The framework — three separate laws offer the same 48% bonus depreciation

Location-linked tax incentives that reach lodging facilities come mainly from four laws: the Act on Special Measures for Support of Sustainable Development of Depopulated Areas (Act No. 19 of 2021, hereafter the “Depopulation Act”), the Remote Islands Development Act, the Peninsula Development Act, and the Act on Special Measures for the Development of the Amami Islands. On the national tax side (income tax and corporate tax), all of them connect to Article 12(4) (income tax) and Article 45(3) (corporate tax) of the Act on Special Measures Concerning Taxation and use the same bonus depreciation framework.

What matters is that the ryokan/hotel business is explicitly listed among eligible industries. This refers to the hotel, ryokan and simple-lodging categories defined in Article 2 of the Hotel Business Act; boarding-house operations and adult-entertainment-related operations are excluded (per published materials from Nagano Prefecture and the town of Yurihama). Eligible assets are machinery and equipment, buildings and their fixtures, and structures — and for buildings this covers not only new construction but also acquisition or construction through extension, reconstruction, repair or alteration work. In other words, renovation of an existing ryokan is squarely eligible.

The acquisition-price requirement is tiered by the operator’s capital size. Under the depopulation scheme, a corporation capitalized at ¥50 million or less (including individuals) qualifies from ¥5 million; above ¥50 million and up to ¥100 million, from ¥10 million; above ¥100 million, from ¥20 million. Where capital exceeds ¥50 million, only new construction or expansion qualifies (per published materials from Yurihama Town and Kyotango City). The peninsula and remote islands schemes draw the brackets slightly differently: from ¥5 million for capital of ¥10 million or less, from ¥10 million for capital above ¥10 million up to ¥50 million, and new construction or expansion of ¥20 million or more for capital above ¥50 million (MLIT, “Bonus Depreciation Scheme for Industrial Machinery in Peninsulas and Remote Islands”).

The procedure has two stages. First, a confirmation application is filed with the municipality, which issues a certificate stating that the capital investment conforms to the industrial promotion provisions of the “municipal plan for sustainable development of depopulated areas.” That certificate is then attached to the filing with the tax office (per guidance from Kanonji City and Yurihama Town). On the local tax side, separate exemption applications must be filed with the prefecture and the municipality. Note carefully that if the municipality has not set industrial promotion provisions in its plan, the national tax incentive cannot be used even if the municipality is a depopulated area.

What bonus depreciation actually does — ¥82.5 million of depreciation pulled forward over five years

Bonus depreciation is not a tax cut but a deferral. The total depreciation is unchanged; it is merely brought forward. But in hotel development, where cash flow is tightest immediately after opening, that acceleration does move investment returns.

Take the building acquisition price of ¥1.323 billion used in the analysis below (reinforced concrete, 2,000 sqm gross floor area). Applying the statutory useful life of 39 years for a ryokan-use RC structure and the straight-line depreciation rate of 0.026, the ordinary depreciation limit is ¥34.4 million per year. Adding 48% brings it to ¥50.9 million, so the ¥16.5 million difference, over five years, totals ¥82.5 million of additional deductible expense.

That said, what really moves an investment decision is the part that is not a deferral — the exemptions from fixed asset tax and real estate acquisition tax. These are outright reductions, never clawed back. In the Shimoda case below, the three-year fixed asset tax exemption alone saves ¥37 million cumulatively, and the real estate acquisition tax exemption saves ¥32.4 million.

Designated areas cover 63% of Japan’s land — yet attract only 3.4% of new rooms

There are 885 depopulated municipalities. According to the Ministry of Internal Affairs and Communications’ “Current State of Depopulation Measures, FY2023 edition (summary),” this is 51.5% of Japan’s 1,719 municipalities and 63.1% of the national land area — while accounting for just 9.3% of the population. The breakdown is 713 fully depopulated, 158 partially depopulated and 14 deemed depopulated (Ministry of Internal Affairs and Communications, “List of Depopulated Municipalities,” as of April 1, 2022. Additional designations follow the results of the 2020 and 2025 national censuses, so the latest status should be confirmed with each municipality).

So does new hotel supply actually go to these vast incentive-eligible areas? We matched lodging facilities of 20 rooms or more whose opening between 2019 and 2025 could be confirmed in our database against this municipal classification.

Depopulation-related municipalities account for 51.5% of all municipalities but host only 10.4% of new rooms. The figure of 3.4% across the 713 fully depopulated municipalities (41.5% of the total) shows directly how limited the power of tax incentives to attract supply is. The reason partially depopulated areas (6.0%) exceed fully depopulated ones is likely that partially depopulated municipalities include many cities with urban centers, and openings concentrate on that urban side — outside the depopulated zone.

“Depopulated equals weak demand” does not hold — 97 fully depopulated municipalities run 30% above non-designated ones

Supply not arriving is not the same as demand being weak. When we take our estimated settled ADR by municipality as a 12-month average (August 2025 to July 2026) and compute medians by depopulation classification, an unexpected picture emerges.

