Aya New York 2025: Manhattan Residential Is Recovering, But the Value Still Gets Stuck Above the Assets

Company Overview

Aya New York in 2025 is not a mature listed company reporting another normal year. It is a new bond wrapper that was only formed on August 22, 2025 with no assets, liabilities, or operations, and only after the balance sheet date, once the bond issuance was completed in February 2026, did it receive the direct holdings in the Manhattan assets. Anyone reading this like an ordinary annual report is likely to make two mistakes: first, to assume the 2025 consolidated numbers tell the full operating history of the company, and second, to assume that every improvement at the asset level automatically reaches the bondholders or the public-company layer. That is not what is happening here.

What is actually working now? The residential book is already showing results that are hard to dismiss. Segment revenue rose to $13.9 million from $11.3 million in 2024, fair-value gains rose to $29.5 million, and profit attributable to the company’s owners in the segment jumped to $14.4 million. Renoir now stands at $78.8 million with annual NOI of $2.848 million and 93% occupancy, Riverside reached $60.1 million after a rehab program of more than $9 million, and W42 finished the year at $33.4 million with 95% occupancy and approved development rights of about 90 thousand square feet. This is not just a marketing deck. There is real asset-level improvement in Manhattan residential.

So where is the bottleneck? It sits above the assets. Part of the economics flows through promote mechanisms, management fees and acquisition fees, a public bond that is effectively secured only on two residential assets, and Lady D, a closed hotel acquired in August 2025 that already carries a $94.9 million value but still no NOI. In other words, the question is not simply what the assets are worth. The question is how much of that value can actually travel upward without getting absorbed by partners, lenders, operators, and fee layers.

There is also a practical actionability constraint. For now this is a bond-only TASE story. In the prepared market context there is no local listed equity, no relevant short-interest data, and the market read will run first through Series A. That means the real question is not whether a clean NAV story can be told, but whether the bond and public-company layer are actually supported by assets that can generate cash rather than revaluations alone.

The right economic map looks like this:

This chart makes the key risk obvious. Lady D is already the largest asset in the portfolio, but it is also the least proven operationally. Renoir and Riverside, by contrast, are already carrying NOI, and they are also the two assets supporting the public collateral package.

Events and Triggers

The year the public layer was born

2025 is almost a technical transition year at the wrapper level, but not at the asset level. The company itself was created in August 2025 with no real activity, and the pro forma statements are designed to show what the numbers would have looked like as if the Manhattan holdings had already sat inside it. That matters, because the investor is not reading a long operating history of a listed company. The investor is reading a package built for a bond issue.

The defining event arrived in February 2026. The company completed an initial issuance of NIS 289.038 million par value of Series A bonds, then expanded the series by another NIS 3 million par value in a private placement to the controlling shareholder, taking the total to NIS 292.038 million par value. The bonds started trading on February 17, 2026, carry a 7.7% annual coupon, are not linked, and repay only 2% of principal at the end of 2027, another 2% at the end of 2028, and the remaining 96% at the end of 2029. That gives near-term relief, but it also creates a very concentrated repayment test at the end of the term.

The bond bought time, but told a narrower story

The more interesting point is not the issuance itself, but how it was designed. In the February 2026 prospectus amendment, the company explicitly updated the use of proceeds so the expected repayment would cover the loans on Riverside and Renoir only. The same amendment also updated the collateral provisions so the liens for Series A relate to those two assets only. That creates a meaningful split: the public bond story is not backed by the full Manhattan portfolio, but by two residential assets that already show NOI.

This is not a technical footnote. Bondholders get a cleaner story than someone looking at the whole portfolio as one block. The pledged portion already includes income-producing residential assets with relatively high occupancy and real improvement. The non-pledged portion holds most of the optionality, but also most of the uncertainty: W42 with its development rights, Little Charlie as a small operating hotel, and Lady D as a closed hotel whose operating thesis still has to be proved.

After the issuance, the company did not just refinance debt, it also bought more rights

Shortly after the issuance, the third-party loans on Riverside and Renoir were repaid in an aggregate amount of about $81 million. At the same time, the company also completed the exercise of an option to acquire a partner’s holdings in Lady D and Renoir. It paid $3 million in cash, while the remaining $18 million of consideration was left in the form of a seller loan. This improves the future value-capture potential, but it is not a one-way positive. On one hand, less partner ownership means more economics if the assets perform. On the other hand, the value did not suddenly become free. Part of the partnership layer was simply replaced by a debt layer.

