IDI Insurance: How Much of 2025 Was Core Underwriting and How Much Came From Assumptions, Financing, and Legal Noise
IDI Insurance ended 2025 with NIS 381.1 million of net profit, or roughly NIS 416 million excluding legal provisions in the presentation bridge. This follow-up shows that the improvement was more real than the headline alone suggests, but not all of it came from clean underwriting: life was mostly operational, health was still moved by financing, and part of the value was pushed forward into CSM under IFRS 17.
What This Follow-Up Is Isolating
The main article already argued that underwriting moved back to the center, while capital still sets the ceiling. This continuation isolates the more forensic question: how much of 2025's earnings really came from ongoing insurance economics, and how much came from legal provisions, financing effects under IFRS 17, and value that was pushed forward into CSM.
That question matters even more in the first full year under IFRS 17. The annual picture looks strong: net profit rose to about NIS 381.1 million, and the company presents roughly NIS 416 million after excluding legal provisions. But profit on its own does not answer the quality question. It has to be broken into insurance service, financing, actuarial assumption effects, and legal noise.
Four points stand out immediately:
- Legal noise distorted the headline, but it does not explain all of 2025. Excluding legal provisions, annual net profit rises from about NIS 381 million to about NIS 416 million, but company-level insurance service result net of operating expenses rises from about NIS 450 million to about NIS 503 million. In other words, the core business improved even after the noise is stripped out.
- The fourth quarter shows the difference between underwriting and financing. Company-level insurance service result net of operating expenses rose to about NIS 113 million after normalizing legal provisions, up from about NIS 94 million a year earlier, but normalized net profit slipped slightly to about NIS 109 million from about NIS 111 million. The service layer improved, the financing tailwind did not.
- In life, the jump was mostly operational. Insurance service result rose to about NIS 132.6 million from about NIS 51.8 million, while insurance finance income rose only to about NIS 16.0 million from about NIS 11.8 million.
- In health, the picture is less clean. Insurance service improved, but part of the annual gap also came from financing, which swung from an expense of about NIS 4.3 million in 2024 to income of about NIS 7.5 million in 2025.
First Strip Out the Legal Noise
The 2025 accounting headline was meaningfully distorted by legal noise. In the presentation, net profit is shown at about NIS 381 million, or about NIS 416 million excluding legal provisions, versus about NIS 306 million and about NIS 328 million respectively in 2024. The same adjustment exists in the fourth quarter, but it is much smaller: about NIS 106 million reported versus about NIS 109 million normalized, compared with about NIS 95 million versus about NIS 111 million a year earlier.
That chart matters because it avoids two common reading mistakes. The first is to assume that all of the 2025 improvement came from removing a one-off line. That is not correct. Even after legal provisions are excluded, company-level insurance service result net of operating expenses still rises by about NIS 129 million. The second mistake is to assume that the fourth quarter was especially strong simply because reported net profit stayed above NIS 100 million. That is not quite right either. On a normalized basis, fourth-quarter net profit was essentially flat versus last year.
One more detail keeps this from being dismissed as old noise. The year-end provision balance for all pending legal claims stood at about NIS 113 million at the end of 2025, up from about NIS 60 million at the end of 2024. So this is not only a line that ran through the P&L and disappeared. It remains a stock of potential volatility.
But it still has to be placed correctly. The board report makes clear that the unusual legal-provision effect sits in general insurance, and the general-insurance segment as a whole carried about NIS 52 million of that effect in 2025 versus about NIS 32 million in 2024. The presentation also explicitly flags the other property and liability line as including legal provisions. So the legal provisions matter for the reported profit headline, but they are not the main explanation for the jump in life and health.
Life: Earnings Jumped, but Mostly Through Insurance Service
In life, the 2025 picture is good and relatively clean. Gross premiums rose about 5% to roughly NIS 373.2 million. Insurance service result jumped to about NIS 132.6 million from about NIS 51.8 million. Even after operating expenses, the presentation shows a rise from about NIS 41.6 million to about NIS 121.6 million. That is not cosmetic. It is a real move in the core economics of the line.
