to the extraordinary dividend: it was a live keyword theme in prior quarters and does not appear anywhere in this quarter's set, and not one analyst on the call asked about it. Alonso Aramburú had pressed on exactly this in August; this time the thread simply dropped. Both readings survive the evidence. If you are earning 19.9% and targeting 22%, retaining marginal capital is defensible arithmetic. If you are staring at a "potentially extreme" weather event with an 8% loan-book exposure, retaining capital is also plain prudence — and the more boring explanation is usually the right one at a bank. What you can say is that Credicorp has stopped promising the payout it was promising in August, and it did so without being asked to explain why.

Yape meets its arithmetic ceiling

The other structural change is quieter and arguably more consequential. Yape closed the quarter with 17 million monthly active users and 71 transactions each per month, generating 9.8% of group risk-adjusted revenue and PEN 11.9 of revenue per MAU against PEN 6.1 of expense. The whole innovation portfolio crossed 10.7% of risk-adjusted revenues — past the 10% ambition the company set years ago. But the ceiling has arrived. Revenue-generating transaction value grew just 0.5% quarter over quarter, and the number of users is flattening, because there are only so many Peruvians. Francesca Raffo's explanation was refreshingly blunt: growth in revenue-generating transactions is slowing “just because the number of customers and the population” — Francesca Raffo, Chief Innovation Officer · 2026-11-10. The CEO's joke about needing to raise Peru's birth rate was the most quotable line of the call and also the most revealing one. So the Yape story now has to move from user acquisition to depth: a PEN 2.3 billion loan book, more than 6 million customers who have borrowed at least once, multi-installment and MYPE lending, cross-border, and a neobank wrapper that now houses Yape, Eo and Tenpo. The strategic update scheduled for November 17 is presumably where the new North Star metric gets set — the old one has been beaten, and the old one was a revenue-share number, not a profit number. That is a real milestone hidden inside a quarter that otherwise looks like more of the same.

The backdrop nobody is fighting anymore

Worth noting what has disappeared from the narrative. For most of 2026 the political risk premium was the lead item; this quarter Congress did not grant executive powers, the Ministry of Finance is drafting investment-authorisation legislation, and the topic barely registers. Domestic demand is growing near 6%, private investment near 14%, and Gianfranco can plausibly describe the economy as carrying "a lot of tailwinds." The risk has migrated from politics to weather — from an event you legislate around to one you provision against. That is also the el niño confluence: it is a genuinely global keyword this cycle, showing up across salmon farmers, banana growers and crop-input names in the recent reporter set. Credicorp is unusual in that it does not take the commodity-price hit — high copper helps Peru — it takes the credit-quality hit through a retail and microfinance book. The underlying cost of risk being 1.6% while the headline reads 2.0% is the whole argument in one number: the portfolio is fine, the forecast is not. Which is why the reaffirmation of the 22% medium-term ROE matters more than the 19.9% print. Management is asking investors to look through two or three quarters of weather, keep the transactional activity that is driving 20.4% fee growth, and wait for the November 17 reset. If the provisions land inside the range and the dividend decision was genuinely about growth rather than defence, the story writes itself. If the sea keeps warming, the ladder has another rung — and the next word up from "extreme" is the one nobody wants to say.