VoxEU Column Labour Markets

Has China’s new labour contract law worked?

Formal contracts are key to enforcing workers’ rights. This column presents evidence that the Labour Contract Law improved worker outcomes in China, especially among migrant and low-wage workers. The proportion of workers with social insurance increased without any significant negative effect on the employment rate.

In 2007 China enacted a new Labour Contract Law (LCL) – the first major labour reform in over a decade. The law sought to pressure firms to give workers written contracts that would help workers enforce their legal rights at the workplace. Because local governments put economic growth and business interests above worker well-being, implementation of labour laws in China has historically been weak (Tang 2008) with the result that many workers in China suffered ill treatment by employers (Lee 2007, Chan 2001). Many firms did not give contracts to the hundreds of millions of workers with rural hukou, nor pay legally mandated social benefits, nor give workers other legally guaranteed rights. Omitting social insurance payments could save firms 20% to 30% of labour costs. Paying migrants and other low-wage workers less than their full wage or legally required overtime pay further contributes to the profit margin. Without a written contract, workers have difficulties making a case to the labour dispute arbitration commissions that help resolve disputes with management. Many workers responded with strikes and public protests.

After a heated open debate, the People's Congress enacted the LCL to take effect on 1 January 2008. The signature features of the law were articles that strengthened the legal requirement that firms give workers a written contract. The government hoped that workers could use the contract to protect their rights themselves rather than relying on local officials or labour bureau inspectors (Cooney et al. 2013). Some analysts believed that the debate surrounding the law increased worker awareness of their rights, which would strengthen demands for compliance (Gallagher and Dong 2011). The support of Beijing and of the official All China Federation of Trade Unions (a major backer of the legislation) meant that local governments could no longer ignore illegal treatment of workers. The law identified groups responsible for enforcement and penalised non-complying employers, requiring them to give a worker denied a legal contract double pay each month that the worker lacked the contract. Still, given the history of firms and local governments ignoring labour law, many analysts doubted that the 2007 LCL would improve worker well-being. Conservative opponents of the law feared that if it were effective, it would cause more harm than good by raising costs of labour and reducing employment.

To assess the effect of the new law on migrant and other low-wage workers, researchers have examined:

  • the likelihood that workers obtained individual contracts before and after the law;
  • the link between having an individual contract and receiving social insurances; and
  • the effect of gaining a contract on receiving social insurances after the law was passed.

Analysing datasets covering different geographic areas and groups of workers, many researchers (including us) have concluded that, contrary to the view that the LCL would be ineffective or harmful, the LCL raised the proportion of workers with individual contracts and with social insurances, while possibly inducing firms to outsource more work to ‘dispatched workers.’

The four points below summarise the evidence behind these conclusions.

1. The law and workers with written labour contracts

The first place we look for impacts of the LCL is on the contract status of workers. Several datasets show that the proportion of workers with contracts increased after the law:

  • The China Household Income Project (Gao et al. 2012)
  • The China Urban Labour Survey (Gallagher et al. 2013)
  • The China General Social Survey (Zhou 2013)
  • The Rural-Urban Migration in China survey (Cheng et al. 2013)
  • The Pearl River Delta (PRD) migrant workers survey (Freeman and Li 2013)

Line 1 of Table 1 gives our before/after contrast in the PRD survey. In 2006 40% of migrant workers had a contract while in 2008 (2009) 59% (60%) had contracts.

Table 1 Percentage of workers with coverage and social insurance in Pearl River Delta surveys, by year

Source: Table 1 from Freeman and Li (2013).

Regression analysis shows that the increase in coverage reflects a genuine before/after difference for workers with the same characteristics. To pin down causality further we estimated the proportion of workers without a written labour contract who obtained a first contract in every month from 2006 through 2008 – the ‘hazard rate’ for getting a contract. Figure 1 shows a jump in the hazard around the January 2008 implementation of the law. The hazard rate increased from 1.1% per month in 2006-2007 to 7.1% per month in 2008.

Figure 1 Monthly hazard rates for workers gaining first contracts

Source: Figure 1 from Freeman and Li (2013). The hazard rates of getting contracts in a particular month is the ratio of the number of workers getting a first contract in a month divided by all workers without contracts that month.

We further decomposed the increased contract coverage into:

  • the part due to a worker employed by the same firm gaining a contract before and after the law; and
  • the difference in coverage between workers who obtained their first job after the law and before the law, and between workers moving from firms without coverage to those with coverage.

The largest factor was a change in policies for workers employed at the same firm. Consistent with this, Cheng and Yang (2010) found in a survey of human resource managers that the LCL ‘significantly raised the signing rate of contracts, extended the term of contracts, and increased the amount of non-fixed term labour contracts'.

