The supply-chain confidence and automation-ready machine story behind five-axis demand.
China CNC Daily Insights · Issue #23
Japan's High-End Five-Axis Export Curbs Take Effect — and Why It Accelerates China's Domestic Substitution
Published 2026-08-21 · Sources: Japanese METI Order, Chinese industry & trade-media reporting (Aug 16–19, 2026) · Prepared for overseas buyers, investors and partners. ← Back to GREE CNC · All Daily Insights →
Contents
A new export-control rule took effect on 2026-08-16 — and it is less a "ban" than a bottleneck that quietly reshapes the global five-axis market.
- What changed: Japan's Ministry of Economy, Trade and Industry (METI) moved high-end five-axis CNC machine tools, core components (high-precision grating scales, rotary tables, motor-spindle parts) and even debugging / upgrading / maintenance & remote support into "Category-2" controls. General blanket licensing is replaced by case-by-case approval: lead time stretches from ~15 days to 45–180 days, with reported rejection rates above 80% for sensitive sectors. Orders under old licenses are only grace-period valid to 2026-12-31.
- Why it matters: Japanese high-end five-axis brands (Fanuc, Mazak, Mori Seiki / Makino) had long held an estimated ~70% share of China's high-end five-axis market. That channel is now functionally throttled — for new machines and for spares, upgrades and remote service on machines already installed.
- The flip side: China's five-axis localization rate climbed from 6% (2015) → 18% (2020) → 59.5% (2025), first surpassing imports. Controls like this are exactly what force the remaining gap shut — turning "we could try domestic" into "we must".
2026-08-16
Japan METI control order effective
45–180d
New approval lead time (was ~15d)
>80%
Reported rejection rate, sensitive sectors
~70%
Japan's prior share of China high-end 5-axis
59.5%
China 5-axis localization rate (2025)
~70%
C919 titanium parts on domestic 5-axis
1 · What actually changed on 2026-08-16
On 2026-08-16, Japan's revised Export Trade Control Order (published by METI on 2026-06-16) came into force. It adds high-end machine-tool technology — especially five-axis CNC — to a "Category-2" control list covering all regions, with China as the obvious focal point. The mechanism changed from a blanket license to case-by-case approval.
The three-layer scope (per trade-media reading of the order)
- (1) Ultra-high-precision five-axis machines — machining centers with special high-precision rotary axes and precision tooling systems.
- (2) Core components — high-precision grating scales, precision rotary tables, motor-spindle key parts.
- (3) Machine commissioning, upgrade & maintenance tech — full commissioning, system upgrades (incl. NC programs, CAM software), and original-spec maintenance incl. remote technical support.
The practical effects, as reported across multiple Chinese trade-media outlets:
- New orders: every new high-end five-axis order now needs one-by-one approval; lead time balloons from ~15 days to 45–180 days, with sensitive-sector rejection reportedly above 80%. New-order delivery becomes effectively uncontrollable.
- Installed base: spares, factory expansion and repairs on existing Japanese machines are also hit. Old-license orders already signed are only grace-period valid to 2026-12-31.
- Remote-lock risk: Japanese high-end machines commonly carry remote-lock capability — parameter changes or relocation can trigger a head-office shutdown. This is now a named concern for Chinese plant operators.
Context: Japan's third round of semiconductor-equipment export controls landed 2026-08-01 (advanced packaging, case-by-case approval). The 2026-08-16 machine-tool rule extends the same "long approval + high rejection = de-facto cutoff" logic beyond chips.
2 · Why this accelerates domestic substitution
The control lands at a moment when the substitution curve was already steep — but the "last mile" was a confidence problem, not a capability problem. Plant managers kept an import fallback, so they were slow to risk domestic machines on critical parts. Forcing the fallback shut flips the incentive.
China's five-axis localization rate — the trend line
- 2015: ~6% — almost entirely import-dependent
- 2020: ~18%
- 2025: ~59.5% — first time domestic share surpassed imports
Reading: the capability existed; the control removes the excuse not to use it. Aerospace, mould, automotive and robotics high-end scenarios have already begun large-scale import replacement.
There is a second-order effect worth noting for buyers: Japan is walking away from a market it dominated for ~20 years. One report notes Makino's share price fell >4% the day the rule hit, because buyers froze new orders. The ~70% share Japan held is now a vacuum Chinese makers are positioned to fill.
