|Title:||China minimum wage elasticities of rural migrant workers in urban labor markets by geographic region in percentages for 2007|
|Source:||Contemporary Economic Policy|
Start of full article - but without data
Minimum Wage Elasticities, Rural Migrant Workers in Urban Labor Markets
Minimum Wage East Central West Measure
Minimum wage X.XXX (X.XXX) X.XXX (X.XXX) X.XX.X * (X.XXX)
Minimum wage X.XXX (X.XXX) -X.XXX ** -X.XXX *** lagged (X.XXX) (X.XXX)
Sum current and X.XXX (X.XXX) -X.XXX (X.XXX) -X.XXX (X.XXX) lagged
Minimum wage X.XXX (X.XXX) -X.XXX (X.XXX) X.XXX ** (X.XXX)
Minimum wage X.XXX (X.XXX) -X.XXX *** -X.XXX *** lagged (X.XXX) (X.XXX)
Sum current and X.XXX (X.XXX) -X.XXX ** -X.XXX (X.XXX) lagged (X.XXX)
Minimum wage -X.XXX (X.XXX) X.XXX (X.XXX) X.XXX (X.XXX)
Minimum wage X.XXX ** X.XXX (X.XXX) -X.XXX * lagged (X.XXX) (X.XXX)
Sum current and X.XXX * (X.XXX) X.XXX (X.XXX) -X.XXX (X.XXX) lagged
Sample size XX XX XX
Note: Standard errors in parenthesis. Statistical significance is
denoted by ** at the X% level, * at the X% level and * at the XX% level
Source: Calculated from the minimum wage estimates of Appendix Table CX.
Issues associated with minimum wages are generating heated debate in China. Supporters argue that in the transition to a market economy, minimum wages can serve an important safety-net role by providing a minimum floor on wages, thereby protecting otherwise vulnerable workers from the most egregious aspects of intensively competitive markets (Sun 2006; Zhang 2007; Zhang and Deng 2005). This is especially the case for the large influx of rural migrant workers coming from the countryside to the city, both from within the province and from other provinces. Furthermore, the higher labor cost may foster managerial efficiency and labor productivity and induce employers to move up the value-added chain and invest in productivity improving technology (Cooke 2005). From these lines of argument, Chinese scholars have often argued in favor of the more proactive use of minimum wages, which they often regard as otherwise too low and ineffective (Han and Wei 2005).
In contrast, opponents of minimum wage laws argue that they inhibit the transition to a market economy by interfering with China's comparative advantage based on an abundance of low-wage labor. Such regulations raise costs and inhibit the growth in demand for Chinese products and hence in the derived demand for their labor. Such growth would otherwise increase both employment and wages. Opponents reiterate the basic arguments against minimum wage legislation in that it will reduce the employment opportunities of low-wage labor and also lead to offsets in such forms as reductions in other elements of the compensation package (Gong 2009; Chen 2001; Xue 2001; Ye 2005). This argument is part-and-parcel of the broader argument that artificially raising wages can jeopardize the growth process that by itself will foster increases in the demand for labor and hence foster higher wages and improved labor standards in general. (X)
China's experience with minimum wages is new, having first introduced them in 1993. As discussed subsequently, however, minimum wages have been altered many times since then such that they can be regarded as a potentially important policy instrument. In spite of this importance, there are very few empirical studies on the effects of minimum wages in China (Luo 2007a, 2007b; Shi 2009) and only one study in English (unpublished) to our knowledge (Ni, Wang, and Yao 2008). This is part of the more general gap of a lack of evidence on labor markets in developing countries (Hamermesh 2002)--evidence that could otherwise facilitate critical evidence-based policy making. This is particularly important in the minimum wage area because minimum wages are obviously subject to policy control and manipulation, and in countries such as China, they are an integral part of the transition to a market economy.
The purpose of this article is to begin to fill that gap by providing evidence of the impact of minimum wages in China. We do so using the novel methodology of a prespecified research design (Levine 2001; Neumark 2001) based on data that provide a large number of minimum wage changes from which to identify the effects of minimum wages.
The article is structured as follows. Section II discusses the theoretically expected impact of minimum wages and Section III outlines the existing empirical evidence from the literature in both developed and developing countries. Section IV provides details on minimum wage legislation and its implementation in China. Section V outlines the methodology of the prespecified research design and provides the estimating equation, while Section VI outlines the data. The empirical results are presented in Section VII, and Section VIII provides a concluding summary.
