Eight in ten South Korean students, a record 80.0%, received private education in 2024, pushing household spending to 29.2 trillion won, about $20.1 billion, even as the country’s student population shrank (Statistics Korea, Private Education Expenditures Survey 2024). Korea is the extreme case of a worldwide pattern: the global private tutoring market reached $131.04 billion to $133.8 billion in 2025, depending on the analyst (The Business Research Company; IMARC Group), and forecasts cluster between 7.1% and 10.4% annual growth through 2034 (IMARC Group; Fortune Business Insights). The product being bought has unusually strong evidence behind it: a meta-analysis of 96 randomized trials found tutoring lifts achievement by 0.37 standard deviations, roughly 14 percentile points (Nickow, Oreopoulos, and Quan, NBER Working Paper 27476). This analysis consolidates data from IMARC Group, Statistics Korea, the NCES School Pulse Panel, the Sutton Trust, and 16 other primary sources to map the private tutoring economy in 2026.
TL;DR
- The global private tutoring market was worth $131.04-$133.8 billion in 2025 on the two largest estimates (The Business Research Company; IMARC Group).
- Growth forecasts range from 7.12% to 10.42% CAGR through 2034 (IMARC Group; Fortune Business Insights).
- Offline delivery still produced 76.5% of revenue in 2025; online accounted for 15.8% (IMARC Group).
- Online tutoring services were a $12.06 billion market in 2025, heading to $23.73 billion by 2030 at 14.5% CAGR (Grand View Research).
- The AI tutors market reached $3.55 billion in 2025 (Mordor Intelligence).
- 42% of US public schools offered high-dosage tutoring in spring 2025, up from 37% in 2022 (NCES School Pulse Panel).
- Across 96 randomized trials, tutoring raised achievement by 0.37 standard deviations, about 14 percentile points (Nickow, Oreopoulos, and Quan, NBER).
- Korean households spent a record 29.2 trillion won ($20.1 billion) on private education in 2024, with an 80.0% participation rate (Statistics Korea).
- 29% of UK secondary pupils have had a private tutor, rising to 45% in London (Sutton Trust, Private Tutoring 2026).
- England’s National Tutoring Programme delivered 16.4 million tutoring hours to 1.07 million pupils in its final year (UK Department for Education).
- Typical US tutoring rates run $25-$80 per hour, with test prep starting above $100 (Tutors.com, 2026).
- New Oriental’s non-academic tutoring drew about 918,000 enrollments in a single quarter of fiscal 2025 (company filings).
1. Global Market Size and Growth
Market researchers disagree on the absolute size of private tutoring but agree on its direction. IMARC Group puts the 2025 market at $133.8 billion and The Business Research Company lands within 2% of that at $131.04 billion, while Fortune Business Insights counts a narrower $66.96 billion but assigns it the fastest growth rate of the three. Every major firm projects the market to roughly double within a decade, converging on a $160-$250 billion range by 2034. Asia Pacific is the center of gravity with a 35.5% revenue share, and the US is no laggard: Technavio forecasts $28.85 billion of incremental US growth by 2029.
| Metric | Value | Source |
|---|---|---|
| Global private tutoring market (2025) | $133.8 billion | IMARC Group |
| Global private tutoring market (2025, alternate estimate) | $131.04 billion | The Business Research Company |
| Global market, narrower definition (2025) | $66.96 billion | Fortune Business Insights |
| Forecast for 2034 | $248.4 billion (7.12% CAGR) | IMARC Group |
| Forecast for 2034 (narrower scope) | $160.5 billion (10.42% CAGR) | Fortune Business Insights |
| Forecast for 2030 | $208.94 billion (9.9% CAGR) | The Business Research Company |
| Long-run forecast, 2030 | $154.8 billion (6.8% CAGR from $91.65 billion in 2022) | Grand View Research |
| Asia Pacific revenue share (2025) | 35.5% | IMARC Group |
| US incremental market growth, 2024-2029 | +$28.85 billion (11.1% CAGR) | Technavio |
Context: the two-to-one gap between the IMARC and Fortune Business Insights baselines reflects scope, not error; Fortune Business Insights excludes several supplemental education categories, so treat cross-firm comparisons as a range rather than a single number.
