Printer Friendly

Chinese minority income disparity in Urumqi: an analysis of Han-Ugyhur labour market outcomes in the formal and informal economies.


Ethnic tensions between Uyghur and Han Chinese culminated in 2009 with the Urumqi riots, one of the bloodiest conflicts in Xinjiang's modern history, which left more than 200 people dead and another 1,700 injured. Following the 2009 Urumqi riots, Chinese policy towards Xinjiang underwent a major shift, breaking away from its previous "gradualist" approach to one of economic development. In its place, China implemented policies meant to spur "leapfrog development" in Xinjiang. These new economic policies are likely to bring about rapid growth to the region; however, the question remains whether this growth will be equitably transferred amongst all ethnic groups and regions within Xinjiang.

To this end, the Chinese literature on Han minority studies is understudied in terms of income inequality, poverty and labour market outcomes. (1) The lack of Han-minority studies is largely because individual "income" is generally not reported by ethnicity in official Chinese statistical sources, and due to political sensitivities, researchers are generally uninterested or unable to carry out fieldwork in minority areas, predominately located in western China.

The small body of research that does exist on income inequality among ethnic groups has produced mixed evidence in terms of Han-minority disparities. On the one hand, Hannum and Xie (2003) contend that both income and occupational attainment gaps between Chinese minorities and Han have widened (2), while Gustafsson and Shi (2003) argue that Hui minorities are not at an economic disadvantage relative to the Han in urban Ningxia. (3)

The purpose of this paper is to contribute to the growing body of Han-minority studies on China. The two main questions addressed in this paper are: 1) How do the roles of ethnicity, gender and migrant status impact wage outcomes and are the respective impacts independent of human capital; and 2) Does the rate of return on key demographic variables (e.g. human capital, ethnicity, gender and migrant status) vary depending on whether a respondent is employed in the formal or informal sectors. Both research questions are timely and relevant in the wake of the 2009 Urumqi riots, especially since economic inequality is considered a major driver of the uprising and Uyghur discontent. Moreover, very little is known about China's informal economy, in part, because the state statistical apparatus largely neglects the informal sector. (4) It is suspected, however, that Uyghur participation in the formal sector is lower compared to Han, coinciding with a rise of Uyghurs engaging in informal sector employment. (5)

The analytical framework adopted in this study was informed by a rich source of empirical survey data, which was collected in 2008 in Urumqi by the author, in collaboration with Chinese affiliates. To the author's knowledge, Zang's study (2011) is the only other quantitative work that uses survey data to examine income disparity between Uyghur and Han. (6) Although many of the general results found in the present paper confirm Zang's findings (i.e., a large Han/Uyghur and male/female inequality gap) there are several important areas of divergence.

First, the current paper offers the first attempt to statistically analyse Hanminority income disparity in both the formal and informal Chinese economy, whereas Zang only examines the formal economy. Second, the current paper extends the analysis by adding a third ethnic group, the Hui. The Hui are included in the analysis as a type of control group, in order to determine whether income disparity exists along minority-majority lines or specifically, only Uyghur-Han. Third, the present research is informed by more recent data collected in 2008, while Zhang's data was collected in 2005.

The outline of this article is as follows. The next section briefly draws on arguments made by human capital theory and labour market segmentation theory, and summarises the recent work carried out on Chinese ethnic-based labour market outcomes and disparities. The third section gives an account of the two types of major government policies--"Open up the West" initiative and "Leapfrog Development" policies--that have and will continue to dramatically affect economic growth and inequality in Xinjiang. The fourth section describes the survey site Urumqi, as well as provides a detailed account of the data collecting process. The fifth section reports the results generated by the descriptive statistics and regression analyses. The final section summarises the key findings of this study and considers their policy relevance.


Human capital theory and labour market segmentation (LMS) theory are two contrasting theories commonly invoked to study emerging labour markets and income inequality. Human capital theory is rooted in neoclassical economics and its adherents argue that wage differentials are primarily the result of differences in human capital: as human capital is acquired, wages will eventually achieve parity. A prominent China scholar and advocate of human capital theory, Nee argues that marketisation and the emergence of local labour markets in China will gradually increase the rate of return on human capital. (7)

Some evidence on Chinese minorities also indicates the importance of human capital during China's transition, and in particular, educational attainment. For example, Hannum and Xie attributed the widening occupational attainment gap to the increased gap in educational attainment between minorities and the Han. (8) Hannum et al. and Hannum and Wang further established a possible link between an education attainment gap and the persistence of income inequality. (9) The authors found that while most minority groups increased their educational attainment over time, they remained at a disadvantage relative to their Han counterparts. Gustafsson and Ding found that the education level of the household head, location and village mean income all played a more prominent role in accounting for poverty than ethnicity. (10)

In contrast to human capital theory, LMS theory argues that vulnerable groups become trapped in lower or informal segments of the labour market because of mobility barriers (e.g., place of residence, poor work histories, and discrimination), which reduce inter-sectoral job transfers while occupational stratification increases. (11) LMS theorists claim that the labour market is comprised of formal and informal sectors and that the market segmentation is the result of institutional rules that differ across labour market segments and have thus replaced the market processes of supply and demand. (12)

LMS theory is well-developed and commonly tested in multi-ethnic countries, particularly in the developed world. Researchers have used empirical evidence to support the existence of wage inequality between members of differing social groups who possess similar human capital characteristics. In other words, even after controlling for human capital factors, minority workers will still earn less money than members of the majority and are less likely to get hired. (13) Many experts conclude that minority groups are over-represented in the bottom and informal tiers of the labour market, face wage and other forms of discrimination, and are oftentimes unable to achieve inter-sectoral mobility. (14)

In the developing world, research on labour market segmentation most often distinguishes between formal and informal sectors. Empirical findings from developing countries show that minorities, women, and in many cases, migrants have limited access to good jobs and are disproportionately placed in the informal sector. (15) For example, in Brazil, Telles' analyses exposed a large degree of minority income disparity and revealed that black and mixed-race peoples are disproportionally represented in informal sector employment. (16)

Since the 1970s, employment in the informal economy has exploded in the developing world, accounting for 65 per cent of the non-agricultural employment in Asia, 48 per cent in North Africa, 78 per cent in Sub-Saharan Africa and 51 per cent in Latin America. (17) China is no exception. From 1978 to 2006, the number of employees employed in the informal sector skyrocketed from 15,000 employees to 168 million, representing 59.4 per cent of total urban employment in China. (18) According to Huang, China's informal economy consists of 120 million rural-urban migrants (nongminggong) and another 50 million regular urban residents. (19) There is, however, no account of ethnic participation in the informal economy and no information on whether minorities engaged in the informal sector are rural-urban migrants or urban natives.

Although research on China's informal economy is largely missing, studies on labour market segmentation within the private economy have flourished, at least in regard to migrant status and gender. Recent research on China's transition reveals that the adverse effects on the labour markets associated with structural changes have varied in scale and scope. In regard to migrant status and gender, the hukou (household registration) system has not allowed for wage convergence between natives and migrants, generating a multi-tiered labour market segmented by migrant status. (20) Fan gives a detailed account of the segmented nature of gendered labour market segmentation in China. (21) Her research showed that as a result of socio-cultural traditions and social networks, male and female migrants are channelled into distinct sectors of the labour market.

