It is July, which in China means the gaokao is frequently featured in the headlines. Outside of the country, the short story about the “highest exam” is often that this is where the fate of millions of students is determined by a single number in a national contest. This isn’t exactly wrong, but the gaokao isn’t one national contest. To get why much of the recurring debates about the gaokao are centred on regional inequalities and perceptions of privilege and unfairness, one has to be introduced to the provincial quota system.
When a Chinese university plans its admissions, it does not open one national pool and take the top scorers. Rather, it allocates fixed admission quotas to each of the 31 provincial-level jurisdictions. Usually universities also have a province-by-province allocation plan for each subject or subject clusters. This means the cutoff score for the same university/subject is different in every province. There are more places reserved for home students, which isn’t usual. Public universities in the US also favour in-state applicants, but that is essentially a locals versus everyone else divide – there isn’t a quota for every state.
Why build it this way? Partly it’s an inheritance of central planning. After the 1952 higher education reform, university enrollment was administered like any other state function, allocated by administrative unit, guided by national economic goals, and tied to the hukou (household registration) system that fixes you to a province. But there is also a much longer lineage. If we trace the gaokao form and its legitimacy to the imperial examination or 科举, then a controversy in 1397 was a pivotal moment. This year, the metropolitical examination (会试) produced a list of successful candidates entirely from the south. It turned into a political scandal.
After executing a bunch of officials, the Hongwu Emperor held another exam himself and this time selected 61 northerners. Later in the Ming dynasty separate exam papers for the north and south (南北分卷) were introduced to counter-balance the exam success of the prosperous, education-rich south. Qing initially inherited this and later replaced it with a full provincial quota system. The imperial exam, as a tool of statecraft and legitimacy, had to consider the rationale of national integration beyond that of merit-sorting.
There is therefore a certain redistributive, integrative logic to the per-province allocation system. It could be a mechanism to cap strong regions and lift weak ones. In the contemporary context, there are a range of redistributive initiatives meant to mitigate the gap – now between east and west, rather than north and south, such as the 中西部协作计划 (reallocating quotas from eastern to middle and western regions), the poverty-targeted special plans 专项计划, and additional points for ethnic-minority students. However, regional inequality in access to higher education – especially elite universities – remains entrenched.
In a 2019 paper, Hamnett, Shen, and Liang argue that the provincial quota system actually reproduces regional inequality rather than ameliorating it. This rests on two structural facts. First, elite universities are absurdly concentrated. By their count, Beijing hosts 34 of the country’s 140 “Double First-Class” universities, and Beijing, Shanghai and Tianjin together hold more than half the country’s elite-university resources. In their “high-quality university relative abundance index”, Beijing sits at 1.56 and Hebei right next door at 0.01, a hundredfold gap. Second, universities reserve a large share of their seats for their own home province. The mechanism thus funnels the most coveted seats to the already-advantaged.
It’s interesting to look at their “enrolment opportunity index” for Peking and Tsinghua (the national Top 2). Beijing, Tianjin, Shanghai unsurprisingly sit at the top. They are followed by four ethnic-minority provinces: Qinghai, Tibet, Ningxia, and Xinjiang. Shandong and Guangdong, which are much richer costal provinces, are at the bottom.
This is produced partly by the redistributive initiatives mentioned before, and partly because of the tiny candidate pool that could easily skew the per-capita rate. Tibet has 3.6 million people and Shandong 101.5 million. When a student in Shandong complains that “students from ethnic minority regions have it easier than we do”, it is true on the calculated Peking/Tsinghua admission rate on a per capita basis. This could lead to a whole genre of majority grievance (see our paper here on meritocratic nationalism and majority male thinking they are the most oppressed). But this obscures the fact that there is vast difference in educational resources and opportunities overall. Those regions have far less qualified teachers (hence the 支教 programmes), far lower average attainment, fewer good local universities, and lower overall rates of getting into any university at all. The quota lowers the bar at a handful of national universities for a small number of candidates, who are typically the very top of a small, stratified distribution.
