中国经济学刊(2016年秋季号/第4卷/第2期)(英文版)(txt+pdf+epub+mobi电子书下载)

作者：(美)黄少敏主编

出版社：社会科学文献出版社

格式: AZW3, DOCX, EPUB, MOBI, PDF, TXT

中国经济学刊(2016年秋季号/第4卷/第2期)(英文版)试读：

文前辅文

A new and distinctive scholarly journal, Journal of Chinese Economics (JCE), provides an innovative forum for economists and thinkers to expose ideas and analytics in key areas of concerns to the China and world economy.

JCE initiates a new dialogue between China and the global economy through authors who discuss the dual perspective of both historical China and its modern economic theories and practices. From their discussions, a multipoint, collaborative approach to solve economic problems and business issues arises. Robust and practical solutions presented in the journal will offer alternative approaches to China’s economic situations in an increasingly connected global economy.

JCE provides an innovative forum for dedicated young economists in five distinct ways:

1. The Journal concentrates on the work of leading thinkers of a younger generation.

2. The Journal’s focus is applied economics at the local, national, or global levels in China. Articles selected for publication will have the potential to be interesting to a wide range of readers.

3. The Journal recognizes and publishes articles on modern China’s economic reform and development and globalization.

4. The Journal provides exclusive editorial services to Chinese authors and facilitates the publication of their important research.

5. The Journal is also interested in the researches on new business models in the process of China’s economic development and globalization.

JCE invites the submission of articles across a broad range of issues related to economic reform and development in China, including its potential impact and challenge to the global economy.

JCE aims to provide international readers a venue to explore China’s contemporary economics and business. The journal invites scholars affiliated to universities or institutions worldwide, and also encourages submissions from independent scholars.

·Research Articles·

Abstract: China’s labor market has undergone significant change since the early 1990’s when the urban labor market was governed by the communist party in a command and control system where both jobs and wages were set by a centralized employment placement system. During the 1990’s China began adopting a more market-oriented model culminating in a major labor market restructuring. Using the China Health and Nutrition Survey (CHNS), our study provides new evidence on the labor supply of married women in urban areas during China’s period of economic reform. As a more market-oriented system is implemented we find evidence that educational attainment is an important predictor of how women labor supply has changed in response to rising real wages. The results show that wage elasticity differs across educational categories and in some instances seems to have risen throughout the transition, possibly signaling greater flexibility in the labor market. In addition, more educated women are choosing to work fewer hours in later years yet participate in the labor market at much higher rates than women with lower levels of education.

Keywords: Labor Supply, China, Women

JEL Classification: J16 J30 O53

1 Introduction

China’s labor market has undergone significant change over the past three decades since the beginning of the economic reforms and the opening up of the economy in 1978. Prior to reform, the labor market was a highly centralized bureaucracy where jobs and wages were allocated by a highly structured employment placement system. In urban areas both men and women had high rates of labor force participation in full time work for state-owned enterprises that also provided a worker safety net. In addition to stable employment, firms would often provide educational opportunities, healthcare, retirement, housing and childcare. In the mid 1990’s, China’s governing authority transitioned to a capitalist model that supported largely profit-oriented private and profitable state-owned enterprises. These reforms culminated in massive layoffs and high rates of unemployment in the late 1990’s (Li and Zax, 2002). The transition toward a more capitalist model not only impacted the Chinese aggregate labor economy, but it also affected the working lives and labor market decisions of China’s men and women. This study focuses on women’s labor outcomes during China’s economic conversion from a highly centralized system to a much more competitive business environment. Our analysis provides new evidence on women’s hours of work and workforce participation over the course of China’s labor market transition.

Our study utilizes five waves of China Health and Nutrition Survey (CHNS) between 1993 and 2006 to better understand married women’s labor supply in urban China during these transition years. Specifically, we attempt to evaluate how married women with different educational attainment respond to a sharp rise in wages in a rapidly changing labor market. Through analysis of sub-periods, we also seek to find evidence of this transition in married women’s labor supply. This study uses a three-step method to estimate a selection bias corrected wage equation along with an hours worked equation that accounts for the endogeneity of wages. In addition, random effect and fixed effect estimations are used to assess the importance of unobserved heterogeneity.

