1. Five-second answer: the pay ladder is real, but “foreign = legally cheap” is not
Japan’s 2025 Basic Survey on Wage Structure reports average scheduled monthly cash earnings—excluding overtime premiums—of ¥190,300 for technical interns, ¥221,400 for Specified Skilled Workers (SSW), and ¥313,200 for the broader professional/technical group excluding SSW.[1]
A separate 2024 survey can isolate the Engineer/Specialist in Humanities/International Services status—commonly shortened here as Gijinkoku. Its average scheduled monthly pay was ¥278,300, while regular monthly cash earnings including overtime-type payments were ¥311,200.[2]
The easy mnemonic is therefore:
Technical Intern ¥190k → SSW ¥221k → Gijinkoku about ¥278k.
That looks like a perfect three-stage rocket. Legally, however, all three systems contain Japanese-comparator pay requirements. “They are foreign, so minimum wage is enough” is not the rule.[3][4][5]
2. These are not three versions of the same foreign-worker category
Technical Intern Training was designed around skills transfer. SSW is a labor status for designated shortage sectors requiring a defined skill level. Gijinkoku covers professional, technical, humanities, and international-service work.
So the gap is not simply “nationality pricing.” The groups differ in jobs, skills, experience, industries, employer size, geography and tenure. Comparing them without those differences is like averaging the price of sushi, canned tuna and a whole frozen fish because all three live in the seafood aisle.
3. Average monthly pay: scheduled earnings are the cleanest starting point
| Group | Survey year | Average scheduled monthly pay |
|---|---|---|
| Technical Intern | 2025 | ¥190,300 |
| Specified Skilled Worker | 2025 | ¥221,400 |
| Professional/technical excluding SSW | 2025 | ¥313,200 |
| Gijinkoku only | 2024 | ¥278,300 |
Do not rename ¥313,200 “the Gijinkoku average.” That broader category also contains other professional statuses and can be pulled upward by higher-paid workers. For the specific Gijinkoku status, ¥278,300 is the cleaner figure available from the 2024 survey.
The common intuition that Gijinkoku pay feels “Japanese white-collar / graduate-hire-ish” is directionally reasonable, but the survey population is not composed only of new graduates. Treat it as a young-professional pay zone, not as a literal starting-salary statistic.
4. Medians: averages can cast payroll illusions
For SSW general workers in the 2024 Survey on Employment Conditions of Foreign Workers, scheduled monthly pay averaged ¥213,000 and had a median of ¥206,000. Regular cash earnings including overtime averaged ¥250,300 and had a median of ¥253,000. Annual bonuses and special payments averaged ¥107,700—but the median was zero.[6]
That last line matters. If you look only at the average bonus, you may imagine “roughly ¥100k for everyone.” Then median zero quietly walks into the room holding a baseball bat.
In the 2023 Wage Structure survey, scheduled-pay medians were ¥175,400 for technical interns, ¥190,100 for SSW, and ¥250,300 for the broader professional/technical group excluding SSW.[7]
The public tables do not always provide the same granularity in the same year. A broad professional/technical median is not an exact Gijinkoku median. We therefore leave missing comparisons missing instead of manufacturing precision.
5. Industry differences: construction pays much more than manufacturing within SSW
2025 SSW scheduled monthly pay by industry includes:[1][8]
| Industry | Scheduled monthly pay | Approx. workers represented |
|---|---|---|
| Construction | ¥265,900 | 21,750 |
| Health and welfare | ¥225,000 | 31,550 |
| Accommodation and food service | ¥218,300 | 9,110 |
| Wholesale and retail | ¥213,900 | 5,770 |
| Manufacturing | ¥207,300 | 72,660 |
| Real estate / goods rental | ¥224,300 | 550 |
Among the larger categories alone, construction exceeds manufacturing by ¥58,600 per month, or roughly ¥700,000 a year on a simple twelve-month basis.
There are lower published rows, such as transportation/postal services, but the represented worker counts are tiny relative to manufacturing or healthcare. Turning those into a dramatic “worst industry ranking” is how sample-size police get called.
