Invest1 distinct publisher3 min readPublished
The business manager visa now demands 30 million yen of capital, six times the old floor, and 953 foreign operators wound up and left in six months rather than test a renewal grace period that runs to October 2028.
The Investor · Invest desk

Compiled by The InvestorSomething wrong?How this is made
Thirty million yen counts as equity that sits on the balance sheet [4], and an operator who has it can in principle spend it on the business, so the part that bites is the gap between that number and what the business actually needs, which for a restaurant or a two-person software shop is most of it. Harder still are the requirements that depend on the founder rather than the balance sheet: three years of management experience or a master's degree, one full-time employee, Japanese at JLPT N2, a business plan verified by a certified small-business consultant, tax accountant or accountant, and an office that is not the applicant's home [5].
Set against that, the agency's concession looks generous on paper. Existing holders can renew through Oct. 16, 2028 even if they fall short, with further permission possible where there is a prospect of compliance [16], which is three full years from the ordinance revision [10]. The half-year exodus still equals 2.0 percent of the 46,781 holders on the books at the end of 2025 [15][3], and that pace implies a prior-year comparison base of about 244 people [2]. Founders with three years of runway do not wind down in month eight unless they have priced the runway as shorter than it looks, and the Nikkei account notes that scrutiny of actual business operations has tightened regardless of the renewal window [17].
The number that makes this structural rather than a tidy-up is on the domestic side: of the 143,367 companies incorporated in Japan in 2024, excluding sole proprietorships and similar entities, 1,491 carried capital of 30 million yen or more [13], which is 1.04 percent [4]. Foreigners must now enter at a scale that 99 of every 100 Japanese founders never reach.
The counter-thesis carries real weight. The Immigration Services Agency argues the old bar was low by international comparison, citing Korea's 300 million won and an American range of $100,000 to $200,000 [8]; 30 million yen is put at about 257.61 million won [9], or 85.9 percent of the Korean requirement [5] and 6.4 percent below the top of the US range [6]. On level, Japan has moved to roughly where its comparators already sat, and Justice Minister Hiroshi Hiraguchi's statement that screening found cases with no actual business substance [14] is what applications running at 4 percent of their prior rate [9] would look like if most of those applications were options on residency rather than plans to trade.
Where the two readings part company is the operators already inside, and the allocation they are now funding: money into paid-in capital instead of headcount, or, in the 19 cases that would hand management to a Japanese national or permanent resident, control paid over as the price of staying [11].
This is probably wrong, but the durable cost here is the discretionary substance test layered on top of the number, because discretion is the one input a founder cannot underwrite. What would refute it: monthly departures falling back to the few dozen that prevailed before October 2025 [6] while applications recover, which would mean the rule cleared out shells and left the operators alone.
Ranked by verification strength, evidence, and original report placement.
Nikkei reported on Sept. 2 that its analysis of Immigration Services Agency data showed 953 holders of Japan's "business manager" residency status departed the country between January and June 2026 without filing for re-entry permission, 3.9 times the year-earlier figure.
The Immigration Services Agency revised the Justice Ministry ordinance setting landing criteria on Oct. 16, 2025, sharply tightening the requirements for obtaining the business manager status.
Previously, applicants needed to meet only one of two conditions: capital of at least 5 million yen (about 42.93 million won) or at least two full-time employees.
Under the revision, capital of at least 30 million yen became a mandatory condition rather than one of two options.
Additional requirements include three or more years of management experience or a master's degree or higher, employment of at least one full-time worker, Japanese-language ability equivalent to JLPT level N2, and verification of the business plan by a certified small-business consultant, tax accountant or certified public accountant; using one's home as an office is no longer accepted.
Departures of business manager status holders had run at a few dozen a month until just before the revision, jumped to 114 in December 2025, and ranged from 100 to 200 a month from January through June 2026.
Distinct publishers with included, body-backed reporting in this cluster.
en.sedaily.com
1 article · September 2, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
product
Japan's instant settlement plan starts by tokenising the Bank of Japan, not the bonds1 distinct publisher
build
China is not banning germanium to Taiwan. It is just not clearing it.1 distinct publisher
invest
Seven-Eleven and FamilyMart Are Merging Their ATM Fleets Because Japan Is Shrinking1 distinct publisher
invest
Tokyo's 3.8% assumption: debt service eats two-thirds of Japan's budget increase1 distinct publisher
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Countable numbers, all borrowed
Four named sources carry this story and en.sedaily.com holds none of them itself: Nikkei's read of Immigration Services Agency departure records, Tokyo Shoko Research's 299-company survey and its 2024 incorporation tally, the agency's own defence of the threshold, and Hiraguchi's testimony in the House of Councillors. That is unusually checkable provenance for a single-outlet report — dates, sample sizes, counts. What keeps it out of the high seventies is that nobody re-derived any of it, and one seam shows: the itemised survey responses cover 152 companies while only about 135 said they were affected.
The rule has already bitten, and it is counted
Policies usually reach us as intentions; this one arrives with a behavioural record. Monthly departures moved from a few dozen to 114 within weeks of the October revision and held at 100 to 200 through June; applications fell to roughly a twenty-fifth of their prior rate; 953 people gave up their re-entry permission, which is the paperwork equivalent of not coming back. Against that, the survey's closure figure is still an intention — 16 firms considering it, not 16 shutters down — and no one has counted how many owners cleared the new bar and stayed.
Timing is doing the work of causation
The arithmetic in the headline is honest — 30 million yen really does sit above 98.96 percent of Japan's 2024 incorporations. The stretch is the join between rule and exodus. Departures rose after October 16 and the reporting says the timing 'lines up', but not one departing owner is identified or quoted, and the agency's grace period through 2028 means nobody was formally forced out during the period measured. Add the survey's 'considering closing' being read as closures-in-waiting, and the frame runs a little ahead of what has actually been demonstrated.
Everyone framing this has a stake, and each is named
The agency needs the change to look like anti-abuse housekeeping, so it supplies both the abuse anecdote and the flattering foreign comparison — while omitting that its new bar is still below Korea's. Tokyo Shoko Research sells company credit data; its survey and its section chief's warning about wholesalers and landlords put its own product at the centre of the harm story. And a Korean outlet converting every threshold into won is writing for readers who might have been the applicants. None of this is hidden, which is why the reading is moderate rather than severe.
Firm on facts, single on telling
I would repeat the thresholds, the dates and the 953 without hesitation; each traces to an identifiable original that a second newsroom could pull. What I would not do is treat the causal story as settled, or the survey percentages as internally clean, on one publisher's account. The gap between those two levels of trust is the whole reading here.