Japan by the Numbers: Check Who Was Counted

Somebody puts Japan's job-to-applicant ratio on a slide: 1.18 openings per applicant. It sounds like a tight market. For IT roles the figure is 1.33. For office work it is 0.29. For construction it is 8.25.

Most widely quoted Japan numbers are correct. They just count a different group than the reader assumes. I wrote this column for executives who make Japan decisions from data, and for anyone who has been handed a Japan statistic and told to act on it. Before you use a number, check the denominator, who reported it and what scope it covers. Once you do, the decision usually changes.

Can I find the people I need, and where?

It depends on the occupation and the place far more than on the national average. Tokyo's openings ratio is 1.70 counted by place of registration and 1.07 counted by place of work. Foreign workers now number 2.57 million, five times the 2008 level.

• Japan Labor Market Report 2026: unemployment at 2.5%, wage deals above 5%, a ¥983.5 billion placement market and AI changing how hiring works.
• Japan's job-to-applicant ratio only makes sense by occupation, from construction at 8.25 to office work at 0.29.
• Tokyo vs the Regions: Japan's Job Openings Ratio: why the workplace-based figure is the one to use when choosing a regional site.
• Foreign workers in Japan: who the 2.57 million are, and where. Vietnam leads with 606,000; China has 432,000 and still tops white-collar visas.
• How Good Is English in Japan, Really? Japan ranks 96th on the EF index, yet a quarter of TOEIC takers score 745 or more. Decide which language is mandatory first.
• Is LinkedIn Useful in Japan? What the Numbers Say: about 3 million members, small against a 70 million labor force, but dense in foreign firms, tech and bilingual roles.
• Japan's 98% graduate employment rate: check the denominator. It counts only job seekers; of all graduates, 77% got jobs.
• Japan's 3-Year Graduate Turnover, Taken Apart: about 3 in 10 leave within three years, and the real spread is by industry and company size.
• Record Tourists, Short-Staffed Hotels: Japan in Numbers: Japan drew a record 42.68 million visitors in 2025 and 36.6% of their spending went on lodging, yet service jobs had 2.31 openings per applicant in August 2026.
• Japan's Labor-Shortage Bankruptcies Hit 427: Who Fell: Teikoku Databank counted 427 in 2025, a third straight record. 77% had fewer than ten staff, and bankruptcies caused by resignations grew fastest.
• Japan Learns Least Outside Work: 52.6% Do Nothing at All: 52.6% of Japanese workers do nothing to learn outside work, the highest of 18 countries. Why the pay-off structure explains it, and what a company in Japan has to design in.
• Three in Four IT Engineers in Japan Work for a Vendor: 73.6% of Japan's IT engineers work for IT companies and only 26.4% inside user companies, the reverse of the US, which is why in-house teams are hard to staff.
• Japan Lost 13.5 Million Working-Age People in 30 Years: Japan's 15 to 64 population is down 13.5 million since 1995 and employment still rose, and the next decade cannot repeat that trick.
• Half of Japanese Aged 65 to 69 Work. Only Korea Comes Close: Japan's labor ministry's own survey of other countries, used to place Japan's 54.9% participation at 65 to 69 and its employer obligations to 65 and 70 in context.
• 90% Feared Japanese. The Real Gap Was Hou-ren-sou: In a 22-company survey, 90% worried about Japanese before hiring Vietnamese engineers, but only 26% found it lacking afterwards; work habits ranked higher.
• Only 20% in Japan Wanted to Manage, Last of 18 in 2022: Only 19.8% of non-managers in Japan wanted a management post, the lowest of 18 markets in Persol's 2022 survey. What that small pool means for employers promoting or hiring middle managers.
• Why Japan Hires Mid-Career: Experience Tops Two Job Types: In Japan's 2025 Survey on Job Changers, ready-on-day-one experience is the top hiring reason only for managerial and professional jobs. For the other four job groups, the top reason is replacing someone who left.
• Under 40% of Foreign Students Stay to Work in Japan: Only 38.8% of international students who graduate in Japan take a job there. About 40,000 status changes are approved a year, mostly into small firms at new-graduate pay.
• 941,000 Chinese Live in Japan. How Many Can You Hire?: Japan's 941,000 Chinese residents are nearly 40% permanent residents. About 432,000 are employees, and the realistic ceiling for full-time hiring is around 316,000 people.

