Mor Gemel Pensia: The Gemel Engine Already Works. Now Pension Has To Stop Eating Capital

Getting To Know The Company

Mor Gemel Pensia is not a classic insurer, and it is not a brokerage-driven investment house living off trading spreads. It is a long-term savings manager whose economics are built mostly on three moving parts: assets under management, the fee levels it can charge on those assets, and the price it pays to gather and retain that money. Anyone looking only at the jump in profit is missing the core point. The gemel engine already works as a scale business. Pension is still in the phase of building mass, distribution, and operating density.

What is working right now is fairly obvious. At the end of 2025 the company managed NIS 121.8 billion, versus NIS 87.1 billion a year earlier. Revenue rose 24.6% to NIS 655.4 million, pretax profit rose 47.6% to NIS 110.3 million, and net profit reached NIS 70.5 million. The fourth quarter was also strong: revenue rose to NIS 182.4 million and pretax profit reached NIS 28.0 million. On the surface, that looks like a clean breakout year.

But the picture is less clean than the headline suggests. Operating cash flow was only NIS 27.9 million, far below net profit, mainly because the company is still loading deferred acquisition costs onto the balance sheet at a rapid pace. At the same time, capital above the regulatory minimum stood at NIS 72.1 million at year-end. That is real cushion, but not the kind that lets investors ignore the additional dividend approved in March 2026 and the repurchase plan of up to NIS 20 million. This is no longer a question of whether the company can grow. It is a question of how it funds that growth without blurring earnings quality and without eating too quickly into its capital buffer.

That is exactly what the market has to measure over the next 2 to 4 quarters. If strong flows and transfers continue, if operating costs really start to bend down on a unit basis, and if pension keeps moving toward profitability without another surge in acquisition costs, the read on 2025 will improve. If not, 2025 may end up looking like a year with very strong numbers but less clean underlying cash economics.

One more early screen matters here. This is already not a small platform. In 2025 the company had 190 employees, versus 151 in 2024, and it managed more than one million accounts. That matters because the key question is no longer whether Mor has a brand, licenses, or market presence. The real question is whether that scale is now translating into higher-quality profit, or mainly into accounting profit that still runs ahead of cash.

Events and Triggers

The Balance Sheet Moved Up A Level

The big 2025 event was the move from a financing structure that leaned more on banks and owner support toward a more market-based funding structure. In November 2025 the company issued Series A bonds with NIS 250 million nominal value, together with 7.5 million listed warrants, at a fixed 3% interest rate. This was not just a fundraising event. It was also a signal of balance-sheet maturity, capital-market access, and a shift toward longer-dated debt with ten equal annual principal payments through 2035.

That also changes the way the story should be read. At the report date the company had no shareholder loans, and the parent-company loan framework of up to NIS 50 million expired in June 2025 without being renewed. At the same time, the parent’s commitment to make available the means required for regulatory capital compliance is still in place. So parental support remains in the background as a regulatory backstop, but the day-to-day funding picture now looks much more public and self-standing than it did a year earlier.

Dividend, Buyback, And What That Really Signals

During 2025 the company paid NIS 57 million of dividends. On March 22, 2026 it approved another dividend of about NIS 14 million. On the same day it also approved a share repurchase plan of up to NIS 20 million running through March 21, 2027. From a market-signaling perspective, this is hard to ignore: the board is effectively saying the shares trade below economic value and that the company has enough sources to support both distribution and repurchases.

But the hard numbers matter. The company itself wrote that, after the March dividend, distributable profits under the profit test stood at about NIS 31.9 million. So the repurchase plan may pass the formal tests, but it already sits inside a profit-and-capital framework that is far from unlimited. This is not a signal of huge excess capital. It is a signal that management believes the next few years should generate new capital quickly enough.

The Pension Engine Just Got More Time

The most important operating trigger actually sits below the headline level. In September 2024 the company signed an amendment to its operating agreement, effective January 2025, under which the operator is supposed to begin providing the full pension operating services from July 2026. In October 2025, the company agreed to the operator’s request to delay that starting point by 2 to 4 quarters.

At first glance that is a negative. Delaying the pension operating transition could mean that future cost savings move further out. The company does say that a credit mechanism was agreed for the delay period, so the economic damage of the delay should be softened. The implication for investors is two-sided: there is some compensation built in, but part of the thesis around improving pension profitability still depends on future operational execution that has not happened yet.

