China since 2012: recentralisation under Xi Jinping

Background

Xi Jinping became general secretary of the CCP in November 2012 and state president in March 2013. He inherited a party that its own leaders described as threatened by corruption, factionalism and loss of discipline; the fall of Chongqing party chief Bo Xilai in 2012 had exposed elite conflict. The new leadership's answer was to strengthen party control over state, economy and society and to concentrate authority at the top.

How it is governed

• Concentration at the centre. Xi chairs most of the central leading groups and commissions that coordinate policy on reform, security, finance, foreign affairs and cyberspace. The party designated him its "core" in 2016 and wrote "Xi Jinping Thought" into its constitution in 2017.
• Constitutional change. On 11 March 2018 the National People's Congress amended the Constitution by 2,958 votes to 2, with 3 abstentions. The amendment added to Article 1 that party leadership is "the defining feature of socialism with Chinese characteristics", created supervisory commissions at every level and removed the two-term limit on the state presidency. Xi began a third term as general secretary in October 2022 and as president in March 2023.
• Party over government. Reforms in 2018 and 2023 moved functions from State Council ministries into party bodies, reversing the reform-era aim of separating party and government.

Susan Shirk calls this "the return to personalistic rule" after decades of institutionalised collective leadership. Others emphasise that it operates through, rather than against, the party's organisation; the classification is debated.

Anti-corruption and discipline

The Central Commission for Discipline Inspection (CCDI) launched the largest anti-corruption campaign in the party's history, starting with the "eight-point regulation" on official conduct in December 2012. According to figures the CCDI released in October 2022, more than 4.6 million cases were filed in the first decade and 553 senior officials managed by the central Organization Department were investigated. Detention under the party's shuanggui procedure was replaced in 2018 by liuzhi detention by the new supervisory commissions, still outside the courts. Melanie Manion argues that the campaign differs from earlier ones in duration and reach and that it changed incentives in ways that are steps towards institutionalisation; critics see it also as a tool for removing rivals and disciplining the bureaucracy.

Surveillance and repression

• Digital surveillance. Camera networks, facial recognition, real-name internet rules and big-data policing expanded rapidly. Xu Xu finds that, across Chinese counties, the introduction of digital surveillance increased public security spending and arrests of political activists while reducing public goods provision, consistent with a shift from broad co-optation to targeted repression.
• Xinjiang. From 2017 the authorities detained large numbers of Uyghurs, Kazakhs and other Muslims in facilities the government calls "Vocational Education and Training Centres". Adrian Zenz, working from official documents, estimated between several hundred thousand and just over one million detainees. The UN Human Rights Office's assessment of 31 August 2022 found that accounts of detention in 2017–19 were marked by patterns of torture or other ill-treatment, described a large-scale surveillance system including the Integrated Joint Operations Platform, and concluded that the extent of arbitrary and discriminatory detention "may constitute international crimes, in particular crimes against humanity". The Chinese government rejects these findings and says the centres were counter-extremism training that has ended.
• Civil society and Hong Kong. Human-rights lawyers were detained in a nationwide sweep from July 2015; a 2017 law placed foreign NGOs under police supervision; and the National Security Law imposed on Hong Kong in June 2020 ended its protest movement and most of its opposition.

Feedback, responsiveness and local experimentation

Bottom-up experimentation has narrowed. Jessica Teets and Reza Hasmath find a notable reduction in policy experimentation under Xi, with disincentives to innovate locally as "top-level design" replaced pilots and anti-corruption scrutiny raised the risk of trying anything new; some pockets of experimentation remain.

At the same time the party has institutionalised channels for complaints. A field experiment published in 2016 found that about one-third of county governments answered citizen requests posted online, and that threats of collective action or of reporting to higher levels made them considerably more responsive. Research published in 2025 examines the national Local Leader Message Board and city petition sites. Howard Wang and colleagues, using 238,835 claims from the national board (2020–21) and 793,119 from Wuhan, find that complaints receive more responsiveness than other requests, driven by superiors' expectations and officials' wish to avoid punishment, and warn that this may deepen stability maintenance rather than support reform. Jieren Hu, Xuan Gong and Yun Zhao, drawing on interviews in 2022–24, describe "mechanical responsiveness": digital systems speed up handling, but officials under pressure often respond performatively to satisfy superiors rather than resolve problems.

Economy

The state reasserted itself through industrial policy (Made in China 2025, announced in 2015), stronger party committees in firms, a regulatory crackdown on large technology platforms in 2020–21 and the "common prosperity" agenda. China declared in 2021 that it had eliminated extreme poverty at its national threshold. "Zero-COVID" controls in 2020–22, including the Shanghai lockdown of 2022, ended abruptly in December 2022 after rare nationwide protests.

Outcomes

• Freedom. Freedom House rates China Not Free, 9/100 in 2026, with a political-rights score of -2/40.
• Centralised capacity. The campaign removed large numbers of officials and imposed discipline; the state can mobilise surveillance and control at unprecedented scale.
• Costs. Mass detention in Xinjiang, the end of Hong Kong's freedoms, reduced local initiative and policy reversals made without public deliberation.

Debates

• Personalism or party rule? Shirk and Minzner see a break with reform-era institutionalisation; others see a party-state reasserting Leninist norms.
• Anti-corruption or purge? The two are not exclusive; evidence on whether corruption has fallen in a durable way is limited.
• Can complaints substitute for accountability? The 2025 studies find genuine responsiveness to individual complaints but no route to contest policies or leaders.

Lessons for institution design

• Rules without enforcement can be repealed. Term limits were removed by a near-unanimous vote.
• Surveillance changes the bargain. Better information about citizens let the state substitute targeted repression for broad public goods.
• Centralisation trades initiative for control. Punishing local failure discourages local experiments.

Further reading

• Susan L. Shirk, "China in Xi's 'New Era': The Return to Personalistic Rule" (2018)
• OHCHR Assessment of human rights concerns in the Xinjiang Uyghur Autonomous Region (2022)
• Xu Xu, "To Repress or to Co-opt?" (2021)
• Howard H. Wang et al., "How institutionalized feedback works" (2025)
• Carl Minzner, End of an Era (Oxford University Press, 2018)