In regards to ethnicity, only a few empirical analyses on Chinese ethnic minorities' labour market outcomes exist. For instance, Howell (2011), (22) using the same survey data as in the present research, investigates the impact of economic reforms and marketisation on labour market segmentation in Urumqi; results show that ethnicity and migrant status are the main drivers of market segmentation, not human capital variables (i.e., education and work experience). The author's finding is supported by other studies that also indicate China's urban labour market is segmented by ethnicity. (23) Similarly, Zang (2008) finds that ethnicity is the main determinant of labour market discrimination, and the returns from education to Uyghur are smaller compared to Han in terms of job attainment in state firms. (24) Zang's finding suggests that although increases in human capital may lead to positive outcomes in job attainment or income, due to market discrimination, these returns will be disproportionately lower for minorities than for Han.


China's transition is noted as being particularly difficult for Uyghurs, who find it hard to achieve economic equality with their Han counterparts. Emerging labour markets in Xinjiang have led to growing wage inequality, deepening occupational segmentation, increasing discrimination and a pervasive flow of Han migrants into the region. (25) For instance, the tremendous flow of Han migrants in search of employment has dramatically reconfigured the local demographic composition of Han Chinese in Xinjiang from 6.2 per cent in 1945 to 39.2 per cent in 2008, leading to growing animosity and conflict between Uyghur locals and Han migrants. (26)

Although Han Chinese play a vital role in Xinjiang's economic development, they are perceived by many Uyghurs as monopolising local natural resources and accumulating wealth by exploiting minorities. (27) Han migrants are viewed as taking good jobs away from Uyghurs, at the expense of the latter's social mobility and earning capacity. (28) Several studies indicate that the Han-Uyghur income gap is large and that the Uyghurs' quality of life has risen at a slower rate than that of the Hans'. (29)

During the 1990s, as a result of economic hardship and inequality, there were thousands of Uyghur-led acts of violence, including bombings, murders and assassinations. (30) In 1991, the newly gained independence of the Central Asian states (e.g., Kazakhstan, Kyrgyzstan, Tajikistan, and Uzbekistan) further fuelled Uyghur aspirations for an independent Uyghur state, called Uyghurstan or East Turkestan Republic. (31) In the face of a resurgence of minority-led uprisings, Chinese experts began to lobby the central government to focus on reducing inequality, and the local populous' well-being became inextricably linked to China's overall social stability and national security. (32)

To help protect minorities during China's economic transition, Beijing enacted over 280 affirmative action laws and regulations on minority preferential treatment between 1979 and 1995. (33) These laws apply to various spheres of life, including education, college admissions, exemptions from the one-birth policy, employment opportunity recruitment, and job placement in the state sector and in leadership recruitment. (34) Many minorities believe that the affirmative action policies have lost meaning and failed to: 1) increase ethnic regional autonomy; 2) offset growing inequality between Han and minority groups; and 3) reduce growing anti-minority bias among Han Chinese. (35)

To help mitigate the rising inequality and ensure social harmony, China introduced the "Western Development" programme in 1999 and later enacted the "Open up the West" campaign in 2001, as part of the 10th Five-Year Plan (2001-2005). In 2000, Prime Minister Zhu Rongji defined the programme with the following objectives: "to stimulate domestic demand, to support sustainable growth and to develop all regional economics to reach a common prosperity but also to reinforce national unity, to maintain social stability and to consolidate border protection". (36) Within only a couple years of implementing the "Open up the West" policy, the central government reportedly invested over 70 billion yuan (USD8.36 billion) in infrastructure-related projects. (37)

China's push to develop Xinjiang is primarily based on its geopolitical significance and abundant natural resources. Xinjiang represents one-sixth of the total area of China, and has both the largest reserves of natural gas and the second-largest petroleum production capacity that has yet to be fully developed. Its proximity to foreign markets gives Xinjiang locational advantage in the region and serves as China's bridgehead, expanding influence westward into Central Asia and beyond. While Xinjiang is used as a conduit to further Chinese geopolitical interests abroad, it is also one of the most restive regions in China, representing one of the largest internal threats to Chinese rule. (38)

Within only a couple years of implementing the "Open up the West" policy, the central government reportedly invested over 70 billion yuan (US$8.36 billion) in infrastructure related projects. (39) Within a decade of implementing "Open up the West" policy in 2001, Xinjiang's GDP more than doubled. By 2008--the point in time that our survey data was collected--Xinjiang's GDP reached 420.3 billion yuan, up 10.6 per cent on average annually. (40) Although the "Open up the West" campaign led to impressive economic growth rates, other factors such as increasing inequality, as well as restraints on religious, cultural and political freedoms, all contributed to ethnic tensions and rioting in Xinjiang.

The recent 2009 Urumqi riots, in combination with the international financial crisis, had a dual impact on Xinjiang's economy, severely affecting foreign trade, tourism and other industries. In response to the Urumqi riots and the economic downturn, Beijing organised the Central Work Conference in 2010 in Xinjiang. A countrywide effort, the work conference promoted a leapfrog development approach to help revitalise Xinjiang's economy. Beijing, Tianjin, Shanghai, Guangdong, Shenzhen and 21 other provinces and cities agreed to support Xinjiang by providing aid to sister cities and counties located within Xinjiang.

As part of this "new deal"--known as yuan jiang (literally "to help Xinjiang")--hundreds of billions of yuan are to be invested in Xinjiang over the next ten years for the promotion of urban construction, industrial development and modern agriculture. According to one report, the "Yuan Jiang" programme offers Chinese provinces and cities a "historic opportunity" to readjust Xinjiang's economic structure strategically. These new, aggressive development policies aim to bring about internal stability, and accelerate the integration of Xinjiang with the rest of China as well as with the global economy by introducing competitive industries to the region. (41) Unlike China's gradualist growth strategy, adopted since the 1978 reforms, the "Yuan Jiang" initiative represents a break from the traditional Chinese growth model and will undoubtedly shape the growth and development trajectory of Xinjiang, and reduce inequality for the foreseeable future.


The survey site for this research was Urumqi, the capital city of Xinjiang Province. Urumqi is an ideal city to conduct a case study on minority labour market outcomes and income disparity. It has one of the largest non-agricultural workforces in western china (70 per cent), and the most mature labour market in Xinjiang that has also experienced a recent boom in the service economy. (42) In particular, the number of private enterprises (siying qiye: eight or more employees) and self-employed units (geti hu: fewer than eight employees) have experienced substantial growth since 2001. (43) The study area included four of Urumqi's seven urban districts, two older and two recently created. Tianshan and Shayibake make up the oldest part of the city and serve as the city core. Shuimogou and Xinshi district, on the other hand, are relatively newer districts that have resulted from the recent in-migration trends and subsequent urban expansion. Unlike cities in eastern china where the ethnic composition is highly homogeneous, Urumqi's ethnic composition is highly diverse. Together, the four districts included in the author's survey represent 77 per cent of Urumqi's total population, 89 per cent of its Uyghur population and 68 per cent of its Han population.

