Korean financial companies are mostly bought for reasons other than their operating performance. Banks trade on governance and value-up programmes. Samsung Life trades on a shareholding it cannot sell. In that context Samsung Fire & Marine Insurance Co., Ltd. (KRX:000810) is unusual, because the operating numbers have been getting better for four years and continue to.
Operating income: ₩1,506.9bn in FY2021, ₩2,044.7bn in FY2022, ₩2,357.3bn in FY2023, ₩2,649.6bn in FY2024, ₩2,659.1bn in FY2025. Up 76% across the period.
The first half of 2026 produced ₩1,746.5bn against ₩1,603.8bn a year earlier, up 8.9%, with the two quarters growing 8.7% and 9.1% respectively. After a flat FY2025 the growth resumed.
The investment side is the cleanest evidence, because it is the least susceptible to accounting judgement.
Interest income by quarter, oldest to newest across the eight on file: ₩569.9bn, ₩578.9bn, ₩583.2bn, ₩586.3bn, ₩593.7bn, ₩608.0bn, ₩624.2bn, ₩635.7bn.
Eight consecutive increases, with no reversals. Annually the line went from ₩1,740.6bn in FY2021 to ₩2,371.3bn in FY2025, up 36.2%.
That is what a bond portfolio does when it is steadily reinvesting maturing paper bought in the low-rate years into higher-yielding paper. It is slow, mechanical and durable, and it will keep working for as long as reinvestment yields exceed the coupons rolling off.
The comparison worth making is with the affiliate. Samsung Life's interest income has been essentially flat since FY2023, at ₩8,437.0bn, ₩8,461.0bn and ₩8,406.6bn. Same country, same rate environment, same three years. Samsung Fire's grew.
The cash flow statement records the same improvement more dramatically.
Cash from operating activities was negative ₩8.5bn in FY2022 and negative ₩562.1bn in FY2023. Then ₩3,127.0bn in FY2024 and ₩3,413.2bn in FY2025.
The first half of 2026 produced ₩2,482.3bn against ₩1,607.4bn in the same period of 2025, up 54.4%, so the trend has continued and accelerated.
An insurer generating ₩3.4tn of operating cash on ₩115.5tn of assets is funding its own growth, its own dividend and its own investment portfolio from premiums and investment income rather than from financing. Financing activities have been an outflow in every year on file.
Dividends paid: ₩375.0bn in FY2021, ₩511.1bn in FY2022, ₩587.8bn in FY2023, ₩681.1bn in FY2024, ₩808.4bn in FY2025. Up 115.6% in four years.
In the first half of 2026 the figure was ₩829.6bn, already exceeding the whole of last year.
Against FY2025 net income of ₩2,020.3bn, the ₩808.4bn payment is a payout ratio of 40.0%. Against the market capitalisation of ₩30.45tn at the August 27 close of ₩682,000, the first-half payment is a yield of about 2.72%.
Retained earnings have grown alongside it, from ₩10,270.7bn at the end of FY2021 to ₩15,076.9bn at June 30, so the distribution is being funded out of earnings rather than out of the balance sheet.
One number in the series looks like a break. Net income fell from ₩2,076.8bn in FY2024 to ₩2,020.3bn in FY2025, a decline of 2.7%, and it was the only year of the five that went backwards.
Pretax income did not fall. It rose from ₩2,744.5bn to ₩2,783.3bn. What changed was tax: ₩667.8bn to ₩763.0bn, lifting the effective rate from 24.3% to 27.4%.
So the entire decline in reported net income was a tax rate move, not an operating one. Anyone screening on net income growth would have seen a deterioration that did not happen in the business. The fourth quarter of 2025 is where most of it landed, with a ₩176.4bn tax charge on ₩410.8bn of pretax income, an effective rate of 42.9%.
Three things temper all of this.
The first is the fourth quarter. Operating income was ₩252.8bn in the fourth quarter of 2024 and ₩391.3bn in the fourth quarter of 2025, against ₩717.5bn and ₩664.1bn in the respective third quarters. Korean insurers update actuarial assumptions annually and the review lands in Q4, so the final quarter is structurally weak and can be much worse than expected. Anyone annualising a strong first half at this company will overshoot.
The second is that Korean motor insurance is a regulated line. Pricing is effectively supervised, and premium rates have been cut under political pressure more than once when the industry looked too profitable. A P&C insurer earning record margins in Korea attracts exactly that attention.
The third is that the summary financials do not carry an insurance service result line for this company, so the split between underwriting profit and investment profit is not visible from what I have been working from. The notes to the half-year report carry it, and anyone wanting to know whether the long-term health book or the investment portfolio is doing the work needs to read them. That distinction matters, because one of those is a franchise and the other is a bond ladder.
Credit loss provisions are also worth tracking. They ran ₩19.3bn in FY2023, spiked to ₩148.8bn in FY2024, and came in at ₩102.1bn in FY2025. A sevenfold jump in one year at an insurer usually means a specific exposure went wrong.
The fourth quarter, reported in February. Two consecutive weak Q4s have been the pattern, and the direction of the third would tell you whether the annual assumption review keeps producing adverse outcomes or whether FY2024 and FY2025 were transition effects working through.
The second marker is the interest income line. Eight consecutive quarterly increases is a strong run, and it continues only while reinvestment yields exceed what is maturing. The first sequential decline would mark the point where the tailwind from the rate cycle stops, and at a company earning ₩2.4tn a year from that line, it matters more than anything happening in motor insurance.
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.