Kardan: How Stable Is Dalian's Occupancy When More Than 80% of Leases Roll by 2028?
Dalian ended 2025 almost full, but the occupancy headline sits on softer terms in hard-to-lease space, more than 80% of leases rolling by 2028, meaningful kids-category exposure, and a very competitive Dalian market. The real question is not how many stores are open, but how durable the mall's NOI really is once the renewal wave starts.
What Exactly Needs To Be Tested Here
The main article argued that Dalian is no longer Kardan's weak operating link. The asset works, but the value is still stuck beneath the debt layer. This follow-up isolates the question that remained open inside that same story: how stable year-end 2025 occupancy really is, and whether the NOI that comes with it is good enough to hold when more than 80% of leases roll by 2028.
That matters because the headline looks impressive. Year-end 2025 closed with 96.5% occupancy, a 95.1% open-store ratio, 220 tenants, and more than 7.4 million visitors. But that same year also included leasing of large problematic spaces on softer terms, a decline in average income per square meter versus 2024, and a local market where asking rents in shopping malls barely moved higher. In other words, occupancy improved, but occupancy quality is the real question.
96.5% Occupancy, Without Clean Pricing Power
Dalian's strongest 2025 datapoint is not just year-end occupancy. It is also the fact that the mall returned to a record footfall level of more than 7.4 million visits. Visitor traffic increased by about 5% versus 2024 and tenant sales increased by 4%. NOI also rose to EUR 6.836 million. The problem is that this improvement did not translate into stronger economics per square meter.
Average income per square meter per month fell to EUR 15.2 from EUR 15.8 in 2024. Average rent and management fees per square meter per month fell to EUR 15.1 from EUR 15.5. The ratio of rent and management fees to average tenant sales per square meter fell to 22% from 23%. So even after better occupancy and stronger traffic, the average tenant is not paying more for the space, and the mall is not showing a real return of pricing power.
The income structure itself is not fully rigid either. More than 90% of tenants pay the higher of base rent and turnover rent, while only 3% and 6% of leases pay base-only or turnover-only rent. That structure gives the landlord some protection, but it also means NOI quality depends on the tenants' sales quality, not only on whether the unit is occupied.
Savills also builds the turnover-rent component of its projection on actual 2025 sales. That is an important detail. The peak-footfall year is already embedded in the valuation model. If 2025 was a relatively strong year for traffic and sales, the room for another positive surprise is smaller than the occupancy headline alone suggests.
The Economic Giveback Sits Precisely In The Hardest Space
The annual report says this almost directly: the relatively low level in new leases signed during 2025 mainly reflected agreements on especially large commercial spaces that were problematic because of location, store shape, or business category. The company's own examples are not marginal.
| Space or category | Share of mall area | Why it matters |
|---|---|---|
| Gym | About 4% | A category that traditionally pays low rent |
| Duty-free store | About 2% | Located in problematic space |
| Children's swimming facility | About 2% | A use that traditionally pays relatively low rent |
| Supermarket expansion | About 1.4% | Another large area signed on relatively low economics |
| 7 stores together | About 40% of the area leased in 2025 | A large part of annual leasing came through a small number of large hard spaces |
The company adds that some of those areas had been vacant for a long time, so leases were signed at low rent terms. Savills describes the same move in slightly different language: in 2025 the leasing strategy became more active, including more favorable commercial terms to attract new tenants, especially in large units that are harder to lease or harder to renew on better terms.
That is the core distinction between occupancy and NOI quality. Early in the life of the mall, concessions came through fit-out support, participation in adaptations, and temporary rent holidays. The company says no such incentives were given in 2023 to 2025. But that does not mean lease economics tightened. The form of the concession simply changed. In 2025 it appears as lower rent levels and softer terms precisely in the spaces that were hardest to fill.
That is also why NOI barely moved. It rose only from EUR 6.814 million to EUR 6.836 million while average occupancy improved to 95.2% and the open-store ratio climbed to 95.1%. In other words, a meaningful part of the additional occupancy did not come with the same economic quality one would expect from a strong renewal or from leasing highly desirable space.
