Isras in 2025: the balance sheet is strong, but the real test has moved to lease-up and capital allocation

Getting To Know The Company

At first glance, Isras looks like one of the easier Israeli income-property stories to read. The company has a large portfolio, moderate leverage, relatively cheap debt, equity attributable to shareholders of ILS 4.72 billion, and annual net profit of ILS 386.6 million. Anyone who stops there will see a stable landlord with dividends and buybacks. That is not the full story.

What works today is easy to identify. In 2025, NOI rose to ILS 475.3 million, FFO under the ISA approach rose to ILS 251.2 million, portfolio occupancy reached 87%, and excluding Brosh it was already 90%. At the same time, Isras holds about 637 thousand square meters of leasable assets, alongside roughly 119 thousand square meters under construction or planning. With equity equal to 44.7% of the balance sheet, this is not a near-term funding-stress story.

But the active bottleneck is no longer the balance sheet, and it is not a question of whether the mature assets can generate NOI. The bottleneck has moved to the company’s ability to turn value that already sits on paper, mainly in a handful of office assets, into stable and accessible NOI without eroding cash flexibility. Brosh in Har Hotzvim, Park Gissin in Petah Tikva, and Ramat Hahayal in Tel Aviv are now the places that will decide whether 2026 reads as another calm year for an income-property platform, or as an operating proof year that requires execution rather than just valuations.

That is also why net profit on its own can mislead. In 2025 the company recorded ILS 224.9 million of fair-value gains on investment property. That does not mean the value is artificial, but it does mean the right economic reading has to separate accounting value from actual NOI, and existing NOI from NOI that still depends on lease-up, tenant improvements, and execution speed.

There is also an important market screen to state early. Against a strong balance sheet and low funding cost, short interest as a percentage of float stands at 6.96%, versus a sector average of 0.55%, with SIR at 14.37 versus 1.562 for the sector. That does not read like fear of immediate balance-sheet pressure. It reads like skepticism about how quickly future NOI will mature, and about how much of the embedded value will actually reach shareholders in the near term.

The Economic Map Right Now

This chart matters because it explains the gap between what looks diversified and what actually drives the current debate. The portfolio spans several uses, but the active friction sits mostly in office assets and in properties that have not yet reached full operating maturity. So the Isras story in 2025 is not a generic real-estate discussion. It is a focused discussion about a few assets that can materially change how the whole company is read.

Events And Triggers

The first trigger: after the balance-sheet date, the board approved both an additional dividend for 2025, ILS 15 per share, about ILS 74.3 million, and an intention to distribute a total of ILS 30 per share in 2026, about ILS 148.6 million, subject to future board decisions. At the same time, it approved a new buyback plan of up to ILS 50 million through the end of 2026. The message is clear. Management is telling the market it sees value in the share and feels comfortable with the capital structure. But that same message also raises the hurdle. If the company is returning capital this actively, the market will want proof that NOI and liquidity support it after real cash uses.

The second trigger: Park Gissin only began recognizing partial income from the government lease in the third quarter of 2025. That matters because 2025 already includes some contribution, but still not a full annual run rate. So part of the improvement in 2026 can come not from new projects, but from the maturation of leases that already exist.

The third trigger: Brosh ended 2025 at 34% occupancy, and after the balance-sheet date the company signed an additional lease for about 6,000 square meters, taking occupancy to above 50%. That is real progress. Still, there is a middle stage between a signed lease and clean NOI in the financial statements. That is exactly where a weaker or slower office market can delay proof.

The fourth trigger: short interest is aggressive relative to the sector. When short float is close to 7% and days to cover are very high, every coming quarter is likely to be judged through the question of whether Brosh, Gissin, and Ramat Hahayal are truly progressing, not through the question of whether the company can meet covenants. Leverage is not the short-term story. Lease-up is.

This does not tell us who is right. It does show where the market has placed the argument. If the debate were mainly about balance-sheet safety, these short-interest levels would look harder to explain against 36.2% net debt to CAP and a wide pool of unencumbered assets. The more plausible explanation is skepticism about the quality and timing of future office NOI.

