G City 2025: The Real Estate Has Improved, but Value Still Has to Pass Through the Balance Sheet

Getting To Know The Company

The core story at G City in 2025 is a gap. On the property side, this was a good year: same-property NOI rose 6.5%, total occupancy improved to 96.0%, lease renewals increased 12.4%, and the company posted a 674 million NIS revaluation gain after several difficult years. On the common-shareholder side, the picture is much less clean: consolidated NOI fell to 1.617 billion NIS, AFFO fell to 368 million NIS, and net profit attributable to shareholders was only 70 million NIS.

That is why G City is no longer just a shopping-center landlord. It is also a holding structure built around a few large value pools: Citycon in the Nordics, G Europe in Poland and Central Europe, G Israel, Gazit Horizons in the US, and Gazit Brazil. Anyone looking only at rent growth, or only at NAV, misses the real issue. What matters here is not just asset quality. It is the parent company's ability to turn asset value into accessible cash, upstream dividends, and a cleaner balance sheet.

What is working now is fairly clear. The portfolio remains operationally resilient even in a high-rate environment. What is still not clean is the path between the asset and the shareholder. G City is still selling assets, refinancing debt, increasing control in Citycon, buying back bonds, and buying back shares. Each of those moves can create value. Each of them also consumes cash, raises the importance of funding access, and keeps leverage at the center of the discussion.

The market, at least for now, is focused on exactly that point. Short interest reached 20.59% of float by late March and SIR rose to 15.42, extremely high even by sector standards. That is not a side statistic. It is the market saying that the debate is no longer about whether the assets are any good. It is about whether the value created at the asset level can actually make its way to G City's common shareholders.

The Economic Value Map

This table explains why G City has to be read on two layers. On one layer, this is a large, diversified, urban real-estate platform. On the second layer, much of the value sits below the listed parent, which means not every operational improvement automatically becomes accessible value for common shareholders.

Events And Triggers

The Citycon tender offer: G City increased its Citycon stake aggressively. It bought roughly 7.7% of Citycon in a block transaction in November 2025, then crossed the 50% threshold and was required to launch a mandatory tender for the minority. After completion in March 2026, its stake reached 86.4%. On paper, the economics are compelling: a net outlay of only about 26 million euro after offsetting dividends declared during the tender period, alongside expected annual FFO accretion of about 49 million NIS, an equity uplift of about 640 million NIS, and 3.57 NIS of equity per share. The problem is that the deal also raises consolidated leverage by about 2.7%, so the value creation is real, but it is not costless.

Orion and the disposal program: In 2025 the company completed the Orion transaction, selling three assets for 456 million euro, after already exiting the Czech market with the 201 million euro Flora sale and selling Turkish land for 53 million euro. Strategically this makes sense. It sharpens the portfolio and improves liquidity. But it also explains why consolidated NOI fell even while underlying property performance improved. If you sell assets to clean up the balance sheet, you pay for it in reported scale.

Double buybacks: G City did not stop with Citycon shares. During the reporting period it repurchased about 13.2 million of its own shares for about 165 million NIS, and after year-end another 7.5 million shares for about 63.7 million NIS. At the same time it kept buying back its own bonds. During the year it repurchased roughly 190 million NIS face value of its bonds for about 211 million NIS, and after year-end another 343 million NIS face value for about 353 million NIS. After year-end, the board also expanded the share buyback plan to up to 200 million NIS and approved a new bond buyback program of up to 500 million NIS. That clearly signals management sees deep discounts, but it also tells you that the balance sheet remains the main arena in which management is choosing to create value.

The ratings signal: At the parent level, the sharp late-2025 warning eased somewhat. The Citycon tender put the company's ratings on negative watch, but after the tender results were published in March 2026, S&P Maalot removed the company from watch and affirmed the existing ilA- issuer rating with a stable outlook. Midroog also said it does not expect a near-term rating change. This is an important external signal, but not a full clearance. At the Citycon level, ratings were cut during 2025, so the platform in which G City increased its exposure is still more leveraged and more execution-dependent than before.

The legal overhang: In January 2026, a request for document disclosure was filed as a preliminary step toward a possible derivative claim against the company and Norstar, relating to allegations that Haim Katzman exploited a corporate opportunity through private US residential holdings. The company rejects the claim. This does not change current operating economics, but it adds governance noise at exactly the time the market is being asked to trust aggressive capital-allocation decisions.