Will people actually move?

Fewer than you would think. In 2025, 10.23 million people in Japan wanted to change jobs and 3.3 million did. The men's job-change rate went from 3.0% in 1990 to 3.8% in 2022. Over three decades, it moved less than one point.

• Japan's job-change rate moved two points in thirty years, and men aged 45 to 54 never passed 2.6%.
• 10.23 Million Want to Change Jobs in Japan. 3.3 Million Did. A record number want to move, yet 65% of them are not even searching.
• How Long Do People in Japan Stay at One Company? The 12.4-year average is a mean; the median is 5 to 9 years. Order your outreach by tenure band.
• Resignation proxies in Japan: is it 2% or nearly half? Estimates run from 2% to 48.7% depending on who was asked.
• Japan's Resignation Proxies: One in Four Went Quiet. Operators are closing while the share of companies receiving proxy resignations rose to 8.7%.
• How People in Japan Actually Find Jobs: Recruiters Are 9%: Only 9.3% of Japan's job changers in 2025 came through recruiters. Ads, personal connections and Hello Work do most of the work, and the mix shifts with company size and occupation.
• Changing Jobs in Japan: 40% Got a Raise, 31% Took a Cut: In 2025, 40.8% of Japan's job changers earned more and 31.0% earned less. The age where raises turn into cuts, and how candidates and hiring managers should each use the number.
• Six in Ten or Nearly Nine: Japan's Post-Hire Gap Surveys: Depending on the question, six to nine in ten job changers in Japan report a gap after joining, yet one in ten is dissatisfied. What employers say in interviews should be treated as claims.
• Only 22% of Japan's New Hires Plan to Stay Until Retirement: Three months into their first job, 22.4% of Japan's new graduates plan to stay until retirement and 21.1% expect to leave within three years, which changes how a graduate role should be designed.

What does a person cost, and how is pay set?

Plan on 1.2 to 1.5 times salary for one hire. Pay itself has been nearly flat for thirty years: average pay rose from ¥111,000 in 1949 to ¥4.67 million in 1997 and then stalled, with real pay peaking in 1992.

