作者：周敏
2026年初，年关将至。山西，一名农民工因讨薪多次无果，选择了最极端的报复方式——将拖欠他工资的雇主全家杀害。同一时期，福建泉州，一名包工头讨要工程款不得，将自己的汽车开到工地，引火自焚；泉州惠安，另一批被拖欠工资的工人，则选择烧掉了欠薪企业的仓库。
这不约而同的疯狂，渐渐成为可以被预测、被统计的社会现象。据民间记录项目”昨天”统计，2026年春节前那段时间，平均每天收集到的集体讨薪事件超过两百起；仅2026年4月，该项目记录的全国抗争事件就有七十四起，其中劳动者维权占比超过三分之一。这一年的讨薪潮，被多家观察者形容为”比往年更惨烈”——建筑、制造、医疗、公交，几乎所有行业都卷入其中，天津、广东、陕西等十余个省市无一幸免。
灭门与自焚，只是这条绝望光谱上最极端的两个刻度。在它们之间，是更大规模、却已经被这个社会”正常化”了的日常图景：工人爬上塔吊、爬上楼顶，以跳楼相威胁；包工头堵住工厂大门；讨薪者挤满劳动局、信访局的接待大厅，场面混乱而无人能够真正解决问题。
这一切，并非发生在某个法治真空、政府缺位的失序地带。而是一个拥有全套劳动法律、专项治理通知、年度考核机制、应急救助基金的国家，年复一年上演的固定剧目。
中国特色讨薪是被制度催熟的的绝望。把视野拉长到过去几年，会发现这套剧本的同一套台词。2024年，四川宜宾一名工人因讨要800元工资多次无果，纵火烧毁工厂，被网民称为”800哥”；当地官方的回应，是将其纵火行为定性为轻生厌世，而拖欠他的5370元工资，”正在走审批流程”。这句”正在走审批流程”俨然成了官方话术的标本。标本下的使用说明写着：在今日之中国，工人讨薪要作为紧急议题被一脸严肃地举起，然后再无限期地正常化、流程化、合法化。
更早之前，湖南耒阳一名农民工因雇主拖欠1560元工资，最终杀害对方一家三口；这桩2012年的旧案，在2024年又一次被网络重新传播——人们之所以反复提起它，并非出于猎奇，而是因为类似的故事，十几年来始终不曾真正停止。
这种以命相搏、以暴制暴的讨薪方式，已经成为一种可以被命名的现象：人们称之为中国特色讨薪——它有固定的季节性（年关爆发），固定的舞台（塔吊、楼顶、工厂大门），固定的诉求方式（喊口号、爬高处、堵大门），以及，几乎固定的结局：工资不一定能拿得到手，讨薪者却可能先一步被警方以扰乱公共秩序、寻衅滋事为名拘留。
国家在场，问题依旧。这是一场”治理”了二十年的顽疾。
2023年，国务院修订印发了《保障农民工工资支付工作考核办法》，将欠薪治理成效纳入对地方政府的政绩考核；几乎每到冬季，国务院就业促进和劳动保护工作领导小组办公室都会下发通知，开展治理欠薪冬季行动，要求摸排风险、公开举报渠道、应急处置。建筑行业还建立了农民工工资监管账户和”代发”制度，理论上由施工方直接发放工资，绕开层层转包中的克扣环节。
个案确实存在成功的例子——安徽滁州一名农民工和五十五名工友,在被拖欠四年、超过八十万元工资后，最终经协调拿到了钱。这些稀罕事会出现在官方媒体的年终总结里，作为治理成绩。
但制度的设计者自己也清楚，这些举措相当地隔靴搔痒。2026年的一篇官方背景调研报道直言不讳：基层人社干部反映，《保障农民工工资支付条例》虽然对压实各方责任有明文规定，但各部门在执行中存在偏差，行业主管部门之间配合参差不齐，”中梗阻”现象普遍存在。换句话说，他们自己都承认，问题不是缺少法律条文，而根本是条文从没真正执行过。
耐人寻味的是建筑工人在网络上的自问自答：”如今2026年了，国家建立了监管账户，实行了代发制度……怎么还有农民工讨要工资呢？”答案是：班组每月上报的工资额度，往往远低于工人实际收入。剩下的部分，还得等老板自己收到工程款后补发。更恶劣的情况下，层层分包的小包头甚至会把已经代发到工人卡里的工资，重新收回去——制度的漏洞，反而成了新一轮克扣的工具。
这套治理体系最值得审视的地方，在于它的功能错位：当局投入大量精力，要求对欠薪引发的”负面舆情”联动处置，对”恶意炒作、挑动社会对立情绪的”内容及时管控，防止发酵蔓延。很明显,一部分本该用于解决欠薪的治理资源，实际上被用于了管控舆论。讨薪问题之所以年复一年地”治理”却始终未能消失,根源正在于此——它压根没被当作一个需要被解决的劳资纠纷,而是被当作一个需要被管控的舆情风险。
