THE STATE OF US-CHINA ACADEMIC EXCHANGE, 2026

For two decades, the story of US-China academic exchange ran in one direction — up and to the right. Enrolment from mainland China at American universities climbed past 370,000 by the late 2010s. Joint research papers between American and Chinese institutions multiplied. Then the line broke. Two trade wars, one pandemic, and a series of visa-policy reversals later, the picture in 2026 looks very different — and considerably more interesting than the headline numbers suggest.

THE NUMBERS, AS WE FIND THEM

The latest Institute of International Education data, paired with our own informal survey of returning study-abroad coordinators, suggests three things. Chinese enrolment at US universities has stabilised around 277,000 — down from the 2019 peak but flat for three consecutive academic years. Graduate enrolment is holding; undergraduate enrolment is the segment that fell hardest and has not recovered. American enrolment in mainland China, meanwhile, sits at roughly 800 students per year, against a pre-pandemic figure above 11,000.

That is the asymmetry that defines the field right now. There is no shortage of Chinese students who want to study in the United States. There is an acute shortage of American students who are willing — or able — to study in China.

WHY THE AMERICAN SIDE COLLAPSED

The collapse of American outbound enrolment to China has four causes, in our reading. None of them is dispositive on its own.

• Programme closures. Several flagship US-run programmes in Beijing and Shanghai paused operations between 2020 and 2022. Some never restarted. The result is a thinned-out pipeline of host institutions.
• Travel-advisory effects. US State Department travel advisories and Chinese exit-ban reporting have made universities cautious about insurance, liability, and parent communications.
• Language attrition. Mandarin enrolments at US colleges fell by roughly a quarter between 2016 and 2024. Fewer language students means a smaller pool of credible study-abroad candidates.
• The substitution effect. Taiwan and Singapore have absorbed much of the Mandarin-language study-abroad demand. A semester at National Taiwan University now does the work that a semester at Peking University did fifteen years ago, from the American student's point of view.

"We are not training the next generation of China specialists. We are training the next generation of Asia generalists. That is a different category, and the field has not caught up with it."

The quote is from a senior university-based China-studies director who asked not to be named. It maps onto every conversation we have had with language-programme staff in the last twelve months.

WHY THE CHINESE SIDE HAS STABILISED

The stabilisation of inbound Chinese enrolment in the United States is the quieter story. After the dramatic drops of 2020 and 2021, the segment that recovered fastest was STEM graduate enrolment — particularly in engineering, computer science, and the life sciences. Undergraduate enrolment recovered more slowly and remains below trend. Master's enrolment in business and finance has come back to roughly 80 percent of the 2019 peak.

Two things are sustaining the figure. First, the US continues to operate the world's largest concentration of research universities, and the global supply of seats in elite STEM doctoral programmes is, in absolute terms, still small. Second, the post-graduation work-visa pipeline — OPT, STEM OPT, and the H-1B lottery — remains, despite years of restrictions, the most permissive among the major English-speaking destinations for the cohorts we track. The UK and Canada have both moved in less welcoming directions; that redirects flow back toward the US.

WHAT THE ALUMNI NETWORKS LOOK LIKE

Behind both flows sits an enormous, mostly invisible China alumni network — Americans who studied in China between roughly 1995 and 2019, and Chinese who studied in the US over the same period. We estimate the combined population at well over a million. The implication for any serious academic exchange program is that the supply of potential mentors, host-family coordinators, and second-stage placement contacts is not the constraint. The constraint is institutional memory and the willingness of universities to invest in matching infrastructure.

For a deeper look at the inbound side, see our companion essay on the quiet return of study abroad in China, which covers the language-programme rebuild now underway. The institutional side — what universities actually do with the alumni they already have — is the subject we plan to take up in the next essay in this series.

THE POLICY ENVIRONMENT IN 2026

Two policy variables matter more than all the others combined. The first is the US student-visa renewal regime — specifically, the length of the F-1 issued to Chinese STEM graduate students. Multi-year renewals stabilised in 2024 after a stretch of one-year reissues that had effectively rationed enrolment by attrition. The second is the Chinese exit-visa landscape for American researchers and students, which has tightened in specific sectors and relaxed in others. The result is a programme-by-programme picture in which the policy weather is genuinely different at the level of individual institutions.

Universities that have adapted to this granularity — by maintaining live legal counsel relationships in both jurisdictions and by training their study-abroad offices in mainland-specific contingency planning — are reopening programmes. Universities that are still using the 2018 operating playbook are not.

WHERE WE THINK THE FIELD GOES NEXT

Three quiet trends are worth watching. First, third-country programmes (Singapore, Taiwan, and a small number of European-Chinese joint ventures) are absorbing the experiential layer that used to belong to mainland programmes. Second, the rise of dual-degree arrangements between US and Chinese universities — paused during the pandemic — is restarting on noticeably more cautious legal terms. Third, the private foundations that underwrite much of US China educational exchange are consolidating; the field is now funded by fewer, larger institutions than it was in 2018, with predictable consequences for editorial range.

None of these trends is catastrophic. None of them is reassuring either. If you are reading this and you run an exchange programme, an alumni association, or a language department — write to us. The institute reads every email and the field is small enough that the conversation still matters.