Sampling Design

In order to obtain a large sample size for both Han and Uyghur, a stratified sampling method was selected. The sample frame consisted only of full-time urban workers employed primarily in private enterprises and self-employed units--although some state-owned enterprises (SOEs) were also surveyed--operating in Urumqi's tertiary sector. A novel data collecting approach that the author introduced was administering surveys on-site at respondents' work locations. This breaks from the conventional norm of collecting data at the respondents' place of residence. The key advantage to this research design is that the enumerators can observationally conclude whether a respondent is employed in the formal or informal sector. When surveys are done at the household, it is very difficult to determine this distinction; it would introduce tremendous respondent bias if the enumerator asked the respondent directly. Another key advantage of the author's research design was that only those who are gainfully employed were surveyed. Administering surveys at the place of residence will naturally include survey data of the unemployed, all of which must be discarded in a study on labour market outcomes. The third advantage of the author's sampling design was that surveys were administered in an open, public place that facilitates easier access to survey respondents, and may also provide a more comfortable environment for respondents to answer survey questions, diminishing respondent bias. This is in opposition to, for example, needing to gain access to predominantly Uyghur residential areas, which can be difficult. Even then, any foreign enumerators entering the private residence of a Uyghur home may lead to an unknown degree of respondent bias.

The author chose to limit the sample frame to Urumqi's tertiary sector, because Pannell and Schmidt reported that minorities, particularly Uyghur, remained excluded from the industrial and energy sectors. (44) In contrast, during the 1980s, the first major sectoral transition occurred, when the share of minorities engaged in the tertiary sector in urban areas increased from 25.8 per cent to 41.2 per cent, a shift that remains evident today. Particular industries of growth include low-skill service industries such as retail, transportation and the restaurant business. (45)

Within the tertiary sector, the author chose to over-sample low-end service industries based on findings from previous research, which contends that minorities in Xinjiang are highly under-represented in high-skill service jobs such as technical, administrative and professional occupations. (46) Therefore, to obtain a healthy sample size for both Uyghur and Han with similar sets of worker characteristics, the author interviewed respondents who were typically employed in visible, commercial and low-scale retail and wholesale jobs, including informal work.

Data Collection

Prior to the start of the survey, approximately 20 survey sites were selected. Looking at an urban district map of Urumqi, the author selected sites that were distributed as evenly as possible across each of the four districts surveyed, with an average of five sites per district surveyed. At each survey site, approximately 30 interviews were conducted. To the greatest extent possible, on-location work places were chosen randomly by selecting every 10th store location. A set of diverse work places were captured in the sample, including various retail stores, hospitals, restaurants, barbershops, massage parlours, banks, insurance agencies, street vendors, repair shops, cultural specialty items; people who worked as taxi and bus drivers were also interviewed, etc. Only one respondent was interviewed per workplace. Respondents from a total of 595 different workplaces were interviewed.

The survey questionnaire included questions related to earnings (monthly income), personal endowment characteristics (gender, age, marriage, education and number of children), mobility characteristics (place of origin, number of places lived, length of stay and household registration status), occupation characteristics (type of employment, length of employment and type of firm), and social capital characteristics (how did one find employment, where did one's father work and the number of languages one speaks).

Variable Definitions

Average monthly income: The measure of income in this survey captured the total monthly income (yue gongsi) generated from the respondent's current place of employment; it included all bonuses but excluded income generated from other activities or family members.

Migrant status: Respondents were delineated into three classifications--natives, intra-provincial migrants and inter-provincial migrants. The author defined natives as individuals born in Urumqi and migrants as those born outside Urumqi. The definition of migrants adopted here is analogous to that of the floating population, a stock measure that largely ignores the timing of migration. For the purpose of this research, a stock measure was appropriate because the author was more concerned with whether an individual was a native of Urumqi than when he/she arrived in the city; the author again selected place of birth rather than hukou--a key dimension of the floating population definition--to determine whether an individual was a native rather than whether he/she had obtained Urumqi hukou.

Employment type: Respondents were classified into the following three employment classifications: self-employed, managers and employees. Self-employed were those who own their own business; they may or may not also be an employer. The self-employed respondents tended to be a diverse group. On the one hand, numerous self-employed respondents operating in the formal sector were surveyed at medium-sized department stores in newly built shopping malls; on the other hand, several self-employed respondents employed in the informal sector were surveyed at very small, often mobile workplaces, such as street stalls. Many of the respondents captured in the survey worked as employees or managers in small work-place units (geti hu). Managers oversaw employees and handled day-to-day responsibilities, and on occasion were also owners, in which case respondents were marked as both self-employed and managers. Employees were neither managers nor owners, and had limited job duties and responsibilities. In general, managers were in the highest socio-economic strata and employees made up the lowest socio-economic strata. For self-employed respondents, their socio-economic status was largely based on whether they were formal or informal workers.

Skilled/Unskilled: Respondents were classified as skilled or unskilled based in part on their job position, industry and work responsibilities. Many of the unskilled workers were employees in the retail industry and sold a variety of items, including snacks and beverages, electronics, home appliances, clothes, home furniture and minority specialty foods or goods (carpets, knives, etc.). Other unskilled respondents worked as servers in restaurants, or taxi and truck drivers. Respondents classified as skilled workers were employed in a variety of industries, including telecommunications, clerical and business. Specific skilled job positions captured in the data included mechanics, security guards, barbers, construction workers, professors, doctors, engineers, computer technicians and clerical workers.

Originally, the "skilled" category was further divided into "semi-skilled" and "skilled" classifications; however, the small number of respondents in professional jobs (i.e., doctors, professors, engineers, etc.) in the survey required respondents in the "semi-skilled" classification to be incorporated together with respondents in the "skilled" classification. To ensure that both skilled and unskilled respondents were captured in both the formal and informal sectors, the following informal positions were classified as skilled: various informal jobs including shoe shiners, keysmiths and bike repairers, etc. Conversely, respondents engaged in retail jobs in the informal economy and those selling clothing, food/snacks, fruit, knives, cellphones, etc., were designated as unskilled informal workers.

Formal/Informal: Respondents were not directly asked if they were formally employed. Instead, if a respondent's official work permit was not clearly visible (the law mandates that it should be), the job was marked as informal. Holding an official work permit is significant because it indicates that the workplace is regulated by the authorities and contributes taxes to the state (the absence of which is the major defining characteristic of the informal economy). Therefore, the distinction between formal and informal sectors in this research was based on the workplace establishment, not the respondent. It is also important to note that not all informal workers are self-employed as is commonly perceived. In some cases, respondents engaged in informal work were recorded as employees if they indicated that they were not the boss. This was common in shanty retail shops selling clothes or shoes in the informal sector; in such cases the respondent surveyed was oftentimes one of the relatives of the self-employed boss.

Data Limitations

While this research attempted to supplement the lack of publicly available data from Xinjiang, the survey design was not without drawbacks. First, all of the respondents surveyed were employed in the tertiary sector, therefore the survey results cannot be generalised to other sectors of the economy, such as manufacturing or extraction industries. Given the regional sensitivities, it is difficult to assess the level of respondent and measurement bias. Approximately 20 per cent of the respondents refused to be surveyed. In the case of refusal, the next store location was visited. The question about taxes elicited the largest no response among interviewees, with approximately 50 per cent refusing to answer.

Given the importance of the informal economy in this research, it is necessary to discuss the potential measurement bias for differentiating between formal and informal employment. Respondents who were actually informal workers but were employed in establishments with visible work permits, were classified as being engaged in the formal sector. The most prominent example of this scenario arises when conducting surveys of waiters and cooks working in official restaurants. The waiters and cooks, many of whom were or may actually be informally employed, were recorded as being engaged in the formal sector because they were employed in a restaurant that is regulated and certified by the authorities, evident in its visible and up-to-date work permit.