那么，有人替工人说话吗？为什么一个拥有如此完整治理体系的国家，会让讨薪沦为一场年复一年的血泪剧？大约有三重结构性的缺位。
第一重缺位，是没有真正属于工人的组织。
中国的工会体系，名义上覆盖了绝大多数企业，但是，它的任命决定了它的立场——基层工会主席往往由企业方指定或默许，而非工人自主选出、对工人负责的代表。当冲突发生时，工人举目无亲，不得不单枪匹马与资方周旋，没有一个独立的、依法享有罢工权与集体谈判权的组织可以为他们撑腰。这与许多市场经济体，比如美国，形成了鲜明对照：在美国等地，工会虽然力量有起伏，但其受劳动法保护的集体谈判权，本身就是一种真实的制衡力量。
第二重缺位，是地方政府的双重身份——既是裁判员，又是利益相关方。地方政府往往与本地企业捆绑着招商引资、税收、就业指标等共同利益，在处理劳资纠纷时，很难做到真正中立。经济下行压力之下，为了保住企业能提供的就业岗位，地方政府在处理纠纷时，时常倾向于优先保企业、而非保工人——这也是为什么，浙江桐乡某光电公司千余名员工因欠薪罢工时，员工反映”市政府、镇政府都不出面，还阻拦我们，抓了十几个人”，而不是出面协调企业偿付欠薪。
第三重缺位，是司法救济本身的失灵。就算工人愿意走法律途径，胜诉也未必意味着能拿到钱——一旦欠薪企业已经破产、无力支付，法院的判决便只是一纸空文，执行照样会卡死。诉讼成本高、周期长、执行难，许多工人根本不敢、也不愿走司法这条路，最终只能回到最原始危险的方式：用肉体去要挟——爬上塔吊，威胁跳楼。
三重缺位叠加之下，工人被推到了一个几乎没有合法、有效救济渠道的境地。极端事件，无论是以自己的生命相要挟的跳楼讨薪，还是转向施害者的灭门与纵火，是制度空转下，被反复证明正常渠道走不通之后，逻辑上几乎必然出现的产物。
而被牺牲的，往往是最沉默的那群。
经济下行的代价，最先扑向的，是产业链条最末端、议价能力最弱的那群人。当房地产企业债务违约、地方政府财政紧张、制造业订单萎缩,这些宏观层面的风险，层层传导，雪花飘落般盖在了具体的某个农民工、某个外卖骑手、某个包工头身上——他们没有能力对冲风险，没有制度为他们兜底，只好用身体去赌一个本不该由他们承担的代价。
2026年的这一连串灭门、自焚、跳塔事件，是一套长期存在的制度性逻辑，在经济下行周期里被进一步放大后的必然显影。当一个社会的治理资源更多地投向“如何让讨薪者闭嘴”，而非如何让欠薪者付钱；当讨薪比欠薪还承担更高的法律风险；当工人除了以命相搏之外，找不到任何一条真正畅通的维权渠道——我们应该再对下一起灭门案、下一次跳楼跳桥感到意外吗？
编辑：李晶
校对：王滨
翻译：戈冰
When Wage-Seekers Become Criminals: The Institutional Impasse of Wage-Claiming with Chinese Characteristics
By Zhou Min
At the beginning of 2026, as the Lunar New Year approached, a migrant worker in Shanxi, having repeatedly failed to claim his back pay, chose the most extreme method of retaliation—murdering the entire family of the employer who owed him money. Around the same time in Quanzhou, Fujian, a subcontractor, unable to obtain his project funds, drove his car to the construction site and immolated himself; in Huian, Quanzhou, another group of workers whose wages were withheld chose to burn down the warehouse of the defaulting enterprise.
