Norstar's Parent Layer: Why ILS 1.9 Billion of G City Value Does Not Become Free Cash

Where This Follow-up Fits

The main article argued that Norstar's discount is not really an argument about the quality of G City assets. It is an argument about the parent layer. This follow-up isolates only that layer. The question here is not whether value exists. It does. The question is why a G City holding carried at ILS 1.907 billion ends 2025 with only ILS 2 million of cash at the parent company.

The misleading number on first read is the ILS 1.9 billion. It sounds like a thick cushion. In practice, that holding sits against about ILS 602 million of net financial obligations at the parent and wholly owned subsidiaries, against a cash-dividend restriction below ILS 1.25 billion of equity, and against a financing structure that already pledges parts of the G City stake to bonds and bank lines.

There is one more forensic point that matters. During 2025 G City repurchased about 13.2 million of its own shares for about ILS 165 million, and after year-end it bought another 8.1 million shares for about ILS 70.1 million. That move lifted Norstar's ownership in G City to about 52.1% and added about ILS 105 million to Norstar's capital reserves. But that was not cash sent up to the parent. It was cash spent at G City. In other words, even part of the value creation in 2025 took a form that increased Norstar's paper ownership, not its free cash.

That is the short answer to the title. Norstar is not short of balance-sheet value. It is short of a clean path to turn that value into free cash.

Parent Cash Flow: Money Came In, But It Did Not Stay Free

The right frame here is all-in cash flexibility, meaning how much cash is actually left at the parent after the real cash uses of the period, not how much value was created underneath it. Under that test the picture is very clear.

Norstar funds itself mainly through dividends from G City, credit lines, loans, and bonds. In 2025 the Norstar group received about ILS 34 million of cash dividends from G City and another roughly ILS 38 million of Orion shares as an in-kind dividend. But once you move to the company's separate cash flow statement, you can see where the money went: operating activity used ILS 4 million, investing activity brought in ILS 79 million net from consolidated companies, financing activity used ILS 83 million, and cash fell from ILS 10 million to ILS 2 million.

What sat inside that ILS 83 million financing outflow? On one side, Norstar raised about ILS 120 million net through Series 14. On the other side, it spent about ILS 194 million on bond repayment and early redemption, and another ILS 9 million on dividends to its own shareholders. So any additional liquidity opened at the parent was first absorbed by liability management.

That is the core of the value-versus-cash gap. Even when the parent receives repayments from subsidiaries and even when it raises new debt, the money does not become free surplus. It is used first to shorten the debt layer, remove an older series, and keep the structure functioning.

The Orion in-kind dividend should also be read correctly. It showed that value can move upward, but it did not solve the cash question. Orion shares are an asset Norstar received, not a cash buffer left in the bank account. That is why the more important line is ILS 2 million of cash, not ILS 38 million of an asset received in kind.

The G City Share Is Both the Engine and the Collateral

To understand why the value does not become free, you have to look at what already sits on the G City share. At the parent layer, that share is the main asset, but it is also the collateral base for several financing layers at once.

At year-end 2025 the Norstar group had revolving credit lines of about ILS 185 million. Of that amount, about ILS 72 million was already drawn on December 31, 2025, and by the report date about ILS 135 million was drawn. Those facilities are secured by part of the G City holding, and their term is three to four years, so most of them expire in 2027 through 2028. To support those lines, Norstar pledged about 49 million G City shares.

At the same time, Norstar's Series 13 bonds are secured by about 33 million G City shares, and Series 14 is secured by about 10 million G City shares. In Series 14 the collateral does not stop at the share itself. It also covers the attached rights, including dividends, sale proceeds, other payments linked to the shares, and rights created in connection with them. In plain terms, even the cash flow coming out of the share is already part of the security package.

The filings describe those pledges separately, so the numbers alone do not prove whether there is full or partial overlap between them. But even without asserting overlap, the message is clear: the G City holding is already being used to support several financing layers at the same time. That is the opposite of an unencumbered asset waiting to become free cash.