Efficiency, Profitability, And Competition

The core insight here is that Isras’s mature portfolio still operates efficiently, but the edges of the portfolio are what will determine the pace of improvement from here. Rental and operating income rose to ILS 560.0 million in 2025 from ILS 547.8 million in 2024, and NOI rose to ILS 475.3 million from ILS 468.9 million. The increase came mainly from CPI linkage in leases, higher income at Har Hotzvim, and first-time income recognition at Park Gissin. In other words, what already worked kept working.

But once the portfolio is broken down asset by asset, the picture becomes much less uniform. GTI Jerusalem is almost fully occupied, and the mature Har Hotzvim buildings remain highly occupied, while Brosh, Park Gissin, and Ramat Hahayal still show a wide gap between current NOI and full-occupancy NOI. That is the heart of the story, because it means most of the upside is not spread across the whole portfolio. It is concentrated in a few office properties.

The meaning of this chart is sharp. GTI and the mature Har Hotzvim assets are already close to full extraction. Brosh, Park Gissin, and Ramat Hahayal together produced only about ILS 62.3 million of NOI in 2025, versus about ILS 133 million at full occupancy. That is more than ILS 70 million of annual upside, at least on paper. On one hand, this explains why the share can still look interesting despite an already strong portfolio. On the other hand, it also explains why the market is not satisfied with balance-sheet strength or dividends alone, because most of that gap still has to pass through real leasing execution.

What Really Supports Profitability

Isras’s main strength in 2025 is that the spread between property economics and debt cost still works. The CPI-linked bonds carry a weighted average interest cost of 1.71%, the equity base is high, and covenants are comfortably far from their limits. That is why the rise in finance expense to ILS 167.5 million did not erase the operating improvement.

But it is important not to confuse stable operating profitability with a frictionless earnings base. Ramat Hahayal is a good example. Occupancy fell over the years from roughly 95% before a major tenant left, to 68% in 2024 and 65% in 2025. The valuation report explicitly says the leasing process for vacant space is taking too long because of market conditions. In addition, one floor leased to a flexible-office operator was still in a stage where expenses exceeded income, so it was not yet contributing profit to the owner at the valuation date. This is not an asset-quality problem. It is a speed-of-maturation problem.

Brosh and Gissin are also very different from GTI. At Brosh, the current cap-rate profile still reflects an asset in the middle of lease-up. At Park Gissin, the government award is clearly supportive, but only part of the income entered 2025. That means consolidated NOI already shows improvement, but the quality of that improvement still depends on execution rather than just contract signing.

What Competition Says About 2026

Management describes an office market where negotiations are taking longer and tenants are asking for broader flexibility, both in space and in lease duration. That wording deserves attention because it appears at exactly the point where the company is trying to fill large office spaces. Ramat Hahayal already shows what such a market looks like in practice, and Brosh still needs to prove that the recent occupancy improvement can translate into a smoother NOI run rate.

This chart adds another important layer. NOI kept rising, FFO recovered nicely in 2025, but management AFFO declined. That was not because the core business suddenly weakened. It was because this metric adds back CPI indexation on debt principal. So it is a friendlier metric than FFO, and it is certainly not the same thing as cash left after all uses. Anyone trying to judge earnings quality needs to move from that line to actual cash flow.

Cash Flow, Debt, And Capital Structure

Cash Flow

The right framework for Isras in 2025 is all-in cash flexibility, not a normalized cash-generation view. The reason is simple. The debate around Isras is not whether the property portfolio can generate NOI, but how much room remains after dividends, buybacks, development, and funding needs. Once that is the question, the analysis has to include the real uses of cash.

There is an important gap here between how the number is presented and how it should be read. Management AFFO stands at ILS 336.9 million, but that is not “cash left over.” In practice, it is FFO plus ILS 85.7 million of CPI indexation on debt principal. So it is a softer metric, not a harder one.