3 · The Chinese makers stepping up
Several domestic players are repeatedly named as the beneficiaries of this "forced substitution window":
| Maker | What's cited | Credibility |
|---|---|---|
| KEDE CNC (科德数控, 688305) | >90% whole-machine localization; ~85% core-component self-sufficiency (CNC system, spindle, motor-spindle, rotary table, swing head, tool magazine, grating scale all self-developed). Supplies AECC / COMAC-linked aerospace users. | High (audited disclosures) |
| Gree Intelligent Equipment (格力智能装备) | Self-developed five-axis reportedly reaches 0.003 mm accuracy; full-chain in-house on linear motors, spindles, servo, rotary tables. Positioned on NEV die-casting & whole-line automation. | Medium (vendor-stated) |
| Beijing Jingdiao (北京精雕) | Cited at "maturity level 8" (described as the highest domestic grade); 0.1-micron feed, 1-micron cutting. | Medium (industry citation) |
| Sichuan Deyang (四川德阳, 2026-06) | World's first all-direct-drive five-axis blade mill; core direct-drive motors & CNC 100% domestic; accuracy within 1 micron; cost ~40% below imported equivalents. | High (named project) |
The honest caveat. Chinese media also openly states the gap is not closed: high-end CNC systems remain ~70% foreign-dependent; precision-bearing localization is only ~15%; high-precision ball screws / linear guides import rate >70%. Mean-time-between-failure is cited around 3,000–4,000 h vs ~8,000 h for imports. Substitution is real and accelerating — but "functionally replaceable" is not yet "process-equivalent" at the very top end.
4 · The C919 proof point
The most concrete evidence that domestic five-axis is production-grade — not just lab-grade — is the C919 narrowbody airliner. Multiple reports state that roughly 70% of the aircraft's titanium-alloy structural parts are now machined on domestic five-axis machines, including aero-engine blisk / structural work. KEDE is named as having signed a co-innovation center with Shanghai Aircraft Manufacturing (COMAC) and begun supplying controllable high-end five-axis machines for large-aircraft part validation.
Why this matters for the substitution thesis: aerospace is the hardest, most certification-heavy end of five-axis. If domestic machines are trusted on a flagship airliner's titanium structure, the "will it hold up in real production?" objection loses much of its force elsewhere.
5 · What it means for overseas CNC buyers
If you are evaluating Chinese five-axis (Gree, KEDE, Jingdiao, Haitian…): the capability floor has risen sharply, and the supply window is widest exactly when Japanese lead times blow out to 45–180 days. For NEV dies, 3C, robotics and general aerospace-tier parts, domestic five-axis is now a legitimate shortlist item — not a gamble.
But verify, don't assume: insist on a real test-cut of your part, a long-cycle (≥6-month) accuracy re-inspection clause, and pinned-down sources/lead-times for the still-import-dependent parts (grating scales, precision bearings, high-end screws/guides). The honest gap is accuracy retention and process databases, not first-article precision.
If you are a Japanese-machine owner: the rule raises spares and service risk on installed bases — a reason to qualify a second-source (domestic) supplier for non-critical and even some critical paths before the 2026-12-31 grace window closes.
6 · Data Credibility Rating
| Information | Source | Credibility |
|---|---|---|
| Japan METI order effective 2026-08-16; case-by-case approval; 45–180d; rejection >80%; grace to 2026-12-31 | METI order + multiple Chinese trade-media readings (ccpityc,chwang,qq, sina, toutiao) | High (primary order + cross-checked) |
| Japan held ~70% of China high-end five-axis market | Multiple trade-media; widely repeated | Medium (industry estimate) |
| China 5-axis localization 6%→18%→59.5% (2015–2025) | Chinese industry media citing sector data | Medium (single-source sector stat) |
| KEDE >90% machine / 85% component self-sufficiency | KEDE disclosures + media | High |
| Gree self-developed five-axis reaches 0.003 mm | Vendor / media | Medium (vendor-stated) |
| C919 ~70% titanium parts on domestic five-axis | Multiple media; repeated figure | Medium (widely cited, not independently audited) |
| Remaining gaps (systems ~70% import; bearings 15%; MTBF 3–4k h) | Chinese media self-analysis | Credible (adversarial self-report) |
Frequently asked questions
Did Japan ban five-axis machine tool exports to China?
Not a flat ban — a throttle. From 2026-08-16, high-end five-axis machines and core components moved to "Category-2" controls requiring case-by-case METI approval. New orders face 45–180 day lead times and reportedly >80% rejection in sensitive sectors, with old-license orders valid only to 2026-12-31.
Why does this help Chinese makers?
Japan held an estimated ~70% of China's high-end five-axis market. Throttling that channel removes the import fallback that kept many buyers from qualifying domestic machines — exactly the "forced substitution window" industry analysts describe.
Is Chinese five-axis actually good enough now?
First-article precision is strong and rising fast (KEDE >90% localized; Gree cites 0.003 mm; C919 titanium parts ~70% on domestic five-axis). The honest remaining gap is long-term accuracy retention, process databases and high-end CNC systems (~70% still imported).
What should a buyer do?
Shortlist domestic five-axis for NEV, 3C, robotics and many aerospace-tier parts, but insist on a real test-cut, a ≥6-month accuracy re-inspection clause, and pinned sources/lead-times for still-import-dependent components. Japanese-machine owners should qualify a second-source supplier before the 2026-12-31 grace window closes.
Related reading
Talk to our application engineers
Send your part material, tolerance and volume — we return a recommended model, spec sheet and quote within one business day.
Request a Quote →