2 A Brief History of the Labor Market in China

During the past two decades, China’s economy has undergone tremendous change as, it has moved toward a more profit driven capitalist structure. Prior to reform, a large portion of production was centrally governed and concentrated in urban areas. Workers were given little leeway to choose their jobs or change their employers. For some urban workers, this time may be fondly remembered as the “iron rice bowl” because their jobs at state-owned enterprises came with extended support from their employers “cradle to grave”.

The command and control system fully integrated women into the urban working environment; China had among the highest rates of labor force participation for women and the Labor Law of the People’s Republic, passed in the mid 1990’s formally established women’s labor rights. Except for a few select state jobs, these policies gave women the right to equal jobs, working conditions and wages, although there are evidences that wages and job opportunities were inferior to men’s (Cooke, 2001; Du and Dong, 2008).

Before 1993 major reforms targeted improvements in worker productivity and labor market flexibility at state-owned enterprises and fostered new foreign private investment and joint ventures which increased economy wide competition. As significant as these changes were, few state-owned firms reformed their employment policies and lay-offs were rare.

Economic reforms in the mid 1990’s greatly impacted urban labor markets in China when new ownership reforms reduced or eliminated subsidies to many of the state-owned enterprises forcing managers to radically alter their employment policies. The Chinese government maintained control of only the largest, most profitable state-owned enterprises, while all others were merged, privatized or allowed to go bankrupt (Dong and Xu, 2009).Unemployment rose rapidly as state-owned enterprises laid-off millions of workers after more draconian reforms were instituted in late 1997. To help mitigate the effects of rising unemployment, the central authority instituted “xiagang” policies that retrained or forced workers into early retirement (Wong and Ngok, 2006).

In the late 1990’s and early 2000’s public sector enterprises declined in size and number while more profit oriented private sector investments and joint ventures expanded. Production moved from large capital intensive firms to smaller private export oriented firms that produced light industrial goods. Because workers were allowed to quit their current employer to seek better jobs, for some, restructuring increased labor mobility and opened up new opportunities. For other workers, increased market competition resulted in high rates of unemployment, lower wages and fewer job opportunities.

Evidence suggests that women may have been more adversely affected than men during this major labor market reallocation (Peng et al., 2009). For one reason, working mothers benefit more than men from employer provided services such as child care, medical care and shorter working hours and working mothers are adversely affected by the elimination of such benefits (Peng et al., 2009).Furthermore, women were more likely to become unemployed than men. One study finds that women were seen as higher cost hires due to maternity expenses and were the first to be laid off (Zhang et al., 2008b).Finally, some authors find that traditional patriarchal Confucius values reemerged and reignited a cultural stereo type of male dominance as unemployment rose and competition for jobs increased (Beaver et al., 1995). As firms became liberated from centrally organized employment relationships, privatization allowed more flexibility to accommodate [3]discriminatory hiring and firing practices.

3 A Review of the Literature

Studies that have examined women’s labor market outcomes in China show a fairly consistent pattern. Over the last two decades the gender wage gap for urban workers has increased and labor force participation has declined (Zhang et al., 2008a; Chi and Li, 2008; Knight and Li, 2006; Dong and Bowles, 2002). Between 1988 and 2004 more women than men left the labor market and the gender wage gap widened, especially among the low skilled workers. Evidence suggests that since 2000, the gender gap has continued to increase especially among higher income urban workers (Zhang et al., 2008a). Researchers attribute a portion of the wage gap to a female skills deficit and skill mis-match in the transition to new industries, but a portion of the wage gap remains unexplained.

