Three Layers Determine Height — Slant Planes, Absolute Limits and Landscape: Which One Binds?

The rules that constrain building height are not a single block. In hotel site development at least three layers overlap. Each has a different legal basis, works differently, and is relaxed through a different counterpart.

In site-development practice, the bottleneck tends to be layer 2. Layer 1 has design escape routes through setbacks and the relaxation provisions for slant-plane limits, and layer 3 allows a landing point to be found through consultation. The absolute height limit in layer 2, by contrast, does not move unless the city-planning cap itself changes. Moreover, the increments differ from one municipality to the next. Setagaya Ward (Tokyo) operates seven caps — 15m, 16m, 19m, 25m, 28m, 31m and 45m — for residential zones other than low-rise exclusive residential zones, quasi-industrial zones and some neighbourhood commercial zones (effective April 1, 2019). Kyoto City, since its New Landscape Policy of 2007, has designated height control districts over about 95.3% of its urbanization promotion area, a designation rate that stands out nationally.

It is no accident that 31m occupies a special position among the seven caps. It corresponds to the “100 shaku” limit dating back to the old Urban Building Act, and is the threshold most widely used in Japanese city planning. Dividing by a hotel floor height of 3.1m, 31m comes to exactly 10 floors. The calculations that follow rest on this conversion.

What Kyoto City Changed on April 25, 2023 — Three Areas Before and After

On April 25, 2023, Kyoto City put into effect changes to zoning, height control districts, fire-prevention districts and related designations (some took effect on September 25 of the same year). The areas covered include Rakunan Shinto, the Sotokanjo-sen corridor on the eastern side of the city, the industrial zones in the western urban area, the areas around Mukomachi and Yodo stations, and the area south of Kyoto Station. The so-called “Tanoji” district, the heart of the city centre that includes Shijo-Karasuma, was not changed. Below we set out the three areas that bear directly on hotel development, based on the before/after table in the revision pamphlet published by the city.

The requirements of the “space-creation height control district” are specific. On the premise that the site abuts the Sotokanjo-sen or Shin-Jujo-dori for at least 2m, (a) at least half of the ground-floor area must be given over, continuously, to activity-generating uses (anything other than houses, apartment buildings, dormitories, boarding houses, garages, car parks and warehouses); (b) a sidewalk-like open space at least 3m wide must be provided along the Sotokanjo-sen or Shin-Jujo-dori; and (c) a 2m setback must be taken from other road boundaries and adjacent-lot boundaries. Only buildings that meet all three are exempt from the maximum height. If any requirement is missing, the 31m ceiling remains.

Height and FAR Multiply — Under a 20m Cap, Usable FAR Tops Out at About 480%

What is easily overlooked in discussions of height limits is that the absolute height limit and FAR are not independent constraints. With building coverage b and floor count F, the maximum total floor area that can actually be built on a site is set by b × F. In other words, there is such a figure as “the maximum FAR that can be used under a given height cap.”

Applying this table to the changes south of Kyoto Station reveals the thinking behind the deregulation. Under the previous 20m (6 floors), only 480% could be used even with the full 80% building coverage. Raising FAR to 600% while leaving the cap at 20m would have left it unusable. The new 600% can, on paper, be reached in 8 floors (about 25m) if 80% coverage is used in full, but in practice the building footprint often falls to around 60% of the site once car-park access and open space are secured, in which case exactly 10 floors — 31m — are required. The required number of parking spaces itself differs by a factor of 2.5 from city to city. The relationship that emerges here is that a height relaxation is meaningless unless it comes as a set with a FAR relaxation.

The maximum FAR of 1,000% set for Rakunan Shinto would require about 17 floors (roughly 52m) at 60% building coverage, or about 25 floors (roughly 78m) at 40% coverage. Under a 31m height control district it is physically impossible to use. That is precisely why this framework is combined with conditions such as a 10,000㎡ minimum site area and a 10m exterior-wall setback, opening up the vertical dimension in exchange for open space on large sites. Talking about development capacity from the FAR figure alone means counting floor area that cannot actually be built.

Three Years On, How Has Kyoto Prefecture’s Room Supply Moved? — Reading It With Lead Times Factored In

Three years have passed since the April 2023 enforcement. Within the scope tracked by MetroEngines Research & Consulting, new openings in Kyoto Prefecture (confirmed by OTA listing) line up by year as follows.

From a peak of 172 properties and 6,321 rooms in 2021, the room count has contracted: 3,298 in 2022, 3,558 in 2023, 2,517 in 2024 and 1,806 in 2025. The 618 rooms for 2026 are a part-year tally, and because OTA listings only appear a few months before opening, the figure will rise as further listings come through.