I took the gaokao in Shandong more than 20 years ago. As far as I remember, back then the loudest complaints came from Shandong and Jiangsu. Shandong’s grievance was about sky-high cutoff scores, and Jiangsu’s was brutal exam paper. Jiangsu had their own paper, different from the national paper and famously punishing. In Shandong, “gaokao migration” used to be a trend, although illicit. Students change their hukou to e.g. Inner Mongolia or Xinjiang in the final year of high school to benefit from both strong educational base in Shandong and low cutoff score elsewhere.
Today, however, the grievance centre seems to have shifted to Henan. I think partly because the prize has changed. Before the enrollment expansion of the last few decades, the fight was to get into any university. Now, the anxiety shifted to getting into a good one, measured by the 985/211 labels (now rebranded as “Double First-Class”, but 211/985 is still widely used informally). Shandong and Jiangsu have a large population and high score pressure, but they also host their own strong universities that have reserve seats for local students. Henan, with roughly 100 million people, has one “211” and no “985” (top-tier) at all. On the account of elite-university numbers relative to population, Henan is the most extreme version of the “losing” type.
In 2016, the Ministry of Education introduced a plan to shift enrolment quotas from 12 provinces to 10 others. It was unclear how these provinces were selected. Hebei and Jiangsu, for example, were considered highly competitive and lost quotas. But unsurprisingly, enrolment quotas of one place that enjoys the best opportunities – Beijing – were untouched. Tens of thousands of parents in Jiangsu, Hubei, and Heilongjiang took to the street to protest; many were arrested and online discussions censored.
A story recently circulating on social media again is about a Henan student named Zhao Siyan. In 2019 he scored 539, just 34 points above Henan’s science first-tier line. He had already decided to repeat the year, and had idly listed Peking University as his first choice under the National Special Plan, the poverty- and rural-targeted admissions track, without any hope. Then something strange happened. That year, Peking’s 8 science-and-engineering places for Henan under the special plan did not fill. Zhao was the 8th. His score was 131 points below the 6th-ranked candidate.
Peking did not want him, claiming that his score was too low to survive its teaching pace. It filed to reject him, sending an official memo “unable to complete the degree” to Henan’s examinations authority (the exam yuan). But the authority refused the rejection. A minute later, Peking refiled. The authority refused again. Somewhere in that exchange the authority sent a message that would travel across the Chinese internet: “Henan’s students are, on the whole, of high quality, with solid foundations. Please consider this” (河南整体生源质量较高，考生基础扎实，请考虑为盼). People online were deeply moved, calling the exam yuan “a mother hen shielding her chicks”.
The plea and the online outrage eventually worked. Peking reversed itself in August and conceded that Zhao should be admitted. The story resurfaced recently as in 2026, Zhao completed his master’s degree from Peking’s School of Earth and Space Sciences. It makes Peking’s initial claim that his score was too low to follow the course all the more absurd. However, in that 2019 online debate, while most were on Zhao’s side, some people also claimed it was not “fair” because Zhao’s score was significantly lower than other Henan students who got into Peking. They say it was procedural but not substantive “justice”.
One thing that struck me about this story is how people praised the bureaucratic authority as a saviour and hero. When the best available remedy for structural disadvantage is a sympathetic bureaucracy negotiating on a case-by-case basis, it only shows how few structural possibilities there are.
The other thing is, nothing else like the gaokao in contemporary China is so tightly bound to the public language of fairness. No other institution is at once trusted to be, and accused of failing to be, the one clean ladder. Roland Bénabou says “there is value-free definition of meritocracy”. A single number produced by a standardized test, however many corrective quotas and special plans are bolted onto it, cannot be neutral, cannot be simply “fair”. Reform, I think, will have to involve not only changing educational disparity on the ground, but also rethinking the myth of meritocracy itself.
Leave a Reply