The company explains most of the move through the claims ratio: in 2025 it was below normal, while in 2024 it was above normal, partly because of the war backdrop. It also says the fourth quarter included the effect of assumption changes following actuarial studies. But even here, the relative weight of the lines matters. Insurance finance income rose only to about NIS 16.0 million for the year, from about NIS 11.8 million the year before. The main driver was insurance service, not financing.
That distinction becomes even clearer in the fourth quarter. Insurance service result rose to about NIS 38.1 million from about NIS 12.3 million, and insurance service result net of operating expenses rose to about NIS 35.6 million from about NIS 9.6 million. At the same time, insurance finance income fell to about NIS 3.8 million from about NIS 15.1 million. The result is that pre-tax profit still rose to about NIS 39.2 million from about NIS 24.9 million, but by much less than the underlying service line itself.
That is exactly the right test for earnings quality. If financing weakens and profit still goes up, the operational engine is stronger than the headline alone suggests. That is what happened in life in 2025.
Health: The Year Improved, but the Quarter Reminds You Financing Still Matters
The health picture is more mixed. On the one hand, there was real operational improvement here too. Gross premiums rose about 9% to roughly NIS 311.2 million. Insurance service result rose to about NIS 104.7 million from about NIS 89.4 million, and the presentation shows insurance service result net of operating expenses rising from about NIS 79 million to about NIS 97 million. The company attributes that to higher activity and lower claims.
On the other hand, financing mattered more here. Insurance finance moved from an expense of about NIS 4.3 million in 2024 to income of about NIS 7.5 million in 2025. That is why pre-tax profit rose to about NIS 104.4 million from about NIS 75.3 million, a faster increase than the service layer alone would imply.
The fourth quarter is where the true quality test shows up. Insurance service result rose to about NIS 22.9 million from about NIS 19.1 million, and insurance service result net of operating expenses rose to about NIS 21 million from about NIS 16 million. And yet pre-tax profit fell to about NIS 20.1 million from about NIS 29.8 million because financing flipped to an expense of about NIS 1.1 million versus income of about NIS 14.5 million in the comparable quarter.
That is the key point. A fast annual reading makes health look like a much cleaner profit engine. The embedded quarter says something more careful: the service result did improve, but the bottom line in health is still materially exposed to the curve and the financing line.
| Metric | Life 2024 | Life 2025 | Health 2024 | Health 2025 |
|---|---|---|---|---|
| Insurance service result | 51.8 | 132.6 | 89.4 | 104.7 |
| Insurance service result net of operating expenses | 41.6 | 121.6 | 79 | 97 |
| Insurance finance income | 11.8 | 16.0 | -4.3 | 7.5 |
| Pre-tax profit | 53.2 | 135.2 | 75.3 | 104.4 |
That table sharpens the difference between the two lines. In life, most of the improvement came from insurance service. In health, the service layer improved too, but the annual gap was also meaningfully helped by financing.
IFRS 17: What Went Through Current Earnings, and What Was Deferred Into CSM
This is where the continuation becomes genuinely useful. The company explicitly says that when demographic assumptions change in life and health under IFRS 17, part of the effect no longer goes straight through current earnings. Instead, it is booked into CSM and spread over the life of the policy. That is why 2025 cannot be read as if every actuarial improvement already became current-year profit.
Note 16 gives the hard breakdown for 2025. Across the segments, assumption changes contributed only about NIS 2.2 million to insurance service result and another about NIS 15.4 million to investments and financing, while at the same time increasing CSM in retention terms by about NIS 50.1 million. In plain language, more value was pushed forward into future years than was left inside 2025 earnings.
The split between life and health is even more telling:
- Life: assumption changes added about NIS 9.3 million to insurance service result and about NIS 9.4 million to financing, while the net CSM movement was close to flat at only about NIS 0.5 million. In life, 2025 was mainly a year of stronger current operating profit, not mainly a year of pushing value into the future.