Efficiency, Profitability And Competition

The Operating Core Improved

Operationally, 2025 was a good year. Same-property NOI rose 6.5%, total occupancy improved from 95.7% to 96.0%, and tenant sales rose 1.4%. This is not the profile of a broad operating slowdown. The group's urban retail assets are still holding occupancy, tenant sales, and rent growth.

The chart matters because the improvement is not coming from a single pocket. Israel, the US, and Citycon all posted roughly 5% to 6% growth. G Europe was the outlier on the upside at 14.3%, which suggests that even after heavy disposals the remaining Polish and Central European exposure is still strong. Brazil is the weak spot at 1.6%, which fits the broader view that this is increasingly a strategic optionality bucket rather than a core earnings engine.

Occupancy tells a similar story. Israel remains very strong. Citycon is stable. Gazit Horizons improved sharply from 92.7% to 96.4%. The only softness is mild slippage in Brazil and G Europe, and it is nowhere near large enough to overwhelm the broader operating improvement.

Net Profit Improved, But That Is Not The Same As Shareholder Economics

This is where investors need to slow down. A surface read shows 390 million NIS of net profit versus 96 million NIS in 2024, and a 674 million NIS revaluation gain versus only 38 million NIS the year before. That looks like a fast recovery.

But the metrics that better capture repeatable economics are NOI, real-estate FFO, and AFFO. Consolidated NOI fell to 1.617 billion NIS, real-estate FFO rose only to 387 million NIS, and AFFO fell to 368 million NIS. In other words, the real estate itself got better, but the part that reaches common shareholders in a cleaner form did not improve nearly as much.

That chart is probably the best summary of 2025. Real-estate FFO kept rising despite asset sales, which is clearly positive. AFFO, however, dropped sharply. The main reason is not asset weakness. It is the disappearance of special financing gains. In 2024 the company reported 115 million NIS of FFO from special financing activities. In 2025 that line turned to minus 19 million NIS. In plain language, last year's shareholder economics were helped much more by financing activity and liability-management gains, and that benefit faded.

That also explains why real-estate FFO per share was basically flat at 2.00 NIS versus 2.03 NIS in 2024. Investors looking for clean per-share growth have not gotten it yet. What they have gotten is a better property base paired with an equity layer that still absorbs financing noise, FX effects, and structural complexity.

The Growth Quality Is Good, But It Is Not Free

The good news is that G City's operating growth looks reasonably clean. The same-property NOI improvement came alongside higher occupancy, stable tenant sales, and higher real rents on lease renewals. That is a very different profile from a year driven mainly by valuation gains or capital-structure moves.

At Citycon, for example, about 88% of leases require tenants to bear operating costs in addition to rent, and a large portion of leases are indexed. The largest Citycon tenants, S-Group and Kesko, matter operationally, but at the group level they accounted for only about 2.7% and 2.0% of consolidated rental income in 2025. That means the company is not living on a single tenant relationship.

The problem is that good growth quality at the asset level does not solve the quality question at the shareholder level. As long as the route from asset performance to shareholder value runs through disposals, upstream dividends, and balance-sheet engineering, the stock will continue to trade more like a financing structure than a plain real-estate operator.

Cash Flow, Debt And Capital Structure

The Right Cash Lens Here Is All-In Cash Flexibility

If you try to understand G City only through AFFO, you get an incomplete picture. The company itself explicitly says FFO and AFFO do not represent operating cash flow and do not reflect cash available for distribution. The right framing here is all-in cash flexibility, meaning how much real financial room is left after actual cash uses.

On that basis, 2025 looks less clean than the operating story. Cash flow from operations was 638 million NIS. At the same time, the company and its subsidiaries used cash for, among other things, net loan and credit-line repayment of about 1.853 billion NIS, transactions with non-controlling interests of 272 million NIS, treasury-share purchases of 165 million NIS, dividends of 101 million NIS, and hybrid bond interest payments of 161 million NIS. That does not mean the company is facing an immediate funding problem. It does mean the year was supported by disposals, refinancing, and capital-structure management, not just by recurring property cash generation.

The company itself says 2025 was funded mainly through net investment-property realizations of about 730 million NIS after investment outlays. That is the key point. Disposals are not a nice extra. They are part of the financial-flexibility engine.

Leverage Improved, But Not To A Level That Ends The Debate

As of year-end, interest-bearing liabilities and bonds fell to 21.628 billion NIS from 22.994 billion NIS a year earlier. On the extended solo view, the net debt-to-assets ratio was 68.4%, down from 69.7% at the end of September 2025. On a consolidated basis, the ratio was 62.4%. That is improvement, but it is still not a balance sheet that lets the market stop asking questions.