2 Contract coverage and social insurance

The government hoped that written contracts would increase the likelihood that workers would gain legally required social insurance and other protections. All of the datasets show that workers with a contract are more likely to have social insurance than otherwise comparable workers without a written contract. Our estimate is that contracts are associated with about a 15-20% higher chance that PRD migrants received old-age insurance/social security, arguably the most important social insurance. We also found that workers with contracts were less likely to suffer wage arrears and more likely to be unionised.

3. Workers with new contracts gained social insurances

The increase in contract coverage and cross section relation between having a contract and social insurance suggests that the LCL increased social insurances as well as contract coverage. Indeed, there was a substantial before-the-law to after-the-law increase in social insurances. First, as lines 2-5 of Table 1 show, the proportion of migrant workers in the PRD who obtained the four forms of legally required insurance jumped by 12 to 19 percentage points between 2006 and 2008-09. Because the survey did not ask workers if they had social insurance on previous jobs, we could not measure whether the increase in social insurances occurred primarily among those who gained a contract. On the basis of the characteristics of persons having social insurance we estimated separately the likelihood that workers who gained a first contract after the law and the likelihood that those who did not gain a contract had social insurances before the law. Using these estimates to measure insurance before the law we found that the increased contract coverage accounts for most of the increase in social insurance. Consistent with our results, changes in contract status and insurances in panel data show that workers who gained first contracts, particularly first term contracts, were especially likely to gain social insurance (Gao et al. 2012, Table 3).

4 To circumvent the law firms outsourced work to dispatched workers rather than cut employment

Since the global recession dominated changes in employment in China in winter 2008-09, it is difficult to find any effect of the law on aggregate employment through a before-after design. Accordingly we rely on surveys of managers to assess potential adverse employment effects. The PRD survey asked:

  • ‘Did your firm carry out large scale layoffs?’; and
  • ‘Did your firm force the termination of existing employment contract and reset new employment relations?’

In the 2008 survey only 4.6% of respondents reported large layoffs in their firm since 2007 and only 1.68% reported that their firm terminated and reset employment contracts. A People's Bank survey of manufacturing firms found that nearly twice as many managers said the new law reduced firing (31%) than said it reduced hiring (16%) (Gallagher et al. 2013, Table 10). Nothing in these reports suggests that the law caused great job losses.

Firms and government agencies seemingly responded to the new law by shifting work to temporary help agencies that dispatched workers to the firms, strengthening an upward trend in such workers. But Yu Zhou’s (2013) analysis of dispatched workers in the 2008 and 2010 China General Social Survey shows that approximately the same proportion of dispatched as of other formal sector workers had contracts. In any case, the People's Congress was sufficiently concerned about the growth of dispatched workers that in December 2012 it added provisions to the law to regulate subcontracting.


Despite the generally weak implementation of labour laws in China, the Labour Contract Law increased contract coverage of migrant and other low wage workers and raised the proportion of workers with legally mandated social insurances, without any apparent adverse impact on employment. Although many migrant workers and others remain without written contracts or social insurances, the Labour Contract Law made some headway toward its goal of spreading the rule of law to China's labour market. In a period when the situation of workers in many countries has stagnated or deteriorated, the LCL helped improve the position of workers in China.


Chan, A (2001), ‘China's workers under assault: the exploitation of labour in a globalizing economy”, ME Sharpe.

Chang, S (2009), China's Economic System, CITIC Publishing Press.

Cheng, Y-y and Y Liu (2010), “The Impact of Labour Contract Law’s Enforcement on Chinese Enterprises’ Human Resources Management,” Economic Theory and Business Management 7, pp. 66-73.

Cheng, Z, R Smyth and F Guo, (2013), “The Impact of China’s New Labour Contract Law On Socioeconomic Outcomes for Migrant and Urban Workers” Monash University Economics Discussion Paper 51/13

Cooney, S, S Biddulph and Y Zhu (2013), Law and Fair Work in China (Vol. 93). Routledge.

Freeman, R and X Li (2013), “How Does China's New Labour Contract Law Affect Floating Workers?” NBER Working Paper No. 19254.

Gallagher, ME and B Dong (2011), “Legislating Harmony Law Reform in Contemporary China,” in From iron rice bowl to Informalization, Cornell University Press (14), pp. 36.

Gallagher, M, J Giles, A Park and M Wang (2013), “China's Labour Contract Law: Implementation and implications for China's workers”, Human Relations, pp. 1-39.

Gao, Q, S Yang, S Li (2012), “Labour contracts and social insurance participation among migrant workers in China”, China Economic Review 23, pp. 1195-1205

Lee, C K (2007), Against the law: Labour protests in China's rustbelt and sunbelt, University of California Press.

Tang, K (2008), "The Evolution of Labor Contract Law and Comments on Two Papers", China Currents 7(3).

Zhou, Y (2013), "A Pluralistic Paradigm of Chinese Employment Models", presentation at NBER, LWP-Harvard U: China Economy Seminar.