also disappearing between 1998 and 1995 (table 7, rows 11 to 17). The convergence of
wages among employment sectors, however, does not occur spatially. The provincial
dummy variables are all jointly significant and their level of significance is increasing
over time (row 18). This could imply increasing wage inequality among provinces, a
result that suggests lagging labor market development. The result, however, may be a
function of the timing of our survey and normal frictions in labor markets. China’s
economy was growing at its peak speed in 1995 and the demand for labor was very high
throughout the country. The wage premiums offered by those fastest growing areas may
reflect temporary rises in wages that will eventually be competed away, a conjecture that
can only be tested with additional data collection and analysis.
Most relevant for our study, the wage regressions provide further evidence that
incoming labor has not pushed the wages received by local workers down to a level of
those villages without incoming workers. Villages with incoming labor (commuters
and/or migrants) have significantly higher wages than other villages in 1995 but not in
1988 (table 7, row 1, column 1 and 2). While we can not rule out the possibility that
continuing in-migration will not dissipate the wage premium (a possibility that is
reinforced by the disappearance of the female wage premium—columns 5 and 6) , at least
in the mid-1990s, the average worker (or more specifically, the average male worker)
who is a resident of a village that has incoming labor, receives a higher wage than
workers in other villages receive. The persistence of higher, or of at least not lower,
wages is another reason why barriers to hiring decisions by local firm managers may be
weakening.
25