On the positive side, liquidity is meaningful. At year-end, the company and its consolidated subsidiaries had around 3.8 billion NIS of liquidity and unused committed credit lines, of which about 2.5 billion NIS sat at the parent and wholly owned subsidiaries. In addition, the company had around 2.2 billion NIS of assets available for pledging at the extended solo level, plus two European assets worth about 2.7 billion NIS securing roughly 1.1 billion NIS of debt, which management says could generate roughly 0.9 billion NIS of excess cash if sold or refinanced.

On the less comfortable side, the company openly says the parent-only accounts show negative operating cash flow, even though the extended solo view is positive. That is exactly the gap between good assets and common-shareholder access. G City is not in immediate liquidity stress, but it is still dependent on the route from disposals and refinancing to upstream cash working smoothly.

The Risk Is Not Just Debt, It Is Also Encumbrance And Cross-Default Structure

G City does not just have debt. It has debt tied directly to assets and subsidiary equity. At year-end, about 54.7 million Citycon shares were pledged to series 21 bondholders, about 232 million G Europe shares were pledged to series 16 bondholders, and additional Citycon and G Europe shares were pledged to credit lines. Part of the financing stack also includes cross-default mechanisms. The company says credit facilities linked this way total about 2.3 billion NIS, while listed bonds with cross-default exposure total about 9.5 billion NIS.

The point is not that the company is close to a breach. It is not. The point is that value here is not free-floating. A meaningful part of it is pledged, rating-sensitive, and dependent on continued refinancing capacity and orderly upstream cash movement.

The maturity profile itself is manageable. Average bond duration at the extended solo level is 2.8 years and the average coupon is 3.1%. This is not an immediate wall. But it does explain why management keeps spending so much effort on bond buybacks, refinancings, and active balance-sheet management.

Forecasts And Forward View

Four Points The Market Cannot Ignore

The asset base is improving faster than the balance sheet. That is the central line. Same-property NOI, occupancy, lease renewals, and revaluation gains all say one thing. AFFO, disposal dependence, and parent-level leverage say another. Both are true at the same time.

The Citycon deal is strong value creation, but not low-cost risk reduction. On paper, the deal is hard to argue with. Expected annual FFO accretion of 49 million NIS, a 640 million NIS equity uplift, and only 26 million euro of net outlay are attractive numbers. But those numbers rest on Citycon actually executing disposals, refinancings, and upstream dividends while holding operations together after a year of rating pressure.

2026 looks like a balance-sheet bridge year, not an earnings breakout year. Management is guiding to 6% same-property NOI growth and real-estate FFO per share of 1.98 to 2.08 NIS. That is reasonable guidance, but it is not a message of sharp per-share acceleration. It is more a message that the company is trying to hold the per-share line while continuing to reshape the capital structure.

Capital allocation remains the main story. When the company is buying back both shares and bonds while also increasing control in Citycon, it is effectively saying that the discounts in its equity and debt are large enough to justify aggressive action. If that works, it can be highly accretive. If it happens before leverage comes down far enough, the market will keep asking whether management is exploiting an opportunity or just deferring a full balance-sheet clean-up.

What Is Supposed To Support 2026

The good news is that management did deliver against its own 2025 targets. Consolidated NOI came in at 1.617 billion NIS, inside the 1.590 to 1.630 billion NIS target range, and real-estate FFO per share reached 2.00 NIS, inside the 1.95 to 2.01 range. That means the company is not just throwing optimistic numbers into the market.

The less exciting point is that 2026 guidance is still restrained. If same-property NOI grows 6% while FFO per share remains broadly flat, the implication is that the company is still running in place at the common-shareholder level while it keeps selling, refinancing, and reworking the structure. That is why 2026 looks more like a proof year for the balance sheet than a breakout year for the equity story.

What Has To Happen For The Read To Improve

First, Citycon has to execute. In the presentation, G City leans explicitly on three Citycon pillars: at least 1 billion euro of non-core disposals over 24 months, an MoU for 215 million euro of financing on an asset group, and a cash policy that allows distributions from excess liquidity. It also points to an OCR target of 12% to 13% versus 9.2% today. These are not secondary details. They are part of the parent company's thesis on how the tender-offer economics become accessible value.

Second, the parent has to keep deleveraging even after the buybacks. If the market sees the extended solo leverage ratio keep moving down, liquidity remain solid, and buybacks executed out of genuine excess rather than instead of deleveraging, the read can improve materially. If buybacks start to look too aggressive relative to leverage, they can work the other way.

