Shinhan's Rate-Hike Math Breaks: NIM Goes Flat as Lotte Insurance Takes the Wheel

A quarter where the deposit side beat the loan side, a KRW 693bn insurance deal jumped to the top of the agenda, and the word 'distribution from reduced capital' vanished entirely

055550.KS · Earnings Call · 2026-10-27

The rate-hike trade stopped working

For two years, Shinhan's pitch was mechanical: Korean rates go up, the bank's loan book reprices, NIM widens. In July the bank CFO put a number on it — another hike and “there can be an additional 4 to 3 bps improvement of the NIM” — Jeong Hoon Jang, CFO · 2026-07-23. Two hikes landed (July and August). The NIM went the other way, down 3bps quarter-on-quarter. The explanation on this call is the most interesting thing in the transcript, because it is an admission that the transmission mechanism has inverted. Funding costs rose 13–14bps while lending rates rose only about 10bps. Deposits that came due in Q3 re-priced into a market that expected higher rates, and competition for them got fiercer. So the funding rate beat the lending rate, and the bank's own index mix is the culprit: roughly 60% of funding is tied to six-month and one-year debentures, with only ~20% or less on the three-month CD. The delay is structural, not accidental.

In the past, there was this classic movement when there is increasing rates, 2 to 3 bp was sort of automatically expected improvement on our NIM. But currently, the funding rate is the larger factor in NIM direction... we are at this point, keeping our guidance on NIM neutral.
Jeong Hoon Jang, CFO · 2026-10-27

That is a quiet but material rewrite of the group's earnings engine. Group NIM guidance is now flat-to-neutral through year-end, with a hoped-for stabilisation only "from next year." Net income of KRW 1,584.6bn fell 12.9% QoQ. ROE and ROTCE did improve 0.9pp YoY to 11.9% and 13.4%, but that is the lagging scoreboard, not the trajectory.

Where the deposits went

The funding squeeze has a plumbing story that management told with unusual candour. Household deposits fell about KRW 9trn (KRW 7.5trn of time deposits plus KRW 1.5trn of liquid money) while corporate deposits rose roughly KRW 25trn — a mix swap forced by liquidity and NSFR compliance, not by choice. The loan-to-deposit ratio drifted from 96% at end-2025 to 92.6% in Q1/Q2, which mechanically drags NIM even with spreads intact. Household deposits has become a genuine constraint rather than a growth lever. This is not a Shinhan problem alone — it is the defining arithmetic of this reporting season. Across the funding cost theme, peers from Hana to European and Turkish banks spent their calls on total deposits, time-deposit migration and core deposits. The global quarter was littered with Fed rate hike chatter and retail deposits anxiety. Shinhan is not idiosyncratically weak here; it is early in a sector-wide repricing cycle where the liability side moves first.

Lotte Insurance: a genuinely new word at the top

What is genuinely company-unique is the strategic pivot. The BOD approved the acquisition of 77% of Lotte Insurance for KRW 693.3bn — a keyword that did not exist in Shinhan's vocabulary a quarter ago, and which now occupies the single top slot in the company's keyword ranking. It costs about 19bp of CET1 (13.62%, up 12bp QoQ) and is funded entirely by capital reallocation. YTD buyback reaches KRW 1.6trn, up 28% YoY, with another KRW 450bn approved; full-year DPS of KRW 2,960 would be a 14.3% increase. The pushback from analysts was immediate and well-founded: 0.9x price-to-book, and how much capital injection follows? Management answered with KRW 900bn of simple-calculated need, of which only ~KRW 300bn is common equity — and notes the group can use instruments individual shareholders cannot. The strategic logic is a fix of a known hole: a life business running about KRW 50bn of losses, plus PPA gains of KRW 30–40bn in a higher-rate environment. Against a KRW 10–12% ROE corridor, that swing is the difference between the bottom and the middle of the range. Importantly, the deal was pre-flagged. In July the CFO was explicit that on non-life M&A “nothing has been finalized at this point” — Jeong Hoon Jang, CFO · 2026-07-23, while also framing the hurdle — that M&A must clear the required return, with “no short-term impact from the M&A deal” — Jeong Hoon Jang, CFO · 2026-07-23 on shareholder returns. So this is not a surprise; it is a promise being cashed. The stock-relevant question is not the price paid but whether the group can keep a 50% total-return payout while also absorbing a non-life balance sheet it has never run.

What management stopped saying

Contrast is where the signal lives. For several quarters the loudest theme in Shinhan's calls was distribution from reduced capital — a top-ranked item in the July 2025 quarter, repeatedly queried as a peer did it. It is simply gone. So is the heavy, repetitive drumbeat on asset quality as a front-of-mind problem, and the detailed debate over the value-up program's payout arithmetic; both have been subordinated to the Lotte deal and the deposit question. Part of that is earned: credit cost came in at 42bp, four basis points better YoY, NPL coverage at 115.48%, and the CRO guided to the “low 40 bp range” — Na Hoon, Group CRO · 2026-10-27 even under extra provisioning. Cost discipline also surprised: G&A fell 5.0% QoQ and the cost-to-income ratio held at 37.6%. But the quarter's weak spot was the other half of the P&L. Noninterest income collapsed 45.5% QoQ, with fee income down 27.7% — a slide in trading value and market securities valuation losses. That is exactly the bucket management has been selling as the escape hatch from a structurally low-growth balance sheet. A group whose NIM is flat and whose capital-markets income just halved is a group leaning on an M&A fix and buybacks for its multiple. One more asymmetry worth noting: the OCI hole. The exchange-rate move cut RWA and should have flattered CET1, but roughly KRW 1.4trn of accumulated OCI losses — about KRW 800bn of overseas business translation, KRW 800bn in life insurance duration marks, KRW 170bn at the bank — netted the benefit to about 22bp. Shinhan's capital ratio is now partly hostage to the won.

The bottom line

Shinhan delivered a steady capital quarter and a flag-planting insurance deal, while quietly conceding that its most reliable earnings lever — rising rates — has stopped pulling. The rate-hike thesis has not been disproved forever; repricing of the debenture-linked book is sequential, and the bank's own LDR was distorted by a deposit mix it could not control. But investors should now treat the 10–12% ROE corridor, and the 50% payout ambition, as functions of a KRW 1.6trn buyback, a turnaround of a loss-making life book, and a capital-markets recovery — in that order. The generic sell-side check depends on which of those three turns first.