chinese-war.com

A 2027 U.S.–China conflict with Australia aligned to the U.S. China trade stops immediately, Taiwan exports are slowed/limited, and China launches large-scale cyber disruption in waves. This is a planning scenario, not a forecast. Planning scenario, not a forecast. No end date is set.

Education & Research — United States (Investor-Grade)

Read the Australia briefingReport date 2026-11-07

What happens

What stops, what keeps enrolling, and what can halt a semester

  1. China trade stopping removes the China enrollment line of 265,919 students and the China scholar line of 25,983. It does not remove India at 363,019. It does not remove the 6 percent denominator's complement, which is the domestic and other-origin majority of U.S. higher education. The 6 percent figure is the printed international share. A headcount for the complement is not calculated here because 6 percent is a rounded print.

    iie.orgopendoorsdata.org

  2. Inside the China line, 23.3 percent are already on Optional Practical Training and 45.2 percent are graduate students. The first cash month is the new intake. The fall 2025 snapshot says new enrollments decreased by 17 percent across the participating institutions. That snapshot is all origins. A China-only fall 2025 count was not on the press release.

    opendoorsdata.orgiie.org

  3. Federal grants keep paying if the drawdown system stays up. Almost 60 billion dollars and 55 percent of academic R&D were federal in FY 2023. HHS alone was 33.1 billion dollars. Foreign funds of 1,619,058 thousand dollars are the slice that could include a foreign sponsor, and the table does not say how much of it is China.

    ncses.nsf.govncses.nsf.gov

  4. Scholars are the joint-lab mechanism that is printed. 75 percent of international scholars pursued research activities. Short-term appointments, less than one year, were 35 percent of scholar appointments in 2024/25. Mid-length appointments were 40 percent and long-term appointments were 25 percent. Those duration shares are all origins. A stop on new Chinese scholars hits the short-term inflow first. Whether a scholar already on campus finishes the appointment is an inference.

    opendoorsdata.org

  5. The domestic break in week 1 is the student system. Approximately 68 percent of exposed U.S. PaperCut servers were in the Education Facilities Subsector, and the early May 2023 attempts were against that subsector. The Idaho exercise found senior leadership was not prepared to run a ransomware scenario once IT staff were removed, and student data was part of the scenario. Days of a real university outage were not printed on those two pages.

    cisa.govcisa.gov

  6. Taiwan slowing is a student fact and an unpublished equipment fact. The Taiwan fact sheet total is 23,263 students, with 86.5 percent at doctoral universities. The Census international-trade API returned HTTP 302 and an empty body, so this report does not state a Taiwan share of microscopes, vacuum tools, or other campus equipment. Other Asia is not used as Taiwan.

    opendoorsdata.orgapi.census.gov

How hard, how soon

0–3 months

  • Short-horizon enrollment analog: new international enrollments decreased by 17 percent in the fall 2025 snapshot. The snapshot covered over 825 institutions. It is a semester start, not a year-over-year peak, and it is not China-only. Average published in-state public four-year tuition and fees were 11,610 dollars. That figure is an academic-year sticker, not a semester fee and not a China fee.

    iie.orgnewsroom.collegeboard.org

  • Short-horizon system analog: exploitation attempts against Education Facilities Subsector PaperCut servers were observed in early May 2023, and approximately 68 percent of exposed U.S.-based PaperCut servers were in that subsector. The opened PaperCut page and the Idaho exercise page do not print a count of days offline or a count of students affected.

    cisa.gov

1–6 months

  • Pipeline analog: graduate enrollments decreased by 12 percent in the fall 2025 snapshot. That is the missed-cohort band. Undergraduate enrollments in the same snapshot were up by 2 percent and are not this band.

    iie.org

12+ months

  • Year path: students from China posted a 4 percent decline, to 265,919, on the 2024/25 release. Federal R&D remained 55 percent of a total that reached 108.8 billion dollars. The 4 percent decline is the long enrollment analog. The 55 percent share is the grant base that does not depend on that decline.

    iie.orgncses.nsf.gov

Wave 1

Wave 1 is the campus that teaches. Research universities and student information systems are direct. International students from China, students from India and other countries, and Taiwanese students are indirect because they stop registering when that system stops. The evidence is the PaperCut education-facilities advisory and the Idaho student-data scenario. The week order is a planning model.

cisa.govcisa.gov

Wave 2

Wave 2 is the enterprise layer that spends the grant. Federal research grants, campus laboratories, and payroll systems are direct. The federal book is almost 60 billion dollars and R&D personnel pay was 47.1 billion dollars. This order is an inference. The fetched pages do not put payroll in week 2.

ncses.nsf.govcisa.gov

Wave 3

Wave 3 is the distributed and external set. Vocational and associate's colleges are direct. Laboratory equipment buyers are indirect. Associate's colleges hold 1.4 percent of students from China, so the commercial China hit is small there. The later tag is a planning inference. The Idaho exercise did include two-year colleges in the same drill as four-year universities.

