The first people you meet in life are your parents. That simple fact makes the family the foundational institution of society: learning how to be a person, and what other people are like, happens first in the family. After that, the next social institution most people encounter, in modern societies anyway, is usually schooling and education. Schools transmit social values to new generations, and help determine the specific social roles that the members of the new generations will occupy. Education is always a window onto society, and the new book The Highest Exam: How The Gaokao Shapes China, by the economists Ruixue Jia and Hongbin Li, is very illuminating on how China’s educational system has mattered for both social and economic outcomes, and on just why education is so important to so many Chinese families.

More profoundly than that, though, it makes a case for how one specific set of educational practices–the college-admissions exam known as the gaokao–serves as a fundamental structuring institution of contemporary Chinese society. Scores on the gaokao (or more precisely, the relative ranking of those scores) are the sole determinant of which students get admitted to which universities. Because college admissions are determined by gaokao scores–and nothing else, not recommendations or ability to pay–the test functions as a meritocratic channel for social advancement. There is fierce competition for the opportunity to attend the best colleges, but the competition is based on merit and merit is measured by exams.

In this basic structure, the gaokao is widely understood as the modern reincarnation of the keju, the imperial examination system that offered a chance, in theory, for any man in China to join the emperor’s civil service. The authors describe how millions wasted their lives in fruitless efforts to score high enough on the imperial exam to achieve a new and higher social status. But the fact that a meritocratic channel for social advancement existed was important; not having one seems to have, historically, not worked out well. The authors observe: “The exam system persisted for over a millennium, but after years of mounting criticism, the Qing government abolished it. Shortly thereafter, the Qing dynasty itself fell in 1911.” The ending of the exam system meant a huge loss of opportunity for the men who had spent years preparing for it, an experience that politically radicalized many of them.

When the Communist Party took over after China’s civil war, they moved quickly to reinstate a meritocratic channel for social advancement. The gaokao was first established in 1952, in the early years of the People’s Republic, as a nationwide exam that would identify talented young people for further education. The new leaders wanted a modern system for testing useful knowledge, not the ability to recite the Confucian classics, so the gaokao was a self-conscious modernization and reform of the old system. But the cultural memory of the keju was broadly positive: the population was already prepared to accept the idea that an examination system is, in fact, a fair way to identify merit. By associating itself with a traditional culture of exams and meritocratic advancement, the new Communist government gained rather than lost legitimacy.

The next Chinese leader foolish enough to mess around with exams was Mao Zedong, who during the Cultural Revolution shut down the gaokao, along with the normal functioning of the entire educational system. The Cultural Revolution for a while turned meritocracy on its head, punishing the educated classes and elevating workers and peasants. Universities did eventually resume classes, but they were mostly about ideology, and admissions were done on the basis of political recommendations. But Mao’s death opened the floodgates, and the leaders in Beijing were inundated by protests and petitions demanding the reinstatement of the gaokao. Although the start of China’s “reform era” is conventionally dated to 1978, when some top-level political meetings were held, some people consider the real start to be 1977, when Deng Xiaoping restored the test and univerisites admitted a new class of students selected on the basis of academic merit.

One possible reading of the last century or so of political history in China, therefore, is that governments who provide a meritocratic channel for social advancement have legitimacy and popular support, and governments who do not provide one, do not. Such a pattern may have held in previous centuries as well: one of the co-authors’ many papers summarized in the book quantifies how the introduction of the keju system, around the 7th century AD, reduced Chinese emperors’ risk of being dethroned by a factor of 10. Certainly, the way more recent governments have behaved suggests they think the gaokao helps keep them in power. The gaokao is not seen as legitimate because it is instituted by the Communist Party; rather, it is the Communist Party that gains legitimacy by administering the gaokao in an even-handed way.

That’s not to say that the exam system is without downsides. For all of its popular support, the gaokao creates an incredibly high-pressure social environment. It’s not just that there’s one test that everyone has to take, but that children’s entire educational trajectory is built around preparing for success on that test. The gaokao is just the culmination of a series of tests and sorting procedures that choose who will be admitted to China’s elite universities, which to a remarkable extent determines who will be in the social and economic elite. Parents work relentlessly to position their children to do well on the gaokao, because, literally, nothing else matters; there are essentially no alternative pathway to success in Chinese society. Because people arrange their entire lives to help their kids succeed on the gaokao, changes can threaten the investments already made, and lead to strong resistance.

In 2013, according to the book, Hongbin Li was asked by China’s government to design a replacement for the gaokao. The initial idea was to switch out the single all-important test administered on only one day for a series of tests in different subjects, which could be taken multiple times throughout a student’s high school career–somewhat analogous to the Advanded Placement (AP) tests in US high schools. That probably would be less stressful for students, and make for a fairer assessment of their achievements. Li designed textbooks and tests and even started training teachers in the new approach, before the whole effort was scrapped. Even smaller changes to the gaokao system to try to address entrenched inequality had generated surprisingly widespread political protests, and been quietly walked back. The impression is very much that the exam system amounts to an untouchable “third rail” of Chinese politics.

One of the more interesting implications of the book is that because the social prestige of the gaokao is so high in China, and because everyone has been trained to understand and work with that kind of system, that it has implicitly become the model for other social institutions outside of education. The authors analyze the underling structure of the gaokao as a “centralized hierarchical tournament”: the competition between students is centralized because there is only one standard of success, and it is hierarchical because success is defined in relative terms, by doing better than those around you. Once the concept is grasped, it is easy to see many other centralized hierarchical tournaments in Chinese society.

The most consequential of these is probably the competition among local governments to generate economic growth, which many scholars have identified as one of the fundamental structures underlying China’s reform-era growth boom. Local officials are always competing for promotion and advancement, and one of their key performance indicators is economic growth in their jurisdiction. That competition is centralized because there is, effectively, only indicator of success, GDP growth, and only one arbiter of success, the central Party apparatus. And it is hierarchical because success is defined relative terms: it’s about where is your GDP growth relative to your peers and predecessors in office.

That system makes sense for some of the same reasons that the gaokao does: GDP growth, like an exam score, is a relatively objective and transparent metric, better than many alternatives. Also, centralized hierarchical tournaments seem to be good at creating very effective incentives. The competition among Chinese students to do well on the gaokao does in fact result in most of them acquiring real skills and knowledge. And the competition among local governments has in fact delivered a lot of GDP growth.

The centrality of the gaokao, and gaokao-like institutions, in China makes it all the more intriguing that the man now sitting on top of the political system seems less committed to them than many others. Xi Jinping belongs to the last generation of “worker-peasant-soldier” students admitted to university on the basis of political recommendations rather than exam scores. (This wasn’t an automatic result of his being the offspring of a senior Party leader. Joseph Torigian has documented that because Xi’s father had been purged during the Cultural Revolution, his initial applications to university starting in 1973 were rejected; but eventually a sympathetic administrator approved his admission.) This has always set Xi apart from other more “technocratic” figures in the government, like his former No. 2 Li Keqiang (see my old post on “The education of Li Keqiang“).

Xi famously oversaw the shuangjian or “double reduction” campaign launched in July 2021, which aimed at easing the “burdens” on students of excessive homework and after-school tutoring–one of the most high-profile attempts to address the downsides of the gaokao system in recent years. This got a lot more attention outside China than most such political campaign, because it essentially outlawed the business model of some publicly traded companies. While the rules remain on the books, recent evidence suggests that enforcement has been relaxed, and businesses are again being allowed to meet parents’ demand for tutoring, which remains very high. The exam system remains a fundamental social institution of today’s China, and as such very difficult to change.