Cash flow from operating activities stood at ILS 340.2 million in 2025. Against that, investing cash flow was negative ILS 248.5 million and financing cash flow was negative ILS 101.3 million. The result is that cash at year-end declined only modestly, to ILS 661.7 million from ILS 671.3 million at the start of the year. That is a reasonable picture, but not one of huge excess cash after every need. It is a picture of a company preserving liquidity while paying ILS 223.8 million of dividends, spending ILS 33.1 million on buybacks, and continuing to develop projects.

What is more interesting is that this quiet cash outcome came in a year when the company also expanded bond series 19 in June 2025 by a gross ILS 565.9 million. So the right conclusion is not that Isras lacks flexibility. It is that the analysis has to distinguish carefully between flexibility that comes from a strong capital structure and good bond-market access, and surplus cash left after all real uses. Those are not the same thing.

Debt And Covenants

This is actually where Isras looks strong. Net debt to net CAP is 36.2%, equity attributable to shareholders equals 44.7% of total assets, and the company has about ILS 6.7 billion of unencumbered assets. Rating agencies reaffirmed stable outlooks in 2025 and early 2026, and the covenant package is comfortably far from stress. The minimum equity threshold is ILS 1.5 billion, the ceiling on net debt to CAP is 70%, the ceiling on net debt to NOI is 16, and the minimum equity-to-assets ratio is 20%. Isras is nowhere near any of those lines.

The maturity ladder does not point to an immediate wall either. Year two is relatively heavy at ILS 843 million, but the company approaches it with a solid liquidity position, a large unencumbered asset base, and demonstrated access to debt markets. Anyone looking for the breaking point in leverage is probably looking in the wrong place.

The real financing risk would emerge only if the company keeps returning capital aggressively while future NOI is delayed, or if the office market forces heavier tenant-improvement spending for longer. Right now that is still a discipline risk, not a survival risk.

Outlook

Before going into detail, four forward-looking findings matter more than any single number:

• First: 2026 looks like an operating proof year, not a harvest year. Most of what can change the market’s reading is already on the map, but not yet fully mature.
• Second: the stable portfolio is already close to being fully utilized, so most of the improvement now has to come from a small set of properties rather than from the whole portfolio.
• Third: dividends and buybacks strengthen management’s confidence signal, but they also reduce the patience the market will give to delays in lease-up.
• Fourth: the development pipeline is meaningful, but it is mostly a 2027-2028 story, so it does not solve the 2026 question. It only creates a horizon beyond it.

What Can Improve The Read In The Next Reports

Park Gissin is probably the asset with the fastest potential for a near-term rerating of the story. Its 2025 NOI was ILS 24.7 million, while the presentation shows an annual estimate of about ILS 47 million including signed contracts and ILS 57 million at full occupancy. That means a meaningful part of the upside is already anchored in contracts, not only in theoretical valuation assumptions. If conversion to income is smooth, this is the kind of asset that can change how 2026 is read.

Brosh is a different case. The upside is very large, ILS 12.7 million of NOI in 2025 versus ILS 36 million at full occupancy, but the gap between contract signing and actual results is larger as well. After the balance-sheet date occupancy rose to above 50%, which clearly matters. Still, what the market will want to see is not just contract volume but how quickly those leases turn into clean NOI, and what price was required to get there in rent, tenant improvements, and time.

Ramat Hahayal is a third kind of test. The asset is in a strong area, but there is clear evidence that the market is not forgiving quickly. Occupancy fell to 65%, negotiations for vacant space are taking longer, and the floor operated through the flexible-office model had not yet generated profit for the landlord at the valuation date. If Ramat Hahayal stabilizes, that would support the case that the weakness is not structural. If it does not, this may remain a more persistent drag.

What Will Not Solve 2026

The development pipeline is impressive, but it is not an immediate answer. The company has projects under construction totaling about 119.1 thousand square meters, with estimated remaining cost to completion of ILS 967 million and full-occupancy annual NOI of ILS 95 million. That sounds attractive on paper, but most of the projects are expected to be completed in 2027 and 2028.

