AFI Properties 2025: Record profit, but 2026 is a lease-up and refinancing year

Understanding the Company

AFI Properties is not just a listed real estate owner sitting on a few trophy assets and waiting for the next fair-value gain. It is a large income-producing real estate platform with a heavy office tilt, a strong Romanian retail base, a rental housing layer that is getting materially bigger, and a development pipeline that still shapes the growth profile. By the end of 2025, income-producing properties, on the company's share, stood at ILS 17.5 billion, gross leasable area reached 1.394 million sqm, and average occupancy was 93%. At the same time, the active and under-construction rental housing stock already reached about 4,400 units. The company is clearly less dependent on one or two flagship assets than it used to be, but it still depends heavily on turning development, lease-up and refinancing into actual cash flow.

What is working right now is clear. Company-share NOI rose to ILS 1.087 billion in 2025, AFFO rose to ILS 485 million, same-asset NOI rose to ILS 946 million, and total occupancy increased from 90% to 93%. Weighted average interest cost fell to 4.4% from 4.8%, 71% of debt is fixed-rate or hedged, and S&P Maalot revised the outlook to positive in February 2026. In plain terms, this was a year in which the operating base improved, not just the balance sheet optics.

But anyone looking only at the bottom line, ILS 933 million of net profit, is missing the core issue. Roughly ILS 716 million came from fair-value gains on investment property and investment property under construction. That does not make the profit low quality by default, but it does mean the market still has to separate accounting value creation from cash generation and funding access. AFI created more value in 2025. The real test in 2026 is whether that value also becomes rent, AFFO and accessible cash.

That matters because the company still sits on a heavy capital structure. Its market cap stood at roughly ILS 9.2 billion in early April 2026, against ILS 13.46 billion of financial liabilities. This is why the stock cannot be read only through NAV or fair value. The correct lens runs through three filters: NOI quality, the lease-up and delivery curve of the 2026 completion slate, and the company's ability to refinance debt without giving back too much of the upside through higher funding cost, tenant incentives or delayed cash conversion.

Quick Economic Map

Events and Triggers

First trigger: equity strengthened quickly. Option exercises injected ILS 572 million into the company and lifted attributable equity to ILS 8.258 billion from ILS 6.854 billion a year earlier. At the same time, debt to CAP fell to 56%. This is not just a balance-sheet headline. It is the reason AFI enters 2026 with better room to maneuver.

Second trigger: debt markets stayed open. During 2025 the company signed financing agreements in Europe totaling roughly EUR 1 billion, including EUR 372 million for refinancing AFI Cotroceni and AFI Loft, and another EUR 148 million for refinancing AFI Brasov and AFI Ploiesti. In January 2026 it expanded series 17 and issued ILS 700 million par, for gross proceeds of about ILS 748 million. In February 2026 S&P Maalot revised the outlook to positive, citing better performance and lower currency risk. That is an important external confirmation, but not a substitute for execution.

Third trigger: portfolio rotation is real. In 2025 the company sold the Concord building in Bnei Brak for total consideration of about ILS 246 million. After the balance-sheet date, in January 2026, it also sold Broadway Palace in Prague for about EUR 6.1 million. At the same time it signed a forward acquisition of Port7 in Prague, a fully leased office and retail asset with about 36,000 sqm, plus land for a 152-unit rental housing project, for about EUR 130 million with expected closing in June 2026. That is exactly the strategy management is presenting: sell assets that have harvested part of the upside, and buy or complete assets that can lift NOI from here.

Fourth trigger: the 2026 completion slate is heavy. The company says it expects to complete 5 office buildings totaling about 115,000 sqm in Jerusalem, Tel Aviv, Brasov and Belgrade during 2026, alongside a 164-unit rental housing project in Bucharest. In the presentation, management states that all office area currently under construction is in advanced negotiations. That is encouraging, but it is still not the same thing as fully signed contracts, actual occupancy and full rent commencement.

Efficiency, Profitability and Competition

The main operating takeaway from 2025 is that the improvement was real, but it still came with a strong accounting tailwind. Net rental and operating income rose to ILS 1.063 billion from ILS 965.9 million. Company-share NOI rose 8%, and AFFO rose 11%. The fourth quarter also showed continuing momentum, with NOI of ILS 284 million versus ILS 271 million in the comparable quarter, and AFFO of ILS 125 million versus ILS 113 million.

What Really Drove the Improvement

The reporting package breaks the improvement into two relatively healthy operating engines. The first is new assets that were completed or added to the portfolio, which contributed about ILS 74 million to net rental income, including about ILS 42 million from Landmark A. The second is better performance in existing assets, which contributed about ILS 44 million. FX moved against the company and reduced European revenues in shekel terms by about ILS 21 million. In other words, the business itself improved, and the euro actually offset part of the picture.

The presentation makes the NOI bridge very clear. Most of the increase came from leasing activity in existing assets, new completions, and CPI linkage, while EUR-ILS was a headwind. That matters because it shows growth was not held together only by one-off concessions or accounting uplift.

