Capital Point: How Much of 2025 Profit Is Actually Realizable

How Much of the Profit Actually Sits on Value You Can Touch

The main article argued that Capital Point's return to profit in 2025 leaned heavily on fair value and on the market backdrop. This continuation takes that claim one step further. Not all fair value is the same. Some of it sits against an observable market price. Some of it still depends on a comparable set or on a model. Some of it is optionality that management itself refused to book as profit. So the right question is not whether the profit is "real" or "not real". The right question is which part of it is already close to monetization, which part still depends on screens, time or execution, and which part was left outside profit until stronger proof exists.

The number that frames the whole discussion is straightforward. Capital Point ended 2025 with NIS 7.279 million of net profit, but the net movement in fair value through the investment book was NIS 11.55 million. In other words, without the fair value contribution the year would still have stayed below zero. That is not an accounting flaw. This is exactly how an investment entity works. The company measures investments at fair value through profit or loss, so earnings pass through the value of the book. But it does mean that 2025 profit is still not the same thing as free cash.

Even the layer that looks less accounting-driven is still linked to the market. Finance income rose to NIS 6.344 million, and NIS 4.311 million of that came from revaluation and interest on marketable securities and investment funds. Another NIS 1.874 million came from interest on bank deposits. Against that, operating cash flow remained negative at NIS 6.232 million. That is the point. There was a real accounting and economic recovery in the book during 2025, but it still did not translate into a cash pace that would justify reading all of the profit as if it were already sitting in the bank.

That chart sharpens why the headline "returned to profit" is not enough on its own. The two biggest engines of the year are not a recurring operating engine and they are not cash conversion. They are fair value changes and finance income, most of which is still tied to the financial portfolio. That is the core of this continuation. The question is not whether the profit exists in the financial statements. The question is how much of it can already be seen as monetizable, and how much still needs the market, time or a transaction.

The second chart ties the three key points together in one glance. Net profit is positive, the fair value contribution is bigger than that profit, and operating cash flow is negative. So 2025 profit is partly realizable, not yet fully cash-converted.

The Book Became More Market-Based, but Not Clean

If the focus is valuation quality, 2025 was actually a year of improvement. At the end of 2024, within the line item for investments in project companies and listed holdings, NIS 24.919 million was classified as Level 3 and only NIS 35.320 million as Level 1. By the end of 2025 that picture had changed: NIS 37.936 million sat in Level 1, NIS 8.201 million in Level 2 and NIS 16.172 million in Level 3. Put simply, the portion of the book that rests on observable or near-observable market inputs increased, while the portion that still depends on models decreased.

That also holds in a broader read of all financial assets measured at fair value. Level 3, including loans to project companies, fell from NIS 25.299 million at the end of 2024 to NIS 17.257 million at the end of 2025, while Levels 1 and 2 combined rose from NIS 66.112 million to NIS 97.211 million. This is not a complete cleanup of valuation risk, but it is a material shift away from paper value that is hard to test and toward a book that sits more on screens, prices and markets.

That chart matters for two reasons. On one hand, it says 2025 earnings are higher quality than 2024 earnings, because a smaller share of the portfolio sits in Level 3. On the other hand, it also explains why the auditor still flagged fair value estimates of unquoted project companies as a key audit matter. NIS 16.172 million of unquoted project investments and another NIS 1.085 million of project-company loans are still large enough to shape how the market reads the year. And the disclosure itself is explicit that Level 3 fair value depends on funding rounds or sale transactions with adjustments and, in their absence, on discounted cash flow and assumptions such as forecasts, discount rates and comparable-company values.

In other words, the book became more monetizable, but it is not clean yet. If the main question in 2024 was whether meaningful value remained in the portfolio at all, the 2025 question is more precise: which part of that value now sits on market anchors, and which part still sits on assumptions.

Nasus and Exithera, Two Very Different Types of Quality

The strongest case inside the 2025 recovery is Nasus. On August 12, 2025, Nasus completed an IPO on Nasdaq of 1.25 million ordinary shares at $8 per share, implying a post-money valuation of roughly $72 million. Capital Point bought 150 thousand shares in the offering for a total of $1.2 million, and after the IPO held about 6.3% of Nasus. At the end of 2025, the Nasus holding was carried at NIS 10.699 million against cost of NIS 5.395 million.