This synchronized madness is gradually becoming a predictable, statistical social phenomenon. According to statistics from the civil documentation project “Yesterday,” during the period leading up to the 2026 Spring Festival, more than two hundred collective wage-claim incidents were collected on an average day; in April 2026 alone, the project recorded seventy-four protest incidents nationwide, with labor rights defense accounting for more than one-third. The wage-claiming wave of this year was described by multiple observers as “more tragic than in previous years”—construction, manufacturing, healthcare, and public transit, almost all industries were engulfed, and more than ten provinces and municipalities, including Tianjin, Guangdong, and Shaanxi, were spared without exception.
Annihilation of entire families and self-immolation are merely the two most extreme markers on this spectrum of despair. Between them lies a larger-scale, yet already “normalized” everyday picture within this society: workers climbing cranes and rooftops, threatening to jump to their deaths; subcontractors blocking factory gates; wage-seekers packing the reception halls of labor bureaus and letters-and-visits bureaus, creating chaotic scenes where no one can truly resolve the problems.
All of this does not take place in some lawless vacuum or an unregulated zone devoid of governance. Instead, it is a fixed drama staged year after year within a state that possesses a complete set of labor laws, special rectification notices, annual assessment mechanisms, and emergency relief funds.
Wage-claiming with Chinese characteristics is a despair ripened by the institutional system. Extending the scope of vision to the past few years reveals the exact same lines from this identical script. In 2024, a worker in Yibin, Sichuan, after repeatedly failing to claim 800 yuan in wages, set fire to a factory and was dubbed “Brother 800” by netizens; the local official response was to characterize his arson as a suicidal aversion to the world, while the 5,370 yuan in wages owed to him was “currently undergoing the approval process.” This phrase, “currently undergoing the approval process,” has virtually become a specimen of official discourse. The instructions for use beneath the specimen read: In today’s China, worker wage-claiming must be held up with a solemn face as an urgent issue, only to be indefinitely normalized, proceduralized, and legalized.
Even earlier, a migrant worker in Leiyang, Hunan, because an employer withheld 1,560 yuan in wages, ultimately murdered three members of the counterparty’s family; this old case from 2012 was recirculated online once again in 2024—people bring it up repeatedly not out of curiosity for the bizarre, but because similar stories have never truly ceased over more than a decade.
This style of wage-claiming, where lives are wagered and violence is countered with violence, has already become a named phenomenon: people call it wage-claiming with Chinese characteristics. It possesses a fixed seasonality (erupting at the turn of the year), a fixed stage (cranes, rooftops, factory gates), fixed methods of appeal (shouting slogans, climbing to high places, blocking gates), and an almost fixed ending: wages are not necessarily obtained, yet the wage-seekers may find themselves detained beforehand by the police under the names of disrupting public order or picking quarrels and provoking trouble.
The state is present, yet the problem remains. This is a chronic illness that has undergone “governance” for twenty years.
In 2023, the State Council revised and issued the Measures for the Evaluation of the Work of Guaranteeing Wage Payments to Migrant Workers, incorporating the effectiveness of wage-arrears governance into the political performance evaluations of local governments. Almost every winter, the Office of the State Council Leading Group for Employment Promotion and Labor Protection issues notices to launch a winter campaign for governing wage arrears, requiring risk screening, the disclosure of reporting channels, and emergency response disposal. The construction industry has even established supervision accounts and “agency payment” systems for migrant worker wages, theoretically allowing the construction party to distribute wages directly, bypassing the deduction links within tiered subcontracting.
Individual successful cases do exist—a migrant worker and fifty-five fellow workers in Chuzhou, Anhui, after being owed more than 800,000 yuan in wages for four years, ultimately received their money through coordination. These rare occurrences appear in the year-end summaries of official media as governance achievements.
But the designers of the system themselves are well aware that these measures scratch the surface from outside the boot. An official-backed investigative report in 2026 stated bluntly: primary-level human resources and social security cadres reflected that although the Regulations on Guaranteeing Wage Payments to Migrant Workers contain explicit provisions on consolidating the responsibilities of all parties, deviations exist among various departments during implementation, coordination between competent industry authorities is uneven, and the “middle obstruction” phenomenon is prevalent. In other words, they themselves admit that the problem is not a lack of legal clauses, but fundamentally that the clauses have never been truly executed.
What invites reflection is the self-questioning and answering by construction workers on the internet: “Now that it’s 2026, the state has established supervision accounts and implemented the agency payment system… how are there still migrant workers claiming wages?” The answer is: the wage quotas reported monthly by work shifts are often far lower than the actual income of the workers. The remaining portion still must wait to be backpaid after the boss himself receives the project funds. Under even worse circumstances, small subcontractors within the tiered subcontracting system will even recollect the wages that have already been paid via agency into the workers’ cards—the loopholes in the system, conversely, become a tool for a new round of deductions.
