A Korean first-grader’s family should budget about ₩433,000 a month for private lessons on the all-student average, or ₩512,000 once the child joins the 84.4% of elementary students who take any. The ₩604,000 in the headlines is neither: it averages lesson-takers across every school level.
Our daughter was born in 2020 and starts first grade in March 2027. I build internal systems at a Korean financial company, my wife is an ENFP and I am an ISTJ, so in our house I am the one who opens the spreadsheet. What sent me there was a proposal, not a price. On Sept. 3, 2026, a National Education Commission forum discussed adding school hours for grades 1–2, and the Korea Times headlined it with a question: would longer school days keep kids out of private academies? I realized I could not say what those academies cost a family like ours.
My post on Neulbom School (늘봄학교, Neulbom hakgyo, the state’s after-school care program) covered the free side, the block that runs until 3 p.m. in grades 1–2. This is the sequel: what happens after 3 p.m. The numbers come from the 2025 private education survey (사교육비조사, sagyoyukbi josa, “private education spending survey”), released March 12, 2026 by the Ministry of Data and Statistics and the Ministry of Education, covering about 74,000 students at about 3,000 schools. One caveat up front: every elementary figure below averages grades 1 through 6, and I found no grade-1 line in the release that I could verify.
Which Average Is Which: ₩433K, ₩512K or ₩604K?
The survey defines private education (사교육, sagyoyuk) as supplementary lessons a family pays for outside the school curriculum. That covers hagwon (학원, hagwon, a private after-school academy), one-to-one and group tutoring, and internet or distance lessons, so the figures below are wider than academy fees alone. Three averages circulate for the same child, and each divides by a different group.
| Number | What it averages | Use it for |
|---|---|---|
| 84.4% | Share of elementary students taking any private lessons | How normal it is |
| ₩433,000 | All elementary students, including those who spend nothing | A family that has not decided |
| ₩512,000 | Only elementary students who take lessons | A family that has joined in |
| ₩604,000 | Only lesson-takers, but across elementary, middle and high school | Nothing for a first-grader |
| ₩458,000 | All students at all school levels | The average that fell 3.5% |
Of the two elementary figures, ₩512,000 is the only one that describes families who actually pay, though it is still an average, not any one family’s bill. ₩433,000 blends in the 15.6% of elementary students (about one in six) who take no lessons, so it is the fair planning number before a decision is made but nobody’s actual bill. I checked that the two describe the same children: ₩512,000 × 84.4% comes to ₩432,100, within ₩900 of the published ₩433,000. The same multiplication reproduces middle school (₩632,000 × 73.0% ≈ ₩461,400 against the published ₩461,000) and the all-level headline pair (₩604,000 × 75.7% ≈ ₩457,200 against ₩458,000). That is my arithmetic, not an official figure. Over twelve months, ₩433,000 is ₩5.20 million and ₩512,000 is ₩6.14 million.
The same identity settles the headline puzzle. Participation fell 4.3 points to 75.7%, so the all-student average dropped 3.5% and total spending fell 5.7% to ₩27.5 trillion, while the students who still take lessons pay 2.0% more, a record ₩604,000 (Xinhua converted that to about $410, and ₩458,000 to about $310). Elementary school splits the same way: ₩433,000 (−2.1%) for everyone, ₩512,000 (+1.7%) for lesson-takers, with elementary spending down 7.9% to ₩12.2 trillion.
What Does a First-Grader’s Afternoon Look Like After 3 p.m.?
Regular classes for a first-grader end around 1 p.m., though the time varies by school and weekday. From then until 3 p.m. the free block applies to grades 1–2. The private window opens at 3 p.m., and by the numbers most children walk into it.
| Time | What happens | Who pays |
|---|---|---|
| Until about 1 p.m. | Regular classes | The state |
| About 1–3 p.m. | Free Neulbom block, grades 1–2 | The state |
| After 3 p.m. | Private lessons: elementary figure of 7.4 hours a week (basis unclear) | The family |
Free and paid do not replace each other. 52.2% of elementary students use school-run programs (Neulbom plus the school’s own 방과후학교, banggwahu hakgyo, “after-school classes”, free or paid), against 84.4% using private lessons. Two groups that add up to 136.6% of the class must overlap in at least 36.6% of students, so by simple arithmetic at least about one child in three uses both. School programs also peak in elementary school, at 52.2% against 17.2% in middle school and 29.5% in high school. And they are small money: ₩779.9 billion in total across all school levels in 2025 (−2%), against ₩12.2 trillion for elementary private lessons alone, about 16 times as much.
Elementary school also leads on time. Its 7.4 weekly hours of private lessons beat middle school’s 7.2 and high school’s 6.6. Spread over five weekdays that is about 1.5 hours a day for whichever group the average covers (my division, and real weeks are lumpy). The release, in the passages I read, does not say whether that group is lesson-takers or all students, so I do not turn the hours into a price per hour.
What Do 12 Years of Private Lessons Add Up To?
