Housing the people who carry out recovery — what lodging takes on immediately after a disaster

According to the Japan Meteorological Agency, the earthquake occurred at 16:27 on July 28, 2026, with its epicenter in the Kumamoto region of Kumamoto Prefecture, a magnitude of 7.1, and a maximum seismic intensity of 7 observed in Uki City and Hikawa Town, Yatsushiro District. The agency noted that four earthquakes of intensity 5-minus or above had occurred by 23:00 on July 28 and called for continued vigilance, citing “past cases in which quakes of similar scale continued for roughly a week.” The Prime Minister’s Office reported that, as of 17:35 on July 28, power outages affected a total of 48,280 households in Kumamoto Prefecture and elsewhere, and city gas supply was suspended to approximately 500 households. The tsunami advisory was lifted at 18:10 the same day. The Ministry of Land, Infrastructure, Transport and Tourism has published damage updates continuously, from its first report as of 18:10 on July 28 through its fifth report as of 14:00 on July 29, and the Cabinet Office (Disaster Management) has compiled the damage situation as of 07:00 on July 29.

In a disaster-struck region, the first role lodging plays is not to host tourism. During the emergency response phase, support units arrive from local governments, police, fire services, the Self-Defense Forces, and the electricity, telecommunications, gas and water utilities. After that the roster turns over — damage assessment, emergency repairs, then full-scale reconstruction work — and visitors from outside the region keep flowing in continuously over months and years. This kind of stay differs in character from both ordinary business travel and leisure travel.

First, it is long. Consecutive stays of several weeks to several months per work zone are the norm, and the same individual stays repeatedly at the same property. Second, it is weekday-centered. Because returning home on weekends is standard practice, demand stacks up heavily from Monday through Thursday. Third, room-only stays and simple meals are what is wanted. Early departures and late returns are routine, so practical features matter more than elaborate in-house dining: light meals available early in the morning, laundry facilities, parking space, and provision for handling work clothes. Fourth, price tolerance follows corporate rules. Rates that fit within per-diem and lodging allowance regulations get chosen, so demand does not flow readily to the very top of the price range.

In other words, a new layer of demand lands squarely on the territory that existing business hotel and city hotel stock is inherently good at serving. Below, we first review how the record from a decade ago captured this demand, then map where the current Kumamoto-area stock sits, at what price points, and in what volume.

The 2016 record — business-oriented properties beat the prior year from two months after the quake

The Japan Tourism Agency’s “Accommodation Travel Statistics Survey” publishes room occupancy rates with properties split into two groups: those where leisure guests account for 50% or more of stays, and those where they account for under 50%. The latter is effectively the set of properties serving primarily business use, and placing these two groups side by side makes what happened in Kumamoto Prefecture in 2016 clearly legible.

Room occupancy at business-oriented properties fell to 51.9% in April 2016, the month of the quake — a decline of 7.4 points from 59.3% in the same month a year earlier. Two months later, however, it had recovered to 68.9% in June, exceeding the prior-year 57.6% by 11.3 points. From there it stayed consistently above the prior year through the end of the year: 72.1% in July (+7.8pt), 76.4% in November (+7.0pt), and 69.9% in December (+6.9pt). The full-year figure came to 68.4%, up 4.1 points from 64.3% in 2015 and 7.0 points from 61.4% in 2014.

Leisure-oriented properties moved differently. May 2016 came in at 40.0%, 5.9 points below the prior-year 45.9%, and the shortfall persisted from summer into autumn — 52.9% in August (-2.6pt) and 43.5% in October (-3.7pt). The full year finished at 43.1%, only marginally above the prior year’s 41.8%, and the gap against the business-use segment widened.

Total guest nights show the same pattern. Guest nights at business-oriented properties were down 3.3% year on year in May 2016, but by June had reached 338,170 — 19.0% above the prior-year 284,200 — and July continued with a 6.3% gain. The full year came to 4,033,170 guest nights (+3.2% year on year). Over the same period, leisure-oriented properties fell 19.0% year on year in May and 10.8% in August, ending the year at 3,242,020 guest nights (+0.7%).