Default Pension Status, But Not At Any Price

The comprehensive pension fund’s status as a default pension fund remains one of the company’s most important growth engines. That status was extended through October 31, 2028, and from June 2025 a new allocation mechanism took effect under which employers who did not run a competitive process must direct employees who did not choose a fund into a selected default fund according to the allocation rules. That is a real tailwind.

The problem is that this engine comes with a price. The company itself explains that pension profitability is pressured by fee erosion because of the pricing levels embedded in these tenders. So growth in pension does not automatically translate into profit. It first creates volume. Only later does the question become whether that volume can really carry itself.

Efficiency, Profitability and Competition

The core story of 2025 is real operating leverage, but not perfectly clean operating leverage. Revenue rose by NIS 129.3 million, while total costs rose by NIS 93.7 million. That spread explains the jump in pretax profit to NIS 110.3 million. The fourth quarter also showed strong acceleration: revenue rose 31.1% and pretax profit rose 56.5%.

The difference between reported profit and adjusted profit matters a lot here. In 2025 adjusted pretax profit stood at NIS 125.3 million, versus reported pretax profit of NIS 110.3 million. The nearly NIS 15 million gap comes from share-based compensation expense. In 2024 that gap was only about NIS 4 million. So anyone looking only at the 59% growth in adjusted pretax profit may be getting an overly rosy picture. Reported profit was strong as well, but part of the adjusted improvement comes from a much more aggressive exclusion of expense than in the prior year.

The other side of the story is that selling and acquisition costs are still rising quickly. Commissions, marketing, and acquisition expense rose 27.7% to NIS 332.5 million. The company attributes that mainly to ongoing agent commissions and to higher amortization of deferred acquisition costs because of strong fund raising. This is exactly where investors need to pause. AUM growth is not free. It is purchased.

Earnings Quality Runs Through Deferred Acquisition Costs

One of the most important numbers in the report is not in the income statement at all. It sits on the balance sheet. Deferred acquisition costs rose to NIS 409.4 million, from NIS 297.4 million at the end of 2024. That increase is not accidental or cosmetic. It reflects the fact that part of the cost of customer acquisition does not hit the P&L immediately, but is spread over years.

This is also a real accounting sensitivity point. The company amortizes deferred acquisition costs over 6 years in gemel and over 10 years in pension, and it reassesses both recoverability and the appropriateness of the amortization pace each year. That leads to a non-obvious conclusion: the bigger earnings-quality sensitivity today sits inside the mature gemel engine, not inside the younger pension engine.

That also helps explain why the external auditor flagged the impairment review of deferred acquisition costs as a key audit matter. This is not an accusation that the company is inflating earnings. The simpler point is that part of the 2025 improvement is clearly real, but it is still being measured through assumptions about customer life and the pace at which acquisition cost is recovered.

Gemel And Pension Are Two Different Economic Engines

Gemel is the company’s scale and profit engine. Its weighted market share rose to 10.41% from 9.10% a year earlier, and the company held 10.4% of industry gemel assets at the end of 2025. More than that, it captured 88.8% of industry net transfers and 35.7% of net accumulation in gemel. Those are unusual numbers. They say the brand, the distribution machine, and investment performance are all working.

Pension is a different story. Weighted market share rose to 1.43% from 1.03%, and assets jumped to NIS 16.4 billion. That is very fast growth, but it is still happening inside an industry where insurance groups control about 90% of assets. Fees reflect that pressure. In the comprehensive fund, average asset-based fees for active members fell to 0.19% from 0.21%, and in the general fund they fell to 0.18% from 0.21%. In other words, pension is growing, but it is growing in a market where price erodes quickly.

There is one real bright spot. In pension, the direct channel’s share of assets rose to 26% from 21%, while the agents channel fell to 74% from 79%. That is not a revolution, but it is movement in the right direction. If a larger part of future pension growth comes through direct distribution rather than only through agents, the economic quality of that growth can improve.

Competition Is Both A Moat And A Dependency

The company works through more than 3,000 pension agents, and the agents channel still controls 90% of gemel assets and 74% of pension assets. That gives Mor a strong and not easily replicated distribution engine, but it also creates high dependency. The company explicitly says that one agency house accounts for more than 10% of premium-fee flows in the operating segments. That is not an existential risk, but it is clearly a yellow flag. In this kind of business, distribution power is both a moat and a pressure point on margins.