2. Online, Platform, and AI Tutoring
Online is the growth engine but still not the revenue center. Offline tutoring generated 76.5% of global revenue in 2025 against 15.8% for online (IMARC Group), yet Grand View Research and Mordor Intelligence both clock online tutoring growth at 14.5-15.6% a year, roughly double the pace of the overall market. The newest layer, AI tutors, was already a $3.55 billion market in 2025 (Mordor Intelligence), though forecasts diverge sharply: Mordor sees $6.45 billion by 2030 while Future Market Insights projects $21.6 billion by 2035. Capital is following the category; our edtech funding statistics track how much of recent venture money is flowing to AI-first tutoring startups.
| Metric | Value | Source |
|---|---|---|
| Offline share of private tutoring revenue (2025) | 76.5% | IMARC Group |
| Online share of private tutoring revenue (2025) | 15.8% | IMARC Group |
| Online tutoring services market (2025) | $12.06 billion | Grand View Research |
| Online tutoring forecast (2030) | $23.73 billion (14.5% CAGR) | Grand View Research |
| Online tutoring, alternate estimate | $12.61 billion (2025) to $26.03 billion (2031), 15.58% CAGR | Mordor Intelligence |
| US online private tutoring forecast (2030) | $8.09 billion | Grand View Research |
| AI tutors market (2025) | $3.55 billion, to $6.45 billion by 2030 (12.69% CAGR) | Mordor Intelligence |
| AI tutoring services, alternate trajectory | $3.72 billion (2025) to $21.63 billion (2035), 19.3% CAGR | Future Market Insights |
Note: the AI tutoring forecasts differ by a factor of three because the category boundary is unsettled; some firms count only standalone AI tutor products while others include AI features inside broader learning platforms.
3. High-Dosage Tutoring in US Schools
The US ran the largest tutoring experiment in its history with pandemic relief money, and the aftermath is a partial retreat rather than a collapse. 42% of US public schools offered high-dosage tutoring in spring 2025, up from 37% in 2022, even as the share offering any tutoring at all fell from 51% to 43% over the same window (NCES School Pulse Panel). Districts committed more than $1.7 billion of ESSER III relief funds to tutoring and coaching (FutureEd), and with that money gone, federal support has narrowed to targeted grants. Parent demand has not retreated: EdChoice finds 35-45% of school parents in any given month either have a child in tutoring or are actively seeking it, part of the same do-it-yourself instinct visible in our homeschooling statistics.
| Metric | Value | Source |
|---|---|---|
| Public schools offering high-dosage tutoring, 2022 to spring 2025 | 37% to 42% | NCES School Pulse Panel |
| Public schools offering any tutoring, Dec 2022 to June 2025 | 51% to 43% | NCES, via EdWeek Market Brief |
| High-dosage tutoring growth by region (Northeast / South) | +10 / +6 percentage points | NCES, via EdWeek Market Brief |
| ESSER III funds committed to tutoring and coaching | over $1.7 billion (est. up to $3.6 billion by deadline) | FutureEd |
| Districts that planned ESSER spending on tutoring | over 40% | FutureEd |
| FY2025 federal EIR grants to scale state tutoring | ~$89 million across 7 states | Accelerate |
| School parents with a child in tutoring or actively seeking it | 35-45% in a given month | EdChoice |
Outlier: tutoring interest exceeds 55% among private school parents and special education families but sits below one-third among rural and small-town parents (EdChoice, May-July 2025 waves), a geography gap that mirrors the UK data below.
4. Does Tutoring Work? The Effectiveness Evidence
Few education interventions carry this much experimental evidence. The Nickow, Oreopoulos, and Quan meta-analysis of 96 randomized controlled trials found a pooled effect of 0.37 standard deviations, one of the largest reliable effects ever measured for a school-age intervention. Not one of the 96 randomized studies reproduced the fabled two-sigma effect, but 0.37 SD still translates to roughly 14 percentile points of achievement (Education Next). Stanford’s National Student Support Accelerator reports that high-dosage programs move an average student from the 50th to the 66th percentile, and that effects survive scaling, at reduced but still meaningful size. The same logic, structured practice with rapid feedback, shows up in adult learning; our corporate training statistics cover that market.
| Metric | Value | Source |
|---|---|---|
| Pooled effect across 96 tutoring RCTs | 0.37 standard deviations | Nickow, Oreopoulos, and Quan, NBER w27476 |
| Percentile-point equivalent of the pooled effect | ~14 points | Education Next |
| RCTs reproducing a two-sigma effect | 0 of 96 | Education Next |
| Percentile movement from high-dosage tutoring | 50th to 66th | National Student Support Accelerator |
| Additional learning from tutoring interventions | 3-15 months of school | National Student Support Accelerator |
| Learning gains in scaled programs (500-7,000 students) | 2-10 months | National Student Support Accelerator |
| High-dosage threshold | 3+ sessions/week, or ~50 hours over 36 weeks | National Student Support Accelerator |
Note: effects are strongest for tutoring delivered by teachers and paraprofessionals, and weaker for nonprofessional and parent-delivered programs (Nickow, Oreopoulos, and Quan).