The survey also excludes agriculture, forestry and fisheries, so it cannot directly show SSW agriculture or fishing wages.[1]
6. Annual pay: “monthly × 12” and realistic total cash are different animals
Simple scheduled-pay annualization gives approximately:
| Group | Monthly | Simple ×12 |
|---|---|---|
| Technical Intern | ¥190,300 | ¥2.284m |
| SSW | ¥221,400 | ¥2.657m |
| Broad professional/technical | ¥313,200 | ¥3.758m |
| Gijinkoku only (2024) | ¥278,300 | ¥3.340m |
Using the 2024 survey’s regular monthly cash earnings including overtime-type pay, multiplied by 12, plus reported annual bonuses gives a rough total-cash proxy:[2]
| Group | Monthly cash | Annual bonus | Proxy annual total |
|---|---|---|---|
| Technical Intern | ¥210,000 | ¥39,400 | ¥2.559m |
| SSW | ¥250,300 | ¥107,700 | ¥3.111m |
| Gijinkoku | ¥311,200 | ¥357,300 | ¥4.092m |
A useful memory aid is ¥2.6m → ¥3.1m → ¥4.1m.
But this is a proxy, not a payroll ledger: the monthly figure comes from a survey month while the bonus figure covers the previous year. Do not present it as an exact realized annual salary.
7. “Technical intern = minimum wage” is legally wrong, even if the market impression is understandable
Japan’s Technical Intern Training Act requires pay at least equivalent to what a Japanese worker would receive for comparable work and skill.[3] SSW rules similarly require pay equal to or above that of a Japanese worker doing equivalent work.[4] Gijinkoku also has a Japanese-equivalent-or-higher remuneration standard.[5]
So the legal rule is not “foreigner discount.”
Why, then, is technical-intern pay so low? Because the comparison group itself is concentrated in younger, lower-paid, on-site occupations and industries, with shorter tenure and different employer profiles. “Equal to a comparable Japanese worker” is not the same as “equal to the average Japanese worker nationwide.”
8. Remittances make the weight of ¥1,000 look different
The 2024 foreign-worker survey found the share sending money to family abroad was 45.1% for Gijinkoku, 81.6% for SSW and 83.5% for technical interns. Among remitters, average annual amounts were ¥1.153m, ¥1.233m and ¥1.063m respectively.[9]
This does not mean every worker remits that amount, and it creates no moral obligation for friends to subsidize each other. It simply reminds us that the same ¥1,000 restaurant charge can have a different household-budget weight.
A thousand yen is physically the same thousand yen; its CPU load on a household budget may not be.
9. Does refusing to treat a lower-paid foreign friend make you terrible? No—do not install a finance ministry inside friendship
If you know a friend probably earns less, it is normal to wonder whether strict splitting feels stingy. But income inequality does not automatically create a private payment obligation.
- Splitting: normal.
- Paying this time because you want to: kind.
- Letting them treat you: also normal if they genuinely want to.
- Paying every time: generous, not mandatory.
The healthier motive is not “you are low-paid, therefore I must pay for you.” It is “I want to get this one.”
10. A treat turns from affection into a weird tax the moment it becomes compulsory
“Dinner was fun; I’ve got this” is a gift.
“If I do not pay every time because their income is lower, I am a bad person” is fiscal policy.
Friendship does not need a progressive taxation bureau. Nor does every meal need double-entry bookkeeping.
If you enjoy being treated, remember the other person may enjoy it too. Some people feel guilty receiving; others happily say “Thanks!” Neither reaction is morally superior.
11. The easiest operating rule: pay when you want to; receive graciously when someone wants to pay
A low-friction rule set looks like this:
- You invite someone to an unusually expensive place: offering to pay more is natural.
- Ordinary meal: splitting is perfectly fine.
- You feel like treating: treat—no economic impact statement required.
- They want to treat you: accept with thanks if they are not being pressured.
- Do not freeze the relationship into one permanent payment rule.
Avoid deploying Friendship ERP with fields such as “unsettled karaage balance: ¥324.”
12. What statistics can and cannot tell us
Statistics clearly show a wage hierarchy across residence-status groups and substantial industry differences within SSW. They also show that the legal framework does not authorize simply paying less because a worker is foreign.
Statistics cannot tell you one friend’s actual disposable income. Housing deductions, overtime, savings goals, remittances and family obligations differ enormously.
Statistics are a map for understanding people, not the person standing in front of you.
13. Data warning: do not mix mean, median, overtime and bonus into one salary smoothie
- Scheduled cash earnings exclude overtime premiums.