Other wage related studies find that returns to education are quite low prior to reform but start to rise after 1990 (Fleisher and Wang, 2005). In China, similar to other transition economies, returns to education increase as the economy becomes more market driven (Chen and Hamori, 2009; Deng and Li, 2009; Zhang et al., 2008a; Yang, 2005). In addition to exploring wage trends, several studies have focused on the labor force participation decisions of women. Empirically, the data show that women’s labor force participation rates in China have traditionally been much higher than in other countries but have fallen since1990. In 1989, 82% of women were in the urban labor force and this dropped to 79% by1996 (Li and Zax, 2002) and 75% in 2005. Studies find that labor force participation rates differ by demographic group (Dong et al., 2006; Maurer-Fazio et al., 2009). Maurer-Fazio et al. (2009) find that household composition impacts whether or not a woman works. The presence of older women in the household increases the likelihood of women participating in market work while the presence of pre-school age children significantly decreases the likelihood of women participating in market work. Consistent with these results, Edwige and Andizio-Bika (2004) find that age, education, and the presence of other adults in the household significantly impact the probability of women participating in the labor force. In another study, Du and Dong (2008) find that urban Chinese women who are better educated and more affluent are more likely to participate in paid work.

Zhang et al. (2008b) use the China Urban Labor Survey and the China Adult Literacy Survey to examine the effect of “marital capital” on labor force participation. They find that married women with children have the lowest labor force participation rate followed by married women without children. Single women have the highest labor force participation rates and these rates are no different than males’. Their study does not analyze hours of work but estimates a logistic regression of employment status and finds that schooling, literacy score, communist party membership and work experience are positively and significantly related to employment status.

Less is known about individual labor supply decisions, in particular, how wages affect the number of hours worked. Yueh (2008) estimates women’s and men’s wage and hours of work equations using a sample of rural migrant workers in 1999. The study finds that migrant women have a negative own wage elasticity, which means that as wages rise, the workers work fewer hours. The author suggests that rural migrant women, who already work an average of 9.2 hours each day are not likely to work longer hours in response to a marginal increase in their wages. Yueh (2008) included only wage, age and measures of worker’s social contacts and capital. The wage elasticity results changed sign, depending upon whether worker’s social capital was included in the regression.

In the only other known study to look at hours worked as a variable of interest, Li and Zax (2002) predict the number of hours worked for a sample of urban workers in 1995 using a two-stage least squares regression that corrects for the endogeneity of wages. Controlling for age, education, assets and non-wage income, the authors find that women have a positive wage elasticity. Although Li and Zax (2002) did not distinguish between rural migrant and urban resident workers, their results are opposite of those in Yueh (2008) and suggest that raising wages will increase a women’s labor supply.

Our study adds to this limited body of literature by providing new information on the labor supply of married women in the urban areas during China’s period of market reform. Previous studies have focused on rural areas. No studies have examined women’s labor supply in urban China over an extended time frame. The results will shed light on whether women’s labor force participation, hours of work, and wage elasticity vary by educational attainment and whether these measures have changed over the course of China’s transition to a more market oriented economy.

4 Data

The data used in this study was obtained from five waves (1993, 1997, 2000, 2004 and 2006) of the China Health and Nutrition Survey (CHNS), a longitudinal study administered by the Population Center at the University of North Carolina at Chapel Hill and the Chinese Center for Disease Control and Prevention. The survey population covers individuals in over 4,000 households from communities with diverse socioeconomic backgrounds in nine provinces of China, including Liaoning, Heilongjiang, Shandong, Henan, Hubei, Hunan, Guangxi, Guizhou, and Jiangsu.

Our sample includes married urban women whose age was between 19 and 50 at the time of the survey. A majority of the women have completed some level of middle school (36% completed lower middle school and 23% completed upper middle school) but only 20% of the women have a college education. Approximately 70% of the sample was employed at the time of the survey and more than half of these women are employed in state owned enterprises and a third worked in professional occupations. On average, each woman has 1.08 children and 30% had grandparents (to the children) living in the household. The property income for women, calculated as household [4]income minus the respondent’s income, averaged 380 RMB per week. The asset index is the first principle component of a number of household assets. A minority of the sample (approximately 5%) live in communities that have only dirt, unpaved roads. The community average wage represents the inflation adjusted average of daily wage rates for ordinary male and female workers in the community.

