Printer Friendly

The promotion logic of prefecture-level mayors in China.


Leadership research is a very important component of China Studies. With Deng Xiaoping's decentralisation strategy, sub-national governments were delegated broad economic powers, and the role of local leaders became more prominent. (1) These new circumstances have prompted observers to note that "in some senses, real politics in China is local politics". (2) Yet these powerful leaders do not operate without constraints. They are appointees by nature, (3) and their political promotion is subject to the will of their superiors. Most previous studies have concentrated on the analysis of national and provincial leaderships. Important as they are, their limited scope leaves many questions unanswered regarding the political elite in the country as a whole. Some systematical studies on the political mobility of local leaders have been published, (4) but few pay attention to the mobility of mayors. (5) By focusing on mayors of prefecture-level cities (diji shi), this article seeks to fill this vacuum. (6)

Cities and Mayors in China

According to the Organic Law of the Local People's Congresses and Local People's Governments of the PRC (Difang geji renmin daibiaodahui he defang geji renminzhengfu zuzhifa), (7) there are three types of cities in Mainland China. The first is a directly-controlled municipality (a municipality directly under the central government, zhixia shi), which is a city having a status equal to that of a province (shengji shi). The second is a city divided into districts (though directly-controlled municipalities are also divided into districts); in reality, this type includes prefectural- and sub-provincial-level cities (fushengji chengshi and diji shi) while the term "prefecture" is never mentioned in the organic law. The third is a city that is not divided into districts; this includes all sub-prefecture-level cities (fudiji chengshi) and county-level cities (xianji shi).

The bureaucratic-ranking structure of cities is much more sophisticated in practice than in theory. There are five levels of hierarchy for Chinese cities: (1) directly-controlled municipalities, as described above; (2) sub-provincial-level cities, which are prefecture-level cities ruled by provinces but administered independently with respect to economy and local by-law; (8) (3) prefecture-level cities, which are administrative divisions ranked below provinces but above counties in China's administrative structure; prefecture-level cities form the second level of the administrative structure (alongside prefectures, leagues and autonomous prefectures); since the 1980s, prefecture-level cities have mostly replaced the prefecture administrative unit (diqu xingshu); (9) (4) sub-prefecture-level cities, which are county-level cities with powers approaching those of prefecture-level cities; (10) and (5) county-level cities, which are the county-level administrative unit of Mainland China.

Due to these different ranks, the rank of mayor also varies in the political hierarchy. Mayors of directly-controlled municipalities enjoy a status equivalent to governors of provinces, while mayors of sub-provincial-level cities have a political status similar to a vice-governor. The same logic applies to mayors of prefecture-, sub-prefecture-, and county-level cities.

To make the comparison more effective, this article focuses exclusively on mayors of prefecture-level cities, and does not include cities at different bureaucratic levels, as existing studies have done. (11) Prefecture-level cities suitable for quantitative study numbered 268 as of 1 January 2006.

Existing Explanations for Political Promotion

There are three approaches to explain the career mobility of Chinese political elite. The first one is a theory based on guanxi (connections), factionalism, localism and/or family background. Although people have defined the concept of factionalism differently, (12) they all agree that the faction could be an independent variable to explain political mobility. Elites came to power partly due to their technical credentials and partly because of their non-technocratic political or family connections. (13) Factors such as technical education, professional occupation, and non-technocratic political or family connections are identified as predictors for political elite recruitment in current China. Li stresses that nepotism and favouritism in elite recruitment have become prevalent at a time when educational criteria and technical expertise are more important than class background and revolutionary experience. (14)

The second approach is a theory based on actual leadership performance: the performance model. (15) In an earlier study, economic performance measured in terms of taxation contribution from a province to the centre was employed as a key variable to predict the political mobility of provincial leaders from 1949 to 1998, other things being equal. The model suggests that the higher-level government uses personnel control to motivate officials to promote local economic growth, (16) implying a trend that new elites in China are more oriented towards economic achievement than the old elites. (17) This model seems to be more easily understandable as the Communist Party of China (CPC) is eager for economic development to maintain its legitimacy and needs to link good governance to political rewards. The data on mayors at the prefecture level and above from 1990-2000 also supported the performance model. (18) However, the observation on the direction of causality between the economic performance of a jurisdiction and the political mobility of its leaders may be interpreted incorrectly. Even though the correlation deepens our understanding of why leaders of wealthy jurisdictions are more likely to be promoted, it is hard to say leaders really contribute much to the local economy and thus deserve political rewards since development economists argue that economic growth has its inherent logic, (19) which means the influence of local leaders on economic growth is largely, if not fully, insignificant. For example, the shuffling of leaders in developed cities since 1978, such as in Shenzhen and Wenzhou, have never changed economic trends in these cities.

