139130 139130

Key data

The business

iM Financial Group is a Daegu-based bank holding company, formerly DGB Financial Group, renamed in March 2025, built around iM Bank, formerly Daegu Bank, which in May 2024 became the first bank in 32 years to win a new nationwide commercial banking charter from Korea's Financial Services Commission, converting from a lender confined to the Daegu and Gyeongsangbuk-do region into one that can compete for deposits and loans across the country. The group also owns iM Securities (formerly Hi Investment & Securities, an 85.3% stake acquired in 2018), iM Capital (leasing and corporate/auto finance), and iM Life (insurance). iM Bank carries the franchise: it earned ₩245.7B of the group's ₩295.6B in first-half 2026 net income, about 83% of the total, with won-denominated loans crossing ₩60 trillion for the first time as nationwide expansion widens the lending base.

What the quarterly print doesn't show directly is how much of the credit story is a completed cleanup versus an ongoing one. The group's real-estate project-financing book, the asset class that forced heavy provisioning through 2023 and 2024, has seen its NPL ratio fall from 11.1% at end-March 2025 to 3.8% at end-March 2026, the product of two years of risk-management tightening since the charter conversion. Bank-level asset quality is now cleaner than the group average: iM Bank's own delinquency ratio was 0.87% and its NPL ratio 0.94% as of June 2026, versus a group-wide delinquency ratio of 1.43% as of March 2026, the gap reflecting residual stress at the non-bank subsidiaries that absorbed most of the PF exposure. The largest shareholder, OK Financial Group, a savings-bank and consumer-lending conglomerate, has built a 9.99% stake across three affiliates, just under the 10% ceiling Korean law now imposes on any single holder of a nationwide bank's parent, a governance overhang covered in Management below.

The numbers

Group net income fell 4.4% year over year in the first half of 2026 to ₩295.6B, with the second quarter down 8.9% to ₩141.1B, even as net interest income rose 4.1% to ₩843.6B and non-interest income rose 5.2% to ₩265.4B; management attributes the net-income gap to a 2026 education-tax-rate change plus higher loan-loss provisioning tied to the balance sheet's own growth, not a reversal of the PF cleanup. Group net interest margin held near 1.92%-1.93% through the first half; at iM Bank alone, NIM was flatter, 1.82% in both the first and second quarters, below the group figure because the securities and capital units run higher spreads. Group CET1 reached a record 12.27% at quarter-end, up 25 basis points quarter over quarter, and the group credit-cost ratio held near 0.41%. The number that moved the wrong way is the NPL coverage ratio: loan-loss reserves against non-performing loans, excluding regulatory reserves, fell to 82.2% at quarter-end from 93.6% at the end of the first quarter (137.7% including regulatory reserves, down from 143.6%), which the company has committed to rebuilding toward roughly 100% by year-end.

FY2024's ₩186.5B trough reflects the PF provisioning that preceded the charter-conversion risk overhaul; FY2025's rebound to ₩457.5B is the base the market is now pricing forward growth from.

FY2025 net income above is the consolidated-basis figure from third-party financial data used for the multi-year trend; Korean press reports FY2025 net income attributable to controlling shareholders at ₩443.9B (+106.6% year over year), a variance of about 3%, likely reflecting minority-interest treatment. Forward ROE carries no consensus estimate; a single broker (Yuanta, Jul 29, 2026) models FY2026E forward ROE near 7.8% against a 10.5% cost of equity.

Peer set: BNK Financial Group (138930.KS), the closest direct comparison as the other major regional-bank-holding-company-turned-nationwide-bank story. Big Four nationwide holdcos referenced in prose below, not in this table. vs peer compares FY26E to FY26E; peer FY26E P/E of 6.2x is BNK per Mirae Asset Securities, Feb 9, 2026.

P/B, ROE and dividend yield carry no forward consensus for either name. P/B and dividend yield use the current price against FY2025 book value and FY2025 dividend per share for both names; ROE is each company's most recent disclosed figure (139130: H1 2026 annualized; BNK: FY2025).

Closest comps beyond BNK are the Big Four nationwide holding companies, KB, Shinhan, Hana and Woori, all larger and all trading at higher trailing P/B, roughly 0.75x to 1.04x on the same current-price-over-FY2025-book method, with trailing ROE in the 9% to 10% range. iM Financial's nationwide ambitions put it in that group's lane over time, but its ₩2.8T market cap is still roughly a tenth the size of Woori, the smallest of the four.