II. EXPECTED IMPACT (X)
Unlike the empirical evidence that is controversial (discussed subsequently), the theoretically expected impact of minimum wage legislation is well established in the literature. Minimum wages should reduce the demand for labor because substitution and output effects work in the same direction (i.e., labor demand curves slope downward). In response to higher wages, firms will substitute away from the higher priced labor into other inputs. The substitution process can be direct, such as with the increased use of tools or capital equipment, or it can be more subtle (especially in the longer run) such as through the increased use of prefabricated components or processes that are simply labor saving. It could even involve the substitution toward using more human capital through using more skilled labor because their wage premium relative to unskilled labor has now been reduced. As well, in response to the higher labor costs, firms may raise their prices and face a reduced demand for their output and hence in their derived demand for labor. In the extreme, some firms may even go out of business. Both these substitution and output effects reduce the demand for low-wage labor that is subject to the minimum wage increase. In growing labor markets such as that of China, the adverse employment effect will likely occur in the form of the counterfactual of slower employment growth, rather than outright reductions in employment. However, in the recent economic crises, it can occur in the form of additional layoffs.
There are, however, a myriad of potentially offsetting factors. Shock effects may exist in that the higher labor costs may shock management into more efficient managerial practices rather than reducing their demand for low-wage labor. They may also absorb the cost increase in the form of reduced profits. Workers themselves may be more committed and provide more discretionary effort in return for the higher wage, induced in part by the greater number of workers who are queuing up for the higher wage jobs. This can be particularly important in developing countries, given the large queue of surplus labor from the countryside. Turnover may be reduced with the reduction in turnover cost offsetting at least some of the rise in wage costs. Firms may also try to offset some of the wage cost increase by reducing other nonwage components (free uniforms and break periods) or even by increasing the pace of work.
Especially in developing countries such as China, employers may also simply not comply with the legislation (Chan 2001; Rawski 2006; Solinger 1999), trading off the expected wage cost if they comply with the (likely low) probability of being caught and fined if caught. Workers who are paid a subminimum wage may be reluctant to complain, especially if there is a large pool of surplus labor willing to take their low-wage jobs. As discussed subsequently, however, efforts at compliance have been strengthened in China.
A potentially important possible offset that has received attention in the literature is that of monopsony (Manning 2003; Bhaskar, Manning, and To 2002). In circumstances where a particular employer is dominant in a local labor market, they may have to raise wages to attract additional workers; conversely, if they lower their wages, they will lose some but not all of their workforce. As they have to raise wages to attract additional workers, they may also have to raise the wages of their existing incumbent workers for reasons of internal equity and harmony. In essence, the marginal cost of expanding their workforce exceeds the higher wage. This clearly inhibits their hiring of new workers. A minimum wage imposed in that situation can actually induce them to expand their employment because they are no longer constrained by having to pay the higher wage to their existing incumbent workers; they have to pay the minimum wage to all their low-wage workers who otherwise would be below the minimum wage.
A modified version of the monopsony argument that is particularly relevant to China is provided by Dong and Putterman (2000, 2002). They argue that the state enterprises in China essentially face a rising supply price or average cost of labor because state enterprises will internalize and consider the increasing social cost of absorbing rural workers into the urban centres. Such rising costs include those associated with such factors as subsidized housing, education, health care, and pensions (emphasized in Offer 1976 and Riskin 1987) as well as the cost associated with reductions in agricultural output and hence rising agricultural prices (emphasized in Dixit 1971; Fei and Ranis 1964; Gillis et al. 1996).
As in conventional monopsony, this will lead to marginal costs rising faster than average costs which will inhibit their hiring of new workers. As is also the case in conventional monopsony, a minimum wage can induce such employers to expand their employment because they are no longer facing a rising supply price of labor; rather, they have to pay the minimum wage to all their low-wage workers who otherwise would be below the minimum wage. Such monopsonistic behavior would apply to state-run organizations because they have the political incentive to internalize or consider the rising average cost of labor. This is especially the case for those in the East where labor shortages would otherwise dictate increasing their wages. Dong and Putterman (2000) provide evidence of such monopsonistic behavior in the state industrial sector.
In a subsequent study, Dong and Putterman (2002) provide additional evidence on the existence of monopsony in state-owned enterprises. They also indicated that it dissipated somewhat in the presence of market-oriented reforms and was strongest where the state's share of output was highest.
With respect to the potential for monopsony to offset the conventional negative impact of minimum wages and to possibly even lead to increased employment, their analysis suggests that such monopsony effects should be more prevalent in state-owned enterprises in the East where such increasing costs associated with integrating rural migrant workers are likely to be most prominent.