This is still fair value, but it is already a different kind of fair value. Once Nasus became public, a larger part of the debate moved from an internal model to a market anchor. That is better earnings quality. The problem is that even here not all of the value was freely monetizable at year-end. The company states that about 425 thousand out of roughly 575 thousand Nasus shares held after the IPO were locked and not tradable for a period set in the underwriting agreement. So Nasus improves valuation quality, but it does not automatically turn that value into free cash.

It is also important to separate the two value streams attached to the same company. On one side there is the equity holding. On the other side there is the business development services agreement signed on August 7, 2025 and running through the end of June 2026 for total consideration of $600 thousand. Those are two different stories. The agreement creates contractual income. It is not the same thing as monetizing the investment. So even in the clearest positive case in the portfolio, 2025 shows why it is dangerous to blur together equity value, ongoing fee income and shareholder cash.

Exithera represents the opposite edge. In July 2024, after Exithera's board decided to stop operating activity, Capital Point recognized roughly NIS 28 million of fair value loss. In September 2025 Exithera completed a sale of its main assets to Cardenal, with total contingent consideration of up to $15 million across six milestones, plus royalties of 2% of net sales outside China and 50% of royalties Cardenal may receive from its China exclusivity arrangement.

But then comes the most important line in the whole note. As of the report date, the company assessed the probability that the buyer would meet each milestone as remote, and therefore recognized no income from the contingent consideration. That looks like a small detail, but it matters a lot. If management had wanted to decorate the 2025 recovery, this was the easiest place to do it. It did not. From an earnings-quality perspective, that is actually constructive. It means not every open option was pushed straight into reported profit.

Those two examples explain why 2025 is not a year of "everything is just paper", but also not a year of "everything has already been monetized". Nasus is an upgrade in valuation quality. Exithera is proof that the company did not treat unproven optionality as booked income.

What Is Actually Close to Cash, and What Still Needs Proof

To understand how much of the profit is realizable, it is not enough to look at net income. You also need to look at the balance sheet layers. At the end of 2025 the company had NIS 25.865 million of cash and cash equivalents and another NIS 5.347 million of short-term deposits. On top of that, it held NIS 38.467 million of marketable securities and NIS 12.607 million of investment funds. So there is a meaningful layer of assets here that can be priced at any moment and, in part, can also be monetized relatively quickly. That matters because it means not all of the 2025 recovery sits inside a sealed private-biotech valuation model.

But discipline is still required. The mere existence of liquid assets does not mean 2025 profit has already become cash for shareholders. During the year the company realized NIS 28.223 million from project-company and listed investments and sold NIS 6.552 million of marketable securities. Against that, it reinvested NIS 18.383 million into project-company and listed investments and bought NIS 22.904 million of marketable securities. The result was a negative NIS 4.096 million investing cash flow. In other words, even when realizations happen, the cash does not automatically move upstream. Part of it simply gets recycled into the next round inside the book.

That is the difference between profit that is realizable and profit that has already been realized. In 2025 Capital Point improved the quality of its marked book. More of the portfolio now sits in market-based layers. Part of the Nasus uplift rests on a company that is already traded. Exithera stayed outside profit until stronger evidence emerges. All of that is constructive. But the profit still is not supported by operating cash flow, and the distance between market-based value and shareholder cash remains real.

Conclusion

The short answer is that a meaningful part of 2025 profit is realizable, but not all of it and not immediately. The quality of 2025 is better than the quality of 2024 because the book moved toward market prices, because Nasus became a case of value that can actually be checked, and because the company did not treat Exithera optionality as if it had already turned into income. But the headline still cannot be "profit became cash". As long as the net fair value contribution remains larger than net profit, and as long as operating cash flow stays negative, the right read of this year is an accounting recovery with a genuine upgrade in quality, not a full proof of monetization.

That also explains the 2026 test. For the discount to narrow in a durable way, Capital Point still has to complete the last step, from value that is more market-anchored to value that actually becomes realizations, dividends or cash that remains available to shareholders. Until then, 2025 is a better year, but it is not yet a year that closes the argument.