The Lease-Roll Wall Arrives Sooner Than The Headline Suggests
Savills includes the most important datapoint in this entire continuation: 42.6% of leases expire within one year, another 30.5% within two years, and another 13.0% within three years. Cumulatively, 86.1% of leases roll by 2028, and more than 95% by the end of year five.
That means year-end 2025 occupancy is not a concrete wall. It is the starting point for a very large renewal cycle. And here the valuation assumptions are not aggressive. Savills assumes only 95% committed occupancy in 2026 to 2029, and 96% only from 2030 onward. In other words, even the appraiser is not underwriting another occupancy jump. The model is mostly underwriting preservation of the current level.
At the same time, once a lease expires, the model assumes the unit is leased again at prevailing market rent. That is a reasonable assumption, but it highlights the real risk: when such a large share of leases reopens within three years, NOI quality depends less on the 96.5% year-end 2025 figure and more on the ability to renew this large wave without another round of commercial concessions.
The rent-growth assumptions do not describe a hot market either. Savills assumes average rental growth of 3% per year in 2026 to 2029, then 2% in 2030 to 2033, and 1% in 2034 to 2036. That is a path built on gradual recovery, not a jump. If competition intensifies or renewals continue to require favorable terms, even this moderate path can start to look optimistic.
The Local Market Gives Dalian Time, But Not Much Room For Error
There is a real positive side here as well. Galleria Dalian is the only shopping mall in the Donggang district, it has no competitor inside the district itself, transport infrastructure around it improved, and a nearby high-end residential project can still add customers. That helps explain why mall traffic managed to recover and even move above the pre-pandemic level.
But the broader Dalian picture is less comfortable. Savills describes a highly competitive retail market in which a few key players hold most of the market share. Asking rent on first floors in prime retail districts slipped slightly in 2025 to RMB 610.2 per square meter per month from RMB 610.8 a year earlier, and citywide retail vacancy improved only marginally to 15.66% from 15.7%. This is not a market in which an owner facing a huge renewal wall can simply assume sharp improvement in terms.
New supply also matters from here. Over the next three years, two projects, CMB Garden City and China Resources MixC, are expected to add 287,000 square meters of high-quality retail space to the Dalian market. Even if future supply is lower than during the 2015 to 2020 construction wave, it is still enough to keep the market competitive.
On top of that sits the demographic issue. Savills says the kids category still accounts for the largest share of leased area, and that more than 20% of tenants were in the kids category as of the valuation date. In the same breath it warns that the declining birth rate could hurt those categories. The company itself says the mall's original positioning was tilted toward the kids segment and that this used to be one of its clearest advantages versus competitors. Now that same advantage can gradually become part of the problem.
This again is a question of NOI quality. If the mall's biggest category is under demographic pressure, and if some of the large spaces signed in 2025 include a children's swimming facility, the renewal question is not just how many stores remain open. It is what price they can pay, what sales base will support them, and how much bargaining power the landlord will still have in the next negotiation cycle.
Conclusion
The right way to read Dalian at the end of 2025 is not "the mall is almost full, so risk has fallen." The right read is more nuanced: the mall is almost full, but part of that fill came through large hard spaces signed on softer economics, and more than 80% of leases will reopen within three years.
That means 2025 occupancy buys Kardan time, not immunity. If the 2026 to 2028 renewal cycle passes without further rent erosion, and if the kids category does not become a heavier drag than expected, Dalian's NOI can prove that it is genuinely stable. If not, the 96.5% headline may later look like a peak-occupancy print that came right before another round of concessions.
Three watchpoints from here are straightforward:
- In 2026 the market needs to see not only high occupancy, but also improvement or at least stability in average rent per square meter.
- The large renewals on problematic spaces need to clear without another drop in price levels or a rise in downtime.
- The kids category needs to stop being a major dependency in a mall that is trying to prove NOI quality, not only activity volume.
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.