- Regular cash earnings include overtime-type pay.
- Annual bonuses may refer to a different reporting period.
- Means can be pulled by high earners.
- Medians depend on the exact population and table definition.
- Tiny industry samples are unstable.
- The 2025 Wage Structure survey does not cover agriculture, forestry or fisheries.
¥278k scheduled pay, ¥311k total monthly cash and a ¥4.09m annual proxy are not interchangeable numbers.
14. Bottom line: pay is a three-stage rocket; friendship does not need rocket equations
A practical wage picture is:
Technical Intern about ¥190k/month → SSW about ¥221k → Gijinkoku about ¥278k in scheduled pay.
A rough total-cash annual proxy is:
about ¥2.56m → ¥3.11m → ¥4.09m.
The intuition “technical interns are near the low-wage end; Gijinkoku is closer to Japanese young professional pay” is broadly compatible with the data, but “foreigner = legally cheaper” is not.
And none of that creates a rule that the higher earner must always pay for dinner.
Use statistics for wages. Use your feelings for treats.
References (9)
- 厚生労働省 / MHLW — 令和7（2025）年 賃金構造基本統計調査, foreign workers. 2025 scheduled-pay averages: technical intern ¥190.3k; SSW ¥221.4k; professional/technical excluding SSW ¥313.2k. e-Stat Table 1 also provides SSW industry rows. https://www.e-stat.go.jp/stat-search/files?cycle=0&data=1&layout=datalist&metadata=1&page=1&tclass1=000001229845&tclass2=000001229849&tclass3=000001229862&toukei=00450091&tstat=000001011429&year=20250 mhlw.go.jp
- 厚生労働省 / MHLW — 令和6（2024）年 外国人雇用実態調査, 概況・第4表. General workers: Gijinkoku cash/scheduled/annual special payments = ¥311.2k / ¥278.3k / ¥357.3k; SSW = ¥250.3k / ¥213.0k / ¥107.7k; Technical Intern = ¥210.0k / ¥177.6k / ¥39.4k. https://www.mhlw.go.jp/content/11655000/001552246.pdf mhlw.go.jp
- 厚生労働省 / MHLW — 技能実習法第9条 and official implementation material. Technical-intern remuneration must be equal to or greater than remuneration for a Japanese worker performing comparable work. https://www.mhlw.go.jp/content/001089480.pdf mhlw.go.jp
- 出入国在留管理庁 / Immigration Services Agency — 特定技能制度Q&A, Q26. SSW pay must be equal to or higher than that for Japanese employees engaged in equivalent work moj.go.jp
- 出入国在留管理庁 / Immigration Services Agency — 技術・人文知識・国際業務 status criteria. Remuneration must be equal to or greater than that received by a Japanese national engaged in the comparable activity moj.go.jp
- e-Stat / MHLW — 令和6（2024）年 外国人雇用実態調査 統計表, establishment survey Table 4. For general-worker SSW: regular cash average/median ¥250.3k/¥253.0k; scheduled pay ¥213.0k/¥206.0k; annual bonus/special payments ¥107.7k/¥0 e-stat.go.jp
- e-Stat-derived Wage Structure table — 2023 distribution statistics. Scheduled-pay medians: technical intern ¥175.4k; SSW ¥190.1k; professional/technical excluding SSW ¥250.3k. The last figure is a broad category and is not an exact Gijinkoku-only median e-stat.go.jp
- e-Stat — 令和7（2025）年 賃金構造基本統計調査, 外国人労働者 第1表. SSW scheduled pay by industry: construction 265.9k; health/welfare 225.0k; accommodation/food 218.3k; wholesale/retail 213.9k; manufacturing 207.3k; real-estate/rental 224.3k; plus worker-count estimates. Agriculture/forestry/fisheries are outside this survey’s industry scope e-stat.go.jp
- 厚生労働省 / MHLW — 令和6（2024）年 外国人雇用実態調査, worker survey remittances. Share remitting: Gijinkoku 45.1%, SSW 81.6%, Technical Intern 83.5%. Mean annual amount among reported remitters: ¥1.153m, ¥1.233m, ¥1.063m respectively. https://www.mhlw.go.jp/content/11650000/001551388.pdf mhlw.go.jp