What this time series shows is a particular shape of demand: both leisure and business demand fall in the first one to two months after a disaster, but the business side reverses within roughly two months and then holds at a high level for at least the remainder of the year. That said, inbound demand in Kumamoto Prefecture was also recovering over the same period in 2016, so not all of the occupancy gain can be attributed to recovery-related demand. Even so, the fact that business-use and leisure-use properties diverged clearly in both the timing and the slope of their recovery indicates that business-purpose stays were structurally thickening during this period. For how far the leisure side came back over the following decade, Kumamoto Quake 10Y × Aso: Data on Recovery and Tourism Brand Rebound traces the longer trajectory.

The timeline in construction starts — recovery-related building activity ran for more than three years

In gauging how long demand persists, the trajectory of construction activity itself is a strong clue. Lining up annual construction starts for lodging-use buildings in Kumamoto Prefecture from the Ministry of Land, Infrastructure, Transport and Tourism’s “Statistical Survey on Building Construction Starts” makes the shift from 2016 onward legible.

Activity had been running at low levels — 11 buildings and 8,283 m² in 2015, and 14 buildings and 7,551 m² in 2016 — before doubling to 31 buildings and 12,413 m² in 2017, then holding high for three consecutive years at 52 buildings and 31,374 m² in 2018 and 51 buildings and 32,256 m² in 2019. Building counts reached 3.6 to 3.7 times the 2016 level, and floor area around 4.2 times.

One caveat matters here: lodging-use construction starts in this period cannot be explained by recovery purposes alone. The years 2017 through 2019 were when inbound demand was expanding nationwide and new lodging development was at its most active. The increase therefore includes both the rebuilding and restoration of damaged lodging properties and new development betting on market growth. Even so, the timeline — starts ramping up from the year after the quake and staying elevated for at least three years — suggests that construction activity tied to recovery does not wind down quickly. The same question of duration has been observed after disasters elsewhere; on the aftermath of the Noto Peninsula earthquake, Noto Earthquake +16 Months: Wakura Onsen Recovery & Summer 2026 ADR sets out how the lodging side has moved.

More recently, 2024 again reached a high level at 59 buildings and 30,219 m². Behind this lies the semiconductor-related industrial cluster discussed next.

Where Kumamoto’s business-use lodging stock sits

Within the scope tracked by MetroEngines Research, operating business hotels, city hotels and capsule hotels in Kumamoto Prefecture total 117 properties and 11,702 rooms (as of July 29, 2026). Across all categories the figure is 476 properties and 18,377 rooms, meaning stock serving business use accounts for just over 60% of the prefecture’s rooms. Note that this tabulation covers properties whose listings can be confirmed on OTAs and similar channels; it is not a complete census, and unlisted ryokan, minshuku and simple lodging houses are excluded.

Three features characterize the distribution. First, concentration in Chuo-ku, Kumamoto City. Chuo-ku alone holds 43 properties and 5,912 rooms, or 50.5% of the prefecture’s business-use rooms. Adding Nishi-ku (930 rooms), Higashi-ku (342 rooms) and Kita-ku (175 rooms) brings Kumamoto City to 7,359 rooms, or 62.9% of the prefecture. The same area is where the prefectural and city government offices and the branch offices of infrastructure operators — the bases for recovery work — are clustered, and it has the greatest capacity to absorb business demand even in normal times.

Second, the weight of Yatsushiro City. With 8 properties and 889 rooms, it is the largest outside Kumamoto City after Ozu Town. Yatsushiro City adjoins Hikawa Town, where intensity 7 was recorded, and is also a transport hub for the south of the prefecture. Its estimated settled ADR was ¥6,500 as of June 2026, up 14.3% from ¥5,700 in the same month a year earlier (N=20 properties) — the clearest increase among the prefecture’s main cities.

Third, the industrial cluster area of Ozu Town, Kikuyo Town and Koshi City. These three municipalities hold 10 properties and 1,339 rooms. Ozu Town in particular has 6 properties and 912 rooms, a room count that stands out relative to its population. Business demand tied to semiconductor-related plant construction and ramp-up has underpinned lodging stock in this area even in normal times.

Prefecture-wide estimated OCC (based on OTA sales inventory) stood at 84.9% for the business hotel category (July 1–29, 2026), 88.0% for the city hotel category (same period), and 83.9% across all categories (same period). These are high levels for a summer average, and they show that there is no large pool of unsold rooms even in normal times. When recovery-related stays are added, it is hard to argue that existing stock has much room to absorb a large increase — and longer stays simultaneously reduce turnover. The following chapters examine this in detail.