On the other hand, respondents marked as informal were clearly engaged in the informal sector, thus preserving the integrity of the informal classification, which again was determined by the lack of a visible work permit and on-site observation. For example, a common on-site observation of this category of respondents was the manner in which their mobile merchandise (i.e., food, snacks, clothing) was displayed on a moveable cart, often attached to a bicycle. These respondents were clearly engaged in the informal sector. Based on unstructured interviews with informal workers, several respondents anecdotally confirmed their informal status as they proudly declared that they had to move around often to avoid being issued offence tickets by the authorities. In one case, one informal worker was observed hurriedly carting away his merchandise on foot in order to escape the authorities.


Descriptive statistics were used to analyse main background characteristics by ethnicity (see Table 2). Of the 593 respondents surveyed, 72 per cent of the respondents were Han, 21 per cent were Uyghur, and 7 per cent were Hui Muslim. 62.1 per cent of Uyghur workers were male, whereas only a slight majority of Han and Hui workers were female. The majority of Han were married (75.6 per cent), compared to only 58.7 per cent of Uyghurs and 63.4 per cent of Hui. Seventy-seven per cent of Han respondents were migrants, 61 per cent of whom come from outside of Xinjiang, many migrating from neighbouring Gansu Province (see Figure 1). The findings confirmed the research of existing literature: a flood of Han in-migrants have entered Urumqi's workforce and compete with Uyghurs for local jobs.

Table 2 captures three variables that were used as a proxy for human capital: education, language skills and work experience. The education attainment level for all three ethnic groups was comparable, with Uyghur averaging 4.8 years of schooling, Han averaging 5.2 years and Hui averaging 5.3 years. (47) Some scholars reported a much larger gap in schooling between Uyghurs and Han Chinese. However their findings relied on census data for all of Xinjiang, which included a large number of poor, rural minority peasants. (48) In contrast, the results reported in this research suggested that the urban education rate for minorities was comparable to that of the other ethnic groups. Almost all of the Han and Hui were conversationally fluent in Mandarin, compared to only 73.8 per cent for Uyghur. The average work experience for all three ethnic groups was almost equivalent, ranging from 4.3 years for Uyghur to 4.95 years for Hui. Respondents from all three ethnic groups had relatively low work experience: 4.8 years for Han, 4.34 years for Uyghur and 4.95 years for Hui. Combined with the low average age, the survey respondents were largely workers who had very recently entered Urumqi's labour market.

Based on the aforementioned findings on human capital, preliminary evidence suggests that the Han-Uyghur income gap could not be explained by a variation in education levels. Rather, the major variation in human capital attainment stemmed from one's mastery of the Mandarin language. This finding lent some support to the argument that ethnic-based income disparity may be the result of language barriers in the labour market. In-depth multivariate analyses were carried out to statistically test the relationship among ethnicity, human capital and income.

Employment type was classified into three non-mutually exclusive categories: self-employed, managers and employees. (49) The results showed that an overwhelming majority of Han were engaged in self-employed work, in comparison with Uyghurs and Hui; although there is a similar representation of managers across all three ethnic groups. The majority of respondents in all three ethnic groups were employees, meaning that they were neither self-employed nor managers, with Uyghur having the highest proportion of employee workers (64.2 per cent).

As described in the sampling design section above, the majority of respondents were unskilled: 71 per cent for Han, 71.4 per cent for Uyghur and 80.4 per cent for Hui. The average working hours clocked per week for all ethnic groups were high, averaging between 65.4 hours per week for Hui and 68.2 hours per week for Uyghur. Seventy per cent of Han were employed in the formal sector, and 10 per cent were employed by SOEs. Similarly, 67 per cent of Uyghur were employed in the formal sector, and 11.1 per cent were employed by SOEs. Despite only a slightly higher percentage of Hui respondents employed in the formal sector (73.1 per cent), there was a noticeable increase in Hui who were employed by SOEs (17.1 per cent).

This finding is interesting because some scholars have noted that compared to Uyghur, Hui do not face as much difficulty in the labour market in terms of market discrimination and income inequality, leading to better assimilation into Chinese society. One reason put forth in the literature to explain the Hui-Uyghur variation in assimilation is based on the fact that the physical features of Hui people resemble more closely those of the Han Chinese. Despite the affirmative action policies that have attempted to protect minorities from discrimination and ensure that minority quotas are met for government positions, these policies may disproportionately benefit some minority groups, such as Hui, over other minority groups, such as Uyghur. At this time, however, this line of reasoning is only exploratory and cannot be confirmed based on the present set of data.

The average monthly earnings varied considerably based on ethnicity. On average, Han Chinese earned almost 28 per cent more than Uyghur and almost 14 per cent more than Hui. More than 36 per cent of Han indicated that their nominal income in 2008 had increased since 2005, compared to only 26.8 per cent for Hui. Shockingly, only 11.9 per cent Uyghur indicated an increase in their monthly earnings since 2005. Correspondingly, 66 per cent of Uyghur expressed job dissatisfaction with their current job, although job dissatisfaction was also high for Han (58 per cent) and Hui (65.8 per cent). These findings indicated a large income gap among the ethnic groups in this survey. With only a small percentage of Uyghurs enjoying nominal income growth from 2005 to 2008, combined with a relatively large majority of them expressing high level of dissatisfaction with their current job, it is easy to argue that economic despair can easily translate into ethnic discontent. This claim offers some support for the literature that cite economic reasons as a main driver behind Uyghur discontent and the recent ethnic violence in Urumqi.

Table 3 further examines the nature of income disparity among ethnic groups by looking at the interaction among ethnicity, migrant status and gender.

The results revealed that, in particular, Uyghur natives experienced the greatest degree of income disparity relative to both Han natives and Han migrants. (50) Compared to Uyghur natives, on average, Han natives were found to earn 66 per cent more, Han intra-provincial migrants earned 69 per cent more, and Han inter-provincial migrants almost 39 per cent more. The high degree of income disparity was especially surprising given that Uyghur natives' average educational attainment (5.6 years) was similar to that of Han natives and Han intra-provincial migrants, and slightly higher than that of Han inter-provincial migrants.

On the other hand, the ethnic inequality gap was considerably reduced when comparing Uyghur migrants to Han workers. At the same time, a widening income gap for Uyghur was introduced along lines of migrant status. The relatively large income gap between Uyghur migrants and Uyghur natives, approximately 33 per cent, has been explained by Howell and Fan. (51) According to the authors, the survey data recorded Uyghur migrants who are largely the benefactors of affirmative action policies in Xinjiang. A select number of successful minorities from surrounding Xinjiang counties choose to migrate to Urumqi, which is a top destination for Uyghurs. The authors warned against generalising the success of Uyghur migrants to all Uyghur living outside of Urumqi, and suggested that these successful Uyghur migrants who participated in the Urumqi surveys were not representative.

Considerable gender income inequality exists as well. Females earned less than their male counterparts in all of the five ethnic-migrant sub-groupings. Only Han native females and Uyghur migrants earned close to equal wages as their male counterparts. Conversely, the largest female-male income gap existed for Han intra-provincial migrants, where females made only 68 per cent of what their male counterparts made.

Multivariate Analysis of Income Disparity: Minority Outcomes in the Formal and Informal Economy

The descriptive analyses thus far only highlight the univariate differences among workers, and may in fact be the result of spurious relationships. To test for statistical differences among respondents' characteristics, the author employs multiple regression analyses to verify the relationship between ethnicity and income disparity.