Since the cadre appointment and promotion system in China is sensitive and complicated, various factors would, officially or unofficially, affect a leader's appointment and promotion. The personnel issue in China could be viewed as two systems: "one that is relatively performance-oriented links rewards to performance through competitive mechanisms; the other one selects on the basis of many different criteria some of which may be irrelevant to the job".20 The economic performance model is too simplistic to reveal a full picture of local leaders' promotion in China.

The third is a theory based on human capital endowments. Classical economists, such as Adam Smith, take skills, dexterity (physical, intellectual, psychological, etc.), and judgement ability as human capital, which can be acquired through formal schooling and on-the-job training. In the view of modern neoclassical economics, human capital is similar to "physical means of production": one can invest in human capital (via education, training, medical treatment, etc.), and one's output depends partly on the rate of return on the human capital one owns. (21) Becker's most famous academic contribution to date is the theory of human capital investment, which is formulated as a set of rate of return functions from human capital investment that determines the correlation between income from work and human capital. (22) Human capital is created when a person's skills and capabilities are augmented. The acquisition of human capital improves the conditions for an individual to act in new ways. (23) Investments in human capital are often measured as years of schooling and work experience. Many empirical studies of inequality rely on the analysis of human capital, such as differences in schooling and training and differences in income profiles and income over time.

In practice, the theory of human capital is highly consistent with the policy context in China. Throughout Chinese history, there have been two sharply contrasting lines on the subject of the use of cadres, one being to "appoint people on their merit", and the other to "appoint people by favouritism". Though the CPC always claims that it would take the former as the honest way, it is difficult to operate the concept of "merit". The term has had different meanings in various eras. "Merit" referred to whether or not a cadre was resolute in carrying out the Party line, kept to Party discipline, had close ties with the masses, had the ability to find his bearings independently, and was active, hardworking and unselfish. With its long period of "left" thinking, the Cultural Revolution then destroyed discipline, removed incentive and emasculated management. "Merit" was considered to be associated with political integrity but not professional competence. Many cadres were promoted because of their political or ideological loyalty but had no professional competence. During the reform era, the CPC has stressed both political integrity and professional competence as criteria in selecting cadres. Deng indicated in 1980 that the new contingent of leaders in the reform era should be better educated, professionally more competent, younger and more revolutionised. (24) Deng's policy reflects China's hunger for a pool of talent that is both well-educated and trained "on-the-job". The implementation of Deng's policy makes the human capital theory more applicable to the case of Chinese mayors. Even though various factors would both officially and unofficially affect the appointment and promotion of local leaders, the official criteria for political promotion still include Deng's four standards.

Some studies actually use functionalism to explain that personnel mobility is likely to be heavily influenced by functional expertise or specific work experience. (25) Empirical data pertaining to Chinese mayors in the 1980s showed evidence of the effect of work experience on their promotions. (26) The concept of human capital can be infinitely elastic, including immeasurable variables such as personality or connections. The prestige of a credential may actually be as important as the knowledge gained when determining the value of an education. For example, Deng included "more revolutionary" as one of his four conditions. While Chinese scholars emphasise the significant impact that family background and personal connections have on elite mobility, the theory of human capital investment attributes income inequality not only to differences in talents but also to family background, legacies and other assets. (27) Family background, such as the amount of attention a child receives from parents, is argued by Coleman to have a significant impact on school performance. Social infrastructure, such as a child's family relationships, is part of the social capital. (28) Therefore, neither of the two approaches, the guanxi model and the human capital model, is a pure one.

