ISA Holdings follow-up: Bond A, the financial covenants, and the gap between bond-market access and working-capital pressure
Bond A gave ISA Holdings access to the public debt market and created a defined security layer over Bialik and Giborim project surpluses, but it did not erase the working-capital squeeze. At year-end 2025 the bond covenants and collateral ratio looked comfortable, while the group still carried a 12-month working-capital deficit of NIS 261.2 million and NIS 846.4 million of short-term bank credit.
What This Follow-Up Is Isolating
The main article already framed 2026 as a year that still depends on financing and on the actual release of project surpluses. This follow-up does not go back to the broad project map or the overall operating story. It isolates Bond A and asks what really changed once ISA moved from bank loans and partner funding into the public bond market.
The answer is not simply that the company raised NIS 120.5 million and therefore solved its financing problem. That is too shallow. Bond A did open a new funding channel, but it did so through a very specific structural trade-off: bondholders were inserted ahead of equity on a defined layer of project surpluses, and the company's ability to expand the series or keep drawing flexibility from it now depends on collateral tests and on banks actually releasing surplus cash.
That matters because the filings present two stories that seem almost contradictory on first read. On the one hand, the bond itself looked comfortable at year-end 2025: equity stood at NIS 255.5 million, the equity-to-balance-sheet ratio at 27.41%, the collateral ratio at 67%, and the company had completed the full security package. On the other hand, the same evidence set shows a 12-month working-capital deficit of NIS 261.2 million, short-term bank credit of NIS 846.4 million, of which about NIS 410.7 million was land-acquisition financing outside construction accompaniment, NIS 58.3 million of interest paid, and NIS 35 million of dividends paid during 2025. So Bond A solved an immediate access-to-cash problem, not the pressure layer that created the need in the first place.
Bond A Changed The Order Of Claim On Project Surpluses
In January 2026, ISA issued NIS 120.453 million par value of Bond A at a fixed 6.24% rate. Principal is repaid in four uneven installments between December 2027 and December 2029. The prospectus supplement estimated net proceeds of about NIS 118.1 million after fees and expenses, but the more important point here is not the price of money. It is the path through which that money was secured.
| Layer | What was pledged | What it means in practice |
|---|---|---|
| Pledged account | The company's full rights in the designated account and its contents | Cash deposited there sits first against the bond debt |
| Trust account | The company's full rights in the trustee account and its sub-accounts | Issue proceeds were controlled through the trustee before release to the company |
| Project surpluses | 100% of Bialik surpluses and 50% of Giborim surpluses | Bondholders sit ahead of equity on the pledged share of surplus cash |
| Remaining Giborim surplus | The unpledged 50% | Can remain available to the project company or be pledged elsewhere |
The critical detail is that the bond did not receive a first-ranking lien over the projects themselves. It received a right over a defined surplus layer, and only after the lending bank has been paid first under the project financing agreements. The trust deed says this explicitly: the assignment over Bialik and Giborim surpluses is subordinated to every lien held by the project lender, including the lien over the project accompaniment account.
That is a real capital-structure change. Equity did not lose the projects, but it did lose exclusive access to the surplus layer. As long as surpluses have not been released from the project accompaniment account, the trustee cannot enforce them without the bank's prior consent. And even after release, the entire released amount is trapped for the bond at Bialik, while only half is trapped at Giborim.
There is another clause that makes the security package less straightforward than the financing headline suggests. The trust deed allows the project companies to take additional credit from the lending bank for project construction, even if that credit reduces the future surplus pool. In other words, bondholders received priority over surpluses, but not control over how large that surplus pool will remain.
The architecture was also not softened between the updated January 5 draft trust deed and the published January 13 deed. The core structure stayed the same: 100% of Bialik surpluses, 50% of Giborim surpluses, an 80% maximum collateral ratio, and a 70% test for series expansion. That is not drafting noise. It means the market was offered a very specific deal from the start: access to public debt, yes, but through a security wrapper that points first at the surplus layer.
The Covenants Are Comfortable Today, But The Ladder Matters More Than The Headline
At the narrow covenant level, year-end 2025 does not read like a near-breach case. Equity under the trust deed definition stood at NIS 255.5 million, far above the NIS 120 million minimum. The equity-to-balance-sheet ratio stood at 27.41%, far above the 13% minimum. The collateral ratio stood at 67%, below the 80% cap.
But the trust deed is built as a ladder, not as a single line:
| Layer | Threshold | Practical meaning |
|---|---|---|
| Rate step-up | Equity below NIS 150 million or equity-to-balance-sheet below 15% | 0.25% extra for each breach, up to 0.5% total |
| Distribution limit | Equity below NIS 180 million or ratio below 18% | No dividend distribution |
| Hard covenant | Equity below NIS 120 million or ratio below 13%, if the breach also remains in the next quarter's test | Immediate-repayment trigger |
| Collateral ratio | Above 80% | 30 business days to cure with cash or an autonomous bank guarantee |
| Series expansion | Post-expansion collateral ratio above 70%, or total series above NIS 150 million par value | No expansion |
The real analytical point is that Bond A is not currently a near financial-covenant story. It is a collateral-headroom story. As long as the company remains far from NIS 120 million and 13%, bondholders are not living on the edge of a hard default clause. But if the collateral ratio rises, or if the company wants to reopen the series, the market test comes much earlier, already at 70% for expansion and at 80% for a breach that requires cure.