To estimate the ethnic wage differential conditioned on ethnicity and individual characteristics (human capital, demographic, geographic), ordinary least squares regression analysis was performed. The estimated regression is of the following form:

log [w.sub.i] = [X.sub.i]B + [alpha] [Uyghur.sub.i] + [e.sub.i]

The dependent variable is the logarithm of monthly income, where B denotes a column matrix of control variables, including demographics, migrant status, employment characteristics, human and social capital; X denotes the corresponding coefficients that measure the magnitude and direction of the relationship between the independent variable and the dependent variable, income. The main variable of interest is Uyghur, and a is the coefficient that estimates the level of income disparity for Uyghur respondents. It is important to note that a thorough analysis of residuals was performed and all three models behaved well. Also, low variance inflation factors (VIFs < 2) reveal that the model did not suffer from high multicollinearity. The r-squared for each model was lower than ideal; however, typical econometric models on the labour market are often around .30. The F-statistics for each model were also statistically significant at greater than the .001 level, indicating that the overall models were each statistically significant.

Model 1 examines the entire Urumqi sample and reveals that the coefficients of "Uyghur", "female" and "migrant" ("intra-provincial", "inter-provincial") are all negatively associated with income, thus reinforcing the labour market segmentation theory. At the same time, the coefficients of the human capital variables--"education" and "work experience"--were both positively associated with income, thereby reinforcing the human capital theory. The coefficient of Uyghur was the largest in magnitude, surprisingly affecting income outcomes more so than either of the statistically significant human capital variables. Taking the exponent of the coefficient, Uyghur respondents were found to earn 29 per cent less their Han counterparts.

The coefficient of "migrant" ("intra-provincial", "inter-provincial") makes good sense and supports the literature on migration in China. The empirical literature on other cities shows that most low-skilled migrants do not have official hukou, or official permission to work in their destination city, and therefore a person's hukou status is extremely important in predicting wage outcomes. (52) Model 1 also shows that skilled workers, formal sector workers, SOE workers and self-employed workers were positively related to income. After taking the exponent of the coefficient of "formal sector", respondents employed in the formal sector earned almost 22 per cent more relative to their counterparts employed in the informal sector.

Models 2 and 3 are subsamples of Model 1 and are delineated according to whether respondents are employed in the formal (N = 411) or the informal (N = 182) economy. Model 2 and Model 3 each show a different set of income-generating processes that exists in the formal and informal sectors respectively. Interestingly, the income gap between females and males in the informal sector is considerably larger than the gender income gap in the formal sector. On the other hand, the Uyghur-Han income gap was about the same in both the formal and informal sectors and remained the largest coefficient produced by either model. These results control the variation in education and work experience, thus indicating that the impact of ethnicity, gender and migrant status on income was independent of human capital acquisition in both the formal and informal sectors. The consistently large coefficient of "Uyghur" in both the formal and informal sectors suggested that market discrimination against Uyghur may be more systemically along lines of ethnicity rather than gender or migrant status.

In addition to inter-group income disparity, the formal sector appeared to reward workers with higher levels of education and work experience; whereas neither education nor work experience was statistically significant in the informal sector. Instead, the informal sector tended to reward workers based on some type of marketable skills. The coefficient of "skilled workers" in the formal market is only .016, compared to .357 in the informal sector. Taking the exponents of these two coefficients, skilled workers in the informal sector earned 43 per cent more than their unskilled counterparts; whereas in the formal sector, skilled workers earned only 2 per cent more income than unskilled workers.

Based on the coefficients of "SOE", state sector workers earned 15 per cent more than non-state workers in the formal sector. Based on this result, combined with the coefficient on "formal sector" in Model 1, it is clear that workers in Urumqi earn the highest income on average in the state sector, followed by the non-state sector in the formal economy, then informal workers. These findings also support Zang's conclusions that the returns to income for being employed as a state worker are greater than the returns being employed in the non-state sector.

The interaction term--"Uyghur-migrant"--in Model 2 confirms the results in descriptive statistics in Table 3, indicating that Uyghur migrants earned substantially more than Uyghur natives, at least in the formal sector. This finding is interesting and adds to the conventional Chinese migrant labour literature, because in other Chinese cities, urban natives with local permanent household registration earn more on average than migrants. (53) However, in Urumqi, the addition of ethnicity complicates this relationship and reconfigures the social hierarchical order, which places Uyghur migrants above Uyghur natives. Based on the coefficients of "ethnicity" and "migrant status", Urumqi's urban hierarchy is as follows: Han natives are ranked at the top, followed by Han intra-provincial migrants, Han inter-provincial migrants, Uyghur migrants and finally Uyghur natives (despite having local urban hukou).


Findings from the descriptive statistics contrasted with previous scholarly work on Chinese minorities, which have suggested that minority income disparity and placement in the labour market are largely a function of inter-group differences in education attainment level. (54) As seen from the results put forth in this article, Uyghur natives have higher education levels than Han inter-provincial migrants, yet still make 39 per cent less on average than Han inter-provincial migrants.

From the regression results, the accumulation of human capital plays an important role in the emerging labour market in Urumqi's formal sector, and macro-economic policy should continue to promote education and training in all minority-concentrated areas. However, respondents in the informal sector earn 22 per cent less than their respondents in the formal sector, thus contributing to the intra-urban income gap; and human capital attainment is not an important determinant in income.

Income disparity existed not only between the formal and informal sector, but also within each sector as well. Within both the formal and informal sectors, being designated as a Uyghur, a female, or a migrant had greater impact on income outcomes than the accumulation of human capital. The degree of income disparity was largest when based on ethnicity. Uyghurs earned 29 per cent less, on average, compared to their Han counterparts. This effect on income was independent of education and work experience, indicating a large degree of market discrimination on the basis of ethnicity in both the formal and informal sectors. Again, policy that only promotes education advancement as a panacea for inequality will likely to be ineffective due to the segmented nature of the labour market along the ethnic, gender and migrant status divides.

Based on this evidence, even if overall education attainment were to increase among urban workers, due to the segmented nature of the labour market, wages would not converge and income disparity would continue to increase, at the very least between the formal and informal sectors. Job growth is vital to diminish the role and presence of the informal sector and achieve inter-sectoral wage convergence. Unfortunately, due to a shortage of jobs, coupled with the influx of Han in-migrants to urban areas in Xinjiang, many of the would-be formal Uyghur workers turn to the informal economy to make a living, not as a first choice, but as a last resort. This view is supported by several unstructured interviews carried out on Uyghur natives, as well as Han migrants, employed in the informal sector; the interviewed respondents confirmed that they first attempted to find formal employment, but were unable to do so, and out of necessity became engaged in informal work.

The implications of this study for future economic development policy in Xinjiang are clear. The consequences of relying on purely economic growth policy to resolve ethnic-based income disparity in a segmented labour market may further lead to increased inequality. This in turn, will likely result in an increase in grievances among minorities--especially in conflict-prone provinces like Xinjiang--who attain relatively high levels of education, but are still unable to secure a good job due to discrimination and exclusion. This reality indeed has significant relevance to China's new "leapfrog development" policy--implemented in 2010 in response to the deadly 2009 Urumqi riots--which favours educational attainment, economic growth and rapid development. It is also noted that China's emphasis on efficiency over equity has led to all kinds of regional, social and economic inequalities across China. (55) It is imperative to ensure that affirmative action policies are robust and effective, especially during the initial stages of growth that will occur over the next few years as a result of the "Yuan Jiang" programme.