Some pertinent studies in China studies in fact have shown evidence that human capital variables, such as educational background and work experience, are contributing factors to elite mobility, even though only a few of these explicitly employ the term "human capital". A couple of articles even state that education variables have greater influence on governmental officials' career advancement while work-experience variables have a stronger impact on Party leaders' mobility. (29)

It is clear that changing the institutional contexts from the Cultural Revolution to the reform era has made existing explanations outdated and beg a new theoretical perspective to study the career advancement of local political leaders in China. Based on the pertinent works discussed above, it is possible to propose a new explanation for Chinese mayoral promotions; this explanation takes human capital as the supplementary rather than alternative model of the performance model. This article provides empirical evidence that the former is a determinant of the latter.

A More Comprehensive Logic: Economic Performance Results from Human Capital Accumulation and Previous Promotion

Correlation is a necessary but not sufficient condition for establishing causation. To draw a causal arrow directly from economic performance to political promotion, but not vice versa, sounds illogical when the statistical correlation between a jurisdiction's economic indicator and the mobility of its leaders has been observed. If local economic performance is more dependent on previous economic records than its leaders (as proposed by economists), a more important theoretical question needs to be answered: why are some officials more likely to be leaders of wealthy jurisdictions than others? The performance model might argue that a mayor may be transferred or promoted from somewhere else because of the economic performance of the locality/jurisdictions under his leadership. Yet, the fact is that almost a half of the mayors in 2006 were transferred or promoted from provincial governments (rather than sub-provincial localities) where the economic performance evaluation was never applied. The performance model therefore fails to reveal the root cause of leaders' promotions.

Logically, serving as mayor of a city with a better economic record might be a result of political nomination. That is to say, a candidate's nomination for mayorship of a rich city may be caused by other factors. Thus, a promotion variable used as a predictor of the city a mayor will govern fits the scenario that the candidate already enjoys career success prior to nomination as mayor of a richer city. Referring to the human capital theory and the Chinese policy context as discussed above, it seems reasonable to hypothesise that governing a city with a better economic record is the reward for its mayor's previous investment in human capital and accumulation of political resources.

Why are some individuals more likely to be placed as mayors of richer regions? A possible theoretical explanation is that officials with better abilities and skills will more likely earn promotions (see flow path at first arrow in Figure 1) and their previously accumulated advantages in promotions, measured as the "speed" of promotion, augment their chances to govern a richer region (see flow path at second arrow). To be sure, the performance model may be applicable after this stage and it is argued that the cities' economic advantage will continually provide better records to support their mayors' further promotion (flow path at third arrow). That is to say a strong starting point somehow leads to every step that follows. This view is grounded on the logic that assumes that strong credentials and political connections guarantee political rewards. But this logic does not necessarily fully dispute the performance model. Landry found that good economic performance did help mayors get promoted and those who had lacklustre performance were not punished; (30) economic performance serves as a necessary rather than a sufficient condition for mayors' further career advancement.



This article is designed to test the first and second steps in Figure 1, namely the two hypotheses:

Hypothesis 1

The more human capital an official accumulates, the faster his promotion to prefecture level or mayor will be.

Hypothesis 2

The faster an official has been promoted, the more prosperous the city in which he serves as mayor will be.

Data collection started in early 2006 and an attempt was made here to include the most updated information available about Chinese mayors. Therefore, the cut-off date for the review period was 1 January 2006, and a mayor who was already in office on that date was accounted for here. According to the China Statistical Yearbook 2007, there were 268 prefecture-level cities in 2006. There is substantial information available on the mayors' background, mainly from the Chinese Military and Political Elites Online website, (31) which includes information on variables such as birthdate; Party age (dang ling); date of first day of work and appointment dates for respective positions; educational background; academic discipline (highest degree attained); gender; number of years worked prior to gaining a full prefecture-level position; work experience in the Communist Youth League (CYL), Organisation Department (OD), and economic sectors; and work experience as secretary (mishu) to senior leaders. The information also covers past work experience, such as whether the mayor has worked in his or her home province; whether the current appointment was his/her first at the prefecture level; and whether the official worked in provincial government or sub-provincial government before promotion to a county-level or full prefecture-level position. Most of these variables are dummy variables. The onset of nomination as mayor serves as the time baseline, and only the work experience before this baseline was included here for statistical analysis. A total of 264 cases were observed here.

