Almost everything written about Samsung Life Insurance Co., Ltd. (KRX:032830) is about the Samsung Electronics shares it owns, and with some reason: those shares turned ₩33.66tn of equity a year ago into ₩146.96tn at June 30. But the company also sells life insurance to about a fifth of South Korea, and that business has been getting smaller for four years.
The number that shows it is one US readers may not know.
Korean insurers report under IFRS 17, which the industry adopted in 2023 and which does something US GAAP does not: it separates the underwriting business from the investment business on the face of the income statement.
The insurance service result is revenue recognised from insurance contracts less the expenses of servicing them. It captures the release of stored profit on policies already sold, plus how actual claims and lapses compared with what was assumed, plus the effect of changing those assumptions. It excludes investment returns entirely, which sit on separate lines.
So it is the cleanest available answer to a simple question: is the act of underwriting and servicing insurance making money, and how much?
The series, from the point where IFRS 17 comparatives begin: ₩1,667.8bn in FY2022, ₩1,448.2bn in FY2023, ₩536.6bn in FY2024, ₩981.0bn in FY2025.
Down 41% from the FY2022 figure, with a very poor FY2024 and a partial recovery last year.
The FY2021 entry in the accounts shows negative ₩7,131.4bn, but that predates the standard's adoption and reflects transition rather than underwriting, so it does not belong in the comparison and I would ignore it.
The quarterly picture is worse and more recent. The second quarter of 2026 produced ₩271.4bn against ₩560.8bn in the second quarter of 2025. That is a decline of 51.6% in the underwriting result, year on year, in a single quarter.
The first quarter of 2026 came in at ₩251.2bn. The third quarter of 2025 at ₩268.4bn. And the fourth quarter of 2025 was negative ₩125.9bn.
So four of the last five quarters on record have produced an insurance service result between negative ₩126bn and ₩271bn, against ₩561bn in the second quarter of 2025 and ₩477bn in the third quarter of 2024. The step down is not one bad quarter.
Anyone reading the half-year headline gets a different impression. Operating income for the first half of 2026 was ₩1,980.4bn against ₩1,669.4bn a year earlier, up 18.6%. Net income was ₩1,967.7bn against ₩1,471.1bn, up 33.8%.
Split it. The first quarter produced ₩1,357.8bn of operating income, the highest of the eight quarters on file and up 80% on the same quarter of 2025. The second produced ₩622.6bn, down 32.0% year on year.
One exceptional quarter carried the half. Since the insurance service result was similar in both quarters, at ₩251.2bn and ₩271.4bn, the swing came from the investment and insurance finance lines rather than from underwriting, which is consistent with a company whose asset side was moving violently in the first quarter of 2026.
Which is the general problem with quarterly numbers at a life insurer. Interest rate moves, equity market moves and assumption updates all land on the income statement in lumps, and none of them tells you whether the policies being sold this year are good business.
Look at Q4 in both years available. Operating income of ₩51.7bn in the fourth quarter of 2024 and ₩95.3bn in the fourth quarter of 2025, against ₩796.2bn and ₩815.8bn in the respective third quarters.
That is not seasonality in the ordinary sense. Life insurers review their actuarial assumptions annually, and the review typically lands in the final quarter. Mortality, morbidity, lapse rates, expense assumptions, and the discount rates applied to long-dated liabilities all get updated, and any adverse movement is recognised then.
Two consecutive fourth quarters near zero operating income says the annual review has been going against this company. The negative ₩125.9bn insurance service result in the fourth quarter of 2025 is the same signal in the underwriting line specifically.
Anyone modelling FY2026 should assume the fourth quarter is small and possibly negative, because that is what the last two did.
Two counters are worth weight.
The first is that IFRS 17 makes the insurance service result depend heavily on how much stored profit is being released in a period, and a lower release can mean the company is holding profit back for later rather than earning less. The measure that settles this is the contractual service margin, the balance of unearned profit on policies already written. A rising CSM with a falling service result means good new business and deferred recognition. A falling CSM with a falling service result means the book is running off faster than it is being replaced.
That number is in the notes to the half-year report and it is not in the summary financials I have been working from. Anyone doing serious work on this company should read it before drawing conclusions from anything above.
The second counter is scale. The insurance business is not why anyone owns this stock. At the August 27 close of ₩307,000 across 200,000,000 shares, Samsung Life is capitalised at ₩61.40tn, against a Samsung Electronics stake worth well over twice that. Whether underwriting produces ₩1.0tn or ₩1.5tn a year matters far less to the share price than what happens to one holding and to Korean insurance legislation.
Which is true, and is also an odd thing to be relaxed about. A company whose operating business is deteriorating while its share price depends on a stake it cannot sell is not obviously a safe place to be indifferent.
The contractual service margin balance and new business CSM in the FY2026 annual report, due next March. Those are the two numbers that tell you whether the underwriting franchise is growing or running off, and they cannot be inferred from the summary income statement.
The nearer marker is the fourth quarter, reported in February. A third consecutive Q4 with near-zero operating income and a negative insurance service result would establish that the annual assumption review keeps producing adverse outcomes, which is a statement about the quality of the in-force book rather than about markets. A normal fourth quarter would suggest the last two were the tail of the IFRS 17 transition working through.
kstock reads DART every morning and writes up what moved — the contract, the buyback, the number that does not add up. The daily post and a Saturday roundup, by email.