The monthly figure climbs with each school level: ₩512,000 in elementary school, ₩632,000 in middle school and ₩793,000 in high school, all per lesson-taking student. I multiplied each by 12 months and by the years in the stage, six of elementary, three of middle and three of high school. A child who takes lessons all twelve years comes to about ₩88.2 million. On the all-student averages (₩433,000, ₩461,000, ₩499,000) the same arithmetic gives about ₩65.7 million.
| Stage | Years | Lesson-taker monthly | Stage total | All-student monthly | Stage total |
|---|---|---|---|---|---|
| Elementary | 6 | ₩512,000 | 36.86 | ₩433,000 | 31.18 |
| Middle | 3 | ₩632,000 | 22.75 | ₩461,000 | 16.60 |
| High | 3 | ₩793,000 | 28.55 | ₩499,000 | 17.96 |
| 12 years | 12 | 88.16 | 65.74 |
The step-ups are +₩120,000 a month from elementary to middle school, then +₩161,000 into high school. Elementary school is half the years but only 42% of the lesson-taker total (₩36.9 million of ₩88.2 million). Treat it as illustrative. No inflation, no skipped months, and no real child follows an average. Only the elementary row is a budget I can act on before March.
How Far Do Income and Region Move the Number?
This is where a national average stops being useful. Among all students at all school levels, households earning ₩8 million or more a month spend ₩662,000, against ₩192,000 for households under ₩3 million, about 3.4 times. By region, the all-student average is ₩663,000 in Seoul and ₩499,000 in Gyeonggi, against ₩458,000 nationally. Seoul runs about 45% above the national figure and Gyeonggi about 9%.
If you are in Seoul, I would read that as a hint that elementary spending sits above ₩433,000, and treat ₩433,000 as a floor rather than a midpoint. Those are all-level figures, and multiplying an elementary number by a Seoul-versus-nation ratio would create a figure that appears in no table. A short list of what I keep apart:
- Lesson-takers versus all students. ₩512,000 and ₩433,000 are the same elementary students on two denominators, an 18% gap.
- All levels versus elementary. ₩604,000 and ₩458,000 span elementary, middle and high school. The elementary pair is ₩512,000 and ₩433,000.
- Subject figures are all-level, and the basis flips them. English is ₩281,000 among students who take it but ₩131,000 averaged over everyone; math is ₩270,000 against ₩128,000. Neither is an elementary price.
- Income and region figures are all-level and all-student. Neither ₩662,000 nor ₩663,000 is an elementary number, however alike they look.
Will the 3 p.m. Debate Change a March 2027 First Grade?
The Sept. 3 forum was run by the National Education Commission with the National Assembly Futures Institute and the Korean Educational Development Institute (KEDI) at the National Assembly Library, as input to the 2028–2037 National Education Development Plan. Its starting point is that grades 1–2 get 581 instruction hours a year, roughly 28–29% below the OECD level for those grades. Across all six elementary grades Korea averages roughly 655 hours against an OECD average of 845. Stage 1 would add 114 periods a year, about 3.3 a week, to bring grades 1–2 up to grade 3–4 levels. Stage 2 would add 216, about 6.3 a week. Teacher groups object that first-graders already struggle with five periods.
Nothing is decided.
Would it help working parents or just push hagwon from 1 p.m. to 4 p.m.? The Korea Times notes that dual-income households use hagwon as childcare, which may be part of why the numbers are stubborn: the state already gives two free hours in grades 1–2, and 84.4% of elementary students still take private lessons. Longer days would have to beat that, and I have seen no data on whether they would. The Ministry of Education’s April 2026 cost plan, as Seoul Economic Daily reported it, goes at the other side. A “one student, one art or sport” program would reach 500 elementary schools in 2027, 1,500 in 2028, 3,000 in 2029 and all roughly 6,000 by 2030, aimed at the arts, sports and hobby lessons that the same report puts at ₩173,000 of the ₩433,000 elementary all-student average, about 40%. It keeps the two free customized hours that hold lower-grade dismissal at 3 p.m., and plans to raise cram-school-law fines from ₩3 million to as much as ₩10 million, which would take an amendment to that law.
For a March 2027 entrant, 500 of roughly 6,000 schools is about one in twelve (my division), and the development plan starts in 2028. I am not budgeting on either one helping us.
What Am I Budgeting Before March?
My plan is the boring one. The free block stays at ₩0, and above it I am adding a line to our household spreadsheet at ₩512,000 a month, ₩6.14 million a year, not at ₩433,000. If she takes lessons, ₩512,000 is what the average family in that position pays. If she does not, the gap becomes surplus rather than a hole. Guessing low would cost ₩79,000 a month, ₩948,000 over a year (my subtraction).
The next data point should land the same month she starts school. The survey has come out between March 7 and March 14 in every year from 2022 to 2026, so the 2026 edition is expected around March 2027, an inference from that history and not a scheduled date. When it does, I will compare its elementary lesson-taker line with my ₩512,000 and move the sheet if they differ. For what comes before first grade, the school enrollment notice for foreign families and English kindergarten costs are covered elsewhere on this blog.
My rule for any number in this corner of Korean life: ask who is in the denominator and which grades are in the numerator before asking whether it is high.
This is a reader’s guide, not financial advice. The 12-year totals, reconciliations, overlap and ratio figures are my own illustrative arithmetic, and the policy proposals are not decided.
Sources: Ministry of Data and Statistics release, Korea Policy Briefing, Xinhua, Korea JoongAng Daily, Korea Times on the Sept. 3 forum, Seoul Economic Daily, Kyunghyang Shinmun.