Mivne: How Valuable Is the Data Center Option, and How Much Capital Will It Still Require?
The partnership with DLR and MedOne makes Mivne’s data-center move a real strategic option rather than a generic land story. But at the public-company layer this is still only a 25% option, with at least NIS 148 million of additional capital still required for Phase One alone and no public budget yet for the remaining capacity.
The main article argued that Mivne was entering 2026 with a stronger portfolio but with an open capital-allocation test still ahead. This follow-up isolates the data-center option because this is exactly the layer that can look too expensive if one focuses only on the capital still to be injected, or too cheap if one focuses only on the names of the partners.
What is already working here is clear. This is not a data-center initiative that Mivne is trying to build on its own. The move already includes DLR, MedOne, a dedicated site in Petah Tikva, a connectivity agreement to MedOne’s existing data center, and a planned total capacity of 18 MW. In other words, this is a real option with an operating and infrastructure layer, not another generic land-enhancement story.
The bottleneck sits somewhere else. At the listed-company layer this is only a 25% option for Mivne, and by year-end 2025 the company already presents it with a NIS 37 million carrying value, NIS 148 million of remaining cost, and expected completion only in the fourth quarter of 2027. More than that, the NIS 148 million relates only to Phase One, 10 MW out of the project’s total 18 MW. So the amount already visible in the filings is a floor, not the full bill.
The right way to read the move, therefore, is neither as an income-producing asset already approaching NOI nor as a free option on land. It is a strategic infrastructure option with strong partners, but also with heavy capital needs, a long timetable, and relatively narrow value capture at the public-company layer.
Finding one: this is no longer a theoretical idea. The transaction has closed, the land has been transferred into the partnership, and the site is already in excavation and shoring works.
Finding two: Mivne does not hold the economics of the whole project, only 25% of them. Even if the project succeeds materially, the value reaching Mivne’s listed-company layer will be narrower than the 18 MW headline may suggest.
Finding three: the filings disclose another NIS 148 million of capital for Mivne’s share only for Phase One, while total project capacity is 80% larger than that first phase. The visible amount is therefore an entry ticket, not the total price tag.
Finding four: management’s presentation places the data-center initiative in a separate strategic category, important enough to get its own slide, but still not close enough to operations to appear on the headline activity slide.
Why This Is A Real Option Rather Than Just A Story
The move began back in 2022, when Mivne and DLR formed the Digital-Mivne JV on a 50-50 basis. In 2025 the initiative moved from strategic framing into transaction reality: the company entered into agreements to add MedOne as a partner in a data-center project in Petah Tikva on a site of about 10 dunams, with planned total capacity of 18 MW. Competition approval arrived on June 25, 2025, and the transaction closed on November 5, 2025.
The structure of the deal matters too. The new partnership acquired the land from a Mivne subsidiary for NIS 90 million plus VAT and received around NIS 100 million of initial equity from the partners. In parallel, the new partnership and MedOne signed a connectivity agreement, which means the project is not standing on its own but is linked physically and operationally to an existing data-center asset in Petah Tikva.
That is the difference between a generic real-estate option and an infrastructure option with a real chance of becoming productive. DLR brings global data-center, cloud, and connectivity expertise. MedOne brings active local backup and storage operations. Mivne brings the land and development layer. So this option starts from a stronger position than a project in which the landowner would still need to find an operator, a technology anchor, and connectivity after the fact.
At the same time, Mivne’s own presentation tells the reader how to rank the story. On the opening slide that summarizes activity during and after the reporting period, the move does not appear among the headline 2026 operating items. On another slide it receives a dedicated page of its own, and on the future growth engines slide the company returns to the standard real-estate pipeline without including it in that table. This is not cosmetic. Management itself is placing the data center in a different bucket: strategically interesting, but still not a near-term NOI engine.
Where The Value Actually Lands For Mivne
The easiest number to miss sits in the small line of the projects table. The MED 1 and DLR project is shown there with a 25% company share, NIS 37 million of carrying value in the company’s books, NIS 148 million of estimated remaining construction cost, and estimated NOI of NIS 20 million at full occupancy. In the same row the company also marks start of excavation, soil-contamination treatment, full permit obtained, and expected completion in the fourth quarter of 2027.
In other words, at the listed-company layer this is not an option on the whole project but on one quarter of it. That is critical for understanding value capture. Strong partners improve the quality of the opportunity, but they also divide the economics. Mivne benefits from not having to build an operating platform alone, but in exchange it does not take home the full return if the project works.
That chart shows why the move still deserves serious attention. Once carrying value and remaining cost are combined, the disclosed frame around Mivne’s share reaches about NIS 185 million. Against that, the company presents estimated NOI of NIS 20 million at full occupancy. On paper, that is an attractive-looking disclosed return on the capital already recorded and the capital still to come.
