Hachshara After Best Invest: Why Fee Income Is Rising While CSM Is Falling
In 2025 Hachshara sharply improved life-segment profitability, but much of the lift came from the Best Invest and investment-contract fee engine while ending CSM fell to NIS 284.8 million and the amount slated for recognition over the next two years slipped to NIS 52.2 million. This continuation breaks down the gap between current earnings and future embedded profit.
The main article argued that Hachshara finished 2025 with better earnings, but not with a genuinely wide capital cushion. This follow-up isolates the second paradox inside the year: the Best Invest and investment-contract fee engine pushed profit higher at the same time that the life segment's stock of future profit, the CSM, moved lower.
That is not an accounting quirk. It is a change in the earnings mix. On one side, pre-tax comprehensive profit in life and long-term savings swung to NIS 45.2 million from a NIS 9.9 million loss. On the other, ending CSM in the segment fell to NIS 284.8 million from NIS 335.2 million, and the amount expected to be recognized over the next two years fell to NIS 52.2 million from NIS 71.5 million.
So 2025 looks better in the income statement, but less generous in the stock of profit already sitting on the balance sheet for future recognition. That is the core of the story. Earnings improved, but a larger share now comes from fees recognized today, and a smaller share comes from deferred insurance margin still waiting to be earned.
The 2025 Earnings Lift Came From Fees
The simplest way to see it is through the life-segment bridge:
| Life-segment line | 2024 | 2025 | What it says |
|---|---|---|---|
| Pre-tax comprehensive profit | (NIS 9.9m) | NIS 45.2m | Sharp swing from loss to profit |
| Net result from insurance and investment | (NIS 29.2m) | NIS 9.0m | Real improvement, but still relatively small |
| Fee income from investment contracts | NIS 155.4m | NIS 202.3m | Up 30.1% |
| Other operating expenses | NIS 136.1m | NIS 166.1m | The cost layer also moved up |
The key number here is simple arithmetic. The improvement in pre-tax comprehensive profit in the life segment was NIS 55.1 million. The increase in fee income from investment contracts alone was NIS 46.9 million. In other words, if fee income had stayed at its 2024 level, the life segment would have ended 2025 roughly around break-even.
That move did not come out of nowhere. Gross inflows into Best Invest rose to NIS 6.19 billion from NIS 5.70 billion, up 8.5%, while redemptions were broadly unchanged. At the same time, assets under management in finance and pure savings rose to NIS 26.8 billion from NIS 20.7 billion, up 29%. When the asset base grows like that, fee income has a strong built-in tailwind.
The chart shows how much the fee engine changed in 2025. The three disclosed fee lines add up to about NIS 252.0 million, versus NIS 188.8 million in 2024, an increase of 33.5%. The sharpest jump came from variable fees in participating policies, which rose to NIS 21.0 million from NIS 1.5 million. That is another important signal on earnings quality: part of the improvement came not only from higher assets under management but also from market conditions that allowed the company to collect far more variable fees.
Management also frames the story at the product level. Fixed management fees collected from Best Invest and Maslulit policies rose to about NIS 231.0 million from NIS 187.3 million. This is a real fee platform, not a one-off line. But it is still a fee platform, not a larger stock of deferred insurance profit.
CSM Moved The Other Way
This is where the follow-up matters. The contractual service margin represents profit not yet realized, profit the company expects to recognize as it provides future services. That is exactly where 2025 does not look like a breakout year.
The amount recognized in profit or loss for services provided fell to NIS 32.3 million from NIS 44.3 million. At the same time, contracts first recognized during the period added only NIS 22.2 million to CSM, versus NIS 15.6 million in 2024. That is a healthy improvement in new business, but it still does not refill the bucket. On top of that, changes in estimates that adjusted CSM took away NIS 47.9 million in 2025, versus NIS 42.8 million in 2024.
What makes the picture more important is that the CSM pressure did not come only from normal release. During the year the company ran mortality, morbidity, and lapse studies, and it states that these studies did not have a material effect on 2025 profit but could have a material effect on future years through CSM. The assumptions table already points in that direction: updated mortality, morbidity, lapse, and discount-rate assumptions together reduced life-segment CSM by NIS 11.5 million.