5. Asia’s Shadow Education Economy
Asia is where tutoring stops being supplemental and becomes a parallel school system, what comparative-education researchers call shadow education, with enrollment in some grades exceeding 70% across parts of Asia, Europe, and North Africa (Bray, European Education, 2025). Korea’s numbers keep setting records in both directions at once: total spending rose 7.7% in 2024 while the student population fell, and private education consumed a record 13.5% of household spending for families with children by Q2 2025 (Statistics Korea, via The Korea Herald). China’s 2021 Double Reduction crackdown shrank visible subject tutoring, but peer-reviewed studies find demand shifted into small-group, one-to-one, and non-academic formats rather than disappearing (Liu et al., SAGE Open, 2024); New Oriental’s pivot to non-academic courses is the clearest market evidence.
| Metric | Value | Source |
|---|---|---|
| Korea private education spending (2024) | 29.2 trillion won ($20.1 billion), +7.7%, record high | Statistics Korea, via The Korea Herald |
| Korea participation rate (2024) | 80.0%, record, +1.5 points | Statistics Korea |
| Korea participation by level | 87.7% elementary / 78.0% middle / 67.3% high school | Statistics Korea |
| Korea average monthly spend per student (2024) | 474,000 won, +9.3% | Statistics Korea |
| Korea spending growth, 2014-2024 | +60.1% | Statistics Korea, via The Korea Times |
| Korea private education share of household spending (Q2 2025) | 13.5%, record | Statistics Korea, via The Korea Times |
| New Oriental non-academic tutoring (quarter ended May 31, 2025) | ~918,000 enrollments in ~60 cities; new-business revenue +32.5% YoY | New Oriental FY2025 results |
| India private tutoring market | $4.4 billion (2025) to $8.2 billion (2034), 6.91% CAGR | IMARC Group, India report |
| Student tutoring participation, China and India | over 55% | IMARC Group |
Outlier: shadow education reaches near-saturation at the extremes; Mark Bray’s research records more than 90% of final-year secondary students receiving tutoring in Egypt and Angola, the highest documented rates in the world.
6. The United Kingdom: Private Tutors and the National Tutoring Programme
Britain shows what happens when a temporary public program meets a permanent private habit. The Sutton Trust’s Private Tutoring 2026 report finds 29% of secondary pupils have had a private tutor, up from 18% two decades ago, but the distribution is the story. 45% of London pupils have been privately tutored versus 27% in the rest of England, and 33% of urban pupils versus 19% of rural ones. Meanwhile the state-funded National Tutoring Programme wound down in 2023/24 after delivering nearly 5 million courses since 2020 (UK Department for Education), and 58% of schools have since cut their tutoring offer.
| Metric | Value | Source |
|---|---|---|
| UK secondary pupils ever privately tutored | 29% (vs 27% in 2019, 18% two decades ago) | Sutton Trust, Private Tutoring 2026 |
| London vs rest of England | 45% vs 27% | Sutton Trust, Private Tutoring 2026 |
| Urban vs rural pupils | 33% vs 19% | Sutton Trust, Private Tutoring 2026 |
| Best-off vs disadvantaged families (2025) | 30% vs 23% | Sutton Trust, Private Tutoring 2026 |
| Pupils receiving school-based tutoring | 20%, down from 22% in 2023 | Sutton Trust, Private Tutoring 2026 |
| Schools that reduced their tutoring offer year over year | 58% | Sutton Trust, Private Tutoring 2026 |
| National Tutoring Programme pupils, 2023/24 (final year) | 1.07 million | UK Department for Education |
| National Tutoring Programme hours delivered, 2023/24 | 16.4 million | UK Department for Education |
| NTP courses started since November 2020 | ~5 million | UK Department for Education |
Context: the NTP did reach its target population before ending; 45.2% of tutored pupils in 2023/24 had been eligible for free school meals and 28.1% had special educational needs (UK Department for Education).
7. What Tutoring Costs
Price is the mechanism that turns tutoring into an inequality machine, and the spread is wide. Typical US rates run $25 to $80 per hour, but SAT/ACT test prep routinely starts above $100 per hour (Tutors.com, 2026 pricing guide). Global platforms undercut local averages: Preply’s internal data puts the average online tutor at $18.30 per hour, while Wyzant’s marketplace averages $35-$60. For language learning specifically, human tutors now compete directly with apps priced at a fraction of one lesson per month, a substitution effect quantified in our Duolingo statistics.
| Metric | Value | Source |
|---|---|---|
| Typical US private tutoring rate | $25-$80 per hour | Tutors.com, 2026 pricing guide |
| SAT/ACT test prep rate | often $100+ per hour | Tutors.com, 2026 pricing guide |
| Wyzant marketplace average | $35-$60 per hour | Wyzant |
| Average online tutor rate on Preply | $18.30 per hour | Preply, internal platform data |
| English lessons on Preply | $15-$50 per hour | Preply |
| US parents’ stated willingness to pay for tutoring | $357 per month (July 2025) | EdChoice |
Note: platform-reported rates are what tutors charge, not what families pay; Wyzant adds a 25% service fee on top of tutor-set rates, so listed and effective prices differ.