- Health: assumption changes reduced current insurance service result by about NIS 7.1 million, added about NIS 8.1 million to financing, and at the same time moved about NIS 49.5 million into CSM. The appointed actuary's statement points in the same direction: excluding the curve effect, the annual studies improved BE in retention terms by about NIS 45 million, while RA increased by about NIS 3 million.
That leads to a very clear analytical conclusion. In health, a meaningful part of the 2025 improvement did not become current-year earnings. It became future profit stock. That is exactly what IFRS 17 is designed to do, but it is also exactly what can mislead a reader who focuses only on the bottom line.
CSM Grew, but Not Because the Company Harvested Past Profit More Aggressively
To understand 2025, you need to look not only at profit, but also at CSM itself. The presentation shows total CSM rising from about NIS 919 million at the end of 2024 to about NIS 987 million at the end of 2025. But that growth was not evenly shared:
- In life, CSM rose only from about NIS 323 million to about NIS 332 million.
- In health, CSM rose from about NIS 596 million to about NIS 655 million.
Here too, it is important not to jump to the easy conclusion. If CSM grew, that does not mean the company simply harvested less of it into current profit. Note 20 shows that CSM recognized in profit or loss for services provided was almost flat across life and health together: about NIS 141.2 million in 2025 versus about NIS 140.3 million in 2024. In life, the recognized amount actually fell slightly, from about NIS 66.7 million to about NIS 64.0 million, while in health it rose from about NIS 73.6 million to about NIS 77.2 million.
That means the jump in 2025 earnings did not come from a much faster harvesting of old stored profit. It came more from stronger current service economics, financing swings, and the fact that part of the actuarial improvement did not pass through current earnings at all and was instead retained inside CSM.
The company also explains why CSM rose in each segment. In life, the increase came from the effect of unwinding discount rates and from new-business additions exceeding the release from the existing book, despite a negative effect from assumption changes. In health, the increase came from new-business additions exceeding the release from the existing book, except in personal accident, together with rate effects and assumption changes. So CSM growth is not proof that 2025 was weak. It is proof that part of the improvement was carried forward.
So How Much of 2025 Was Really Operational
The short answer is that 2025 was a better operational year, but not a perfectly clean year.
In life, most of the improvement was clearly real. Insurance service improved far more than financing, and in the fourth quarter financing actually weakened while profit still rose. That is a good quality signal.
In health, the answer is more cautious. The insurance service line improved there too, but part of the annual improvement came from the interest-rate swing, and the fourth quarter showed that the bottom line can still weaken even when underwriting looks better.
And above both segments sits IFRS 17. In 2025 it did not only change presentation. It also changed timing. Part of the actuarial improvement ran through financing, part stayed out of current earnings and was deferred into CSM, and the company itself says in both the board report and the actuarial statements that the first year of application still involves systems, processes, and controls that remain in implementation.
So the right reading of 2025 is not "it was all market" and not "it was all underwriting." The right reading is more precise:
- Legal noise weighed on the headline, but it does not explain the core improvement.
- In life, most of the improvement was underwriting and operating quality.
- In health, service improved but financing still shaped the bottom line.
- Part of 2025's value did not land in the year's earnings at all and was retained in CSM for future periods.
That is why the next test is not simply whether the company can print another high profit number. The real test is whether insurance service result in life and health stays strong when the curve stops helping, and whether the larger CSM balance actually turns into cleaner earnings over the next few quarters without another big round of assumption changes or another heavy legal distortion.
Disclosure: Deep TASE analyses are general informational, research, and commentary content only. They do not constitute investment advice, investment marketing, a recommendation, or an offer to buy, sell, or hold any security, and are not tailored to any reader's personal circumstances.
The author, site owner, or related parties may hold, buy, sell, or otherwise trade securities or financial instruments related to the companies discussed, before or after publication, without prior notice and without any obligation to update the analysis. Publication of an analysis should not be read as a statement that any position does or does not exist.
The analysis may contain errors, omissions, or information that changes after publication. Readers should review official filings and primary sources before making decisions.