Reading this trend as “deregulated, yet nothing is being built” mistakes the time axis. As confirmed in our earlier article Hotel Build Times, lodging facilities take an average of 2.80 years from construction start to completion (1.96 years for single-use, 4.00 years for mixed-use). On top of that comes the preceding period for site acquisition, design and building-confirmation application. Projects that secured sites on the basis of the April 2023 city-planning change will complete in 2026 or later even for single-use buildings, and around 2028 for mixed-use. The properties that opened in 2024–2025 were all planned before the deregulation. The lead time for deregulation to convert into room supply is still only at its halfway point.

Indeed, extracting Kyoto Prefecture projects from construction plans (50 rooms or more) based on MLIT’s “Building Dynamics Statistics Survey” yields five: Higashikujo, Minami Ward, Kyoto City (270 rooms, 9 floors, completion scheduled April 2026); Kamigyo Ward (135 rooms, 3 floors, completion scheduled September 2027); Higashiyama Ward (55 rooms, 7 floors, completion scheduled April 2026; and 53 rooms, 4 floors, completion scheduled December 2028); and Kumiyama Town, Kuze District (224 rooms, 9 floors). The count is limited, but because building-confirmation applications are normally filed one to two years before opening, this is the lower bound of the confirmed pipeline at this point. The number of projects and rooms is expected to increase as further applications are added; it should not be read as an upper bound.

The actual price level also needs to be pinned down. We line up monthly estimated settled ADR for Minami Ward, which is covered by the deregulation, the city-centre wards outside it (Shimogyo and Nakagyo), and Higashiyama Ward, a tourist-destination location.

On an 11-month average, Minami Ward is ¥14,500, Shimogyo Ward ¥16,000, Nakagyo Ward ¥17,400 and Higashiyama Ward ¥24,100. Minami Ward’s level is 90.6% of Shimogyo’s, 83.4% of Nakagyo’s and 60.0% of Higashiyama’s. As the area targeted by the deregulation, Minami Ward sits only 10–20% below the city centre, and given its proximity to Kyoto Station the market has sufficient depth. Note that Yamashina Ward has only one property within the scope of our estimated settled ADR, so we do not cite it as an area level. Fushimi Ward is also thin at five properties and is treated the same way.

On the land-price side, too, Minami Ward’s standing is changing. In the Official Land Price announcement (Chika Koji) released by MLIT in March 2026, the average price of commercial land in Kyoto Prefecture exceeded ¥1 million per square metre for the first time in 32 years, and the highest growth among commercial land in the prefecture was recorded in Minami Ward, Kyoto City. On an all-use average by ward, Minami Ward is ¥571,000/㎡ (+11.05% YoY), Yamashina Ward ¥166,000/㎡ (+3.84%), Shimogyo Ward ¥1,882,000/㎡ (+9.27%) and Nakagyo Ward ¥1,729,000/㎡ (+9.31%).

Measured Floor Counts of 70 National Plans — 82% of Rooms Sit in Buildings Above 10 Floors

To generalize beyond Kyoto, we measure the floor counts of construction plans nationwide (50 rooms or more, N=70 plans, 15,517 rooms in total). The floor count is known for 66 plans and 14,965 rooms.

The median is 14 floors, the maximum is 63 floors, and 20 plans have 20 floors or more. Low-rise buildings of 1–3 floors number only six. The 14-floor median converts to about 43m at a 3.1m floor height. In other words, the standard form of the hotels being planned across Japan does not fit within a 31m height control district.

Viewed on a room basis, the implication becomes clearer still. For each height ceiling, we show cumulatively the share of total rooms accounted for by plans that fit within that floor count.

Only 9 plans with 1,234 rooms, 8.2% of the total, fit within the equivalent of 20m (6 floors), and even within the equivalent of 31m (10 floors) it is just 18 plans with 2,697 rooms, or 18.0%. Only when the ceiling is widened to the equivalent of 45m (14 floors) does coverage reach 41 plans with 8,409 rooms, or 56.2%. Put the other way, if a 31m absolute height limit applied uniformly nationwide, 82.0% of currently planned rooms would have to be redesigned.

Differences in use mix also matter. The 30 plans whose principal use is “hotel” alone have a median of 12 floors, whereas the 36 mixed-use plans combining offices, residences or retail have a median of 20 floors. In mixed-use developments the hotel sits on the upper floors, so the building as a whole grows taller. This structure is why mixed-use schemes are hard to assemble in areas with strict height limits.

Rooms per floor (rooms ÷ floors) has an overall median of 15.0; by floor band it is 22.4 for 1–9 floors, 18.2 for 10–14 floors, 15.4 for 15–19 floors and 6.7 for 20 floors or more. The sharp drop at 20 floors and above arises because the hotel occupies only part of the building; limited to the 30 hotel-only plans, the median is 15.6 rooms per floor. For the 29 plans that are hotel-only and have a known total floor area, the median total floor area per room is 26.9㎡ (first quartile 17.4㎡, third quartile 63.2㎡).