Summary: Private Tutoring by the Numbers
| Metric | Value | Source |
|---|---|---|
| Global private tutoring market (2025) | $131.04-$133.8 billion | The Business Research Company; IMARC Group |
| Forecast market size by 2034 | $160.5-$248.4 billion | Fortune Business Insights; IMARC Group |
| Long-run growth rate | 7.12-10.42% CAGR | IMARC Group; Fortune Business Insights |
| Asia Pacific revenue share | 35.5% | IMARC Group |
| Offline share of revenue (2025) | 76.5% | IMARC Group |
| Online tutoring services market (2025) | $12.06 billion, 14.5% CAGR | Grand View Research |
| AI tutors market (2025) | $3.55 billion | Mordor Intelligence |
| US incremental market growth, 2024-2029 | +$28.85 billion | Technavio |
| US public schools offering high-dosage tutoring (spring 2025) | 42% | NCES School Pulse Panel |
| ESSER III committed to tutoring and coaching | over $1.7 billion | FutureEd |
| Pooled effect of tutoring across 96 RCTs | 0.37 SD (~14 percentile points) | Nickow, Oreopoulos, and Quan, NBER |
| Percentile movement from high-dosage tutoring | 50th to 66th | National Student Support Accelerator |
| Korea private education spending (2024) | 29.2 trillion won ($20.1 billion), record | Statistics Korea |
| Korea participation rate (2024) | 80.0%, record | Statistics Korea |
| Korea share of household spending (Q2 2025) | 13.5%, record | Statistics Korea, via The Korea Times |
| New Oriental non-academic enrollments (Q4 FY2025) | ~918,000 | New Oriental filings |
| UK secondary pupils ever privately tutored | 29% (45% in London) | Sutton Trust, Private Tutoring 2026 |
| NTP tutoring hours in final year (2023/24) | 16.4 million to 1.07 million pupils | UK Department for Education |
| Typical US tutoring rate | $25-$80 per hour | Tutors.com, 2026 |
| US parents’ willingness to pay | $357 per month | EdChoice, July 2025 |
Methodology and Sources
Figures were gathered directly from primary reports, government statistical releases, peer-reviewed studies, company filings, and market-research publications from 2024-2026, cross-referencing market sizes across two or more firms and flagging figures with narrower scope or older base years.
- IMARC Group, Private Tutoring Market Report (2025 base year)
- IMARC Group, India Private Tutoring Market (2025 base year)
- The Business Research Company, Private Tutoring Global Market Report (2026 edition)
- Fortune Business Insights, Private Tutoring Market (2026-2034 forecast)
- Grand View Research, Private Tutoring Market Report (2023-2030)
- Grand View Research, Online Tutoring Services Market and US Online Private Tutoring Market (2025-2030)
- Technavio, US Private Tutoring Market (2024-2029)
- Mordor Intelligence, Online Tutoring Market and AI Tutors Market (2025 base year)
- Future Market Insights, AI Tutoring Services Market (2025-2035)
- Statistics Korea, Private Education Expenditures Survey 2024, via The Korea Herald (Mar 2025)
- Statistics Korea decade analysis and household-spending data, via The Korea Times (Jan 2026)
- Sutton Trust, Private Tutoring 2026 (Feb 2026)
- UK Department for Education, National Tutoring Programme statistics 2023/24 (final edition, Dec 2024)
- NCES School Pulse Panel (2022-2025 waves), with regional analysis via EdWeek Market Brief (Oct 2025)
- FutureEd, Georgetown University, ESSER spending analyses
- Accelerate, Beyond Recovery: Funding High-Impact Tutoring for the Long Term
- Nickow, Oreopoulos, and Quan, The Impressive Effects of Tutoring on PreK-12 Learning, NBER Working Paper 27476, with context from Education Next
- National Student Support Accelerator, Stanford University, research overview
- EdChoice, A Kaleidoscope View of K-12 Tutoring in America (2025)
- New Oriental, FY2025 fourth-quarter and full-year results (Jul 2025)
- Liu, Morgan, Zhang, and Wu, Private Tutoring Before and After the Double-Reduction Policy in China, SAGE Open (2024)
- Bray, When the Shadow Threatens, European Education (2025); Bray, Shadow Education in Africa (2021)
- Tutors.com, 2026 tutoring prices guide; Wyzant pricing guide; Preply tutor earnings data
Data watch: Statistics Korea publishes its Private Education Expenditures Survey each March, so 2025-year figures are due to be folded in at the next update; the NCES School Pulse Panel reports through the 2025-26 school year; the Sutton Trust refreshes its tutoring polling annually; and IMARC, The Business Research Company, and Fortune Business Insights all issue new market editions on annual cycles.
Last updated: July 18, 2026. We update this page quarterly as new survey waves and market editions are published.