But the same line also tells the reader why the market should not capitalize the upside too early. First, that NOI belongs to full occupancy, not to the current stage. Second, completion itself still sits out at the end of 2027. Third, there is still no customer disclosure, no pre-lease disclosure, and no commercial pace disclosure that would let the reader know how quickly that number can move from a line in a table into actual NOI.
| What Strengthens The Option | What Limits Value Capture |
|---|---|
| DLR and MedOne bring operating depth, connectivity, and credibility that a normal land project would not have | Mivne captures only 25% of the project economics |
| The project has already moved through signing, competition approval, land transfer, and work commencement | Completion is expected only in Q4 2027, so capital arrives long before NOI |
| The table presents NIS 20 million of estimated NOI at full occupancy against NIS 148 million of remaining Phase One cost | There is still no customer, pre-lease, or public budget disclosure for the remaining capacity |
Why NIS 148 Million Is Only The Entry Price
This is the most important point in the capital discussion. The company explicitly says that its share of the expected cost to complete Phase One is about NIS 148 million. At the same time, the presentation makes clear that Phase One is only 10 MW out of the project’s total 18 MW.
The implication is direct. The NIS 148 million disclosed in the filings is not the price tag for the whole option, only for Phase One. The remaining 8 MW, almost 44% of total planned capacity, does not yet receive a detailed public budget in this period. If Phase One works and the partnership wants to build out the full 18 MW, additional capital beyond what is already disclosed will be required. The filings do not say how much. So the most precise conclusion available today is not “Mivne will need NIS 148 million,” but “Mivne will need at least NIS 148 million for Phase One, while the rest is still not publicly priced.”
This is also why the move should not be read as a free option on land. At the project level the site has already been sold into the partnership for NIS 90 million, the partners have already put in about NIS 100 million of initial equity, and the company is now reporting another NIS 148 million for Mivne’s share of Phase One alone. In other words, the option is already consuming real capital before it has reached NOI.
The timing layer matters as well. When the company presents excavation, soil treatment, and expected completion only in the fourth quarter of 2027, it is effectively saying that capital comes in now while income only arrives after another long path of construction, commercialization, and occupancy. Anyone buying the story only through the NIS 20 million of NOI is likely to miss that the road is still long, and that the full cost of all 18 MW is not yet on the table.
What The Market Should Measure From Here
The first question is execution pace through the end of 2027. The filings already provide enough evidence to say the move is real, but not enough evidence to read it like an operating engine. So updates on work progress, Phase One funding, and timetable discipline matter more for now than any general statement about the data-center market.
The second question is commercialization. There is still no disclosure here on customers, pre-leasing, or revenue already locked in. That does not mean demand is absent. It does mean that the next signal the market should look for is not another description of partner quality, but a first commercial proof point that closes part of the gap between potential infrastructure and real NOI.
The third question is disclosure on the remaining capacity. As long as the 8 MW beyond Phase One remain without a public budget, the market cannot know whether Phase Two will be a relatively comfortable expansion or another large capital step-up. That is a material disclosure gap, not a technical footnote.
| Checkpoint For The Next 2 To 4 Quarters | Why It Matters |
|---|---|
| Update on construction pace, funding, and timetable discipline | This will show whether the Q4 2027 target still looks firm or starts to slip |
| First disclosure of commercialization, customers, or pre-leasing | Without it, the NIS 20 million of NOI remains largely theoretical |
| Clarification on Phase Two economics and the remaining 8 MW | This will show whether NIS 148 million is most of the investment or only the opening step |
Bottom Line
Mivne’s data-center option deserves attention because this is one of the rare cases in which a real-estate company brings not only land but also a strong global partner and a relevant local operator to the table. That materially increases the chance that the move turns into productive infrastructure rather than staying at the level of vision.
But at the shareholder layer the reading still has to stay precise. This is a 25% option, not a 100% one. It requires at least NIS 148 million of additional capital just for Phase One. And it still offers no public budget for the remaining 8 MW of the project.
Current thesis: this is a real and higher-quality option than usual, but it is still a capital-heavy option with a long timetable and only partial value capture at Mivne.
What changed versus the broader read: the data center no longer looks like a side note inside Mivne’s land bank. It is already a concrete move. But the numbers also show that the path from here to NOI and accessible shareholder value is longer and more expensive than the partnership headline alone may suggest.
Counter-thesis: one can argue that the caution is overstated because the right partners are already inside the project, the full permit has already been obtained, and on paper the relationship between the presented NOI and the visible Phase One capital already looks attractive enough to justify the spend.
What could change the market interpretation in the short to medium term: orderly Phase One funding, timetable discipline, and first commercial disclosure that would stop the market from reading the NIS 20 million of NOI as nothing more than a distant target.
Why this matters: the data-center project will test whether Mivne can turn an established land bank into a new value infrastructure without letting the required capital consume too large a share of the value that Mivne itself can actually capture.
Disclosure: Deep TASE analyses are general informational, research, and commentary content only. They do not constitute investment advice, investment marketing, a recommendation, or an offer to buy, sell, or hold any security, and are not tailored to any reader's personal circumstances.
The author, site owner, or related parties may hold, buy, sell, or otherwise trade securities or financial instruments related to the companies discussed, before or after publication, without prior notice and without any obligation to update the analysis. Publication of an analysis should not be read as a statement that any position does or does not exist.
The analysis may contain errors, omissions, or information that changes after publication. Readers should review official filings and primary sources before making decisions.