• What one hire in Japan really costs: 1.2x to 1.5x salary: a ¥6 million hire, costed line by line with and without an agency.
• Japan's wages: up forty times, then flat for thirty years, and what that history means when you set pay today.
• Japan's job-based pay: how many companies really do it. 78.5% say managers get role or job pay; 26.4% have or plan job-based HR; 2.4% of those cover all staff.
• Overtime in Japan: 13 hours a month, or 21? The official figure, the workers' figure and the unpaid hours in between.
• Japan's M-Curve Is Gone. The L-Curve Isn't. Regular employment for women falls from 62.8% in their late twenties to about 40% in their forties.
• 100,000 Centenarians: Will Working at 70 Become Normal? 36.2% of people aged 70 to 74 worked in 2025, mostly part-time or self-employed.
• A third of ex-managers in Japan stepped down unexpectedly: 35.3% in a 2026 Persol survey, 81.0% of them after reorganizations.
• Japan sees 3,400 labor tribunal cases a year. That's not small. Half are over dismissals, and the article prices what one dispute really costs.
• Japan's Labor Inspectors Find Violations at 70% of Visits: Japan's labor inspectors find violations at 70% of the 142,000 workplaces they visit each year. Who gets visited, how targets are chosen, and what the rate does and does not mean.
• Unpaid Wages in Japan: ¥12.9 Billion Found in One Year: Japan's inspectors handled ¥12.86 billion in unpaid wages in 2025. Why the series changed in 2022, why claims arrive three years at a time, and the three mistakes that build them.
• Japan's Union Rate Is 16%. Why the Shunto Still Sets Pay: What Japan's 16.0% union density counts and leaves out, who the members are by company size, and the three routes by which shunto results still reach companies with no union.
• How Far Behind US Pay Are Japan's Software Engineers?: Japan's software developers earn a quarter of the US median at market rates and under half at PPP, and the gap is widest at the top, which shapes both hiring and job moves.
• Japan's Labor Share: 47% at Big Firms, 82% at the Smallest: Japan's 64.2% labor share taken apart by definition and company size, from 47% at large firms to 82% at the smallest, with the benchmark a Japan subsidiary should actually use.
• Japan Job Ad Pay Averages ¥5.13M. Experience Adds ¥1.15M: Japan's record ¥5.13M average for job ads is an asking price; ads requiring experience pay ¥1.15M more, and part of the rise is a shift in mix.
• 10, 50, 101: Japan's Headcount Lines for Employers: Japanese labor duties switch on at 10, 37.5, 50, 100, 101 and 301 people; whether a line counts per site or per company decides whether you have crossed it.
• Why Japan's Recruiter Fees Run 30 to 35% of Salary: Japan's 30 to 35% recruiter fee is a filed price, not a legal rate; the capped fee is about 7% of salary and the FY2024 average was ¥1.03 million.
• Japan's Section Chiefs Average 49.5, Promoted at 38.6: Japan's sitting section chiefs average 49.5 and were promoted at 38.6, ten years later than in China. Why reading Japanese candidates by age and title misgrades them in both directions.
• A Promotion to Kacho in Japan Adds 26% in Monthly Pay: With overtime counted, a promotion from subsection chief to section chief in Japan adds 26% to monthly pay, 18% for engineers. Why importing head-office pay multiples misprices both managers and specialists.
• Cost per Hire in Japan: About 1 Million Yen, Seven Years Old: Japan's widely quoted cost per hire, about 1 million yen, dates from fiscal 2019. This piece sets it beside 2025 spending data and counts what is sunk when a hire goes wrong.
• Foreign Workers in Japan Earn 25% Below the Overall Average: Foreign workers in Japan average 25% less than all full-time workers, but professional-status foreigners match their age group. Most of the gap reads as visa mix, age and tenure.
• Retirement Pay in Japan: Leave at 15 Years, Get a Fifth: At large Japanese companies, someone who leaves after 15 years gets about a fifth of the full model retirement allowance, so a mid-career hire leaves the steep back half of the curve behind.

Who are you selling to?

Mostly to small companies run by older owners. Japan has about 10,000 large companies and 2.85 million micro ones, the average president is 60.8, and half of companies have no successor.

• SMEs are 99% of Japanese companies. Decide who you're selling to first: pick the layer, then price the cost of reaching each company in it.
• The average company president is over sixty. What that means if you're selling. A decision-chain problem, with succession as the window.
• The country with the most century-old companies: why newcomers can't move customers: half the world's century-old companies are here, and they hold the channels. Borrow them.
• Eighty percent of Japanese companies use the cloud. For what? Seventy percent just store files, which limits what the market can absorb from you.
• A good market that is hard to enter: how low Japan's inward FDI is: 5 to 9% of GDP, the lowest in the G7, and step one is a map.
• Japan's Convenience Store Count Has Barely Moved Since 2019: 55,620 stores at end-2019 and 56,054 at end-2025, up 0.8%. Sales still rose 8%, because spend per visit rose 15.4% while visits fell 6.4%.
• Japan Trails on Generative AI Policy, Mostly by Not Deciding: in Japan's 2026 ICT White Paper, 68.9% of Japanese firms have a generative AI policy against 82.8% in the US. Only 3.5% ban it and 20.2% haven't decided, so for anyone selling AI tools the main rival is indecision.
• Japan's no-successor bankruptcies hit 264 in six months: Tokyo Shoko Research counted a record 264 in the first half of 2026. In 85.6% of them the president died or fell ill, which matters if you plan to buy a Japanese company.
• Japan's Outward FDI Is 6.6 Times Its Inward FDI: at end-2025 Japan held ¥384.5 trillion of direct investment abroad, against ¥58.6 trillion held by foreign firms in Japan. What that gap means when you hire here.
• Chinese Mobile Games in Japan: 16% of Overseas Revenue: Japan gave Chinese mobile games 16.35% of their overseas revenue in 2025, about $3 billion by my estimate, while BYD sold 2,223 cars there in 2024. Distribution explains the difference.
• How many Chinese companies are in Japan? 317, or 25,000: Chinese companies in Japan number 317, 3,146 or 24,840 depending on who is counted, so check the ruler before you draw them on a market or talent map.
• Japan's Business Manager Visa: ¥30 Million, One Local Hire: Japan's Business Manager visa now needs ¥30 million, one Japanese or permanent-resident employee and B2 Japanese. The money is the easy part; the local hire is the gate.