To test hypothesis 2, the economic variables of cities in various years prior to nomination of an official to mayorship would be adopted as the predictor, since economic data after the appointment of a mayor could be viewed as the outcome, more or less, affected by the mayorship. As mayors who were already in office on 1 January 2006 were appointed at different times from 1999 through 2006, economic statistics--including population, gross domestic product and economic growth--were collected from the 1998-2005 Zhongguo chengshi tongji nianjian (Urban Statistical Yearbook of China, 1998-2005).

Statistical Description of Chinese Mayors

An understanding of how the Chinese local government works at the prefecture level requires some knowledge of the background of the 264 people who held mayoral posts at the beginning of 2006. This section analyses the available data, and discusses their age, gender, nationality, Party age, working age, educational background and work experience.

Table 1 suggests that the average age of the 264 mayors in 2006 was 49.78 years. They were younger than their counterparts in 1990 to 2000 period (50.44 years) but slightly older than the mayors in 2000 (49.6 years). (32) The retirement rule (60 for male and 55 for female cadres at full prefecture level) has been enforced effectively. Age limits have achieved the hoped-for rejuvenation of the Chinese leadership in the reform era. (33) This finding is consistent with Landry's study, and the average age of mayors has continued to decline since 1990.

If mayors in 2006 were comparatively young, they were obviously even younger when they were nominated as mayors (47.53 years). If a comparison is drawn between the two ages, the data shows that the mean number of years served in their current mayoral position was 2.26 years. Yet this does not reflect the mayoral term directly. Assuming that the time distribution of nomination and demission is random, the figure could be interpreted as half of the actual duration served as prefecture-level mayor, that is, 4.52 years. It is still shorter than the formal term for prefecture-level mayors, which is five years according to Article 6 of the Organic Law of the Local People's Congresses and Local People's Governments of the PRC.

To maintain stability in leadership tenure, the OD of the CPC Central Committee issued the Provisional Regulations on Terms of Cadres of the Party and Government (Dangzheng lingdao ganbu zhiwurenqi zanxing guiding) in August 2006. Articles 3 and 4 are in accordance with the regulation stipulated in the Organic Law of the Local People's Congresses and Local People's Governments of the PRC that mayors and officials of county level and above should serve five-year terms and that their terms should be relatively stable. An added regulation also stipulates that cadres may not serve in the same position for more than two terms (Article 6) and that they may not serve in positions of the same rank for more than 15 years (Article 7). (34) Empirical data suggests that, in practice, the duration of a term served in a leadership post has not always followed the stipulated formal rules, and this may explain why the centre decided to issue this document. Historically, however, the leadership viewed personnel shifts as a useful tool for achieving control and guaranteeing adequate political consciousness. (35) There will be a high frequency of personnel shifts if the regime feels it is necessary to prevent cadres from becoming too familiar with their posts and too attached to parochial interests. The regulations should create a proper balance between political control and administrative continuity. Nevertheless, the general question of the stability of mayors deserves further research.

Table 2 compares the age distribution in the 1980s with that in 2006 and further reflects on the influence of Deng's policy since the 1980s. Due to the retirement rules, it is not surprising to find that there were no mayors aged over 60 in 2006. (36)

Table 2 presents the distribution of gender and nationality among Chinese mayors. The trend of male dominance has never changed, though the percentage of females who occupied mayoral positions rose from 1.6 in the 1980s to 2.0 in 1990-2000, and 3.8 in 2006. It will still be some time before women achieve numbers equal to their male counterparts in Chinese leadership circles.

With respect to representation of ethnic groups, it is confirmed from the 1980s data set that most mayors are of Han nationality. Moreover, most minority cadres are limited to serving in minority areas. Otherwise, it is likely that minority mayors would serve in cities outside of their native provinces, including Han areas.