In order to assess the potential positive impacts of the "Yuan Jiang" programme, the author returns to a bold statement made in regard to China's "Open up the West" policy. Lai proclaimed back then that the economic and political benefits reaped by Chinese leaders may be limited due to governmental inefficiency, ethnic division, and low economic returns. (56) A decade later, similar apprehensions can be made about the "Yuan Jiang" programme. If the outcome of the new "leapfrog development" policy implemented in the 12th Five-Year Plan (2011-2015) parallels the outcome of "Open up the West" initiative implemented during the 10th Five-Year Plan (2001-2005), economic growth in Xinjiang will undoubtedly increase; at the same time, intra-urban and intra-rural inequality may persist or even worsen, especially if market-correcting mechanisms are not enacted to offset growing market segmentation between the formal and informal sector, as well as within each sector, along the lines of ethnicity, gender and migrant status.

Anthony J. Howell ( is a PhD candidate in the Geography Department at University of California, Los Angeles. His research interests include economic geography (agglomeration, trade, knowledge spillovers and productivity growth), development geography (emerging labour markets, minority studies and income inequality), as well as a strong regional focus on China and a commitment to the use of (spatial) statistical methods and GIS techniques to studying these topics.

(1) Li, Shi, and Sai Ding, "An Empirical Analysis of Income Inequality Between a Minority and the Majority in Urban China: The Case of Ningxia Hui Autonomous Region", Global COE Hi-Stat Discussion Paper Series 022, 2009.

(2) Hannum, Emily and Yu Xie, "Ethnic Stratification in Northwest China", Demography, 35 (1998): 323-33; Gustafsson, Bjorn and Shi Li, "The Ethnic Minority-Majority Income Gap during Transition", Economics Development and Cultural Change 51 (2003): 805-22.

(3) Gustafsson, Bjorn, and Sai Ding, "Assessing Ethnic Disparities in Income and Poverty in China: The Case of Han and Hui in Ningxia", Paper prepared for the 32nd General Conference of The International Association for Research in Income and Wealth, 2012.

(4) Huang, Philip, "China's Neglected Informal Economy: Reality and Theory", Modern China 35 (2009): 405-38.

(5) Vicziany, Marika and Zhang Guibin "Minority Entrepreneurs on the Move: The Uygurs" in Globalization and labour mobility in China, eds. Ingrid Neilson, Russell Smyth and Marika Vicziany (Melbourne, Monash Asia Institute 2006): 54-72.

(6) Zang Xiaowei, "Uyghur-Han Earnings Differentials in Urumchi", The China Journal, no. 65 (Jan. 2011): 141-155.

(7) Victor Nee, "Social Inequalities in Reforming State Socialism: Between Redistribution and Markets in China", American Sociology Review 56 (1991): 267-82.

(8) Hannum and Xie, "Ethnic Stratification in Northwest China".

(9) Emily Hannum, Jere Behrman, Wang Meiyan and Liu Jihong, "Education in the Reform Era", in China's Great Economic Transformation, ed. Loren Brandt and Thomas Rawski (Cambridge: Cambridge University Press, 2008); Emily Hannum and Wang Meiyan, "China. A Case Study in Rapid Poverty Reduction", in Indigenous Peoples, Poverty and Development, ed. Gillette Hall and Harry Patrinos (Cambridge: Cambridge University Press, 2010).

(10) Gustafsson and Ding, "Assessing Ethnic Disparities in Income and Poverty in China".

(11) Harold Bauder, "Culture in the Labor Market: Segmentation Theory and Perspectives", Progress in Human geography 25, no. 1 (2001): 37-52; Ian Gordon, "Migration in a Segmented Labour Market", Transactions of the Institute of British Geographers 20, no. 2 (1995): 139-55.

(12) Marianthi Leontaridi, "Segmented Labour Markets: Theory and Evidence", Journal of Economic Surveys 12, no. 1 (1998): 103-19.

(13) Gary Becker, The Economics of Discrimination (Chicago: University of Chicago Press, 1971); Zang Xiaowei, "Labor Market Segmentation in Urban China", The Sociological Quarterly 43, no. 1 (2002): 27-44; and Zang, "Uyghur-Han Earnings Differentials in Urumchi".

(14) Joseph Altonji and Rebecca Blank, "Race and Gender in the Labor Market", Handbook of Labor Economics, ed. O. Ashenfelter and D. Card, first edition, vol. 3 (1999): 3143-4359; Bauder, "Culture in the Labor Market"; Amelie Constant and Douglas Massey, "Labor Market Segmentation and the Earnings of German Guestworkers", Population Research and Policy Review 24, no. 5 (2005): 489-512; Gordon, "Migration in a Segmented Labour Market"; Roger Hayter and Trevor Barnes, "Labour Market Segmentation, Flexibility, and Recession: a British Columbian Case Study", Environment and Planning C 10 (1992): 333-53; Daniel Hiebert, "Local Geographies of Labor Market Segmentation: Montreal, Toronto, and Vancouver", Economic Geography 75, no. 4 (1999): 339-69; Kenneth Hudson, "The New Labor Market Segmentation: Labor Market Dualism in the New Economy", Social Science Research 36, no. 1 (2007): 286-312; Sara McLafferty and Valerie Preston, "Spatial Mismatch and Labor Market Segmentation for African-American and Latina Women", Economic Geography 68, no. 4 (1992): 406-31; and Michael Reich, David Gordon and Richard Edwards, "Dual Labor Markets: A Theory of Labor Market Segmentation", The American Economic Review 63, no. 2 (1973): 359-65.

(15) E.M. Beck, Patrick Horan and Charles Tolbert, "Industrial Segmentation and Labor Market Discrimination", Social Problems 28, no. 2 (1980): 113-30; James Coverdill, "The Dual Economy and Sex Differences in Earnings", Social Forces 66, no. 4 (1988): 970-93; Hiebert, "Local Geographies of Labor Market Segmentation"; McLafferty and Preston, "Spatial Mismatch and Labor Market Segmentation for African-American and Latina Women"; Lesley Reid and Beth Rubin, "Integrating Economic Dualism and Labor Market Segmentation: The Effects of Race, Gender, and Structural Location on Earnings, 1974-2000", Sociological Quarterly 44, no. 3 (2003): 405-32; Ross Stolzenberg, "Ethnicity, Geography, and Occupational Achievement of Hispanic Men in the United States, American Sociological Review 55, no. 1 (1990): 143-54.

(16) Edward Telles, "Urban Labor Market Segmentation and Income in Brazil", Economic Development and Cultural Change 41, no. 2 (1993): 231-49.

(17) International Labour Organization, Women and Men in the Informal Economy: a Statistical Picture (Geneva: International Labour Organization, 2002).

(18) Hu Angang and Zhao Li, "Woguo zhuanxingqi chengzhen feizhenggui jiuye yu feizhenggui jingji (1990-2004)", (Urban Informal Employment and Informal Economy in our Nation's Transition Period [1990-2004]). Qinghua daxue xuebao (zhexue shehui kexue ban) 21, no. 3 (2006): 111-19.

(19) Huang, "China's Neglected Informal Economy".

(20) Cindy Fan, "The Elite, the Natives, and the Outsiders: Migration and Labor Market Segmentation in Urban China", Annals of the Association of American Geographers 92, no. 1 (2002): 103-24; Dorothy Solinger, Contesting Citizenship in Urban China: Peasant Migrants, the State, and the Logic of the Market (Berkeley: University of California Press, 1999).