The aspect of this governance system most worthy of scrutiny lies in its functional dislocation: the authorities invest a vast amount of energy requiring the joint disposal of “negative public opinion” triggered by wage arrears, and the timely management and control of content that “maliciously hypes or instigates social antagonistic sentiments” to prevent it from fermenting and spreading. Clearly, a portion of the governance resources that should have been used to resolve wage arrears is, in reality, utilized to manage and control public opinion. The root cause of why the wage-claiming problem is “governed” year after year yet never disappears lies precisely here—it is not treated at all as a labor dispute that needs to be resolved, but rather as a public opinion risk that needs to be managed and controlled.
So, is there anyone speaking up for the workers? Why does a country with such a complete governance system allow wage-claiming to degenerate into a year-after-year drama of blood and tears? There are roughly three layers of structural absence.
The first layer of absence is the lack of organizations that truly belong to the workers.
China’s trade union system nominally covers the vast majority of enterprises; however, its appointments dictate its stance—primary-level trade union chairs are often designated or tacitly approved by the enterprise management, rather than independently elected by the workers as representatives accountable to them. When conflicts occur, workers find themselves utterly helpless and are forced to deal with capital single-handedly, with no independent organization that legally enjoys the right to strike and collective bargaining to back them up. This stands in sharp contrast to many market economies, such as the United States: in places like the US, although the strength of trade unions fluctuates, their collective bargaining rights protected by labor law are inherently a real countervailing power.
The second layer of absence is the dual identity of local governments—acting as both the referee and a stakeholder. Local governments are often bound together with local enterprises by shared interests such as investment promotion, tax revenue, and employment targets, making it very difficult to remain truly neutral when handling labor disputes. Under downward economic pressure, in order to preserve the jobs that enterprises can provide, local governments, when handling disputes, frequently tend to prioritize protecting the enterprise over protecting the workers—this is also why, when more than a thousand employees of a certain optoelectronics company in Tongxiang, Zhejiang, went on strike over withheld wages, the employees reflected that “the municipal government and the township government did not show up at all, and even obstructed us and arrested over a dozen people,” instead of stepping forward to coordinate the enterprise’s repayment of the back pay.
The third layer of absence is the failure of judicial relief itself. Even if workers are willing to take the legal route, winning a lawsuit does not necessarily mean they will get the money—once a defaulting enterprise has gone bankrupt and is unable to pay, the court’s judgment is merely a scrap of paper, and enforcement will still be stuck completely dead. With high litigation costs, long cycles, and enforcement difficulties, many workers simply dare not and do not want to take the judicial path, ultimately leaving them no choice but to return to the most primitive and dangerous method: using their physical bodies as leverage—climbing up cranes and threatening to jump to their deaths.
Under the compounding effect of these three layers of absence, workers are pushed into a predicament where almost no legal or effective channels of relief exist. Extreme incidents, whether it is jumping from buildings to claim wages by wagering one’s own life, or turning upon the perpetrators through annihilation of families and arson, are the products that emerge with almost logical certainty under the idle spinning of the institutional system, after normal channels have been repeatedly proven to lead nowhere.
And those who are sacrificed are often the most silent cohort.
The price of the economic downturn is the first to pounce upon those at the absolute end of the industrial chain with the weakest bargaining power. When real estate enterprises default on debts, local government finances become tight, and manufacturing orders shrink, these macro-level risks transmit downward layer by layer, falling like snowflakes upon a specific migrant worker, a specific food delivery rider, or a specific subcontractor—they have no ability to hedge against the risks, and no institutional system to provide them with a safety net, so they have no choice but to use their bodies to gamble against a price that should not have been borne by them in the first place.
This succession of family annihilations, self-immolations, and tower-jumping incidents in 2026 is the inevitable manifestation of a long-existing institutional logic further amplified during an economic downward cycle. When a society’s governance resources are cast more toward “how to make the wage-seekers shut up” rather than how to make the wage-defaulters pay; when wage-claiming carries higher legal risks than withholding wages; when workers can find no truly unblocked channel to defend their rights other than wagering their lives—should we still feel surprised by the next family annihilation case, or the next jump from a building or a bridge?
Editor: Li Jing
Proofreader: Wang Bin
Translator: Ge Bing