The CPC's unwritten rule is that all mayors should be Party members and should have served as Deputy Party Secretary prior to being appointed as mayor. CPC membership is not a proper measure here. Zang adopted Party seniority to measure political credentials:
   Seniority is the political life of a communist official, which
   starts when he or she joins the CCP. It refers to the total number
   of years of CCP membership a cadre has. CCP membership indicates
   merely whether a party member possesses a political credential,
   whereas CCP seniority measures the amount of political credentials
   he or she has accumulated over the years of party service. (37)

He also argued that CPC seniority is a better measure and more important to career advancement than CPC membership. Thus, Party age is adopted here to indirectly measure the concept of "being revolutionary", which is far beyond the scope of measurability even as an ideological concept.

In practice, working age is also used as a measure of seniority but it does not relate directly to any political or ideological qualification. From the profiles of Chinese political elites that indicate the date they joined the CPC and date they began their working career, some leaders joined the CPC very late in their professional lives. For example, Xu Kuangdi, the former mayor of Shanghai, joined the CPC in July 1983, 14 years after taking his first job. Thereafter he was promoted to deputy president of a university and then moved to the position of head of the Shanghai Bureau of Higher Education. From this case, it is reasonable to infer that Xu joined the Party because his higher posts required CPC membership. It seems working age, rather than Party seniority, can accumulate a wider social network that positively affects upward mobility. Table 3 presents details of Party ages and working ages computed from different baselines.

Li and White stress that the new elites in China have come to power in part because of their advanced educational backgrounds. (38) Table 4 shows the educational background of the mayors. Most (87 per cent) held university degrees, through both part-time and full-time study, with 28 per cent holding full-time university degrees and 34.6 per cent holding master's degrees earned on a part-time basis. More importantly, 10 mayors were full-time master's degree-holders and 14 had earned doctoral degrees before serving as mayor. It is not surprising that the mayors in 2006 were significantly better educated than those in the 1980s and in the 1990 to 2000 period. Deng's policy seems to have worked in practice as well as in theory.

Regardless of educational level, more than 60 per cent of the mayors had part-time study experience, reflecting the impact of Deng's policy. It can be hypothesised that a higher level of education would accelerate their promotion.

Among Chinese elites, technocrats play significant roles, and previous studies showed supporting evidence that academic disciplines systematically affect the promotion opportunity of officials (39) (particularly in the case of governmental officials). (40) It is logical to expect that an academic background in the sciences would bolster a mayor's chances for promotion. Some 80 per cent of the mayors majored in arts, while 46 (20 per cent) majored in sciences. However, when the form of study, be it part-time or full-time, was assigned as a control variable, it was observed that most mayors (138 of 181, or 76.2 per cent) who majored in arts studied part-time, whereas 89.1 per cent (41 of 46) of those in sciences studied full-time (Table 5). This transformation in the elite may echo the current needs of China's economic transformation.

Factionalism identifies that experience as a CYL cadre exerts a positive influence on political promotion. (41) Table 6 shows that 84 of the mayors had worked in a CYL township committee or above, accounting for 35 per cent of the 240 mayors (see Table 6), indicating that CYL cadres constituted an important faction (tuanpai) among Chinese mayors.

Scholars have also found that being a personal assistant or secretary to a senior official allows juniors to accumulate useful personal ties that will push future promotion. (42) Of the mayors whose work experience data is available, 83 had served as the chief of staff, secretary or assistant of higher-ranked officials, accounting for 34.7 per cent of all mayors under study.

Good knowledge of local conditions helps officials win consent and approval from the local population. (43) This may be more significant for those holding governorships. There has been a conscious effort to place local people in highly visible government posts in order to secure popular support. (44) If the extent of localism is measured in terms of the percentage of officials serving in their native province, the conclusion is replicated here. Table 6 shows that 75.5 per cent of all mayors served in their home provinces.

Guanxi, factionalism, and localism as described above was included in the regression analysis in order to explore whether the theory based on these three variables could explain the promotion speed of mayors.