So even while new business improved, and even while current earnings improved, the future margin stock was still being hit from several directions at once. That is exactly the point a quick read of net profit can miss.
The second chart matters as much as the first one. It shows that the issue is not only a smaller CSM balance. The near-term part of the balance shrank faster than the total. The amount expected to be recognized within one year fell to NIS 27.5 million from NIS 38.1 million. Within two years, as noted above, it fell to NIS 52.2 million from NIS 71.5 million. By the end of 2025, 60.7% of life CSM sat beyond five years, versus 55.0% a year earlier.
One nuance matters here. This is not a forecast for total life-segment profit. It is only the release schedule of the CSM that already existed at the balance-sheet date, without future new business and without future CSM changes. Even with that limitation, the conclusion is clear. The remaining margin stock is both smaller and pushed further out.
Distribution Cost Still Takes A Large Cut
To understand why fee income can rise while CSM falls, it helps to look at distribution. In life, including Best Invest, sales are made mainly through agents. Some commissions are paid as advances in the first year of the policy, and the company explicitly discloses that the average commission rate as a percentage of new annualized premium, excluding Best Invest, reached 99.8% in 2025, after 97.2% in 2024 and 56.3% in 2023. At the same time, the company is dependent on one agent whose premiums account for more than 10% of life and long-term savings premiums.
| Distribution-cost layer | 2024 | 2025 | Why it matters |
|---|---|---|---|
| Average commission as % of new annualized premium, excluding Best Invest | 97.2% | 99.8% | Almost the full first-year premium is absorbed by commission |
| Acquisition expenses for insurance contracts in life | NIS 21.6m | NIS 32.9m | Up 52.1% |
| Asset for costs to obtain investment-management service contracts | NIS 192.3m | NIS 245.2m | A cost layer deferred onto the balance sheet |
| Additions to that asset | NIS 46.2m | NIS 72.8m | Growing the fee engine requires more upfront spend |
The last line is one of the more important note disclosures in this continuation. The asset for costs to obtain investment-management service contracts rose to NIS 245.2 million from NIS 192.3 million. The accounting policy makes clear that this asset includes commissions to agents and acquisition supervisors, together with related administrative and general costs tied to new contracts, and that it is amortized over the contract life. In other words, part of the selling cost of the fee engine does not hit the income statement immediately. It accumulates on the balance sheet and gets recognized over time.
The subtler point is recoverability. The company states that the recoverability test relies on assumptions that include lapse rates, operating expenses, and asset returns. That does not mean there is currently an accounting problem. It does mean today's fee story also relies on future persistency and the ability to keep earning management fees over time. Put differently, the fee engine is working, but it is not a cost-free margin stream and it is not detached from future portfolio behavior.
That is exactly where the gap with CSM becomes easier to understand. CSM is created when there is future insurance profit still waiting to be recognized. Best Invest, by contrast, generates current-period income based on assets under management, contributions, and market conditions, while requiring a heavy distribution layer from day one. So it is entirely possible to see rising fee income and a falling stock of future profit in the same year.
Bottom Line
The 2025 paradox is real. Hachshara earned more in the life segment, but not because the stock of future profit became larger. It earned more because the fee and asset-accumulation engine became stronger, because the investment and finance loss narrowed, and because insurance service profit improved. Those are all good engines. They are just not the same thing as a growing CSM.
That does not cancel the improvement. It simply puts it in the right place. Anyone looking only at the swing from loss to profit could conclude that life is back on a broad and durable track. The more precise reading is more restrained: the fee engine became larger and faster, while the contractual service margin became smaller and less near-term.
That is why the 2026 test will not be only whether life stays profitable. It will be much more specific: can Best Invest keep expanding assets under management without another sharp acceleration in distribution cost, can CSM stop shrinking, and can more of life earnings come from insurance services rather than from a fee engine that depends on inflows, persistency, and markets.
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.