Cash Flow, Debt and Capital Structure

If investors want to understand Mor Gemel Pensia without getting carried away by the profit line, they need the all-in cash flexibility view. On that test, 2025 was a good year, but much less shiny than the profit headline.

Operating cash flow came in at NIS 27.9 million, versus NIS 70.5 million of net profit. That gap is mainly explained by a net NIS 112.0 million increase in deferred acquisition costs. Put simply, the company books profit today on the basis of acquisition cost that will be spread over future years, while the cash goes out now. That is exactly why, in a company like this, accounting profitability and capital flexibility cannot be treated as the same thing.

At the same time, end-2025 liquidity was decent. The company had NIS 42.9 million of cash and cash equivalents, NIS 237.2 million of financial investments, and positive working capital of NIS 116.5 million. It also ended the year with NIS 247.0 million of equity against a minimum capital requirement of NIS 174.9 million. That left a surplus of NIS 72.1 million.

But this is where created capital and actually available capital need to be separated. On March 22, 2026 the company approved another dividend of about NIS 14 million. On the same day it also approved a buyback of up to NIS 20 million. If both moves are executed in full before fresh 2026 profits accumulate, the year-end surplus cushion will narrow meaningfully. That does not mean the company is in trouble. It does mean that NIS 72.1 million is not the kind of buffer investors should call fat.

Debt also improved, but it did not disappear. At the end of 2025 the company still had NIS 121.4 million of bank loans and NIS 228.6 million of bonds on the balance sheet. At the same time, the covenant picture looks far from stressed. Net financial debt to CAP stood at 22%, versus a 67.5% threshold for distributions. Equity stood at NIS 247 million, versus a NIS 160 million threshold for distributions. Coverage stood at 4.11, versus a minimum 1.1 threshold. Managed assets were roughly NIS 122 billion, versus a NIS 40 billion covenant floor. So the company’s problem today is not refinancing pressure. It is capital-allocation discipline.

There is also a subtler point. The parent company’s commitment to provide means for regulatory capital compliance still exists, but the parent loan facility of up to NIS 50 million expired in June 2025 and was not renewed. In parallel, the board froze the work of the independent committee that had been examining possible moves around the capital notes. This is a two-sided development: the company looks more self-standing, but it is also no longer leaning on the same immediately available parental liquidity pipe.

Outlook and Forward View

Four points matter before looking at 2026:

• The very strong 2025 profit did not translate into cash with the same force, because deferred acquisition costs rose sharply.
• The capital cushion looked comfortable at end-2025, but the March dividend and repurchase plan reduce the margin for error.
• Pension is growing faster than the company as a whole, but it still does not generate adjusted pretax profit, and management is marking 2027 as the target year for that turn.
• The 2026 guidance looks ambitious because it requires not only continued AUM growth, but a cleaner translation of scale into operating efficiency.

Management estimates 2026 adjusted pretax profit at NIS 140 million to NIS 160 million. That is an ambitious target, not just because of the headline number but because it requires continued asset growth, better cost control, and a larger contribution from scale to bottom-line profitability.

The important framing point is that management itself is not really presenting 2026 and 2027 as years in which pension is already a fully mature profit engine. In the investor presentation it marks 2027 as the year in which the pension segment is expected to cross into adjusted pretax profitability. That makes 2026 look less like a breakout year and more like a proof year. It is the year in which the company has to show that the scale it has already accumulated actually lowers service, operating, and distribution cost per unit of assets.

What could work in its favor? First, growth is still continuing. In the investor presentation, the company already showed about NIS 128 billion of AUM as of March 10, 2026, implying growth of roughly NIS 41 billion since the start of 2025. Second, gemel operations already look more mature, while pension benefits both from default-fund status and from a stronger direct channel. Third, the company is no longer in the licensing and set-up phase. It is at the stage where even relatively modest improvements in efficiency should start showing up in profit.

What is still not clean? Two things. First, the forecast still depends on market conditions and investment returns. In this kind of company that is part of the model, but it means not all of the improvement sits under management control. Second, the future operating savings in pension have not been proven in practice, and the transition to the expanded operator model has in fact been pushed beyond July 2026.