The most important detail is the form of that cure. A collateral-ratio breach is not fixed by promising stronger future surpluses. The deed allows cure only through depositing cash into the trust account or posting a highly rated autonomous bank guarantee. In other words, the safety valve itself depends on liquidity or on more banking capacity. That is exactly where the public bond market reconnects with working-capital pressure.
There is also an important distinction between the equity covenants and the collateral test. A breach of the equity covenant or equity-to-balance-sheet ratio does not become immediate default right away. It must still exist in the next quarter's test. A collateral-ratio breach, by contrast, gets a 30-business-day cure window, but not a wait-for-the-next-report approach. It is a more direct test, and even that test still depends on the ability to bring in cash.
The All-In Cash Flexibility View, What Bond A Solved And What It Did Not
This is where the analysis has to move from the deed back to the annual report. On an all-in cash-flexibility view, Bond A solved three immediate problems, and did not solve the fourth.
First, it gave the company access to the public capital market. Until January 2026, the group was funded mainly through equity, partner loans, customer receipts, and bank credit. The bond added a public funding pipe.
Second, it allowed full repayment of a NIS 35 million bank loan that had been used to finance a dividend. That is an important line item, because it shows that part of the issuance did not go only to growth. It also refinanced a financing burden that had already been created by a cash distribution to shareholders.
Third, it supplied project equity. Under the disclosed use-of-proceeds framework, about NIS 20 million was earmarked for equity in existing projects, about NIS 20 million for ongoing expenses, mainly planning costs, in urban-renewal projects before they enter bank accompaniment, and the remainder mainly for equity in new projects. By the time of publication, the NIS 35 million bank loan had already been repaid, about NIS 10 million had already been used as equity for existing projects, and about NIS 8 million had already been directed to a new project.
| Use of proceeds | Intended purpose | What had already happened by the report date |
|---|---|---|
| Bank loan tied to a dividend financing need | NIS 35 million | Fully repaid |
| Equity for existing projects | About NIS 20 million | About NIS 10 million already injected |
| Planning and other current project expenses before accompaniment | About NIS 20 million | No separate executed amount disclosed |
| New projects | Remainder, mainly equity | About NIS 8 million already directed to a new project |
But there is a fourth problem that the bond did not solve: the pressure layer created by working capital and short-term bank debt across the group.
As of December 31, 2025, the 12-month working-capital deficit stood at NIS 261.2 million. Short-term credit from banks and other lenders stood at NIS 846.4 million, of which about NIS 410.7 million was land-acquisition financing outside project accompaniment. During 2025, the group paid NIS 58.3 million of interest and NIS 35 million of dividends. Those are exactly the numbers that explain why Bond A was needed, and why it cannot be read as a clean fix for the full picture.
There is also a crucial gap between the consolidated layer and the issuer itself. In the separate parent-only numbers, ISA ended 2025 with only NIS 1.7 million of cash against NIS 75.2 million of short-term bank credit. During 2025 it paid NIS 35 million of dividends and NIS 4.8 million of cash interest, while receiving NIS 12.5 million of dividends from equity-accounted holdings. That is the real link to Bond A: the parent layer finally gained access to the public market, but it still depends on dividends and project surpluses moving up from below at a pace that justifies the new debt structure.
The practical takeaway is that Bond A bought time and flexibility, not independence from the banking system. It also did not turn project surpluses into free cash for equity. On the contrary, it formalized a path in which part of those surpluses now meets bondholders first.
Bottom Line
Bond A is far more than a financing headline. It changed ISA's capital structure in three layers at once: it brought the company into the public bond market, it created a security package that places bondholders ahead of equity on 100% of Bialik surpluses and 50% of Giborim surpluses, and it introduced a covenant ladder that starts well before any immediate-default event, through rate step-ups, distribution limits, and the collateral-ratio test.
The positive side is clear. At year-end 2025 the bond metrics themselves looked comfortable, the collateral ratio stood at 67%, and the company proved it could raise NIS 120.453 million at a 6.24% fixed rate as a new bond issuer. That is not trivial.
But it is not the end of the story. The central gap is still open: the issuance addressed an immediate financing need, including repayment of a NIS 35 million bank loan that had financed a dividend, but it did not erase a 12-month working-capital deficit of NIS 261.2 million, it did not remove dependence on NIS 846.4 million of short-term bank credit, and it did not free the surplus layer from the priority of the project lenders.
The current thesis in one line is straightforward: Bond A improved ISA's financing flexibility, but it did so by pledging the surplus layer and adding collateral discipline, not by cleanly resolving the working-capital squeeze.
The strongest counter-thesis is that this reading may be too cautious. One can argue that the covenants are still far away, the collateral ratio is comfortable, and the ability to repay a NIS 35 million bank loan and supply equity to projects is exactly the value of a public bond market. That is a fair argument. The problem is that it remains true only if Bialik and Giborim surpluses are actually released at a pace that keeps collateral headroom comfortable, and if the need for more equity does not pull the company back too quickly into the same short-term funding layer.
What can change the market reading from here is not the mere existence of the series. It is three practical tests: the pace of surplus release at Bialik and Giborim, the ability to keep the collateral ratio below 70% if expansion is needed, and whether the new capital genuinely reduces dependence on short-term credit or merely buys more time.
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.