Stacking 100 Rooms at 31m, 45m and No Height Cap — Required Site Area and Break-Even ADR

Using the measured values so far, we work backward to the site area needed to make a 100-room hotel work. The assumptions: total floor area per room at the measured median of national plans, 26.9㎡, giving a required total floor area of 2,690㎡ (813.7 tsubo); a 3.1m floor height; and a simple stacking calculation that does not treat the ground floor as common space.

The higher you build, the smaller the site can be, but the FAR needed on that site jumps. Stacking 100 rooms on a 336㎡ site at 31m requires FAR of 800%. South of Kyoto Station the post-deregulation FAR is 600%, so it does not work on this site. We therefore re-solve for the solutions that actually work, with both FAR and height as constraints.

South of Kyoto Station, the site needed for 100 rooms shrinks 33.3%, from 672㎡ to 448㎡, saving roughly ¥128 million in land cost. Along the Sotokanjo-sen corridor it shrinks 42.9%, from 672㎡ to 384㎡. From the standpoint of how hard it is to acquire a site, this difference is large. Securing a consolidated commercial plot in the 700㎡ class within Kyoto City and securing one in the 400㎡ class involve entirely different pools of properties for sale.

This calculation also explains why the height ceiling was removed in the “space-creation height control district.” Factoring in the 3m sidewalk-like open space and the 2m setback from adjacent lots, the building footprint that can actually be built on a 400㎡ site (assumed to be a 20m square) is 240㎡, and effective building coverage falls to 60%. Using up FAR of 700% under these conditions requires 12 floors, or about 37m. Even on a 500–600㎡ site it takes 11 floors, about 34m. Only at 800㎡ does it fit within 10 floors, about 31m. In other words, on small and mid-sized sites, 700% cannot be used up while the 31m ceiling remains. If open space is to be demanded while FAR is granted, the height ceiling had to be removed.

Finally, we work backward to the ADR needed for each case to pencil out. For construction cost, we start from the 2025 steel-frame (S) unit cost of ¥2.405 million per tsubo based on MLIT’s Building Starts Statistics, assume a 2029 opening, factor in two years of inflation at +5% per year up to the midpoint of construction, and arrive at ¥2.65 million per tsubo. We assume a target NOI yield of 5.0%, a GOP margin of 35%, full-year occupancy of 80% (a modelling assumption, not measured occupancy) and room revenue at 90% of total revenue.

Minami Ward’s estimated settled ADR averages ¥14,500 over the latest 11 months. That is 36% above the ¥10,600 required in the post-deregulation case. The compression of land cost from the height relaxation lowers the required ADR by only about ¥600, from ¥11,200 to ¥10,600, because land accounts for a small share — around 15% — of total investment. The effect of deregulation shows up less in a lower required ADR than in a wider range of workable site sizes — that is, in a larger population of candidate sites.

Because construction cost makes up about 90% of total investment, the unit-cost assumption moves the conclusion more than anything else. We vary the inflation built into the starting unit cost three ways and show the resulting range of break-even ADR.

Even at a construction unit cost of ¥2.8 million per tsubo, the required ADR is ¥11,200, 30% below Minami Ward’s actual level. Conversely, the required ADR only reaches the actual ¥14,500 at around ¥3.7 million per tsubo (+54% versus 2025), a level that is hard to envisage from current inflation rates. In this location, swings in construction cost are not the variable that decides whether the investment works.

The sensitivity to the assumptions is also worth a look. Changing the floor area per room moves the required ADR considerably.

The required ADR also moves with the target yield and the occupancy assumption. The table below varies both along two axes, each as a full-year modelling assumption.

Even under a stricter target yield of 6.0% and full-year occupancy of 75% (a modelling assumption), the required ADR is ¥13,600, within Minami Ward’s actual ¥14,500. This is a simplified calculation, and an actual investment decision requires a detailed feasibility study, but the direction is clear: the change in the combination of height and FAR has widened the range of sites worth considering.

Three Points for Site-Development Teams to Take Away

One more fact from the national plan data is worth adding. The median floor count of hotels under construction or planned is 14 floors, which does not fit within a 31m height control district. On a room basis, 82.0% sit in buildings above 10 floors. The 31m height limit is a line that collides head-on with the standard form of hotel development today. Put the other way, in areas where height control districts of 31m or less are widely designated, product designs that work at low rise — limited-service hotels that hold floor area per room to around 22㎡, or regular-shaped sites that can secure 20 or more rooms per floor — become the investment opportunity. Where you cannot build up, you solve with width and floor area per room.

In Kyoto City’s case, the deregulated areas are south of Kyoto Station, Rakunan Shinto and the Sotokanjo-sen corridor on the eastern side, among others; the Tanoji district in the city centre has not been changed. With the height environment of the city centre left as is, the scope for assembling new hotels of scale within Kyoto City tends to concentrate, in relative terms, in the deregulated areas. That Minami Ward’s estimated settled ADR has reached 90.6% of the city centre (Shimogyo Ward) is a figure worth keeping in mind as the premise for that.