Altair is an option, not the base case

In early 2026 the company also disclosed that it had signed a financing letter of intent for the Altair acquisition, and the closing deadline was extended to April 30, 2026. That could become another growth engine, but at this stage it is still an option, not the core read. Anyone building the 2026 story on another acquisition before the current platform proves it can generate cash is skipping a step.

Efficiency, Profitability and Competition

Residential really did improve

Residential is the real operating core of 2025. Segment revenue rose to $13.918 million from $11.321 million in 2024. Fair-value gains rose to $29.518 million from $20.972 million, and profit attributable to the company’s owners in the segment jumped to $14.357 million from $5.134 million. This is not one accounting event. It is a portfolio of assets moving in the right direction.

Renoir is the cleanest example. The asset was acquired in November 2024 in an off-market deal. The company converted 34 studio units into one-bedroom units and 6 one-bedroom units into two-bedroom units, cut operating costs, and ended 2025 at a value of $78.8 million versus $53.9 million a year earlier. Revenue reached $6.857 million, NOI reached $2.848 million, and average rent on leases signed in 2025 rose to $112 per square foot versus a full-year average of $74. That means the value creation did not stay at the appraisal level. It also moved into new lease economics.

Riverside is less clean, but arguably more interesting. At acquisition, more than 25% of the free-market units were vacant or unlettable because of serious façade and leakage issues. The company invested more than $9 million in rehabilitation, and within six months about 50% of the free-market units were upgraded and relet at premium rents. By the end of 2025, revenue had already risen to $4.924 million and NOI had turned positive at $1.008 million, after negative NOI of $888 thousand in 2023 and near breakeven in 2024. But this is exactly the kind of asset where the headline can mislead. Representative NOI in the valuation already stands at $3.542 million. In other words, part of the improvement is still ahead rather than behind.

W42 is smaller on current NOI, but not necessarily on option value. Revenue in 2025 rose to $2.137 million, NOI to $897 thousand, and value to $33.4 million. All residential units at the property are free-market, average occupancy rose to 94%, and the asset has approved development rights of about 90 thousand square feet. The implication is straightforward: current operating income is still fairly small, but the property carries an option layer that can create additional value if and when it is turned into an actual planning or business move.

This chart tells the real 2026 story. Renoir is already relatively close to its representative NOI. Riverside and W42 are not. That means anyone buying the residential story is not just buying what already happened. They are also buying the assumption that the transition from rehab to stable NOI is still underway.

Hospitality looks strong in the statements, but not in cash

The hospitality segment finished 2025 with revenue of $2.921 million and fair-value gains of $29.068 million. That is exactly why the numbers need to be unpacked. Little Charlie is genuinely operating: a small 30-room hotel, 100% occupied according to the property table, with annual NOI of $1.296 million and revenue of $2.921 million. But it is also operated by a company controlled by the controlling shareholder, so even the cleaner hotel asset does not sit entirely outside the related-party layer.

Lady D, by contrast, is almost the whole hotel story in valuation terms, yet almost none of that story has been proven operationally. The hotel has been inactive since 2020, was acquired through a non-performing debt purchase and foreclosure process, and entered year-end 2025 with a $94.9 million value. The question is not whether there is upside. There is. The question is who is paying for it today. Right now, the answer is the appraisal, not the operating statement.

This chart makes clear how much of Aya New York’s hotel story still sits in valuation rather than cash flow. Residential already has real revenue and improving NOI. In hospitality, most of the 2025 weight comes from revaluation.

The competitive edge here is not brand, but execution

The report does not present a classic moat of anchor tenants, very long leases, or a fully independent operating platform with a large workforce. In fact, many of the residential leases run for 12 months, and the company does not have any single tenant accounting for 10% or more of total revenue. So the edge visible in 2025 is different: the ability to buy assets in troubled situations, execute fast rehabilitation, improve unit mix, fix building envelopes, and push rents higher.

That is a real edge, but it is also one that has to be re-earned all the time. It does not lock in customers, it does not guarantee cash generation, and it depends heavily on the company continuing to operate better than the prior partner, the prior borrower, or the prior operator. That is exactly why 2026 is a proof year.