Third, the operating improvement has to travel beyond the asset level. A 6% same-property NOI target is good. The question the market will measure is whether that begins to show up in FFO per share, accessible cash, and real equity breathing room.

Risks

The Main Risk Is Not The Mall, It Is The Parent Layer

This is the central risk. G City is not sitting on a weak property base. It is sitting on a complicated equity layer. The company remains dependent on its ability to refinance, dispose, pledge, and upstream cash from subsidiaries. That is why even after a good operating year, value can still remain trapped or get diluted on the way to common shareholders.

Citycon Is Both The Value Engine And A Risk Concentration

Increasing control in Citycon strengthens the upside case, but it also increases dependence on a platform whose ratings weakened during 2025. At year-end Citycon was still within all covenants, with interest coverage of 2.40 and net debt-to-assets of 42%, but the market will not focus only on covenant compliance. It will focus on execution: disposals, refinancing, and dividend capacity.

FX And Hedging Can Still Distort The Equity Story

Part of the gap between property economics and reported equity runs through FX. In 2025 the group posted a negative translation adjustment of 316 million NIS in other comprehensive income, while the shekel strengthened 1.3% against the euro and 12.5% against the US dollar during the reporting period. The company only gradually rebuilt hedging activity during 2025. That does not break the business, but it can meaningfully distort how the market reads reported results.

Governance Is Still A Yellow Flag

The disclosure request tied to a possible derivative action is not an operating event, but it is not meaningless noise either. When a company asks the market to trust sophisticated balance-sheet and capital-allocation decisions, every governance cloud gets a higher weight. At this stage it does not change the thesis, but it does add avoidable friction.

Development Delivery Still Matters

In Israel, the Rishon LeZion office tower is expected to be completed by the end of 2026, and the company has already signed office sales totaling roughly 356 million NIS. That is positive, but any delay, weaker pricing, or higher completion costs would affect the pace at which value is released. In Brazil, the Internacional parking expansion and the Gazit Malls rights issue are still execution items, not completed proof points.

The Short Sellers' Position

The dissonance between better property operations and the market read is very visible in short data. By the end of March 2026, short interest had reached 16.8 million shares, equal to 20.59% of float, while SIR rose to 15.42. For context, the sector average sits at only 2.45% of float and 3.22 SIR. That is not a small disagreement over valuation. It is an unusually aggressive stance.

This should not be overread. High short interest is not proof of failure, just as buybacks are not proof of cheapness. But in G City's case it does say something important: the market is not arguing with occupancy, tenant sales, or rents. It is arguing with the capital structure, the Citycon dependence, and the path from asset value to common-shareholder value.

Conclusions

G City exits 2025 with stronger real estate, better occupancy, and its first positive revaluation movement in several years. That is the part supporting the thesis. The core blocker is still the same blocker: the move from asset value to accessible common-shareholder value still has to go through disposals, refinancing, upstream dividends, and aggressive capital allocation. In the near term, the market is likely to focus much less on mall quality and much more on whether the balance sheet can carry all of that cleanly.

Current thesis: G City is no longer a story about weak assets. It is a story about good assets trying to force their way through a still-heavy equity and funding structure.

What changed versus the prior read is that the debate has moved from operating quality to value accessibility. The real estate has already given a better proof point. That is still not enough for shareholders as long as the balance sheet keeps setting the pace.

The strongest counter-thesis is straightforward: the company has already done most of the hard work, Citycon was acquired at a deep discount, liquidity is ample, and the group is back to positive value creation. If that is true, the stock is simply too cheap relative to the business. That is a serious argument. The problem is that it still needs a few more quarters of execution before it becomes cash and deleveraging rather than just a persuasive spreadsheet.

What could change the market read over the next few quarters is also clear: clean Citycon execution, continued deleveraging at the extended solo level even after buybacks, and the ability to keep FFO per share at least toward the upper end of guidance without reintroducing balance-sheet stress. If those things happen, skepticism can ease quickly. If one of them stalls, the shorts will remain comfortable.

Why does this matter? Because this is no longer a debate about mall quality. It is a debate about whether a real-estate company can convert real asset value into shareholder value without loading the balance sheet again.

What has to happen over the next 2 to 4 quarters for the thesis to strengthen is fairly obvious: Citycon needs to show disposals, financing, and dividend capacity; G City needs to keep leverage moving lower while protecting liquidity; and the operating improvement needs to start showing up in per-share economics. What would weaken the thesis is just as clear: slower disposals, weaker upstream cash access, or a sense that buybacks are being prioritized ahead of enough debt reduction.