III. EXISTING EVIDENCE
There is no consensus in the existing literature from developed countries on the magnitude of the adverse employment effect associated with minimum wages or, indeed, even if there is an adverse employment effect. (X) The earlier literature based mainly on time-series data from the United States over the 1950s, 1960s, and 1970s did arrive at a "consensus range" that a XX% increase in the minimum wage led to a statistically significant l%-X% reduction in employment of youths. Later time-series studies incorporating data from the 1980s, as well as aggregate cross-section studies using states or metropolitan areas as the unit of observation tended to find smaller and often statistically insignificant effects, more in the neighbourhood of a X% reduction in employment associated with a XX% increase in the minimum wage. Studies over the more recent period since the 1980s have often combined aggregate data across states and over time, finding conflicting results ranging from large adverse employment effects that are greater than the former X%-X% range to effects that are negative but statistically insignificant to effects that are even positive albeit generally statistically insignificant when positive.
The more recent studies that have generated the most controversy use the "difference-indifference" methodology to compare "treatment group" jurisdictions that increased their minimum wage with "comparison group" ones that did not increase their minimum wage. Those studies tend to find no adverse employment effect and sometimes even a positive employment effect, although it is generally statistically insignificant when positive. (X) Those studies tend to suggest that the adverse employment effect is statistically insignificant and close to zero.
British evidence, based on the earlier minimum wages set at the industry level by the Wage Councils as well as the more recent evidence from Britain's first-ever national minimum wage of 1999, also suggests no overall adverse employment effect (summarized in Metcalf 2008; Steward 2004). However, a small negative effect was found in the more labor-intensive and low-wage home-care sector (Machin and Wilson 2004).
Canadian data are generally regarded internationally as superior for testing the impact of minimum wages (Hamermesh 2002, XXX-XX) because minimum wages in that country are set at the provincial level and there is considerable variation across jurisdictions and over time from which to identify minimum wage impacts. Minimum wages are increased frequently and coverage is nearly complete. This is in contrast to the United States, where minimum wages are set at the federal level and there is limited variation in minimum wages coming mainly from three sources; state "top-ups" over-and-above the federal minimum, variation in the regional average wage that thereby gives rise to variation in the federal or state minimum relative to the average, and the slow erosion of the real minimum wage over most recent time periods. Recent Canadian studies (X) based on different data sets and methodologies tend to find that a XX% increase in the minimum wage leads to a statistically significant X%-X% reduction in employment, with the effect being larger in the longer run than in the short run.
The relevance of the existing evidence from developed countries could be questioned as to its applicability for developing countries with their large informal sectors, large pools of surplus labor in the countryside, and difficulties in ensuring compliance with such legislative initiatives Although the evidence from developing countries is more limited (Gindling and Terrell 2007, p. XXX), it is also not in agreement. Some studies find no effect, while others find a substantial negative effect. (X)
Clearly, there is no consensus in the existing literature as to the magnitude of the adverse employment from minimum wages or, indeed, even if there is an adverse effect because zero effects are common. Card and Kruger (1995b) also argue that zero effects may be more common than they appear in the literature because there may be a "publication bias" against studies where effects are statistically insignificant, especially when a negative effect is unambiguously predicted by basic theory. The possibility of this publication bias can lead not only to studies that yield insignificant effects not being published but also to researchers continuing to run alternative specifications until they get the theoretically expected negative effects. As the expression goes, "if you beat the data long enough you can get it to tell you what you want." Based on a meta-analysis Card and Kruger (1995b) find evidence of such "publication bias" in the minimum wage literature.
Even if there was a consensus in the literature on the effect of minimum wages, it may not apply to developing countries such as China with their informal sectors, large pools of low-wage labor, and compliance problems. The purpose of this article is to provide direct evidence for China and to begin to fill the gap from the lack of evidence in this area. It does so using a prespecified research design (detailed subsequently) that circumvents the concern of manipulating the specifications until the preferred results emerge. Prior to setting out the methodology and discussing the data, a brief descriptive picture of minimum wages in China is provided.
IV. MINIMUM WAGES IN CHINA
In 1993, China issued its first minimum wage regulations, and in July 1994, they were written into China's new version of the Labor Law. Prior to then, in 1984, China had simply recognized the minimum wage principle that was embodied in the ''Minimum Wage Treaty" of the International Labour Organization (ILO) in XXXX but had not implemented its own legislation.