iM Financial's 0.51x P/B and sub-6x forward P/E already price a bank the market doesn't expect to earn much above its cost of equity: at a 9.4% ROE, the stock trades close to where a high-single-digit-return bank should, not where a growth re-rate candidate would. The multiple isn't compressed because the market missed something; it's compressed because regional Korean bank holdcos, iM Financial included, have carried single-digit P/E and sub-1x P/B for years, a return profile the 2024 government-driven value-up payout push is only beginning to lift. Is the FY2025 earnings rebound to ₩443.9B real or a one-time lever? Mostly real: the driver was the PF book actually healing (NPL ratio 11.1% to 3.8% in a year), not a tax credit or an asset sale, and FY2024's ₩186.5B was itself the one-time-provisioning trough, not a normal run rate. The open question is whether the first-half 2026 dip is a smaller repeat of that same provisioning cycle or genuine noise; management's account, a tax-rate change plus loan-growth-driven provisioning rather than a PF relapse, is plausible given the PF ratio kept improving through March 2026, but the coverage-ratio slide argues for some caution before calling it fully resolved. This reads as a multi-year re-rate case, not a one-quarter trade: the charter conversion widens the loan-growth runway nationwide, and the capital-return program, a CET1-tiered payout framework targeting 40% to 50% of net income by 2027 and an estimated 42% total shareholder return for FY2026, compounds independently of any multiple re-rating at all. Bulls are paying for a cleaned-up PF book and a rising payout ratio at half of book value; bears point to the first-half income decline, an NPL coverage ratio that fell below the company's own 100% comfort level, and an ownership structure, OK Financial Group's 9.99% stake sitting at the regulatory ceiling, that could cap how fast buybacks retire shares.

Management

The numerical record here is thinner than usual for a bank this size: insider-trading and compensation disclosures didn't surface in searchable Korean coverage this pass, and no single-figure FY2026 net-income or ROE guidance was found; the roughly 9.6% FY2026 net-income growth figure circulating in broker notes is analyst-attributed, not company-issued. What is verifiable is the capital-allocation pattern: the diluted share count has fallen from 169.1 million in FY2021 to about 158.4 million now, a 6.3% reduction funded by buybacks and cancellations management has said it will keep prioritizing over cash dividends, consistent with the CET1-tiered payout framework. The governance record's most consequential fact is ownership, not management: OK Financial Group, a savings-bank and consumer-lending group, built a 9.99% stake and became the largest shareholder in early 2026, sitting just under the 10% ceiling Korean law imposes on any single owner of a nationwide-bank holding company, ahead of the National Pension Service. On the July 27, 2026 call, the H1 2026 print, CFO Cheon Byeong-gyu framed the priority as managing the capital ratio stably through year-end while leaning on tax-exempt dividend restructuring and buyback-driven cancellation, without volunteering a coverage-ratio timeline beyond the roughly 100% year-end target already in the earnings materials.

The linchpins

The most immediate risk is the NPL coverage ratio itself: reserves against non-performing loans fell from 93.6% to 82.2% (ex-regulatory-reserve basis) in a single quarter, below the roughly 100% level management has publicly targeted rebuilding by year-end. This is a quarterly-horizon signal, visible in the next one to two prints, and it matters because a bank letting coverage slide while loan growth accelerates, won loans just crossed ₩60 trillion, is choosing growth over conservatism at the exact moment its PF cleanup story needs the market's trust. It is falsified if the third and fourth-quarter 2026 prints show coverage continuing to fall short of the 100% target or trending down further, which would read as under-reserving into a fresh credit cycle rather than a one-quarter blip; it is confirmed if coverage visibly rebuilds toward or above 100% by year-end as management has committed to.

A structural risk sits in the ownership table: OK Financial Group's 9.99% stake is parked just under the 10% ceiling that would trigger a formal major-shareholder-suitability review, and every share iM Financial cancels via buyback mechanically raises OK's percentage of a shrinking float, creating friction between the stated ₩150 billion cumulative buyback target and the ownership cap. This is structural, not a single quarter's news, because it caps how aggressively the buyback engine, the primary lever behind the 40%-to-50% shareholder-return framework, can run without forcing a regulatory reckoning or a sale by OK. It is falsified if the overhang visibly slows or halts buyback and cancellation activity, or if OK is compelled into a formal suitability review that unsettles the stock; it is confirmed if buybacks continue on the stated cadence toward ₩150 billion by 2027 without OK's position becoming a binding constraint, or if OK's stake is resolved cleanly.

The cyclical question is whether the first half's income decline reverses: broker commentary treats roughly 9.6% full-year 2026 net-income growth as the implicit bar, and the ₩295.6 billion first-half print, down 4.4% year over year, means the second half needs to swing back to double-digit growth to hit that level, let alone the roughly ₩492 billion consensus full-year net income. It matters because the entire capital-return and re-rate case assumes FY2025's earnings normalization was a new, higher base, not a one-year bounce; a soft second half would undercut that assumption directly. It is falsified if second-half 2026 net income fails to clearly outgrow the second half of 2025 and full-year 2026 lands meaningfully below the roughly ₩492 billion consensus; it is confirmed if second-half growth reaccelerates and the full year comes in at or above that consensus level, validating the post-PF-cleanup earnings base as durable.

Closing

The evidence leans toward the first half being normalization noise on top of a real repair, not a fresh credit air pocket: the PF book's NPL ratio kept improving through March 2026 even as headline income dipped, and management has already named both the cause of the dip and a concrete year-end target for rebuilding coverage, rather than staying vague about it. The clearest tell arrives on the NPL coverage ratio: watch whether it visibly rebuilds toward the roughly 100% level management has targeted in the third and fourth-quarter 2026 prints, due over the following few months. This works if coverage rebuilds on schedule, second-half net income reaccelerates enough to clear the roughly ₩492 billion full-year consensus, and the OK Financial ownership overhang stays passive enough not to constrain the buyback program; it fails if coverage keeps sliding, the second half stays soft, or the overhang visibly caps the capital-return engine that is doing half of this thesis's work.