As the CPC Organisation Department is in charge of Party personnel issues, there is a popular saying among Chinese cadres that "following the Organisation Department, you will be promoted once a year". (45) By virtue of their role in appointments, dismissals and personnel shifts, officials in the OD have far-reaching influence over the entire political hierarchies. (46) Of all the mayors, 19 per cent had experience in an OD or personnel section at the county level or above. The same proportion was also reported in Teiwes' study of provincial Party personnel. (47)

Leaders in China, especially those in the economic reform era, have had to achieve good performances in order to stay in power. (48) This is reflected in the official nomination procedures. For example, on 10 April 2006, Lan Jun, the former mayor of Songyuan City in Jilin Province, was promoted to Party boss of the city. When the executive deputy head of the Jilin Provincial Party Committee OD announced the nomination, he emphasised the economic performance that Songyuan achieved under Lan Jun's leadership and identified this as the main reason for his promotion. The impressive economic credentials of Sun Hongzhi, who was to become Lan's successor, were also presented in support of his nomination. (49) (However, it is difficult to claim which of the two leaders contributed more to the economic performance.) Of the 239 mayors whose profiles are available, 105 had worked in the economic sector, such as macroeconomic administration and state-owned enterprises.

A dummy variable was used here to code "officials promoted from provincial or sub-provincial units" to identify any difference in influence between provincial and sub-provincial experience on mayoral nomination. Thus, an official, who takes up mayorship as his/her first full prefecture-level position, is promoted from sub-provincial governments or provincial governments depending on his/her first county-level position. An official is defined as a cadre promoted from sub-provincial government if his/her first county-level position is in sub-provincial government. For those who took a mayorship as the second or third full prefecture-level position, the definition depends on the official's previous position. For example, an official, who was secretary of a CYL provincial committee, will be considered to have been promoted from provincial rather than sub-provincial government regardless of his/her work experience in sub-provincial government. Table 6 reveals that 136 mayors (54.2 per cent) were promoted from sub-provincial governments while 115 (45.8 per cent) were promoted from provincial governments.

In this article, we differentiate those officials whose first prefecture-level position was mayor from those who had served in other prefecture-level positions prior to becoming mayor. Table 6 shows that out of 250 mayors, 157 (62.8 per cent) said that the position of mayor was their first prefecture-level position. This means that 37.2 per cent were transferred from other equivalent positions rather than promoted directly from deputy-prefecture-level positions.

Table 7 displays the cross-tabulation of the last two variables. It is no surprise to see that 75.3 per cent of the mayors, whose mayorship was a subsequent prefecture-level position, were promoted from provincial governments, while 72.0 per cent, whose mayorship was their first prefecture-level position, were promoted from positions in sub-provincial governments. This shows that those with work experience in sub-provincial governments were more likely to be directly promoted to mayor, while those with provincial government experience were more likely be promoted to other prefecture-level positions.

Since the mayor in a given city is responsible for the overall coordination of all aspects of government work, particularly economic development, it is expected that the mayoral posts would be rotated among officials with experience in different sectors. Furthermore, this arrangement reflects the relative importance of on-the-job training in various sectors (e.g., 44 per cent of all the mayors had work experience in economic sectors). In order to examine the on-the-job training aspect of the human capital theory, the above-mentioned four variables were adopted as independent variables in the regression analysis.


In this section, the speed of promotion was adopted as a dependent variable to measure how fast mayors climb the hierarchy. To some degree, skills and qualities are attributes most highly valued by the top Party leadership. (50) DiPrete and Soule argued that a supervisor may interpret an employee's slow progress in the past as an indication of limited future potential. Thus, the advantages that empower an employee to move speedily to her/his present position will generally continue to benefit her/him in the future. (51)

However, scholarly opinions on the operational definition of promotion speed range widely. Farmer defined the speed of promotion as the length of "wait" time between an individual's first acceptance of a Party or government position and the time he/she is first elected as a member of the Central Committee of the Communist Party of the Soviet Union. (52) Chinese scholars adopted "fastest runner (climber)" to measure the speed of promotion in their study of Chinese mayors in the 1980s by considering both the administrative level and population of each city and the age of its mayor. (53) All of these studies explicitly assumed that promotion speeds are determined by personal attributes such as education and work experience. Li and Bachman reported that seniority in the CPC, age and especially training in engineering were important predictors of the speed of promotion. (54) Alternatively, Gil Eyal and Eleanor Townley proposed that the year a respondent first advances to a nomenklatura position could be operationalised as the year he/she first advances to a position of authority in the elite. (55)

