Who has Borne the Brunt of the Economic Downturn?

Another interesting issue to examine is who among the rural–urban migrants have been affected the most by the economic downturn. In this section we use our 2008 survey data combined with the tracking records to investigate this issue.

In the 2008 survey, we collected a rich set of information about the households and individual migrants. Using these data we could identify the individual and household characteristics of the group that we were not able to track previously. We estimated a linear probability model to tell us whether a household had been previously tracked. The independent variables were household and household head characteristics such as household size; household head's age, gender, marital status, years of schooling, years since migration, number of cities worked in since the first migration, job in 2008 (workplace size, whether the workplace was in manufacturing or construction, and whether the individual was self-employed); and a subjective variable to indicate whether the household would prefer to stay on in the city if allowed to do so. City-level dummy variables were also included. The results are reported in Table 6 [ PDF 97KB | 2 page ].

The results were largely consistent across the equations for the three different sets of tracking data. In general, age had a positive effect on tracking but it peaked at around 35–40. By 55 the probability of being tracked reduced to the same level as for those aged 16 (see Appendix B [ PDF 43.3KB | 1 page ]). Males, married individuals, and migrants who had other family members living in the same city with them were more likely to be tracked. In addition, those with higher education and longer migration experience were more likely to be tracked, while those who often changed places and did not want to stay on in the city even if allowed to do so were less likely to be tracked. These results are quite intuitive. If we believe that the attrition rate, to a certain extent, indicates the adverse employment impact of the economic downturn, the above results can be taken to mean that individuals who are single, are not in the primary working age, are less educated, and have less migration experience are more likely to be adversely affected by the economic downturn. These findings are largely consistent with the findings reported in the literature on unemployment in most countries (Devine and Kiefer 1991; Svejnar 1999).

With respect to employment, it seems that those who worked in manufacturing and construction in 2008 were more likely to have left their previous job than those who worked in tertiary industry. The effect was much stronger for construction than for manufacturing, especially in the third tracking, where the effect on a construction job was twice as high as that on a job in manufacturing. Working in the construction sector in 2008 made it 15% more likely that the individual would leave the job by February 2009 . The corresponding probability for manufacturing was 7%. Neither self-employment nor firm size had an effect on the individual's leaving the job.

Contingent on all of the above variables, relative to Guangzhou, almost all cities had a higher probability of being tracked, except Dongguan in October 2008. By the second and third tracking Dongguan had a higher probability of being tracked, while the probability for Nanjing had dropped dramatically to below the level of Guangzhou.

Download this Paper [ PDF 387.9KB| 28 pages ].

Comment(s)

There are [0] comment(s) for this entry. Post a comment.