opendoorsdata.orgcisa.govapi.census.gov

Country exposure

origin sharepercent

Cost and price shock by time bucket

price changepercent

sticker feedollars

Downstream scale

sitesinstitutions

studentspeople

dollarsbillion dollars

Operating prints

ratespercent

What to do

Day 7

  • Separate the China enrollment file from the India and domestic files. The baselines are 265,919 students from China, 363,019 from India, and a 6 percent international share of all higher education.

    iie.org

  • Put the student-information system and exposed print servers on the week-1 watch. The PaperCut advisory is the education-facilities path, at approximately 68 percent of exposed U.S. servers. Confirm payroll can run one cycle if that system is isolated.

    cisa.govncses.nsf.gov

Day 30

  • Count Chinese scholars whose appointments end inside the year. Short-term appointments were 35 percent of all scholar appointments. China was 25,983 scholars. A campus-level split of that 35 percent was not printed.

    opendoorsdata.org

  • Watch the fall new-enrollment print against the 17 percent snapshot decline, and keep federal drawdown tested. The federal baseline is 55 percent of R&D.

    iie.orgncses.nsf.gov

Day 90

  • Rank movable equipment already on campus against the unpublished Taiwan share. The only equipment total opened is 3.5 billion dollars, combined with software, and it has no country.

    ncses.nsf.govapi.census.gov

  • Keep vocational colleges on a separate plan. Associate's colleges were 1.4 percent of students from China. Do not staff a community college as if it were a doctoral China cohort.

    opendoorsdata.org

What to watch

By role

Research university finance teams

  • Split tuition receivables into China, India, and domestic before week 1, using 265,919 and 363,019 as the national baselines.

  • Keep federal drawdown independent of the student-information system.

  • Do not book the 14,587,000,000 dollar Commerce impact as tuition.

iie.orgopendoorsdata.orgncses.nsf.gov

Registrars and international offices

  • Run a term-start case at the 17 percent new-enrollment analog.

  • Flag Chinese graduate students against the 120,229 baseline and Optional Practical Training against 61,981.

  • Keep India and other origins in the continuing admit pool.

iie.orgopendoorsdata.org

Grant offices and laboratory directors

  • List Chinese scholars against the 25,983 baseline and mark appointments under one year.

  • Protect the 47.1 billion dollar R&D pay process and the 55 percent federal drawdown.

  • Do not hold a purchase for a Taiwan equipment share that was not published.

opendoorsdata.orgncses.nsf.gov

Vocational colleges

  • Plan on the 1.4 percent associate's share of students from China, not on the doctoral 89.1 percent.

  • Keep the two-year domestic book, which added 67,336 undergraduates in fall 2023, as the operating load.

opendoorsdata.orgnces.ed.gov

Weekly checks

What is not known

Working papers: node summary
NodePrimary impactTime horizonCyber waveEvidence
Research universitiesA China stop removes a doctoral enrollment line, not the federal grant book that pays the lab.0-3 monthsWave 1 (direct)Open Doors China fact sheet; NSF HERD; CISA PaperCut and Idaho exercise
Student information systemsRegistration, grades, and student records stop when the campus system or its print and identity layer is down.0-3 monthsWave 1 (direct)CISA PaperCut advisory; CISA Idaho exercise
International students from ChinaChinese enrollment is the stopped line, and it is one line inside a much larger student body.0-3 monthsWave 1 (indirect)IIE Open Doors 2025; China fact sheet
Students from India and other countriesIndia and other origins keep enrolling when China stops.0-3 monthsWave 1 (indirect)IIE Open Doors 2025; NCES fall 2023
Federal research grantsFederal award dollars continue unless the university cannot draw them down.1-6 monthsWave 2 (direct)NSF HERD FY 2023; HERD Table 15
Campus laboratoriesVisiting researchers from China are the joint-lab exposure, and the grant that pays the lab is federal.1-6 monthsWave 2 (direct)Open Doors scholars; NSF HERD
Payroll systemsPay for research personnel stops if the payroll system is the later enterprise hit.1-6 monthsWave 2 (direct)NSF HERD personnel pay; CISA Idaho exercise
Vocational and associate's collegesA China stop barely touches associate's colleges, where few students from China are enrolled.1-6 monthsWave 3 (direct)China fact sheet; NCES; Open Doors community-college note
Taiwanese studentsTaiwanese enrollment is a real doctoral cohort and a small one next to China.0-3 monthsWave 1 (indirect)Open Doors Taiwan fact sheet
Laboratory equipment buyersMovable equipment keeps being bought, and a Taiwan share of that equipment is unpublished.12+ monthsWave 3 (indirect)NSF HERD equipment line; Census API HTTP 302

Sources