1,277 driving schools, 1.505 million graduates — 55 fewer schools in 10 years, graduates holding in the 1.5 million range

First, the size of the demand pool. According to the National Police Agency’s Driver’s License Statistics (2025 edition), there were 1,277 designated driving schools (designated by prefectural public safety commissions) at the end of 2025, and they produced 1,505,044 graduates during 2025. Of these, 1,033,525 (68.7%) were standard licence (Class 1) graduates. Designated-school graduates made up 97.8% of those who passed the driving test, so nearly every licence in Japan is obtained via a driving school.

Compared with ten years earlier, the number of schools has fallen by 55 (4.1%) from 1,332 at the end of 2016. Graduates, on the other hand, peaked at 1.72 million in 2021 and have since declined, but have stayed in the 1.5 million range since 2023. Graduates per school were essentially flat: 1,172 in 2016 and 1,179 in 2025. Within this distribution of schools, residential courses have become an established way of taking in students from far away.

So where do those graduates get their licences? Dividing prefecture-level graduate counts by population from the Ministry of Internal Affairs and Communications’ Population Estimates (as of October 1, 2024) shows clear regional differences. Nationwide, standard-licence graduates number 8.35 per 1,000 residents, but Yamagata reaches 29.8, 3.57 times the national rate, followed by Tottori at 24.3 (2.91×), Fukui at 16.9 (2.02×), and Niigata and Shimane at 15.1 (1.81× and 1.80×). At the other end, Tokyo is 5.1 (0.61×), Kanagawa 5.5 (0.65×) and Osaka 6.3 (0.76×) (the bottom three per capita are Tokyo, Kanagawa and Nara at 0.67×).

Graduates are counted where the driving school is located. A prefecture far above the national average per capita is therefore most naturally read as one that takes in students from other prefectures. Other factors matter too, of course: car ownership and licensing rates are higher in rural areas, and the share of young people differs. But some regional prefectures, such as Akita (0.90×) and Aomori (0.89×), sit below the national average, so the size of the multiple is hard to explain by rurality alone.

Scale — Yamagata exceeds the national average by 22,000 graduates; at 13 nights each, that is 280,000 guest nights a year

To translate this intake into lodging terms, we subtracted from each prefecture’s standard-licence graduates the value “population × national graduation rate” and called the remainder excess graduates. Multiplying by nights per person gives a sense of the scale of residential-course demand. For nights per person we used 13 nights from enrolment day to graduation day, since the median minimum course length for an automatic standard car across the 108 listed schools is 14 days.

Yamagata’s excess graduates come to 21,723; multiplied by 13 nights, that is about 282,000 guest nights. Yamagata recorded 4,739,560 total guest nights in 2025 according to the Japan Tourism Agency’s Overnight Travel Statistics Survey, so this is equivalent to 6.0% of that total. Tottori has 8,485 excess graduates, about 110,000 guest nights, or 4.3% of its total guest nights. Populous Shizuoka also has a large excess of 20,577, about 268,000 guest nights.

These figures should be read as an indication of scale, not as a precise estimate of residential-course demand. The excess may include students commuting from neighbouring prefectures, and whether students staying in school dormitories are counted in the Overnight Travel Statistics depends on how the dormitory is licensed. Even so, it is clear that demand on the order of hundreds of thousands of guest nights arises every year in specific prefectures, driven by factors unrelated to swings in tourist numbers.

Treating all of the excess as residential-course demand at 13 nights is a conservative assumption on nights but a strong one on attribution. We therefore varied two axes: the share of excess graduates attributed to residential courses (100%, 75%, 50%) and nights per person (13 nights for the automatic minimum, 15 for a manual-transmission course, and 18 for using the full extension guarantee).

Even in the most conservative case — half attributed to residential courses, at the minimum 13 nights — Yamagata retains about 141,000 guest nights (3.0%) and Tottori about 55,000 (2.1%). Conversely, if students stay up to the full extension guarantee, Yamagata expands to about 391,000 guest nights (8.3%). Under every assumption the demand equals 3.0–8.3% of total guest nights in Yamagata and 2.1–5.9% in Tottori — not negligible at the prefectural level.

Within 3 km of 108 schools — median 135 rooms; 16 schools with none, 14 with 1,000 or more

Next, we measured how many rooms there are around the driving schools offering residential courses. The scope is the 108 schools listed on one private booking site: Niigata has 11, Yamagata and Gunma 7 each, and Iwate and Miyazaki 6 each — concentrated on the Sea of Japan side of eastern Japan and in northern Kanto. Aggregating lodging facilities within a 3 km radius of each school gives 940 facilities and 64,353 rooms across the 108 catchments (deduplicated), and a median of 135 rooms per school.

The spread is wide. Sixteen schools have not a single room on sale within 3 km and 30 have 1–99, so together 46 schools (42.6%) have fewer than 100 rooms. At the same time 14 schools have 1,000 or more — schools in the urban areas of prefectural capitals such as Hiroshima, Niigata, Kochi, Miyazaki and Nagasaki fall into this group. Because driving schools need large courses, they tend to locate in suburbs or farmland, and most schools sit away from lodging clusters.

Looking inside the catchments, 74.0% of the rooms are in business hotels. For the 56 schools with three or more facilities whose selling prices were observed within 3 km, the median estimated settled ADR is ¥7,200 (interquartile range ¥6,400–¥8,400), and 39 of the 56 schools are below ¥8,000. Schools near hot-spring resorts (such as Akayu Onsen) see the level rise as ryokan enter the mix, but around most schools what you find are economy-priced business hotels in regional cities.

How students are housed — 59 schools use hotels and ryokan; median distance to partner lodging 3.65 km

Who houses residential-course students varies by school. Classifying the lodging described on each school’s listing page, 49 schools rely solely on their own dormitories or dedicated lodging, while 59 schools include regular lodging — hotels, ryokan, pensions, guesthouses and the like — among their housing options. Schools using regular lodging have a median of 186 rooms within 3 km, more than the dormitory-only schools (median 112). The more rooms cluster around a school, the easier it seems to be to bring outside lodging into its housing.

Still, many schools with no rooms within 3 km use regular lodging too: 11 of the 59 schools that include regular lodging have zero rooms within 3 km. We therefore matched the property names each school lists as student housing against our lodging database and identified 57 schools and 114 school–partner pairs. The straight-line distance from school to partner lodging has a median of 3.65 km: 51 pairs (44.7%) are within 3 km, 20 are 3–5 km, 36 are 5–10 km and 7 are more than 10 km away. Using lodging in a neighbouring town centre or hot-spring resort, with shuttle buses assumed, is evidently common practice.

Of the 114 partner pairs, 90 are business hotels, with a median of 99 rooms. Residential-course students keep the same room on weekdays and weekends alike and spend their days at the driving school. Blocks of singles and twins used two weeks at a time fit business-hotel rooms and operations as they are. Many listing pages say they guarantee an extension of 3–5 nights beyond the minimum course length (44 of the 47 schools that state a guaranteed number of nights), so stays extended by retaking tests also become lodging demand.

Another distinctive feature is that the demand structurally comes from outside the area. Of the 108 schools, 75 state that they do not accept applicants whose residence registration or family home is in the surrounding municipalities. Locals can commute to get their licence, so residential places are channelled to students from far away. From the lodging side, this demand moves for reasons unrelated to the popularity of local tourism or the presence of events.

13 schools in the top 4 receiving prefectures — urban and suburban schools use lodging differently

For the four prefectures with the highest graduates per capita — Yamagata, Tottori, Fukui and Shimane — we lined up the surroundings of their 13 listed schools. Schools near town centres (Tsuruoka, Fukui, Yonago) have more than 1,000 rooms within 3 km and can readily fold business hotels into their housing. Suburban schools, even with only a few dozen rooms within 3 km, can be seen using town-centre hotels 5–9 km away via shuttle.

In Yamagata, four of the seven schools have their own dormitory or dedicated lodging but also use regular lodging. Even within the same school group, some schools house students entirely in dormitories while others fold in town-centre hotels, so how lodging is used differs school by school. Across all 108 schools, several of the facilities that schools describe as on-campus or dedicated housing also call themselves “hotels” or “pensions” — meaning there is also a route by which lodging buildings end up serving as driving-school housing.

Seasonal overlap — in February, the busiest month for residential courses, business-hotel occupancy in the five receiving prefectures is below the annual figure

Residential-course booking sites describe February–March enrolments, when high-school seniors graduate, as the busiest period, followed by the August–September summer holidays. What about the lodging side’s seasonality? Using the Japan Tourism Agency’s 2025 final Overnight Travel Statistics, we lined up monthly business-hotel room occupancy in the five prefectures with the largest intake.

February occupancy was below the annual figure in all five prefectures. Niigata recorded 53.8% in February 2025, 11.1 points below its annual 64.9%; Shimane was 57.5% (annual 67.7%), Tottori 60.6% (68.5%), Fukui 68.8% (74.0%) and Yamagata 67.1% (69.0%). Nationwide, business hotels ran at 74.3% in February, so the pattern is one of winter occupancy dropping in the regional Sea of Japan prefectures. The peak season for residential courses coincides with the period when hotels in these prefectures are most likely to have empty rooms.

Summer, August–September, is different. August occupancy was above the annual figure in all five prefectures, so this is when the demand collides with tourism. For a property considering residential-course demand, a natural annual mix is to take in blocks of consecutive nights from winter into early spring and keep the summer tourism peak for regular sales.

Three location types for lodging operators

We organised the findings so far into three types from the perspective of lodging operators and investors. The classification, by rooms within 3 km and how housing is used, is the HotelBank Editorial Team’s own.

Around Type A schools, business-hotel plans that package singles and twins for long consecutive stays work as housing as they are. For Type B, the entry point is for properties on the shuttle route to connect directly with the school. In every type, demand arises at almost the same time and on almost the same scale every year. Its predictability, unlike tourism or business travel, is probably this demand’s single biggest feature.

Note that 11 of the 16 schools with no rooms within 3 km already include regular lodging among their housing. The host capacity already exists in the next town, so anyone placing new rooms near these schools needs to estimate demand against current operations, including the shuttles.

Frequently Asked Questions

Q. How many nights does a residential-course student stay?
A. Across the 108 listed schools, the median minimum course length for an automatic-only standard car licence is 14 days, or roughly 13 nights from enrolment day to graduation day. Manual courses take about two days longer. Many schools guarantee a 3–5 night extension if the stay runs longer, for example due to retaking a test.

Q. How many driving schools in Japan offer residential courses?
A. No public statistic gives the full count. There are 1,277 designated driving schools nationwide (end-2025); this article covers the 108 of them listed on one private residential-course booking site. Combining multiple sites would increase the number of schools covered.

Q. Can a property capture residential-course demand without a driving school nearby?
A. The median distance across 114 partner pairs is 3.65 km, and 37.7% are more than 5 km away. Lodging in a neighbouring town centre or hot-spring resort often serves as student housing with shuttles assumed, so properties outside the 3 km radius can also act as hosts.

Q. Which prefectures have the largest residential-course demand?
A. By standard-licence graduates per capita, the top prefectures are Yamagata (3.57× the national rate), Tottori (2.91×), Fukui (2.02×), Niigata (1.81×) and Shimane (1.80×).

Related Articles

• B.League Premier: rooms within 2 km of 11 arenas measured — up to a 103× gap, and opening-weekend lodging supply and demand