This also matters because of the LTV mechanics. In both secured bond series the company has to maintain a debt-to-collateral value ratio of no more than 1 at certain testing points, and if it does not, it has to post additional collateral. So even higher G City value does not automatically become flexibility. It first protects the collateral structure.

The Distribution Restriction and the 2027-2028 Wall

From the bondholders' perspective, Norstar looks acceptable for now. As of December 31, 2025, equity excluding non-controlling interests stood at ILS 1.305 billion versus an ILS 1.0 billion threshold, and net interest-bearing debt to total consolidated assets stood at 65.1% versus an 82.5% ceiling. That is comfortable covenant room.

But for equity holders the important test is not the bond covenant. It is the cash-distribution restriction. Norstar committed not to pay cash dividends if equity is below ILS 1.25 billion, or if a dividend would push it below that level. With equity at ILS 1.305 billion, that leaves only ILS 55 million of room. That is a very narrow cushion for a holdco that still has to refinance debt and rebuild market confidence.

The maturity schedule explains why 2026 is a bridge year rather than a freedom year. Parent-level principal and financial-institution obligations stand at ILS 47 million in 2026, ILS 212 million in 2027, and ILS 234 million in 2028. After that the pace drops to ILS 36 million in each of 2029 through 2031 and ILS 10 million in 2032.

That chart still does not include the fact that most of the revolving credit lines, already drawn by ILS 135 million by the report date, also mature in 2027 through 2028. So 2026 is not a year designed to release cash upward. It is a year designed to buy time, reduce debt, roll facilities, and make sure the cushion above the ILS 1.25 billion equity floor does not get thinner.

That is exactly why the bond market and the equity market read the same company differently. A bondholder sees comfortable covenant compliance. An equity holder sees that the room for cash distribution is barely there.

The Post-Balance-Sheet Buybacks Show Where the Surplus Goes

The Series 13 buyback thread after year-end may be the clearest signal about parent-level priorities. Between January 1 and February 6, 2026, the company repurchased ILS 1,693,396 nominal amount of Series 13 for a total consideration of about ILS 1.75 million. Prices ranged from ILS 1.03 to ILS 1.06 per ILS 1 nominal amount.

That is rational capital allocation. If there is a bit of room, using it to shorten debt makes sense. But it is also a direct answer to the title question. When the company gets a bit of breathing room, it does not translate it into free cash for shareholders. It translates it first into debt reduction.

The same pattern was already visible during 2025. Over the year Norstar repurchased about ILS 1.8 million nominal of Series 13 for roughly ILS 1.8 million, repurchased about ILS 15.2 million nominal of Series 12 for about ILS 17.9 million, and then paid another roughly ILS 60.7 million for the full early redemption of the remaining Series 12. This is not the behavior of a parent sitting on abundant spare cash. It is the behavior of a holdco trying to get ahead of its debt before the 2027-2028 wall arrives.

Conclusion

Norstar's G City stake is worth a great deal on paper, but it does not sit above the capital structure. It sits inside it. Before shareholders can call it free cash, it has to pass four checkpoints: about ILS 602 million of net financial obligations at the parent layer, a collateral system built on G City shares and the rights attached to them, a cash-dividend restriction that leaves only ILS 55 million of headroom, and a refinancing wall that starts in 2027 and gets heavier in 2028.

That also explains why moves that look excellent at the G City level, such as buybacks of G City shares that lift Norstar's ownership to 52.1%, do not create the same effect at the parent. They can increase equity and paper value without opening cash.

The bottom line is simpler than the title. ILS 1.9 billion of G City value does not become free cash because it first protects debt, collateral, and the parent equity cushion. Only if 2026 widens that cushion through more dividends from G City, less debt, or released collateral will it make sense to talk about genuinely free value at the parent.