Where Profitability Is Still Less Clean

This is the core of the story. Net profit jumped to ILS 937.3 million, but ILS 715.8 million of that came from fair-value gains on investment property and investment property under construction. Anyone trying to judge earnings quality has to look through AFFO and operating cash flow, not only the bottom line. That is not a flaw. It is what a company still in an active development phase looks like. The question is whether 2026 and 2027 will translate that accounting value into rent, cash and funding spread.

Part of the future improvement also requires real landlord spending. The Landmark valuation shows that many tenants in Tower A received fit-out budgets, and that the office agreements in Tower B include fit-out budgets plus a 6-month rent-free period from occupancy. That is not unusual for a prime office project in Tel Aviv, but it does mean the path from commercial progress to cash NOI is not immediate.

Competition, Occupancy and Concentration

Average occupancy rose to 93% from 90% at the end of 2024. Europe reached 93%, and Israel stood at 92%. Romania was already at 94%, Czechia at 96%, and both Serbia and Poland at 92%. Those are good numbers, especially in a market where office demand still rewards quality and location rather than every new project equally.

But concentration has not disappeared. S&P Maalot notes that AFI Cotroceni still accounts for about 14% of the portfolio, even if dependence on that single asset has come down materially versus the past. That is clearly better than before, but it is not irrelevant. At the same time, competition around South Kirya in Tel Aviv has not gone away. The Landmark valuation itself references a dense high-quality office cluster and additional competing projects under development. So even if Tower B is progressing well, the market will not assign a clean premium until it sees signed leases, occupancy and rent collection.

Cash Flow, Debt and Capital Structure

Two Cash Lenses, and They Should Not Be Mixed

On a normalized recurring cash-generation basis, 2025 was good. AFFO rose to ILS 485.2 million from ILS 438.3 million, and cash flow from operations rose to ILS 845.9 million from ILS 793.8 million. That says the operating base is getting stronger.

On an all-in cash flexibility basis, the picture is less clean. Investing cash flow was negative ILS 899.7 million, which means the company was about ILS 53.8 million short before external financing. The year-end liquidity increase came because financing cash flow added ILS 476.8 million, including ILS 572 million from option exercises, ILS 298 million from bond issuance, ILS 110 million from commercial paper, and ILS 3.633 billion of new loans and credit drawdowns, against ILS 2.588 billion of bank and institutional loan repayments, ILS 634 million of bond principal repayments and ILS 164 million of commercial paper repayment. The business is generating more cash, but the development cycle still needs active external funding.

The Debt Structure Says Two Different Things at Once

The good news is that the debt stack itself is less fragile than a quick glance suggests. Total financial debt stands at ILS 13.458 billion, 42.3% of it is European bank debt that is non-recourse, 36% is bonds, 16.2% is Israel asset-backed debt, and 71% of the debt is fixed-rate or hedged. Weighted average interest cost fell to 4.4%. Public covenant headroom still looks comfortable.

The less comfortable news is that short-term debt is the active 2026 issue. Current liabilities rose to ILS 4.217 billion, mainly because ILS 1.142 billion of Landmark loans moved to short term, ILS 634 million of Czech office loans moved to short term, ILS 206 million of the Hod Hasharon office park loan moved to short term, bond current maturities reached ILS 1.037 billion, and commercial paper added another ILS 110 million. This is exactly where a company can look strong in a year-end report while still being highly dependent on execution with banks and debt markets over the next 12 months.

That table matters because it sharpens the key distinction. Public covenants are not the problem. The active problem is execution around asset-level refinancing. That means the risk is primarily operational and market-based, not covenant-based.

Negative Working Capital, Yes, But

The group ended the year with a working capital deficit of ILS 1.922 billion, and the standalone company had a solo deficit of ILS 1.614 billion. The board concluded that this does not amount to a liquidity warning, citing ILS 905 million of undrawn credit lines, the January 2026 bond raise, and the expectation that debt will be refinanced before final maturity. That view is reasonable on the current evidence. But the simple point still stands: the deficit does not disappear because management says it is manageable. It remains manageable only as long as refinancing execution continues.

Outlook

Before getting into the details, these are the four non-obvious findings of the current cycle:

• 2025 was a record year in the accounts, but not yet a full proof year in cash economics. Accounting value creation still ran ahead of accessible cash.
• Landmark B is closer to commercial closure than to cash generation. Leasing progress is real, but it still comes with fit-out budgets, rent-free periods and completion cost.
• Public debt looks manageable, asset-level refinancing is the real test. The company does not look close to a covenant edge, but it is clearly dependent on refinancing several large property loans in 2026.
• AFI Home is beginning to change the mix, but not fast enough to replace offices yet. Rental housing improves diversification, but offices will still determine most of the 2026 read.

2026 Looks Like a Bridge Year, Not a Breakout Year

Management is presenting a path to representative NOI of ILS 1.179 billion and future NOI of ILS 1.443 billion, alongside representative AFFO of ILS 524 million and future AFFO of ILS 699 million. Those are meaningful numbers, and they help explain why the debt market and the rating outlook remain supportive. But they are also not formal 2026 guidance, and the company explicitly says so.