Put differently, if 2025 was the year of proving scale, 2026 will be the year of proving quality. The market will not be looking only for more growth. It will be looking for profit that starts to show up in cash, for capital cushion that stays healthy, and for pension to stop being only a volume story and become a contribution story.

Short Interest

Short interest in the stock is not extreme, but it is not zero either. As of March 27, 2026, short interest as a percentage of float stood at 1.58% and SIR stood at 3.25. That is a bit above the sector averages of 1.29% short float and 1.833 SIR, but still far from distress territory.

The more interesting point is the pattern. At the start of January 2026, short float jumped to 2.25% and SIR to 6.53, and then eased part of the way back. The read here is not that the market is pricing a collapse. It is pricing an argument. One side sees a company that may have grown too fast, with earnings partly helped by market conditions and DAC accounting. The other sees a company that has already crossed the scale threshold needed to show real operating leverage.

Risks

The first risk is continued price erosion. In pension this is already visible in average fee levels, and the company itself says the pressure comes from market structure and the selected-default-fund tenders. If flows keep arriving mainly at lower pricing, the company may continue to grow without moving toward higher-quality profitability at the same pace.

The second risk is distribution dependence. Ninety percent of gemel assets and 74% of pension assets are linked to the agents channel, and one agency house accounts for more than 10% of premium-fee flows. Any agent pressure on commissions, any shift in behavior by a large distributor, and any performance gap over time can roll fairly quickly into weaker gross inflows or a higher cost of gathering assets.

The third risk is earnings quality versus cash. As long as the deferred acquisition cost asset keeps growing quickly, investors have to live with a company that can look stronger in the income statement than in the cash-flow statement. That does not make the profit fake, but it does make it more sensitive to assumptions and to the economic life of the customer relationship.

The fourth risk is capital discipline. Capital above the regulatory minimum exists, but it is not huge. If the company keeps paying dividends, potentially executes buybacks, and still funds very fast pension growth, the margin for error can narrow quickly. The parent support commitment is helpful, but it is not a substitute for preserving internal capital buffer.

The fifth risk is pension operating execution. The company delayed the start of full pension operating services by the operator by 2 to 4 quarters, even if it also agreed on a credit mechanism for the delay period. That leaves part of the story dependent on another execution stage that still has not happened.

The sixth risk is markets, regulation, and operational incidents. The company’s revenue model depends on managed assets, which means it depends not only on net flows but also on market returns. The company also flags exposure to macro conditions, war-related disruption, regulatory change, cyber events, human error, and fraud. That is not background noise for a savings manager. It is part of the core business model.

Conclusions

Mor Gemel Pensia exits 2025 as a much larger, more profitable, and better-funded company than it was at the start of its public life. The gemel engine is already built and proven, and it now provides real scale. The main bottleneck has shifted elsewhere: can pension keep growing without leaning too heavily on cash and capital. In the near term, the market is likely to focus less on the profit headline itself and more on the link between flows, DAC, capital cushion, and the real progress of the pension segment.

Current thesis: gemel is already generating profit and capital, while pension is moving toward an inflection point, but 2026 still has to prove that this shift will show up in cash as well as in capital.

What changed versus the earlier understanding of the company: 2025 no longer looks like a pure growth story that endlessly consumes capital. The company moved into public debt, built a capital surplus, and showed meaningful profitability. What has not changed is that pension still does not fully pay for itself.

Counter thesis: 2025 was helped both by supportive market conditions and by DAC accounting that smooths part of the acquisition cost forward, so profit still looks better than actual capital flexibility.

What could change the market’s interpretation in the short to medium term: proof that 2026 adjusted profit also flows through to cash, that pension keeps growing without another blowout in commissions and acquisition expense, and that the capital cushion remains comfortable even after dividends and repurchases.

Why this matters: this is the exact point where a savings platform has to move from a story of scale to a story of quality. If that shift happens, the company starts looking more like a mature platform. If not, the market will go back to focusing on the price of growth and on who is really paying for it.

What has to happen next is fairly clear. Over the next 2 to 4 quarters the company needs to show three things: continued AUM growth without meaningfully worse pricing, better cash conversion of profit, and a healthy capital buffer even after shareholder distributions. If all three happen together, 2025 will look like the transition year from a capital-consuming growth platform to a more mature savings business. If one of them breaks, the profit headline will look less convincing.