So anyone building the 2026 thesis mainly on future development is getting ahead of the timeline. Development is an important value layer, but it belongs to the years after the operating proof year, not to the year that still has to prove itself.

What Kind Of Year 2026 Looks Like

2026 looks like an operating proof year backed by a strong balance sheet. The balance-sheet support is already in place. The operating proof is not. What would strengthen the thesis is a clean translation of signed leases at Brosh and Park Gissin into running NOI, a halt to occupancy erosion at Ramat Hahayal, and a comfortable cash position even after active capital returns. What would weaken the thesis is a situation where NOI improvement is delayed while the company keeps returning capital as though the maturation phase were already behind it.

Risks

The first risk is lease-up risk, not asset risk. Brosh, Park Gissin, and Ramat Hahayal show that the gap between a quality property and full NOI can remain open for longer than the market likes. The company itself describes a market with longer negotiations and greater tenant flexibility demands, and the valuation reports explicitly describe a slower office market. If those conditions persist, accounting value can remain stable while operating maturity drags.

The second risk is capital-allocation risk. In 2025 the company declared and paid ILS 223.8 million of dividends, spent ILS 33.1 million on buybacks, and after the balance-sheet date added another ILS 74.3 million dividend plus a new ILS 50 million buyback plan. As long as NOI progresses, that can read as confidence. If lease-up slows, the same policy can look much less comfortable.

The third risk is reliance on revaluations as an important part of the reading. In 2025 the company recorded ILS 224.9 million of fair-value gains. That adds real accounting value, but it is still sensitive to cap rates and lease-up pace. In Har Hotzvim alone, the valuation shows that a 0.5% increase in the cap rate reduces value to ILS 1.181 billion, versus roughly ILS 1.254 billion at the current valuation level. So even without a crisis, a modest shift in assumptions can trim value.

The fourth risk is concentration hidden inside what looks like stability. In the mature Har Hotzvim buildings, about 26% of area in buildings B, C, and D is leased to government housing administration tenants through 2032-2033, but most of that area includes an early-exit clause in 2028. That is not a reason for alarm, but it is a reminder that even the more stable assets have a renewal and continuity question embedded inside them.

The fifth risk is sector-level macro friction exactly where Isras still has to prove itself. Management describes a higher-rate environment, a cooler tech market than the peak years, and more flexible tenant behavior in the office market. If that backdrop remains, even quality assets in demand areas may take longer to fill.

Conclusions

Isras ends 2025 as an income-property platform with a solid core, comfortable leverage, and a funding cost that still works in its favor. That is the part supporting the thesis. The main constraint is that the next layer of value the market wants to see will no longer come from the mature portfolio, but from the company’s ability to lift NOI in three central office assets while keeping capital discipline. That is also what should determine the market reading in the near to medium term.

Current thesis: Isras looks strong on the balance sheet, but the 2026 test has shifted from proving stability to proving lease-up and capital discipline.

What changed: in 2025 the company no longer rests only on a strong existing portfolio. Gissin started contributing, Brosh moved above 50% occupancy after the balance-sheet date, and management chose to keep capital returns active at the same time. So the debate has moved from whether value exists to when it becomes accessible.

Counter thesis: the market may simply be too cautious. With ILS 6.7 billion of unencumbered assets, 36.2% leverage, stable ratings, and a mature portfolio that keeps generating NOI, Isras can afford time. If Brosh and Gissin mature even at a moderate pace, today’s skepticism may look excessive in hindsight.

What could change the market reading: a report showing cleaner conversion of Brosh and Park Gissin leases into NOI, or alternatively a quarter in which Ramat Hahayal keeps lagging while cash shrinks alongside continued distributions.

Why this matters: because for an income-property company the key question is not only whether value exists in the portfolio, but whether that value turns into recurring income and capital flexibility that actually remains with shareholders after development, funding, and distributions.

What has to happen over the next 2-4 quarters: Brosh needs to show a sustained jump in NOI, Park Gissin needs to prove a fuller annual run rate, Ramat Hahayal needs to stop losing occupancy, and the company needs to show that dividends and buybacks are not coming at the expense of flexibility.