Cash Flow, Debt and Capital Structure

The all-in cash picture was very tight even before the bond

On an all-in cash basis, 2025 was not a comfortable year. Pro forma cash flow from operations was $2.998 million. Against that, investments in equity-accounted investees and a loan to a related company consumed $18.801 million. The year was funded through $15.843 million of financing inflow, and the company ended with just $159 thousand of cash.

That matters more than almost any profit line. Even if asset-level improvements are creating value, they consumed nearly all available financing capacity in 2025. Anyone reading the year mainly through pro forma comprehensive income of $26.704 million is missing the fact that the company did not exit 2025 with excess cash. It exited with almost no cash.

The gap between the two working-capital views reinforces the point. At the standalone company level, before the pro forma look-through, the statements showed a current-asset surplus of $77 thousand. But in the pro forma consolidated view there was already a $3.928 million current deficit. That is not a technical contradiction. It is a reminder that the issuance shell looked cleaner than the full economics of the underlying assets.

The public bond solved Renoir and Riverside, not the whole platform

After the February 2026 issuance, a new debt layer was created with reasonable near-term duration, but it did not clean up the whole map. The debt on Renoir and Riverside was repaid, which is clearly positive. But the bond itself is structured as an almost full bullet to the end of 2029, and it sits in shekels above dollar assets. So the public debt did not eliminate the test. It simply moved it from an immediate financing question on two assets to a broader question of debt service and value capture at the company level.

At the property level, leverage still looks fairly reasonable at Renoir and Riverside, both at 42% LTV. W42 already sits at 63%, and Lady D at 40%. Those numbers do not scream immediate distress. But they also do not tell the whole truth, because the public bond sits above the assets, while promote mechanisms and management fees sit between property-level NOI and the cash that can actually move upward.

This chart makes clear that W42 is actually the more leveraged residential asset, while Lady D looks less leveraged on the surface. But in Lady D’s case that can be misleading, because the real test there is not static year-end LTV. It is the ability to turn a closed property into one that produces enough NOI to support expensive financing.

Lady D is the real financing test

Lady D carries a senior loan with an effective interest rate of 9.25%, a structure of SOFR plus 5.25% with a 4% floor, and final maturity on August 28, 2027. There is an option to extend to August 2028, but only if there is no default, if construction and renovations are completed, and if the property meets a debt-service coverage ratio of at least 1.05, a debt yield of at least 15%, and additional conditions including LTV of 60% or less and reserve deposits.

This is where value and debt can separate. The Lady D appraisal can look very attractive if the hotel opens and performs. But the lender is not underwriting a story. The lender is underwriting opening, NOI, covenant compliance, and completed works.

On top of that, the asset stack sits under a management layer that is not free. From the bond issuance onward, the company is supposed to pay the controlling shareholder annual management fees of $1 million plus expense reimbursement of $200 thousand per year. There are also asset-management and acquisition-fee layers inside the asset entities, and Little Charlie is operated by a company controlled by the controlling shareholder. So even when NOI rises, it does not move upward friction-free.

Outlook

Four non-obvious findings should drive the forward read:

• 2025 did not prove Lady D. It only brought it into value.
• The public bond bought the company time mainly on Renoir and Riverside, not on the whole portfolio.
• Residential already shows real asset improvement, but at Riverside and W42 a material part of the economics still sits in representative NOI rather than actual NOI.
• The issuer layer does not sit directly on all created value because of promote, debt, seller-loan, and fee structures.

From here, 2026 looks like a bridge and proof year. It is not a breakout year, because too much of the story has not yet cleared proof points. But it is not a reset year either, because unlike an empty shell, this company already has working assets, improving residential results, and a public bond that opened capital-market access.

The first trigger is convergence between actual NOI and representative NOI at the residential assets. At Renoir the gap is already relatively narrow. At Riverside it is still wide. Until management shows that the rehab is translating into stable NOI and not just a higher appraisal, the read will remain incomplete. W42 adds another layer: the market has to see whether the asset stabilizes enough for the development rights to be treated as real bonus value rather than another capital-consuming option.

The second trigger is Lady D. The attached valuation assumes a hotel that has been closed since the COVID period, is going through a $10.5 million renovation, and reopens around June 2026 under an independent flag. That point needs discipline: this is a valuation assumption, not an operating achievement already delivered. Over the next 2 to 4 quarters, the test is not only whether the hotel reopens, but whether the reopening starts building NOI that can support the expensive financing package.

