China | A new Silicon Valley

Behind Deepseek lies a dazzling Chinese university

It models itself on Stanford, and is in the tech hotspot of Hangzhou

A HUGE STATUE of Mao Zedong still stands near the entrance to Zhejiang University, surveying the transformation of the eastern city of Hangzhou,175km (110 miles) south-west of Shanghai. Such statues look anachronistic wherever they linger across China, but especially so here, and especially after the events of the past few months.

在调查了上海西南175公里(110英里)处的城市杭州, 尤其是在最近几个月的事件的转型后, 中国毛主席的雕像仍然位于浙大的入口处, 这让人们感到这座雕像的过时

The recent revolution in Hangzhou has not been a Maoist one, but a technological one. DeepSeek, the AI company whose groundbreaking models stunned foreign competitors last month, was developed in Hangzhou. It was founded by Liang Wenfeng, an alumnus of Zhe Da, as the university is known (from its Chinese name, Zhejiang Daxue). Zhe Da is still largely unknown outside China, but it is at the heart of a dynamic new ecosystem of researchers and entrepreneurs in the city that models itself on California’s Silicon Valley.

That ecosystem came into sharp focus on February 17th, when China’s leader, Xi Jinping, met a group of tech entrepreneurs in Beijing. Among them was Jack Ma, who founded Alibaba, an e-commerce giant, in Hangzhou in 1999, but then disappeared from view in 2020 after criticising the financial regulator. Another was Mr Liang, who studied artificial intelligence at Zhe Da and made his first forays into business there. China’s leader apparently bringing Mr Ma in from the cold so publicly, and meeting with Mr Liang, will only fuel Hangzhou’s growth.

Zhe Da has already grown fast in recent decades, buying high-end equipment and hiring top-tier scientists. Some 70,000 students and faculty now live and work across its seven campuses, in blocky buildings overlooking lakes and plum trees. Hangzhou and other cities have said they want to transform their universities into “innovation ecosystems”. Zhe Da has not only become a research powerhouse but, even more importantly, has been adept at turning bright young students into business leaders. It says it will be a “world-class” university by 2027. Alumni say it is modelling itself on Stanford University rather than China’s most prestigious colleges in Beijing and Shanghai.

This may sound fanciful, and on overall measures of reputation and student experience by the QS World University Rankings, Zhe Da ranks 47th globally. Yet by some metrics, the university has already eclipsed many of the world’s best. It now produces more scientific papers than any other university, according to the latest Leiden ranking, a measure of the volume of research output. It is behind only Harvard in producing papers deemed to be in the top 10% of their fields

Alumni are among the wealthiest entrepreneurs in China, according to rankings by Hurun, a research firm, of those with more than 5bn yuan ($700m) in assets. They include Colin Huang, the founder of Pinduoduo, an e-commerce giant, and Duan Yongping, an electronics tycoon.

In recent months the university’s reputation for enterprise has reached new heights. China is abuzz with talk of the “Six Little Dragons” of Hangzhou, a clutch of zippy startups, three of which were founded by Zhe Da alumni. DeepSeek is one. Another is Manycore Tech, a 3D-design software firm, which on February 15th announced a plan to list in Hong Kong this year. Then there is DEEP Robotics, which specialises in dog-like robots (pictured), used for patrolling and rescue operations.

Three factors lie behind the university’s success. The first is its ability to attract and foster talent. Although many students in China yearn for a stable government job, Zhe Da has long drawn more daring souls, who throw themselves into startup competitions where they can pitch business ideas to get university funding. Their professors are unusual, too, encouraging students to dabble across disciplines in order to spark ideas, and trying to foster what they call a “mistake-tolerant” atmosphere.

When Huang Chaoyu was in high school he built a temperature-controlled tank for his pet fish. Now a 21-year-old undergraduate studying materials science at Zhe Da, he has set up a company with his supervisor aiming to produce a kind of biological “glue” to help heal wounds. Pure research is important for breakthroughs, he says, but “to truly change society you need industry”. Mr Huang is part of a special class, first offered in 1999, that helps science students learn about entrepreneurship. It meets in a shared workspace where a sleek white robot stands near a poster calling on students to “dare to innovate”. On average, one fifth of the class starts a company within five years of graduating, say university officials. Alumni help them find funding, internships and contacts.