According to Zang, the merit of the definition was the coincidence of the baseline for calculating the speed of promotion and that of a political hierarchy. (56) However, his definition of the rate of mobility (as the inverse of the sum of age and CPC seniority) failed to fully reflect such a consideration. He demonstrated that while CPC membership is useful in a wide range of upper-level political careers, it is inappropriate to equate the date of Party entry with the rate of mobility, as some of the political elite joined the CPC at a rather late stage in their lives (due to their college-educated background which was politically unfavourable under Mao). Zang also argued that one's political career is not inherited by birth. Both concepts of CPC seniority and CPC membership, however, play a role in Zang's definition of the rate of mobility.

In contrast to existing studies (all of which use both biological age and position attainment to define the speed of promotion), we preferred to use working age or length of service, which is the time difference between the date an official began to work and the time when he/she attained a full prefecture-level position. It assumes that the race towards the summit of the political hierarchy begins when the official enters the public sector, rather than when he/she enters the CPC. This is because "the Chinese nomenklatura is more complicated than the prototype of the political bureaucracy in state socialism". (57) In China, especially before reforms of the state-owned enterprises and the tertiary sector, public institution employees such as public school teachers, managers of state-owned enterprises, CYL secretaries, and many others were considered to be candidates for political positions. Working age seemed a better measure than CPC seniority or biological age to reflect the accumulation of political resources an official generates for promotion.

As students of China Studies know, there is a conceptual difference between full prefecture-level position and prefecture-level mayor, even though the latter is also at the full prefecture level. Table 6 showed that 37.2 per cent of the mayors rose to a position at the full prefecture level before being nominated as mayor. Thus, for effective comparison, attainment of both full prefecture-level position and mayor is used here as the baseline. We defined promotion speed A as the inverse of the number of years worked prior to attaining full prefecture-level position; and promotion speed B equalled the inverse of the number of years worked prior to attaining the position of mayor. Therefore, by employing this definition, biological age and CPC seniority need to be included as predictors in the following regression models. The regression did not include any variable that proxies the effect of economic performance during their pre-mayoral tenure because 45.8 per cent of the mayors under study were promoted from the provincial government (see Table 6). Thus, for their career advancement, the performance model was not applicable.

Model 1 reports that Party age had a slightly positive effect on the promotion speed, but statistically it was not significant. This is hardly surprising as statistically it is highly correlated to age and so its impact would necessarily be reflected by age. Therefore, Party age is not included in Models 2 or 4. In all four models, the 99.9 per cent confidence level was used to argue there is a non-linear relationship between age and the promotion speed. Age had a negative effect on mobility whereas age squared had a positive effect. So the total effect of age on mobility would vary among the various age groups. The effect looks like the concave segment of a U-shaped curve (see Figure 2), other things being equal.

This finding broadly coincides with a previous finding that seniority affects the mobility rate negatively. (58) In Great Britain, one study showed that those who became Members of Parliament at a young age, particularly in their 30s, had a greater rate of movement to a higher position. (59) Yet the finding here was much more precise.


Model 1 also shows that the years of schooling and an academic background in sciences can significantly increase promotion speed (sciences = 1, arts = 0), not only fully supporting the findings of existing studies, (60) but also implying that the influence of academic discipline on promotion is more powerful than years of schooling. As the theory of technocracy contends, technical training is especially useful for modern leaders. (61) Technically-trained mayors can therefore be distinguished from those trained in arts. (62) The impact of a change from arts to sciences was equivalent to 5.5 additional years of education. However, as presented in Table 5, more and more mayors who originally trained in sciences later pursued, on a part-time basis, advanced degrees in arts, implying that the transformation among technocrats deserves further research.