The third trigger is the quality of value capture. After the issuance, the company did not just repay loans. It also bought partner interests in a deal that combined cash and seller financing. If that leads to cleaner value flowing to the issuer layer, the read can improve materially. If it simply replaces a partner layer with a new liability layer, the improvement at the assets will keep looking bigger than the improvement at the company.

The fourth trigger is expansion discipline. Altair may be a good move, but it should come only after the existing platform proves it can generate NOI, meet financing obligations, and carry the management and related-party overhead already sitting above it. Otherwise the company risks looking like it is adding another option before the current base is stabilized.

From a near-term market perspective, this story is likely to be read much more through milestones than through a single earnings line. Credible progress at Lady D, better actual NOI at Riverside, or evidence that the seller loan and fee layers are not absorbing most of the created value could strengthen the read quickly. Delays at the hotel or slowing residential improvement would do the opposite.

Risks

Lady D may remain value on paper. Until the hotel actually reopens and produces NOI, the $94.9 million value remains execution-dependent. The financing there is expensive, the extension conditions are explicit, and the gap between accounting value and debt-service capacity can quickly turn from a theoretical strength into a practical yellow flag.

The partner and related-party layer can absorb part of the value. The three residential assets all carry promote mechanisms. Some of the agreements include 2% to 3% acquisition fees, asset-management fees, and the public company itself is supposed to pay the controlling shareholder annual management fees of $1 million plus $200 thousand of expense reimbursement. Little Charlie is managed by a company controlled by the controlling shareholder. At the same time, the controlling shareholder has no non-compete restriction and can hold similar assets outside the company. That does not make the business uninvestable. It does mean the value has to be read much more carefully.

The company received the assets as is, without indemnity. The documents explicitly say the transferred rights were transferred to the company as is and without any indemnification from the transferors. That means any historical operating, financing, legal, or structural problem that was not neutralized at the asset level now sits with the public wrapper.

The public-company governance layer is still new. As of the report date, no internal auditor had yet been appointed. That is not necessarily a fatal issue, but it is a reminder that this is not a long-established public system. It is a newly public structure still building its control layer.

The final bond test was deferred, not removed. A structure in which 96% of principal is repaid only in December 2029 is an advantage in the short term and a constraint in the long term. If by then the company still has not learned how to turn revaluations and property NOI into value capture at the issuer level, it will arrive at the large repayment point with a story that is still unfinished.

Conclusions

Aya New York brings the Israeli market a Manhattan real-estate story that is both real and structurally messy. The residential assets are already showing sharp recovery, the public issuance solved a specific pressure point on two assets, and the hotel portfolio includes one small operating property and one large asset with meaningful upside. That is the part supporting the thesis.

The main blocker is still the distance between asset value and accessible value. Lady D is still closed, part of the economics runs through promote and fee layers, the bond is supported by only part of the portfolio, and the public wrapper itself is still too new and too thin to benefit automatically from every improvement at the assets. In the short to medium term, the market is unlikely to focus mainly on NAV. It will focus on proof.

Current thesis: Aya New York has already shown that Manhattan residential can improve, but it has not yet shown that asset-level improvement can reach the company layer at a sufficient pace and with enough cleanliness.

What changed: In 2025 and early 2026 the story moved from a private portfolio being prepared for issuance to a public bond issuer with collateral, refinancing, and partner buyouts that make the value-capture question much sharper.

Counter-thesis: It is possible that the market is still being too harsh. Renoir and Riverside have already improved, W42 carries planning optionality, Lady D was acquired out of foreclosure at an opportunistic basis, and if the hotel reopens successfully the market may discover that the value gap is much larger than currently assumed.

What could change the market read soon: Proof that Renoir and Riverside are moving closer to representative NOI, demonstrated progress on the Lady D renovation and reopening, and signs that the public debt and seller-loan layers are not consuming most of the value created at the properties.

Why this matters: This is a classic example of the difference between a good asset and a good structure. If the company can move value upward from the assets, it will look like a public Manhattan real-estate platform with improved properties in the core market. If not, 2025 will be remembered mainly as a year of revaluations and repackaging.

If over the next 2 to 4 quarters the company shows more stable NOI at Renoir and Riverside, a credible reopening at Lady D, and cleaner value capture above the partner and fee layers, the read can improve quickly. If those points are delayed, the company will still have interesting assets but a public wrapper that has not yet proved it knows how to turn them into accessible value.