Many faculty members start companies, too. Chinese universities are typically uneasy about distractions from academic pursuits. But Zhe Da has been helping its scientists commercialise their findings for decades, says Jin Yiping, a university administrator. In 2009 it set up an institute dedicated to this purpose. DEEP Robotics is run by Zhu Qiuguo, a professor at the school of engineering. His colleague, Gao Chao, runs a company which makes textiles from graphene, an advanced material.

The second factor helping Zhe Da is location. Hangzhou is a liveable canal- crossed city just 45 minutes by train from Shanghai, but a long way from the politicians in the capital. After the Communist Party took power in 1949, state planners largely ignored it, leaving space for private firms to re-emerge when reforms began in 1978. Of the top 100 companies in Hangzhou, 82 are private, a high proportion for a large Chinese city. Its GDP per person is nearly double the national average.

That Mr Ma’s company, Alibaba, was founded here has also had an impact. The university describes the firm as a “good neighbour, partner and friend”. In 2017 Mr Ma made a big donation to a university hospital. In 2023 Alibaba donated its quantum-computing lab to the university. For nearly a decade Alibaba and the university have jointly run a research centre for what they call “frontier technologies” such as computer vision. It takes interns and post-doctoral students from the university and helps them find jobs in industry.

Third, Hangzhou officials are known for getting things done without asking for favours or fancy dinners, says Zhang Jie, an investor and Zhe Da alumna. This makes it easier for young graduates to start companies. Most government services can be used through an app, notes a local entrepreneur. Officials love tech firms. They offer startup founders with PhDs up to 15m yuan in funding if they move to the city.

Other universities are trying to emulate Hangzhou’s success. Tsinghua in Beijing is producing AI talent, much of which gets hired by DeepSeek. The South China University of Technology in Guangzhou has close links with China’s electric-vehicle industry. But Hangzhou’s mix is hard to replicate, notes Yao Yang, an economist at Peking University. And talent tends to cluster in a few spots, not disperse to many.

Going globel

Zhe Da still faces challenges in competing on a global level. One problem is that it relies largely on the government for funding. That has served it well so far, but leaves it at the mercy of changing official priorities and fiscal constraints. Alumni are generous but building up an endowment to match, say, Stanford’s $36bn is unlikely. And even then, the university must defer to officials on how to spend it.

Compared with its overseas peers, few of Zhe Da’s faculty or students are from outside China. It has poached elite researchers from American universities (including, last year, Sun Song, a star mathematician from the University of California, Berkeley) but they are typically of Chinese descent. Political tensions with the West do not help. Nor do the barriers to free speech common in China.

For all that, Zhe Da shows that “the tectonic plates of global higher education are shifting very dramatically”, says William Kirby, a China expert at Harvard Business School. In January China released a plan to become an “education power…with global influence” by 2035. Not long ago that goal would have looked overly ambitious. Now, places like Zhe Da have made it look surprisingly likely.

China | Bare branches

China’s alarming sex imbalance By 2027 one in six young Chinese men won’t be able to find a partner

“Of course I want to get married,” says Fu, a lorry driver in Yiyang, a far- flung county in Jiangxi province. Once a migrant worker, the 36-year-old returned to the village to live with his ageing parents. They are anxious for him to tie the knot. “But there are few women,” he sighs. The eligible girls around him are all spoken for; others have left to work in cities.

Fu’s plight is not uncommon. Men like him are known as guang gun—bare branches, unable to bear fruit. Their numbers began to increase more than a decade ago. But the scale of the problem is now becoming clear. The Economist has analysed data from the UN’s World Population Prospects, a biennial report, and from China’s 2020 census. The data reveal that the sex ratio—the number of men for every 100 women—among men aged 23-37 and women aged 22-36 will hit a peak of 119 by 2027. (Those are the ages between which 80% of each sex gets married—see chart 1.) It is then predicted to remain high for decades. In 2012 the ratio was just 105.

That means that in 2027 there will be 22.5m more men than women in those cohorts, by far the largest number of “surplus” young males ever recorded anywhere. What is more, the share of unmarried men aged 25-39 shot up from 13% to 30% from 2006 to 2022. This is an issue of huge concern for China’s rulers. It was brought about by the arrival in the 1980s of cheap ultrasound machines, which allowed parents across Asia to tell the sex of their unborn child. The widespread preference for sons opened the door to sex-selective abortions. In South Korea the sex ratio at birth hit a brief peak of 117 in 1994, before falling to 106 in 2012, where it has roughly remained. In India it was 109 as late as 2010 (in 2024 it was 107). In developed countries like America and Britain, it was around 105 in 2024.