(21) Cindy Fan, "Permanent Migrants, Temporary Migrants, and the Labour Market in Chinese Cities" in Resource Management, Urbanization and Governance in Hong Kong and the Zhujiang Delta, ed. Wong Kwan Yiu and Shen Jianfa (Hong Kong: The Chinese University Press, 2003): 55-78.

(22) Howell, Anthony, "Labour Market Segmentation in Urumqi, Xinjiang: Exposing Labour Market Segments and Testing the Relationship between Migration and Segmentation" Growth and Change 42, no. 1 (2011): 200-26.

(23) Li and Ding, "An Empirical Analysis of Income Inequality between a Minority and the Majority in Urban China"; Gustafsson and Ding, "Assessing Ethnic Disparities in Income and Poverty in China".

(24) Zang, Xiaowei "Market Reforms and Han--Muslim Variation in Employment in the Chinese State Sector in a Chinese City", World Development 36, no. 11 (2008): 2341-352.

(25) Robin Iredale, Naran Bilik, Su Wang, Guo Fei and Caroline Hoy, Contemporary Minority Migration, Education and Ethnicity in China (Northampton, MA: Edward Elgar Limited, 2001); Gardner Bovingdon, "Autonomy in Xinjiang: Han Nationalist Imperatives and Uyghur Discontent", Policy Studies 11 (2004): 1-46; Dru Gladney, Dislocating China: Muslims, Minorities, and Other Subaltern Subjects (Chicago: University of Chicago Press, 2004); and Clifton W. Pannell and Phillip Schmidt, "Structural Change and Regional Disparities in Xinjiang, China", Eurasian Geography and Economics 47 (2006): 329-52.

(26) Statistical Bureau of Xinjiang Uyghur Autonomous Region (SBX), Xinjiang tongji nianjian (Xinjiang Statistical Yearbook 2000) (Beijing: China Statistics Press, 2009).

(27) Preeti Bhattacharji, "Uighurs and China's Xinjiang Region", Council on Foreign Relation, 2012, at <> [3 June 2012]; Gardner Bovingdon, "The Not-So-Silent Majority: Uyghur Resistance to Han Rule in Xinjiang", Modern China 28, no. 1 (2002): 39-78; Colin Mackerras, "Xinjiang at the Turn of the Century: The Causes of Separatism", Central Asian Survey 20, no. 3 (2001): 289-303; Michael Webber, "The Places of Primitive Accumulation in Rural China", Economic Geography 84, no. 4 (2009): 395-421.

(28) Ildiko Beller-Hann, "Temperamental Neighbors: Uyghur-Han Relations in Xinjiang, Northwest China", in Imagined Differences: Hatred and Construction of Identity, ed. Gunther Schlee (London, UK: Palgrave, 2002): 57-81; Pannell and Schmidt, "Structural Change and Regional Disparities in Xinjiang, China"; and Iredale et al., Contemporary Minority Migration, Education and Ethnicity in China.

(29) Nicholas Becquelin, "Staged Development in Xinjiang", The China Quarterly 178 (2004): 358-78; Bovingdon, "The Not-So-Silent Majority"; Gladney, "Dislocating China: Muslims, Minorities, and Other Subaltern Subjects"; Mackerras, "Xinjiang at the Turn of the Century"; Pannell and Schmidt, "Structural Change and Regional Disparities in Xinjiang"; Iredale et al., Contemporary Minority Migration, Education and Ethnicity in China; Stanley Toops, "The Demography of Xinjiang", in Xinjiang: China's Muslim Borderland, ed. S. Frederick Starr (Armonk, NY: M.E. Sharpe, 2004), pp. 241-63; Herbert Yee, "Ethnic Relations in Xinjiang: A Survey of Uygur-Han Relations in Urumqi", Journal of Contemporary China 12, no. 36 (2003): 431-52.

(30) Nicholas Becquelin, "Xinjiang in the Nineties", The China Journal 44 (2000): 65-90.

(31) James Milward, Beyond the Pass: Economy, Ethnicity, and Empire in Qing Central Asia, 1759-1864 (Stanford: Stanford University Press, 1998).

(32) Jessica Koch, "Economic Development and Ethnic Separatism in Western China: A New Model of Peripheral Nationalism", Asia Research Centre, Working Paper no. 134 (Perth, Australia: Murdoch University, 2006).

(33) Sautman, Barry, "Ethnic Law and Minority Rights in China: Progress and Constraints", Law and Policy 21 (1999): 283-314.

(34) Gillette, Maris, Between Mecca and Beijing: Modernization and Consumption among Uyghur Chinese Muslims (Stanford: Stanford University Press, 2000); Zang, Xiaowei, "Ethnic Representation in the Current Chinese Leadership", The China Quarterly 153 (1998): 107-27; Zang, "Market Reforms and Han-Muslim Variation in Employment in the Chinese State Sector in a Chinese City.

(35) Sautman, "Ethnic Law and Minority Rights in China: Progress and Constraints"; Zang, "Market Reforms and Han-Muslim Variation in Employment in the Chinese State Sector in a Chinese City".

(36) Gael Raballand and Agnes Andresy, "Why Should Trade between Central Asia and China Continue to Expand?", Asia Europe Journal 5, no. 2 (2007): 235-52.

(37) Becquelin, "Staged Development in Xinjiang".

(38) Mathew Moneyhon, "China's Great Western Development Project in Xinjiang: Economic Palliative, or Political Trojan Horse?", Denver Journal of International Law and Policy 31, no. 3 (2003): 491-523.

(39) Becquelin, "Staged Development in Xinjiang".

(40) China "White Paper, 2009, "White Paper on Development of Xinjiang" (China Daily) http://www. [3 Jun 2012].

(41) Li Xue, "Xinjiang bintuan jinnian jiang jianshe mitan meihua gongdeng lin gyu 50 ge zhongdian xiangmu" (Xinjiang Corps will Build 50 Key Projects of Coal and Coal Chemical Fields), Bingtuan Net, at <> [3 Jun 2012].

(42) Pannell and Schmidt, "Structural Change and Regional Disparities in Xinjiang, China".

(43) H arlan Tyler, "Private Sector Development in Xinjiang, China: A Comparison between Uyghur and Han", Espace, Populations, Societies 3 (2009): 407-18.

(44) Pannell and Schmidt, "Structural Change and Regional Disparities in Xinjiang, China".

(45) Hannum and Xie, "Ethnic Stratification in Northwest China".

(46) Iredale et al., Contemporary Minority Migration, Education, and Ethnicity in China.

(47) Five years of education is almost equivalent to having a primary school education.

(48) Zang reports similar findings for Han-Uyghur education attainment level; see Zang, "Uyghur-Han Earnings Differentials in Urumchi".

(49) The three categories do not add up to 100 per cent because some respondents were classified as both self-employed and manager. The both self-employed and manager dual classification was 15 per cent for Han, 11 per cent for Uyghur and 7 per cent for Hui.

(50) Due to the limited number of Hui respondents, they were not included in the disaggregated analysis by migrant status and gender.

(51) Anthony Howell and Cindy Fan, "Migration and Inequality in Xinjiang: A Survey of Han and Uyghur Migrants in Urumqi", Eurasian Geography and Economics 52 (2011): 119-39.

(52) Liu Zhiqiang, "Institution and Inequality: the Hukou System in China", Journal of Comparative Economics 1 (2005): 133-57; Lu Zhigang and Song Shunfeng, "Rural-Urban Migration and Wage Determination: The Case of Tianjin, China", China Economic Review 17, no. 3 (2006): 337-45.