Can I trust this number?

Often only after you have checked the scope. One Hitachi securities report lists four female manager ratios, 17.4%, 22.7%, 8.9% and 6.8%, all correct for different scopes. Only 1.8% of Japanese firms published their accounts in the official gazette in 2022.

• Why you can't find most Japanese companies' financials: mark the gap as not disclosed and use three legal proxies.
• How to read human capital disclosure in Japanese annual reports. Six numbers, including the pay gap and men's parental leave, that sketch a listed company's workforce.
• One Company, Four Female Manager Ratios: align the scope before you compare companies.
• In Japan, "neither agree nor disagree" is an answer. Where a stance is expected, neutral reads as negative.
• Psychology Studies Managers Quote: Only 36% Replicated: where Stanford Prison and the marshmallow test stand, and how HR should check a claim.
• Does Monitoring Raise Performance? Data on 23,461 People. Across 94 samples, monitoring and performance correlate at about zero while stress rises.
• Muir's Super-Chicken Study: What the Papers Actually Say: the mortality figures from 1996, and where the popular TED version drifts.
• Terracotta Army fingerprints: who did the work? 79 fingerprints, an estimate of who left them, and what it says about credit at work.
• Japan's Crime Fell 30% in a Decade. Why 80% Say It's Worse: Recorded crime in Japan fell 30% in a decade, yet 80% say safety got worse. A worked example of checking what each official number actually counts.
• Japan's 2024 Logistics Crisis: The 14% Gap Was a Projection: the 14% trucking gap was an NX Soken estimate built on FY2019 freight and a 3,300-hour cap. Demand fell 12%, the gap largely closed, and rates kept rising.
• Did China's Robot Density Really Overtake Japan's?: the IFR put China at 470 robots per 10,000 factory workers for 2023. With fuller employment data its 2024 figure is 166, against Japan's 446. China leads on volume.
• Japan's 790,000 IT Worker Shortage: Where It Comes From: METI's 2019 study put the 2030 IT talent gap anywhere from 164,000 to 787,000, depending on assumptions. The famous figure is the worst of six scenarios.
• Japan's 570,000 Care Worker Gap by 2040 Is a Projection: MHLW estimates Japan needs about 2.72 million care workers by FY2040, 570,000 more than in FY2022. The measured count fell for the first time in 2023, to about 2.126 million.
• China's FDI into Japan halved in 2025: check the dates: direct investment into Japan from mainland China fell 52.5% in 2025, but most of that year came before the deterioration in relations reported from November 2025. Check the dates before reading it as a reaction.
• Japan Works Fewer Hours Than the US. Only on Average: Japan's 1,617 annual hours are below America's 1,796, but the average mixes part-timers in. Full-timers work 1,926, and 2.5 million people still do 60-hour weeks.

A habit for any Japan statistic

When a number lands on your desk, ask who was counted, who answered, and whether a company's own figure has an independent source behind it. The graduate employment rate, the overtime figures and the resignation-proxy surveys all change meaning on those questions. Government series move slowly, so some figures here may be a year old by the time you read them.

The rule I use for company figures is in the glossary under claim vs. fact. To see how these numbers feed actual decisions, read Launching in Japan.