In China the problem was made worse by demographic engineering. In 1973 the country began trying to reduce its population with the “later, longer and fewer” campaign. This was followed by the draconian one-child policy in 1979. Being given just one chance dramatically lowered couples’ chances of having a boy naturally, and further incentivised sex selection. As a result, China has suffered by far the worst imbalance in its sex ratio at birth (see chart 2).

Now that those boys are grown up, migration is adding to their woes. The 2020 census showed the sex ratio for young adults was 106 in urban areas and 120 in rural ones. Young rural women who move to the city often marry richer urban men. But changing social mores around marriage mean many better-educated urban women do not want to get married at all, let alone to rural men, shrinking the dating pool yet further. Statistics released on February 8th showed that the number of marriages in China in 2024 fell by 20% from the previous year, to 6.1m. That is less than half the number registered in 2013, and the lowest number since the 1980s.

Some guang gun are resigned to their fate. Guo, a 38-year-old musician in Shangrao, near to Fu’s village, has been on a couple of blind dates. On one, he met an overachiever who studied in Germany and worked as a manager for a car company. It was “useless”, he says. On another date, the woman “scrutinised my family background as if it were a business deal”.

As women grow scarcer they also become more valuable. Bride prices (a payment from the groom’s family to the bride’s to seal the marriage) have soared. One survey in Liaoning, a north-eastern province, found that the bride price in rural areas jumped from 68,000 yuan ($9,000) in 2016 to 176,000 yuan in 2020, adjusting for inflation. Another, across 11 provinces, found that costs of marriage (including expenses like housing and matchmaking) for rural males were 7.6 times higher after 2010 than before 2000.

The shortage of women has had other side-effects. Between July and December of 2018 the Chinese government, working with police in Myanmar, Cambodia, Laos, Vietnam and Thailand, jointly rescued 1,130 foreign women who had been trafficked into China to be married. The Communist Party, obsessed with social stability, also worries a lot about rootless young men with no family prospects.

The ratio in young adults will remain above 115 well into the 2040s. The main hope is that, since peaking in the mid-2000s, the sex ratio at birth has declined. Cultural changes and female economic empowerment have chipped away at couples’ preference for sons. Eventually, the numbers of young men and women could equalise. But that will come as little consolation for today’s rural bachelors.

China | Cheating in exams

Chinese authorities try to stop parents gaming the exam system (again) They will go to great lengths to try February 20th 2025

An ancient Chinese saying states that meng mu san qian: the mother of Mencius, a sage, “moved three times” to find him the right study environment. Tiger mums the world over know the feeling, but few push harder than the Chinese; the game of cat and mouse the Communist Party plays with parents to stop them trying to game the system is legendary. The party has now released new rules to try to stop it.

In theory, any of the 13m students who take the annual gaokao entrance exam can qualify to attend university if they achieve a good score. In practice, because of the skewed way places are allocated by region, the competition is fiercer in populous provinces. So a small number of parents try to move

That is not always easy. To enroll your child at a school in another province might require buying a house, finding a new job, or bureaucratic wrangling to change your hukou, or household registration. So there is a more temporary way by which some parents try to cheat the system. They keep their child attending the same school but, using bribes or connections, arrange for them to sit the gaokao exam in a province where they do not live. In one notorious case in 2021 the child of a head teacher in populous Hebei province (with about 600,000 test-takers) took the gaokao in Tibet (competing with just 40,000 locals). Public outrage ensued.

A grey industry has emerged to help with these tricks, which are known as “gaokao migration”. Consultancies offer to smooth things over with local officials and schools, for a fee. One such firm in Henan province told The Economist that for 60,000 yuan ($7,700) they could arrange for a student to sit the exam in Hainan, a rural southern province.

Worried about the anger that cheating and educational inequality creates, on February 7th the education ministry issued new rules demanding schools send a report on each student twice a year, to make sure all are studying and taking exams where they should be.

It would help if China relaxed regional quotas for universities or scrapped them altogether. But resistance from places which benefit from the current system makes that all but impossible, says Li Hongbin of Stanford University. One such place is Beijing, home to many rich and powerful figures. In 2012 reform-minded officials proposed allowing more people to take the gaokao in the capital. But Beijingers protested that it would make the exam more competitive for their children. After much ado, the reform was stifled. Parents will continue to act like the mother of Mencius.