(53) Cindy Fan, "The Elite, the Natives, and the Outsiders"; Zang Xiaowei, "Labor Market Segmentation and Income Inequality in Urban China; Xu Wei, Tan Kok-Chiang and Wang Guixin, "Segmented Local Labour Markets in Post Reform China: Gender Earnings Inequality in the Case of Two towns in Zhejiang Province", Environment and Planning A 38, no. 1 (2006): 85-109.

(54) Hannum and Xie, "Ethnic Stratification in Northwest China"; Zang, "Market Reforms and Han-Muslim Variation in Employment in the Chinese State Sector in a Chinese City".

(55) There is a long-standing body of research on the growth-inequality linkages. Some evidence suggests that prior short-term growth in China leads to inequality, while sustainable and long-term growth leads to a reduction in inequality. See Dustin Chambers, Wu Ying and Yao Hong, "A Tale of Two Provinces: How Growth can Help or Hinder Equality in China", International Research Journal of Finance and Economics 12 (2007): 214-20.

(56) Lai Hongyi, "China's Western Development Program: Its Rationale, Implementation, and Prospects", Modern China 28 (2002): 432-66.


Population of Urumqi by Major Ethnicity and District, 2008

                           Total Population   Han     Uyghur

Urumqi City (million)       2.41               1.75    0.31
Tianshan District (%)      22.8               20.2    40.1
Shayibake District (%)     21.8               23.0    22.2
Xinshi District (%)        21.5               24.0    16.4
Shuimogou District (%)     10.9               12.6     8.5
Tou Tunhe District (%)      5.7                5.6     6.3
Da Bancheng District (%)    1.9                1.3     0.82
Midong District (%)        11.6               10.9     3.5

                           Hui     Kazak   Other

Urumqi City (million)       0.24    0.07    0.04
Tianshan District (%)      15.40   23.4    29.0
Shayibake District (%)     15.0    11.9    24.6
Xinshi District (%)        13.3     7.6    23.9
Shuimogou District (%)      3.9     4.1     5.4
Tou Tunhe District (%)      6.4     2.5     5.4
Da Bancheng District (%)    5.8     9.7     0.55
Midong District (%)        29.3     6.7     7.3

Source: Statistical Bureau of Xinjiang Uyghur Autonomous Region,


Main Background Characteristics for Urumqi Workers by Ethnicity

Variable                  Han                  Uyghur      Hui
                          (N = 426)            (N = 126)   (N = 41)

                          Demographics (%)

Male                       49.4                 62.1       48.4
Age (Years)                33.7                 30.1       32.4
Married                    75.6                 58.7       63.4

                          Migrant Status (%)

Native                     23.7                 46.8       56.1
Intra-Provincial           16.0                 53.2       31.8
Inter-Provincial           60.3                  0.0       12.1

                          Human Capital

Education (Years)           5.2                  4.8        5.3
  Illiterate (%)            2.8                  7.1        2.4
  Primary (%)              15.5                 19.8       17.1
  Secondary (%)            28.9                 29.4       24.4
  High (%)                 27.0                 24.6       26.9
  Technical (%)            17.8                  7.1       22.0
  University (%)            8.0                 12.0        7.3
Mandarin Fluent (%)        96.7                 73.8       97.5
Work Experience (Years)     4.8                  4.3       4.95

                          Employment (%)

Employment Type (%)
  Employee                 52.6                 64.2        60.7
  Employer                 13.4                 11.9        14.6
  Self-Employed (%)        49.0                 34.9        31.7
Skill Level (%)
  Unskilled                71.0                 71.4        80.4
  Skilled                  29.0                 28.6        19.6
Hours/Week                 67.4                 68.2        65.4
Formal (%)                 70.0                 67.0        73.1
Soe (%)                    10.3                 11.1        17.1
Job Dissatisfaction (%)    58.0                 66.0        65.8


Monthly Income (Yuan)     1,612                1,285       1,424
Earn More Than 2005 (%)      36.3                 11.9        26.8


Average Income Earnings by Ethnicity, Migrant Status and Nationality

                                   Number of    Average
                                  respondents   monthly
Respondent Group                   (Average)    income

All respondents                       593        1,554
Total Uyghur                          126        1,285
Uyghur natives                        59         1,076
Uyghur migrants                       67         1,428
Total Han                             426        1,612
Han natives                           101        1,782
Han migrants (intra-provincial)       68         1,822
Han migrants (inter-provincial)       257        1,494
Total Hui                             41         1,423
Hui natives                           --          --
Hui migrants                          --          --

                                     Years of       Female-Male
                                     Education      earnings (%)
Respondent Group                  Earnings (yuan)

All respondents                         5.1              77
Total Uyghur                            4.8              85
Uyghur natives                          5.6              70
Uyghur migrants                         4.3              96
Total Han                               5.2              82
Han natives                             6.3              96
Han migrants (intra-provincial)         6.2              68
Han migrants (inter-provincial)         4.5              69
Total Hui                               5.3              81
Hui natives                             --               --
Hui migrants                            --               --


Sectoral Multivariate Analyses of Income Disparity

                                 Model 1: Urumqi     Model 2:
Coefficients:                    Tertiary Sector   Formal Sector

(Intercept)                        7.56 ***          7.65 ***


Uyghur                            -0 339            -0.312 ***
Hui                               -0.114            -0.078


Female                            -0.073 *          -0.052 *
Age                                0.027 *           0.039 **
Married                            0.024             0.047

Migrant Status

Migrant (Intra)                   -0.178 **         -0.255 ***
Migrant (Inter)                   -0.115 *          -0.126 *
Native (R)                         -                 -

Human Capital

Education                          0.029 **          0.038 **
Mandarin Fluent                    0.041             0.16
Work Experience                    0.038 *           0.06 **
Work Experience [conjunction]2     0.001             0.001


Skilled                            0.003 *           0.016 *
Formal                             0.197 ***         -
Soe                                0.129 *           0.143 *
Self-Employed                      0.128 **          0.051
Manager                            0.145             0.073

Interaction Terms

Uyghur-Migrant (Intra)             0.177             0.284 **
Uyghur-Female                     -0.005             0.034

Model Performance

[R.sup.2]                          0.272             0.211
F-Statistic                       10.83              2.743
N                                593               411

                                    Model 3:
Coefficients:                    Informal Sector

(Intercept)                        6.97 ***


Uyghur                            -0.326 ***
Hui                               -0.120


Female                            -0.139 **
Age                                0.015
Married                           -0.105

Migrant Status

Migrant (Intra)                   -0.241 *
Migrant (Inter)                   -0.062 *
Native (R)                         -

Human Capital

Education                          0.002
Mandarin Fluent                    0.148
Work Experience                    0.02
Work Experience [conjunction]2     0.001


Skilled                            0.357 **
Formal                             -
Soe                                -
Self-Employed                      0.314 ***
Manager                            -

Interaction Terms

Uyghur-Migrant (Intra)             0.217
Uyghur-Female                     -0.04

Model Performance

[R.sup.2]                          0.201
F-Statistic                        2.243
N                                182

Notes: *** signifies P < .001; ** signifies P < .01; and
* signifies P < .05.
COPYRIGHT 2013 East Asian Institute, National University of Singapore
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:RESEARCH
Author:Howell, Anthony J.
Publication:China: An International Journal
Article Type:Report
Geographic Code:9CHIN
Date:Dec 1, 2013
Previous Article:Resources for China-ASEAN relations.
Next Article:China's technology policy change: how effective has it been?

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |