G.P. Global Power: How IPM's Value Gets Trapped on the Way to the Parent
The asset layer and the intermediate holdings are generating visible accounting value, but the listed parent ended 2025 with only NIS 2.651 million of cash against debt, pledges, and assigned upstream distributions. This follow-up isolates the real bottleneck: not asset quality, but lender priority on the way up.
The main article argued that the operating asset is getting stronger, especially after IPM's partial refinancing and the broader shift into bilateral sales. This follow-up isolates one narrower question: if IPM's value is indeed improving, why is so little of that value visible in the listed parent's cash position.
The short answer: value has to clear four separate gates before it reaches G.P. Global Power's public shareholders, and each gate has a different claim on the cash. First come project debt and covenants at IPM. Then comes free cash flow availability at IPM Holdings. Then comes debt service and approval mechanics at Israel Power. Only after that does cash reach the listed parent, which is already carrying its own debt stack, pledges, and assigned upstream distributions.
That gap is already visible in the 2025 numbers. The company ended the year with net profit of NIS 32.627 million and NIS 42.513 million of income from equity-accounted investees, but operating cash flow was still negative NIS 2.498 million and year-end cash was only NIS 2.651 million. At the same time, the balance sheet carried NIS 150.780 million of investments in associates and another NIS 17.964 million of loans to an associate. There is value in the structure, but very little immediately accessible cash.
Where the value sits, and where the debt sits
At first glance, the 2025 balance sheet looks far better than the 2024 one. Equity rose to NIS 74.844 million from NIS 40.686 million a year earlier, and the accumulated deficit narrowed sharply to NIS 4.765 million. But the layer-by-layer view is less comfortable. On the asset side, most of the balance sheet sits in holdings and in a loan to an associate, not in liquid cash. On the liability side, there are NIS 61.579 million of related-party loans, NIS 36.100 million of loans from others, and a NIS 2.5 million on-call bank loan.
The key point is not just the size of the debt, but the nature of the relief in equity. Of the increase in equity, NIS 1.531 million was credited to capital reserve from transactions with shareholders after a shareholder-loan extension letter. That is accounting relief created by moving maturities, not new cash entering the parent. So even the cleaner year-end equity number does not change the fact that the listed parent still had a very thin liquidity cushion.
Gate one: IPM is creating value, but it still serves its own debt first
At the power-plant level, the business direction did improve. Starting in July 2025, IPM increased the share of capacity allocated to bilateral sales to 36%. That allocation rose to 41% on January 1, 2026, and is set to rise to 51% on April 1, 2026. In parallel, IPM signed a partial refinancing package of roughly NIS 840 million in May 2025, completed in early June, with final maturity on June 30, 2040.
But that does not make plant-level cash freely distributable. IPM's financing documents still require a minimum debt service coverage ratio of 1.05 for each calculation period and a minimum loan life coverage ratio of 1.08. The covenants, security package, and acceleration triggers remained unchanged after the refinancing, and the new package also added a bilateral reserve mechanism that may be required from 2030 under certain conditions. In addition, shareholder loans at the project level are subordinated and pledged behind senior debt. So even if the asset's economics improve, the first cash out of the project still belongs to senior lenders and to the project's protection layers.
Gate two: IPM Holdings only distributes what is left after free cash flow tests
IPM Holdings has a seemingly generous dividend policy: 100% of distributable profits, but only subject to applicable law and to the existence of available cash flow. In a leveraged holding vehicle, that condition matters more than the headline payout ratio.
The bottleneck is even clearer once you look at the shareholder loan that G.P. Global Power itself extended to IPM Holdings. That loan bears annual interest of 7% and is repaid gradually out of the full free cash flow of IPM Holdings on a full cash-sweep basis from the commercial operation date of the plant. In other words, even cash that moves up from the project level into the intermediate holding layer does not automatically become a free dividend. It first enters a loan-repayment channel. That is why the NIS 17.964 million loan asset at year-end 2025 is a claim on future cash flow, not cash waiting at the parent.
Gate three: Israel Power is another debt layer, not a clean pass-through
The next stop in the chain is Israel Power. G.P. Global Power owns 45.98% of it, and Israel Power owns about 16% of IPM. Again, this is not a clean upstream pipe. Israel Power has material decisions that require a 75% special majority, and its dividend policy says that all cash received from IPM will be distributed to shareholders only subject to certain debt repayments at Israel Power itself.
That is not a theoretical restriction. At year-end 2025, Israel Power reported NIS 87.131 million of non-current liabilities, including about NIS 43.2 million of CPI-linked debt bearing annual interest of 6.85%. At the same time, from G.P. Global Power's perspective, Israel Power drove a sharp rise in the carrying value of the investment, from NIS 44.783 million to NIS 79.149 million, and in total comprehensive income as seen by the parent, from a loss of NIS 2.681 million in 2024 to a gain of NIS 74.631 million in 2025. The accounting value moved sharply higher, but upstream cash still has to clear another debt layer first.
Gate four: the parent has already pledged the way up
If the earlier layers could still be dismissed as timing, the parent-level debt stack shows that this is a true priority problem. The core bottlenecks are summarized below:
| Layer | Year-end 2025 balance | Key term | Why it blocks access to value |
|---|---|---|---|
| Shareholder loans to the company | NIS 51.093 million | 4% annual CPI-linked interest; August 2025 extension letter to December 2, 2027; March 2026 extension letter to June 30, 2028 | Buys time, but does not create cash |
| Israel Power loans to the parent | NIS 10.486 million | NIS 6.8 million loan from March 2024 and NIS 3 million loan from September 2025, each with a 24-month extension option | The intermediate holdco is lending to the parent instead of simply upstreaming clean cash |
| Noy loans used to finance the Israel Power stake | NIS 34.690 million | Final maturity is April 27, 2027 | From final maturity onward, distributions and other payments from Israel Power are assigned to Noy |
| Keystone loan used to finance the IPM Holdings investment | NIS 1.410 million in the report, NIS 5.25 million under the March 2026 settlement | The loan is repaid out of distributions the company is entitled to receive from group companies | Cash that should move up from IPM Holdings is also a repayment source for an older financing layer |
The security package completes the picture. In Noy's favor, the company granted an irrevocable instruction letter and a first-ranking pledge over its Israel Power shares, including distributions and other payments due from Israel Power. In Keystone's favor, the company granted a first-ranking specific pledge over its IPM Holdings shares and over its rights to receive cash and distributions from IPM Holdings. So even when value does move up through the chain, it arrives pre-allocated to someone else.
What really changed on March 22, 2026
The most misleading figure in the annual report is a small one. As of December 31, 2025, the balance sheet showed only NIS 1.410 million of current debt under the loan that financed the IPM Holdings investment. On a quick read, that makes the layer look almost resolved.
That is wrong. After the balance-sheet date, on March 22, 2026, the company and Keystone Power signed an agreement and mutual irrevocable waiver under which the outstanding Keystone balance as of December 31, 2025 was NIS 5.25 million, not NIS 1.4 million as shown in the report. In other words, the year-end debt picture understated this layer by roughly NIS 3.84 million relative to the final settlement between the parties.
This is not just a technical correction. The Keystone financing was built to be repaid out of distributions the parent is entitled to receive from group companies, and the dispute history shows how quickly a value layer that looks liquid on paper can become yet another argument over payment priority. That makes the parent's real margin of safety smaller than the December 2025 balance sheet alone suggests.
Why this matters now
The constructive part of the thesis has not disappeared. IPM is clearly a stronger operating asset than it was, and the value that flows through Israel Power is already visible in the parent's 2025 numbers. But the real bottleneck is no longer asset quality. It is the speed at which cash can move up the structure without being intercepted by lenders, pledges, and repayment conditions on the way.
That is why the next three checkpoints are not just more profit at the plant. Checkpoint one: IPM must keep generating surplus cash under the existing covenants even after the broader bilateral-sales mix. Checkpoint two: Israel Power has to become an upstream distribution vehicle again, not a lender to the parent. Checkpoint three: the listed parent must begin to reduce the actual debt and pledge stack, not merely push maturities outward through extension letters.
Until that happens, IPM's value is real, but for G.P. Global Power's public shareholders it remains conditional value. It still has to pass through coverage ratios, free-cash-flow filters, intermediate repayments, pledges, and only then reach the listed parent.
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.