The median across the 97 fully depopulated municipalities that clear the sample threshold is ¥12,300 — 30.1% above the ¥9,400 of the 222 non-designated municipalities. Depopulated areas include many hot-spring and resort destinations, while the non-designated side includes many clusters of regional-city business hotels. On rate level alone, depopulation status is no proxy for weak demand.

Looking at the top of the ranking individually, the areas where tax incentives and demand coexist come into focus.

The most practically important caveat here is “partially depopulated” status. In a partially depopulated municipality, only the area of the pre-merger former municipality is deemed a depopulated area. Seven of the top 15 are partially depopulated, and the conclusion flips entirely depending on whether the center of tourism demand falls inside the zone.

The Izu Peninsula as a natural experiment — one peninsula, three tiers of tax treatment

No place is better suited to testing the impact of these measures than the Izu Peninsula. Within the same peninsula and the same hot-spring market, the applicable scheme splits into three tiers by municipality.

• Depopulation tax tier: six municipalities — Shimoda, Izu, Nishiizu, Minamiizu, Matsuzaki and Kawazu — are designated fully depopulated (Ministry of Internal Affairs and Communications list, as of April 1, 2022). They are also Peninsula Development Areas, but from FY2023 the depopulation scheme applies where designations overlap.
• Peninsula tax tier: Higashiizu Town does not appear on the depopulated-areas list. It appears in MLIT’s “List of districts certified by the relevant ministers as districts formulating an industrial promotion plan under the Peninsula Development Act” (as of July 1, 2025) as part of the Central-Southern Izu district, with a plan period from July 1, 2025 to March 31, 2030.
• Non-designated tier: Atami and Ito appear on neither the depopulated-areas list nor the peninsula district list above.

Set the measured supply-demand figures alongside the tiers and they do not line up neatly. Izu (¥22,000) in the depopulation tax tier exceeds Ito (¥20,700) in the non-designated tier and sits at almost the same level as Higashiizu (¥20,900) in the peninsula tax tier. Meanwhile Shimoda, also in the depopulation tax tier, sits at ¥15,100 — the spread within the market is wide. What supply is currently entering this market is broken down by segment in Shizuoka’s 67 new 2026 hotels: 1,089 rooms, 64% one-room villas.

It is worth checking the macro occupancy figures too. According to the Japan Tourism Agency’s Accommodation Travel Statistics Survey (May 2026, second preliminary report), Shizuoka Prefecture’s room occupancy was 55.1% overall (-0.5pt year on year), with ryokan at 45.3%, resort hotels at 57.4% and business hotels at 70.6%. Our estimated occupancy runs higher because its denominator is only the inventory that listing properties place on OTAs — a different basis from a property’s true overall occupancy. The scenarios below use the Japan Tourism Agency levels as the premise.

Land prices move the other way — Atami commercial land +77% in five years, Shimoda -9%

Assessing the impact of the tax measures requires looking at how the non-tax investment costs are moving. Using MLIT’s Real Estate Information Library, we tracked official land price benchmark sites on the Izu Peninsula at identical locations over time.

The benchmark site in Ginza-cho, Atami (commercial district, FAR 500%) rose from ¥155,000/sqm in 2021 to ¥275,000/sqm in 2026 — +77.4% over five years — and is still climbing at +14.6% year on year in the latest year. The site in Wadahama-minami-cho in the same city (commercial district, FAR 400%) also rose from ¥100,000 to ¥111,000 (+11.0%, +2.8% in the latest year). By contrast, the benchmark site in Shimoda 2-chome (commercial district, FAR 400%) fell from ¥67,600 to ¥61,600, -8.9%, and remains on a downward path at -1.1% in the latest year.

In other words, on the Izu Peninsula land acquisition costs are rising sharply on the side without tax incentives and falling on the side with them. This inverse movement in land prices is essential context when evaluating what the incentives are worth.

Development scenario — on a 40-room new build, the incentives move IRR by 0.55pt

Now to the core question. We model a hotel of identical specification in Shimoda (depopulation tax tier) and in Atami (non-designated tier), and test how far IRR and NPV move with and without the tax incentives. For Atami we also run a counterfactual “if it were designated” case, purely to isolate the impact of the measures (in reality Atami is not designated).

Three things stand out.

First, the impact is stable at +0.55 to +0.73pt. In the Shimoda case, NPV (5% discount rate) rises by ¥66 million. That is 107% of the ¥61.6 million land acquisition cost for the assumed site — enough to cover the land outright. Put another way, it is economically equivalent to raising ADR by 6.8%, or occupancy by 4.4pt. Translated into the day-to-day reality of hotel operations, that is far from trivial.

Second, that margin does not overturn the demand gap between locations. Even using every incentive, Shimoda returns -0.95%, nowhere near Atami’s 4.38% without any incentives. The gap in estimated settled ADR (¥15,100 vs. ¥24,600, a 63% difference) dominates. The fact seen in Table 3 — that the 713 fully depopulated municipalities captured just 3.4% of new rooms — is consistent with this arithmetic. Tax incentives are not a device for making a location work where demand does not reach.

Third, and ironically, the impact grows the stronger demand is. The +0.73pt in the Atami case exceeds the +0.55pt in Shimoda because the enterprise tax exemption presupposes taxable income. The schemes are designed for disadvantaged regions, but the size of the benefit scales with earning power. This asymmetry is likely one reason the measures have not moved supply.

What deserves attention in the sensitivity analysis is that Δ — the impact of the incentives — barely changes under demand stress. Even with ADR down 10% and occupancy down 5pt, the Shimoda Δ holds at +0.54 to +0.55pt. Tax incentives are a high-certainty source of return that is independent of market risk. Their power to lift the absolute level, however, is limited.

Where the measures really work is renovation — from ¥5m, IRR moves 7.94% to 8.57%

Everything above assumed new construction. But both the bonus depreciation and the tax exemptions explicitly cover building renovation (extension, reconstruction, repair, alteration). And for a corporation capitalized at ¥50 million or less, or an individual, they apply from an acquisition price of ¥5 million — a level reachable by renewing the wet areas in ten guest rooms.

Depopulated areas hold a deep stock of aging lodging properties. As Table 7 shows, Shimoda alone has 107 properties and 1,765 rooms on OTA listings. Even in a market where a new build’s IRR does not work, value-add — buying an existing property at a reduced price and refurbishing it — is a different proposition. On the prior question of whether to renovate or rebuild, Japan’s pre-1981 hotel stock: 236k rooms, 19.1% — retrofit or rebuild tests the tipping point on a building-count basis.

The renovation scenario clears the bar at 7.94% even without incentives. What produces the gap against the new build’s -1.51% is that total investment lands at ¥586 million — 39% of the new build’s ¥1,517 million — while revenue holds at ¥173 million, 75% of the new build’s ¥230 million. The year-1 NOI yield of 7.14% exceeds the 6.0% exit cap rate.

Layer the tax incentives on top and it rises to 8.57%. The +0.62pt is economically equivalent to raising ADR by 3.8%. It is smaller than the new build’s 6.8% equivalent because the depreciable building investment is smaller — but what changes the meaning is whether the absolute level already works. Moving 7.94% to 8.57% and moving -1.51% to -0.95% carry entirely different weight in an investment decision.

The two-axis view makes the conditions for the renovation scenario clear. Even if ADR falls 15% to ¥18,000, securing 60% occupancy delivers a year-1 NOI yield of 6.31%, above the 6.0% exit cap rate. Conversely, if the downside hits both rate and occupancy (¥18,000 × 55%), the result is 5.78% and falls below the cap rate. On the upside, ADR of ¥24,300 at 75% occupancy reaches 10.65%. This two-axis grid reveals a structure invisible in the one-axis sensitivity of Table 11: the deal survives losing either rate or occupancy, but not both.

Practical risks and checkpoints — national tax and local ordinance deadlines do not align

Procedurally, the national and local taxes have different filing destinations and different deadlines. For national tax, the municipal certificate must be obtained and then submitted to the tax office together with the tax return. For local tax, separate exemption applications go to the prefecture (enterprise tax, real estate acquisition tax) and the municipality (fixed asset tax). Some local governments set the fixed asset tax application deadline at “January 31 of the year in which the first day of each fiscal year falls” (Kyotango City), meaning an application is required every year — not just in year 1 but in years 2 and 3 as well. That is an operational burden worth budgeting for.

Conclusion — tax incentives are not the deciding factor in site selection, but they lift turnaround deals

The bonus depreciation and tax exemptions under the Depopulation Act, the Remote Islands Development Act and the Peninsula Development Act lift lodging investment returns by +0.55 to +0.73pt of IRR. That range equals +3.8% to +6.8% in ADR terms, or +3.6pt to +4.4pt in occupancy terms, and in NPV (5% discount rate) it is ¥66 million to ¥86 million on a 40-room new build — comparable to the land acquisition cost of the assumed site. And under stress from ADR -10%, occupancy -5pt or construction cost +10%, the impact barely changes. It is one of the few high-certainty sources of return that is independent of market risk.

At the same time, that margin does not overturn the demand gap between locations. On the Izu Peninsula measurements, Shimoda in the depopulation tax tier (estimated settled ADR ¥15,100) does not reach the no-incentive case for Atami in the non-designated tier (¥24,600) even using every incentive available. The measured fact that the 713 fully depopulated municipalities (41.5% of all municipalities) captured just 3.4% of new rooms (20+ rooms) opened between 2019 and 2025 reflects that structure directly.

The practical conclusion, then, is this. Tax incentives will not decide site selection, but in the regeneration of existing stock in designated areas where demand can be measured, they can move the tipping point of an investment decision. The fact that the median estimated settled ADR across 97 fully depopulated municipalities runs 30.1% above the 222 non-designated ones shows that the intuition of “designated depopulated area equals weak demand” does not hold statistically. A list of areas where designation and demand coexist — Minamioguni, Kyotango, Minamiaso, Izu, Miyazu — can genuinely be assembled.

When evaluating a site acquisition or the purchase of an existing property, three questions are worth checking with the same priority as land price and floor area ratio: which designation the municipality holds, whether industrial promotion provisions are written into the municipal plan, and how long the ordinance remains in force.