The Cheapest Door Into Japan Isn't in Tokyo: Yokohama Will Put Up to ¥500 Million Against a Foreign Company's Office — and This Summer's Arrivals Are Walking Through It

JETRO maintains a public newsroom listing foreign companies that establish or expand operations in Japan. It is not a press channel anybody games; it reads like a shipping manifest. Which is exactly why it is useful. Run down the entries for the six weeks to early September 2026 and the addresses start to repeat.

On August 31, ShibuLA Ventures, an American ecosystem developer and venture studio, expanded into Yokohama, Kanagawa. On August 24, Eventus Hong Kong Limited established a Tokyo subsidiary for its healthcare IoT platform. On August 5, two arrivals on the same day: Planar Motor, the Canadian maker of magnetic-levitation transport robot systems, opened a sales office in Kanagawa Prefecture, and Nexus Events, an American online-events platform for startup support, took a Tokyo office. On July 30, BrainSpoke, a Thai developer of brainwave-measuring wearables paired with AI analysis, incorporated a Japanese subsidiary in Setagaya, Tokyo. Earlier in the year, iPS Bio established a Japanese subsidiary in Yokohama, TGL Solutions set up in Yokohama to expand construction DX, and Global Unichip Japan — the Japanese arm of the TSMC group's design specialist GUC — relocated to a new Yokohama office as part of its expansion.

Count them. Of the entries where the address is specified, more land in Kanagawa Prefecture — Yokohama above all — than in the Tokyo wards that a market-entry brochure would recommend. And the Tokyo entries that do appear are not in Marunouchi or Roppongi. BrainSpoke chose Setagaya, a residential ward with almost no corporate signalling value and rents to match. VMFi, the Taiwanese real-time speech-translation company, took Chuo-ku, which is central but decidedly unglamorous.

The obvious reading is that these are small companies priced out of the capital. That reading is wrong, and the presence of Global Unichip on the list is the tell: it belongs to the TSMC group, a company that could rent whatever floor in Tokyo it wanted. It chose Yokohama on purpose. So did the others. What they responded to is a set of offers that most foreign companies planning a Japan entry have never been shown.

Start with Yokohama, because Yokohama is the most aggressive. The city offers foreign companies tax incentives worth up to ¥500 million against the cost of leasing space for a major headquarters or R&D base, and a subsidy of up to ¥5 billion toward purchasing or constructing one. A separate growth-industry location subsidy runs to ¥2.5 million for companies in targeted sectors. On top of the cash, the city runs a Proof-of-Concept support programme and the Yokohama Matching Program, which connects international startups to Japanese business counterparts and drew strong interest in its first year.

Kanagawa Prefecture, which contains Yokohama, adds its own layer. Working with JETRO Yokohama, it operates IBSC Kanagawa, a startup office that foreign firms can use free of charge for 50 business days, and it reimburses half of a foreign company's setup costs up to ¥2 million. Those two instruments are less headline-grabbing than a ¥5 billion construction subsidy, but for the kind of company that actually enters Japan they matter more.

Consider what ¥2 million and 50 free business days actually cover. A Japanese subsidiary — a kabushiki kaisha or a godo kaisha — requires notarised articles of incorporation, a registered physical address, a judicial scrivener to handle the filing, certified translations of parent-company documents, and a corporate seal. That is the bill that stops most first entries, and it is denominated in exactly the range these programmes reimburse. Meanwhile the 50 days of free office space solve the chicken-and-egg problem that catches every foreign entrant: you need a registered address to incorporate, and most landlords want an incorporated entity before they will sign a lease.

The upper end of Yokohama's offer is aimed at a different animal — a manufacturer or a research operation making a capital commitment. But note what the ceiling signals even if you will never approach it. A city that authorises ¥5 billion for a foreign company's building is a city whose economic affairs bureau has been told, in writing, that attracting foreign firms is a priority. That translates into something more valuable than the subsidy: a bureaucracy with an incentive to say yes, and a one-stop desk staffed by people whose performance is measured on your arrival.

Subsidies of this kind are not new. What is new is the exchange rate they are denominated in. The yen has been trading near ¥157 to the dollar, a level it last touched in early August. For a company funding a Japan entry out of dollars or euros, every yen-denominated cost — the scrivener, the deposit, the first year of salaries, the office itself — is historically cheap. A ¥2 million reimbursement is worth the same in yen as it was three years ago; the expense it offsets is not.