How semiconductor construction demand overlaps — competing for the same rooms

Business demand in the Kumamoto area has already thickened by one notch over the past few years. The semiconductor-related industrial cluster centered on Kikuyo Town, Ozu Town and Koshi City has made extended stays tied to plant construction and production ramp-up a permanent fixture. This demand closely resembles recovery-work demand in terms of the kind of room it needs. Four conditions — long stays, weekday-centered, room-only, corporate price bands — align almost exactly, so the two compete for the same room inventory. For how this industrial demand has actually moved ADR in Kikuyo, Koshi and Ozu, TSMC Fab 2 & Kumamoto’s Chip Corridor: Kikuyo–Koshi–Ozu ADR 2024–2026 examines 32 months of data.

Estimated settled ADR in Chuo-ku, Kumamoto City held steady in the ¥8,000s through 2025, then became more volatile entering 2026, with June at ¥7,700, down 5.1% year on year. Looking ahead, the listings observed at the time of this survey imply levels above ¥10,000 for September through November 2026. Yatsushiro City shows a clearer move: after running in the ¥5,700–¥6,300 range in the first half of 2025, it rose to ¥6,500 in June 2026 (+14.3% year on year), with rates in the ¥7,900–¥8,700 range listed for July onward. Ozu Town is similar, moving from ¥8,000 in June 2026 to levels above ¥10,000 from September.

Note, however, that figures from July onward are estimates derived from selling rates listed on OTAs and similar channels at the time of the survey, and they shift as the check-in date approaches. It should also be kept in mind that the effects of this earthquake are not necessarily reflected in these figures.

Movement on the supply side

Supply additions continue. Within the scope tracked by MetroEngines Research, Kumamoto Prefecture saw 37 openings and 823 rooms confirmed in 2025 and 33 openings and 1,164 rooms in 2026. The main 2026 additions are Via Inn Prime Kumamoto (232 rooms), Tabino Hotel Aso Kumamoto Airport (213 rooms), Workers Hotel Kumamoto Ozu (202 rooms) and Hotel Amanek Kumamoto (176 rooms). In 2025, Toyoko Inn Kumamoto Airport (208 rooms), Super Hotel Premier Aso Kumamoto Airport (203 rooms) and Onyado Nono Kumamoto Natural Hot Spring Higo-no-Yu (191 rooms) came on line.

Notably, a property explicitly designed around extended stays has opened in Ozu Town. That is a sign that demand generated by the industrial cluster calls for a product specification different from a conventional business hotel.

Against the prefecture’s 11,702 business-use rooms, new openings in 2025 and 2026 combined total approximately 1,987 rooms. That figure includes resort and ryokan categories, however, so the net addition to business-use stock is smaller. When recovery-related stays are layered on, supply additions will not necessarily keep pace with the increase in demand.

The price-band map — white space above ¥10,000

Breaking down the 88 business-use lodging properties and 10,420 rooms for which an estimated settled ADR could be calculated as of June 2026 by price band brings the market structure into focus.

The ¥6,000-to-under-¥8,000 band concentrates 42 properties and 5,254 rooms, or 50.4% on a room basis. Adding the under-¥6,000 band brings the total to 70 properties and 7,799 rooms — fully 74.8% falling within these two bands. By contrast, the ¥10,000-to-¥12,000 band holds just 5 properties and 717 rooms, the ¥12,000-to-¥15,000 band just 2 properties and 378 rooms, and above ¥15,000 there is nothing at all.

The scatter plot shows properties packed into the mid-scale range of 100 to 250 rooms at ¥6,000 to ¥8,000. This is the standard template common to business hotel markets nationwide, and the Kumamoto area is no exception. Above it, in the territory of ¥10,000 to ¥15,000 at around 100 rooms, almost no properties exist apart from a handful in Chuo-ku, Kumamoto City.

Room to grow in product design for consecutive and extended stays

Taking the analysis so far together, Kumamoto-area lodging stock has room to grow in three directions. All three combine raising the capacity to house the people carrying out the region’s recovery over an extended period with the properties’ own profitability.

