Abstract
The economic restructuring in China over the past decade has resulted in displacement of millions of workers who had been employed in the state sector. This has posed tremendous challenges economically, socially, politically, and culturally. For several years, Chinese policies attempted to cushion the shock by requiring state-owned enterprises to provide living allowances and reemployment services to workers who had been displaced. There have been few empirical studies that have tracked the experiences of these displaced or xiagang workers. This study uses survey data from two large industrial cities covering the period 1998–2000 to analyze the labor market situation of over 2,000 workers 2 years after they had been observed as displaced and unemployed. The findings point to the high rates of labor force withdrawal and a low proportion who find another wage job in the formal sector. It also documents the large number of workers who find work in the informal sector which seems to act as an important safety net. Not surprisingly, education is an important determinant of post-layoff labor market outcomes. Active labor market interventions do not seem to make a substantial difference although there is some evidence from the duration analysis that training does help workers find employment more quickly than they would have otherwise.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
The set of variables and their definitions that are used for the analyses in this paper are listed in Appendix A in Betcherman and Blunch (2006).
In an earlier note, we had found that the complete sample assembled for the training evaluation had been contaminated by a substantial fraction of workers who actually were working in July 1998 (Betcherman et al. 2002). The initial sample restriction we have imposed for this analysis, then, was to exclude all those who were employed in July 1998. This sample restriction was essential for undertaking a time-bound analysis of the labor market experiences of unemployed workers. However, it should be recognized that, by excluding those workers who had been laid-off but had been successful in finding work before t 0, the estimation sample does not include a group of xiagang workers who apparently were more employable than the group of laid-off workers as a whole. As a result, the figures reported later in this paper underestimate the true reemployment probabilities of the population of laid-off workers in Shenyang and Wuhan over the period covered by the data.
For more detail, including the means and t-tests of the variables in the included and excluded observations, see Betcherman and Blunch (2006).
Retirement ages are 55 for males and 50 for females. At this age, xiagang workers move from the SOE to the formal pension system. Thus, in principle, they would not be included in the original sample frames. It should be noted that early retirement was part of the retrenchment strategy of SOEs.
Only 0.3% reported they were employers. Since such a small fraction would lead to very imprecise estimates, we will disregard this group in our econometric analysis.
Since in this sub-section we are concerned with the overall quality of work between these two sectors, we use the full sample for this part of the analysis, i.e. disregard the issue of whether workers were really displaced in July 1998.
The regression results on employment and self-employment are presented in Betcherman and Blunch (2006).
There may potentially be problems of endogeneity for several of the explanatory variables used in the models estimated in this section. Accordingly, we undertook sensitivity analyses by estimating a restricted version that excludes variables that are potentially endogenous—including training related variables (where applicable), whether there are other employed members in the household, presence of children in the household, and job search methods. The results confirm that the results are robust both in terms of direction and magnitude and, hence, indicates that endogeneity should not be too much of a concern for these analyses. These results are not presented here but are available upon request.
As suggested by a referee, this isn’t really a problem if one is studying the reemployment of laid-off workers per se. Rather, it merely becomes a matter of interpretation: the real problem with having a mix of laid-off workers who may or may not want to return to work enters most heavily when one considers the role of training (discussed later).
For example, while the CPHM imposes the proportional hazard (PH) assumption, the parametric Weibull models impose both the PH assumption and the assumption that the baseline hazard is monotonic. The CPHM therefore effectively allows the form of the duration dependency to be free, which we prefer to imposing a restrictive—and possibly wrong—assumption on the baseline hazard (although this may come at a cost of slightly less precise estimates, due to the looser structural form assumptions). Indeed, it has been suggested that only rarely would there any reason for social scientists to prefer a parametric model over the CPHM, mainly due to the strong theory required to impose the added structure from a parametric specification (Box-Steffensmeier and Jones 2004, pp. 66–67). As pointed out by a referee, the discrete logit hazard has actually no distributional or parametric assumptions and would place even fewer restrictions on the data. Since the proportional hazards assumption appears to be satisfied here, we prefer the slightly more restrictive CPHM—especially due to the more precise estimates, it produces.
It should be noted that the dataset in Giles et al. (2006a) contains all dislocated workers rather than, as here, only workers laid off from state sector employment. Also, the focus of Giles et al. (2006a) is specifically on the role of extended family networks, while we are focusing on other types of networks, including other personal contacts and re-employment centers, as well as on the role of training. The two studies are therefore not fully comparable.
The results are not presented here but are available upon request.
With the caveat that the earnings analyses were only included in Bidani et al. (2004).
References
Appleton S, Knight J, Song L, Xia Q (2004) Contrasting paradigms: segmentation and competitiveness in the formation of the Chinese labour market. J Chin Econ Bus Stud 2:185–205
Betcherman G, Blunch N-H (2006) The limited job prospects of displaced workers: evidence from two cities in China. Social Protection Discussion Paper No. 0613, Human Development Network, The World Bank, Washington, DC
Betcherman G, Dar A, Blunch N-H (2002) China retraining evaluation: some further analysis. Mimeo, Human Development Network, Social Protection, The World Bank, Washington, DC
Bidani B, Betcherman G, Blunch N-H, Dar A, Goh C, Kline K, O’Leary C (2004) Has training helped employ the xiagang in China? A tale from two cities. Report No. 24161-CHA, Poverty Reduction and Economic Management Unit, East Asia and Pacific Region, World Bank, Washington, DC
Bidani B, Blunch N-H, Goh C, O’Leary C (2008) Evaluating job training in two Chinese cities. J Chin Econ Busi Stud (forthcoming)
Box-Steffensmeier JM, Jones BS (2004) Event history modeling—a guide for social scientists. Cambridge University Press, Cambridge
Breslow NE (1974) Covariance analysis of censored survival data. Biometrics 30:89–99
Cai F (ed) (2002) Employment in rural and urban China: issues and options. Social Sciences Documentation Publishing Housing, Beijing
Dong X (2004) China’s urban labor market adjustment: a literature review. Mimeo, World Bank, Washington, DC
Giles J, Park A, Cai F (2006a) Reemployment of dislocated workers in urban China: the roles of information and incentives. J Comp Econ 34:582–607
Giles J, Park A, Cai F (2006b) How has economic restructuring affected China’s urban workers? Chin Quat 185:61–95
Huber PJ (1967) The behavior of maximum likelihood estimates under nonstandard conditions. In: Proceedings of the fifth Berkeley symposium on mathematical statistics and probability, vol 1, University of California Press, Berkeley, CA
Lin DY, Wei LJ (1989) The Robust inference for the Cox proportional hazards model. J Am Stat Assoc 84:1074–1078
Peto R (1972) Contribution to the discussion of paper by D. R. Cox. J R Stat Soc Ser B 34:205–207
White H (1980) A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica 48(4):817–830
Acknowledgment
We thank Amit Dar, David Ribar, Bert Hofman, and participants at a World Bank Seminar for helpful comments and suggestions. Comments and suggestions from a referee helped greatly improve this paper. Remaining errors and omissions are our own. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. The authors would like to acknowledge financial support from the World Bank’s Social Protection Unit and China Labor Market Program.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Betcherman, G., Blunch, NH. The limited job prospects of displaced workers: evidence from two cities in China. Econ Change Restruct 41, 187–207 (2008). https://doi.org/10.1007/s10644-008-9048-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10644-008-9048-2
Keywords
- Displacements
- Determinants of activity and labor market status
- Formal employment and self-employment
- Unemployment duration
- China