The second change is the direction of official policy, which has moved from polite encouragement to a number on a page. Japan's inward FDI stock reached ¥53.3 trillion at the end of 2024, up 4.5% year on year and a record. Greenfield investment — new operations rather than acquisitions, which is the category most of the companies on JETRO's list fall into — hit $31.6 billion, up 15.4%, with data centres and logistics facilities prominent. In June 2025 the government raised its 2030 target for inward FDI stock from ¥100 trillion to ¥120 trillion, and signalled it wants ¥150 trillion as early as the early 2030s. Roughly speaking, Tokyo has committed to more than doubling the foreign corporate presence in Japan inside a decade. Prefectural budgets are the mechanism.

The third change is the one that decides whether any of this is worth acting on: demand. Incentives are worthless if Japanese companies are not buying. In the June Tankan, large non-manufacturers registered a diffusion index of +37 — a level last seen in 1991 — and large manufacturers +22, a fifth consecutive quarterly improvement. Corporate Japan is not merely solvent; it is confident. Regular readers will recall that we spent last week's article on the gap between that confidence and actual capital expenditure, which fell 1.2% in the second quarter. That gap is real and it still governs how you should structure a proposal. But it is a gap in the timing of spending, not in the existence of buyers.

Put the three together and the window has a specific shape. Setup costs are cheap in foreign-currency terms, the public sector is actively subsidising the fixed costs, and the customer base is in the best mood it has been in for three decades but slow to sign. That is close to an ideal configuration for a low-burn, patient entry — and a poor one for an expensive flagship that needs to justify itself in four quarters.

Here is the part the incentive brochures leave out. Japan's friction has never been primarily regulatory. Foreign ownership is unrestricted across most of the economy, profits repatriate freely, and the corporate forms are well-understood. The friction is administrative: documentation-intensive processes, filings that must be in Japanese, physical incorporation requirements, and compliance procedures detailed enough to add weeks to a timeline that looked like days on a slide. A ¥2 million reimbursement pays the invoice for that work. It does not shorten it, and it does not do it for you.

Nor does a subsidy generate demand. This is the failure mode we see most often: a company optimises its entry for the lowest landed cost, sets up in a prefecture that pays generously, and then discovers that its buyers all sit in Tokyo and expect to be visited. Yokohama is immune to this problem — it is under half an hour from Shinagawa, and a Yokohama address costs a foreign company essentially nothing in access or credibility. Kyoto, with its concentration of manufacturers and universities, is similarly defensible for the right sector. A location three hours from your customers is a different proposition entirely, however good the grant, and the further the address sits from your buyers, the more the subsidy is quietly paying you to be in the wrong place.

Finally, the money is conditional in ways that reward reading the fine print. These programmes carry eligibility criteria — sector targeting, headcount thresholds, minimum occupancy periods, and in some cases a requirement that the entity be newly established rather than relocated. The published ceilings are ceilings, not offers, and Yokohama's own materials direct applicants to the Business Development Division for the current terms. Any figure in this article should be treated as the start of a conversation with an economic affairs bureau, not as a line in your budget.

None of this argues against taking the money. It argues for sequencing: decide where your customers are, confirm the entity type and timeline with someone who has filed before, and only then ask which of the available programmes fit the decision you were going to make anyway. A subsidy that changes where you land is a subsidy worth scrutinising. A subsidy that pays for a landing you had already chosen is close to free money.

Start from the buyers, not the map. Write down the twenty organisations you most want as customers in year one and find their head offices. If fifteen sit in Tokyo's central wards, your search radius is the Tokyo–Yokohama corridor and nothing further, because the commute is the product. If they cluster around Osaka, Kyoto and Kobe — common in manufacturing, materials, precision components and food — then the Kansai programmes deserve the same scrutiny, and a Tokyo address is the one that costs you access.

Then approach the region before you approach the landlord. JETRO's regional desks and the prefectural investment bureaux exist to be contacted by companies that have not yet decided, and the early conversation is where you learn which programme you actually qualify for and what the current terms are, as opposed to what a 2022 brochure says. Ask three questions: what is the eligibility criterion I am most likely to fail, what is the realistic timeline from application to disbursement, and can you introduce me to a foreign company that went through this last year.

Structure the entry so the subsidy is upside rather than load-bearing. Use the free incubation period — Kanagawa's 50 business days is a genuine head start — to complete incorporation and hire the first local employee, and assume nothing about the grant until it is confirmed in writing. Entries that survive in Japan are the ones with a low enough monthly burn to still be present in the third year, when the relationships built in year one finally convert.

This is the work we do at Medusa Japan, and the reason we are in Osaka rather than Tokyo is the argument this article makes. The choice of where to land in Japan is usually made by default — a serviced office in a district somebody has heard of — and it is one of the few early decisions that is cheap to make well and expensive to reverse. The programmes are published, the desks answer their email, and this year the exchange rate is doing part of the work for you. The companies on JETRO's August list are not smarter than yours. They just read the offer before they signed the lease.