First, rate design built around consecutive nights. Both recovery-related and industry-related stays run in weekly and monthly units. With a per-night rate structure left unchanged, extended-stay guests see the price as expensive while properties accumulate booking-acquisition costs and cleaning frequency. Designing tiered consecutive-night discounts and making cleaning frequency optional lowers the guest’s effective burden while compressing operating costs. Returning the cost saved through adjusted cleaning frequency into the rate creates room to improve profitability while holding the displayed price steady.

Second, adding stay functionality. Laundry, a kitchenette or shared kitchen, storage space for work clothes and tools, light meals available from early morning, and parking bays that accommodate large vehicles. For extended-stay guests these are reasons to choose a property that outweigh price, and they provide the basis for placing product in the ¥8,000-to-¥12,000 band where the gap has opened. That band currently holds only 16 properties and 2,243 rooms, so even converting a portion of rooms at an existing property to this specification can shift its price-band position.

Third, inventory allocation matched to the time structure of demand. As the 2016 record shows, business-use demand ramps up roughly two months after a disaster and then continues for years. Leisure demand, meanwhile, falls during the same period and returns gradually afterward. This asymmetric recovery curve strengthens the logic of an inventory design that combines period and day of week — allocating weekdays to extended stays and weekends to leisure. A two-tier structure that builds an occupancy base on weekday consecutive stays and layers higher-rate demand on weekends contributes to stabilizing RevPAR across the year.

Estimating absorption capacity — three scenarios and a two-axis sensitivity on stay length and occupancy ceiling

Let us bundle the figures so far into a single measure: absorption capacity. Against the prefecture’s 11,702 business-use rooms, estimated OCC for the business hotel category was 84.9% (July 1–29, 2026). Since occupancy never reaches 100% in practice, we set a practical occupancy ceiling that accounts for cleaning, renovation and out-of-order rooms, and treat the gap between that ceiling and current occupancy as “spare rooms.” On top of that, we layer an assumption that a certain share of non-business-use rooms (the 6,675 rooms left after subtracting 11,702 business-use rooms from 18,377 rooms across all categories) is converted to business use. All of the following is derived mechanically from market structure and is not actual performance.

Even in the mid-case scenario, absorption capacity is about 1,160 rooms and new arrivals absorbed per month come to roughly 2,500 people. Given that the prefecture’s business-use stock exceeds 10,000 rooms, that margin is by no means large. The longer stays become, the lower the turnover per room, so the same room count houses fewer people. The table below develops this relationship along two axes: the practical occupancy ceiling and average stay length.

Two things stand out. First, as average stay length extends from 7 days to 30 days, monthly absorption falls to less than a quarter. The longer recovery-work stays run, the more sharply the number of people the same stock can rotate through declines. Second, pushing the occupancy ceiling up 4 points from 88% to 92% multiplies spare rooms by roughly 2.3 times. A few points of upside from better cleaning and inventory management has an impact on absorption capacity comparable to shortening stay length. Consecutive-night discounts and optional cleaning frequency ease the guest’s burden and, at the same time, are levers for raising that occupancy ceiling.

Conclusion — lodging as regional absorption capacity

The Kumamoto area’s business-use lodging stock comes to 117 properties and 11,702 rooms, half of it concentrated in Chuo-ku, Kumamoto City. Estimated OCC for the business hotel category is already high at 84.9% (as of July 2026). If recovery-related extended stays are added on top, it is hard to argue that existing stock holds much spare capacity. Because longer stays lower turnover per room, the total number of people the same room count can house actually moves in the opposite direction — down.

The record from a decade ago shows this demand ramping up roughly two months after the quake and holding at a high level for at least the rest of that year. The construction-starts statistics show lodging-use building activity staying elevated for more than three years from the year after the quake. The timeline is not short.

On top of that, the price-band distribution shows a structure that is thin above ¥10,000. That also means product with features suited to extended stays barely exists in the prefecture. How much accommodation can be prepared where the people carrying out the recovery can stay for months while maintaining their quality of life? That is an investment question, and at the same time a condition that sets the very pace of the region’s recovery. What is needed most right now is a shift in perspective — seeing lodging not only as a receptacle for tourism, but as the region’s absorption capacity.