There are some interesting dynamics occurring in the labor market during the study period. As shown in Figure 1 (a), all education groups had gains in their average real wage rate during 1993-2006. Women with a college education saw the largest wage gains, resulting in a widening wage gap between college educated and women with less education. In 1993 real wages were fairly equal across all education levels but by 2006, female workers with higher education were earning 67% more than those with primary or middle school educational attainment.

The employment rate seems to have a strong and positive association with the educational levels; highly educated workers have consistently higher employment rates than those with low or no educational attainment throughout the sample period 1993-2006 (Figure 1 (b)). Employment rates for all groups except the most highly educated declined between 1993 and 2000. Between 2000 and 2006 the least educated groups’ employment rates rose while employment rates declined for women with middle and upper middle school educational attainment. By 2006, employment rates were equal to about 53% for all but the most highly educated women which were employed at significantly higher rates closer to 90%.

In 1993, those with primary and lower middle school education had the highest working hours on average, about 48.5 hours per week (Figure 1 (c)). Weekly working hours declined between 1993 and 1997 for all women, except the least educated. The married women with no education worked the most hours between 1997 and 2000, but hours for this group dropped drastically from 46 to 39 per week between 2000 and 2006. In contrast, during 2000-2006, women who completed lower middle and upper middle school increased weekly working hours. College educated workers weekly work hours did not increase much between 1997 and 2006, their work hours hovered close to 41 hours per week on average throughout this timeframe. By 2006, workers with lower middle and upper middle school education had the highest average working hours, 47 and 49 hours per week respectively, while women with no education, presumably those who need work the most, had significantly fewer weekly working hours. These patterns suggest that educational attainment may be an important determinant of urban women’s labor market outcomes over this period of market transition.Table 1 Variable Sample Means for Urban Women Ages 18-50Table 1 Variable Sample Means for Urban Women Ages 18-50-Continued

(a) Inflation Adjusted Wage

(b) Employment Rate

(c) Weekly Working Hours per WorkerFigure 1 Major Trends in Labor Market 1993-2006

5 Methodology

From a typical utility maximization exercise, we derive the reservation wage rate and statistically specify it asw = Xθ + ε　　　　　　(1)riiri

where the X is a vector of variables including educational level, inumber of children, property income, and some community characteristics observed by the individual.

Under the same utility maximization framework, the hours of work can be expressed as a choice resulting from the comparison between one’s reservation wage and the observed market wage (w). In general, ithe hours of work can be specified as proportional to the received market wage rate in excess of the reservation wage rate, orh = β (w – w) = 0　　if w ＞ w otherwise　　　　　　(2)iiriiri

When the observed wage rate is lower than the reservation rate, an individual will not work. But since the market wage is not observed for those whose h = 0, a selection bias exists. The presence of the iselection bias implies that estimating (2) by an OLS will yield a biased estimate of β, the wage effect on the number of work hours. One approach widely applied in the literature for obtaining a consistent estimate of the wage effect is to perform a three stage procedure with the selection bias being assessed as and controlled for by an inverse Mills ratio estimated from a selection (participation) equation. The use of inverse Mills ratio was suggested by Heckman, (1979, 1980) and have had applied extensively by others (e.g. Mroz, 1987). In this paper we also implemented this strategy.

By explicitly specifying the market wage as w = Zγ + ε, where Z iimiis a vector of variables including age, educational level, and community average wage rate, we can derive through (2) the likelihood of working as2Pr (working) = 1 – F ［(Xθ – Zγ) / σ］, where the σ is var (ε – ε)iiwwmiri

The likelihood of working can be estimated using a probit. The probit estimates are used to calculate the inverse Mills ratio:

试读结束[说明：试读内容隐藏了图片]

下载完整电子书