So 2026 looks more like a bridge year than a breakout year: not a year in which the company needs to prove that it owns quality assets, but a year in which it needs to prove that deliveries, lease-up and refinancing are moving quickly enough to justify the next step up in NOI and AFFO. If that happens, the presentation numbers will start to look credible. If not, 2025 may end up looking like a record reporting year rather than a durable cash inflection point.

Landmark Is the Engine, and Also the Friction

This is where the 2026 reading will be set. AFI's share in Landmark A was valued at ILS 2.019 billion at year-end 2025, with 100% office occupancy and NOI of ILS 103.2 million. Landmark B, on the company's share, was valued at ILS 459 million, and the valuation report put the full project at ILS 4.955 billion, of which ILS 4.038 billion came from Tower A and ILS 917 million from Tower B.

But anyone reading only the valuation misses the operational and financing friction. The Landmark valuation states that, as of the valuation date, heads of terms had been signed for all office area in Tower B, but full lease agreements were still nearing signature. Tenants are expected to receive fit-out budgets and a 6-month rent-free period from occupancy. Remaining completion cost for Tower B, excluding the residential component, stood at about ILS 226.5 million. At the same time, in December 2025 the company signed a voucher arrangement for the 116 apartments in phase B, with Law of Sale guarantees of up to ILS 240 million until February 2028, and made an early repayment of about ILS 72 million out of the building B construction facility.

This point is critical. On one side, Landmark has already created a major step-up in value and NOI. On the other side, the move from accounting uplift to effective cash conversion is still unfinished. That is why 2026 will be judged less on whether Landmark is "worth a lot" and more on whether NOI, rent collection, apartment sales and bank financing move along the exact path the company is presenting.

What Has to Happen Over the Next 2 to 4 Quarters

What the Market Is Likely to Measure

The first item the market will measure is not the headline net profit but refinancing execution. The second is the speed at which completed projects move into fully paying occupancy. The third is whether rental housing is genuinely starting to reduce dependence on offices and on Cotroceni, or whether that diversification still remains mostly a 2027 to 2029 promise.

Risks

Financing risk remains the first item on the list. The company has cash, credit lines, market access and stronger rating momentum. Even so, negative working capital and the reclassification of debt into short term mean 2026 does not allow for meaningful slippage in refinancing execution.

FX risk is still real, even if it is being reduced. The company itself notes that euro weakness versus the shekel hurt both reported revenue and translation reserves. The presentation says every one-agora move in EUR-ILS changes equity by about ILS 21 million. S&P Maalot also states that the company began implementing a program in early 2026 to hedge shekel debt against the EUR-ILS rate. That is the right move, but it is also a clear signal that FX had become important enough to require an active fix.

Office lease-up risk has not gone away. The Landmark valuation itself points to a highly competitive South Kirya market. At the same time, the presentation points to expected Q1 2026 occupancy of around 85% at AFI Timisoara and around 84% at ZMAJ A-B, while also noting negotiations on about 16% of ACB Business Avenue 1. Demand is clearly there, but it still requires work, incentives and landlord spending.

Fair value remains sensitive. In the Landmark A valuation, a 0.25% increase in the office cap rate cuts value to ILS 3.879 billion from ILS 4.038 billion. In addition, the valuation report states that the issue with the Israel Land Authority regarding construction deadlines has not yet been resolved, and that potential payments are not included in the valuation. That does not mean an immediate problem is about to hit. It does mean value is not the same thing as accessible cash.

Conclusions

AFI Properties ends 2025 in a stronger position than it was in a year ago. The operating base is broader, equity is larger, rating momentum has improved, and the company has shown it can both recycle assets and refinance debt. But 2026 is not a year in which fair value or a strong slide deck will be enough. It is a year in which the company has to prove that deliveries, lease-up and refinancing actually close the loop.

The current thesis in one line: AFI Properties is now less dependent on a single asset and more dependent on executing a dense wave of deliveries and refinancings well.

What changed versus the older understanding of the company is that there is now a much broader NOI and AFFO base, plus better rating support. What has not been solved is the same old question in a new shape: does development value actually turn into cash, or does it still arrive in the balance sheet faster than it reaches the shareholder layer.

The strongest counter-thesis is that the market may still be too conservative. Under that view, the company has already proven access to financing, reduced funding cost, strengthened equity, and assembled a completion slate that can lift NOI faster than the current market caution implies.

What can change the market's interpretation over the short to medium term is mostly a sequence of events: more successful refinancing, stronger lease-up updates at Landmark B and the Europe completions, and evidence that rental housing is starting to carry a bigger share of the portfolio economics.

Why it matters: AFI Properties is moving from a stage where the story was mostly development and fair-value gains to a stage where the story is about the speed at which those gains become NOI, AFFO and accessible capital.

What has to happen for the thesis to strengthen is straightforward: close new financings before maturities, deliver projects on time, turn Landmark B progress into signed leases and occupancy, and keep NOI rising even without unusually strong fair-value support. What would weaken the read is just as clear: delayed refinancing, slower lease-up, or signs that new rent still depends on economic concessions that are too expensive.