The data shows the "provincial/sub-provincial" experiences influence mayors' mobility substantially, indicating leaders with on-the-job training in provincial governments are likely to be promoted speedily. This reinforces the human capital theory. The same cannot be said regarding the "economic sectors" which decreased promotion speed significantly. However, the significant 44 per cent of all the mayors equipped with such experience in the "economic sectors" drives home the point that sufficient accumulation of on-the-job training in economic sectors is necessary for a leader to be promoted to a mayor (this would inevitably decrease the promotion speed), and this finding reinforces--though not so strongly--rather than weakens the human capital theory. Otherwise, an insignificant effect should be observed. However, the statistical insignificance of ODs is far from weakening the human capital approach, as the implication is that on-the-job training in Party organisations is not as important for mayors as it is for primary governmental leaders.

Nevertheless, such related findings can also be found in other studies. (63) Li and Bachman reported that mayors working in the "Party as main career" had a negative correlation with the "fastest runner". (64) The variable, "mayor as the first prefecture-level position", also has a significantly negative impact on promotion speed, suggesting other prefecture-level positions could be attained more speedily than mayorships. Put differently, those who first attained other prefecture-level positions needed to accumulate seniority as a prefecture-level cadre before becoming a mayor. This is consistent with the human capital theory since such seniority accumulation could be viewed as a kind of on-the-job training.

Model 1 also demonstrates, unexpectedly, that work experience in the CYL significantly decreases promotion speed. This defies the common sense that individuals in the CYL faction are promoted more speedily than others and implies that factionalism is an ineffective approach when other predictors are controlled. Work experience as a mishu (secretary) has an insignificant effect on promotion speed, signalling that the theory based on guanxi has definitely failed to explain promotion speed. (65)

Furthermore, a positive correlation between native province and political promotion did not occur here. On the contrary, the regressive model suggests that promotion speed would decline if officials serve in their native province, weakening the concept of localism.

Simply put, an official becomes a fast climber by his previous accumulation of human capital including education and some specific on-the-job training.


Table 9 presents the results of the ordinary least squares (OLS) regression used to test Hypothesis 2 that assumes the faster an official is promoted, the more likely he/she would be nominated to the mayorship of a city with a better economic record.

In this model, the promotion speed A was entered as the main predictor of per capita GDP of the city that the official would govern. Here, the per capita GDP of the year prior to the nomination of the official as a mayor was employed as the measure of the economic situation. Mayoral nomination was independent of political nomination before the personnel decision was made. However, the official's promotion speed existed in the system before her/his nomination.

A pattern emerges from Table 9, suggesting that the promotion speed can significantly affect the city in which the official will govern in the coming promotion exercise. The information in the table is definitely conclusive (sig. =.0005) and fully supports Hypothesis 2, other variables being equal.

This finding provides a more comprehensive explanation to the performance model. Officials who previously climbed the political hierarchy faster will more likely be nominated as the mayor of a richer city. The better economic performance could be viewed as the reward for faster promotion. Although there is a correlation observed between the mayor's further promotion to a higher level and the economic performance of the city he/she governed (as the performance model would predict), the article--based on the current finding--contends that the official's further promotion essentially results from the extension of his/her former career advantage measured as the speed of promotion, rather than from economic performance. The finding lends support to Landry's argument that the economic performance of cities has little substantive impact on political promotion.

Although work experience in the economic sector and sub-provincial units decreases an official's promotion speed, Table 10 shows that both experiences are beneficial for leaders aiming to govern a richer city, indicating the possible correlation between specific on-the-job training and economic performance. There is strong evidence that the accumulated human capital (in terms of on-the-job training) deserves some kind of reward, which is highly consistent with Hypothesis 1 in the previous section. Moreover, the data implies that younger leaders are more likely to serve in a wealthier city than other leaders. These findings deepen our understanding of the debates on the performance model.

In sum, promotion speed could be viewed as a measure of a leader's accumulation of human capital and political resources, and thus a predictor of his further promotion. It is an intervening variable in explaining the political promotion of local leaders and the correlation observed by the performance model. Logically, the wealth of human capital and political resources leads to a higher promotion speed and in turn places officials as mayors of richer cities. Through governing richer cities, mayors will have a greater chance to be promoted on the pretext of better economic records under their leadership. Since the Party always officially cites economic records to justify an official's promotion, leaders in China need this nominal reason through governing richer cities, while only those who are well endowed with human capital and political resources enjoy greater odds to become the mayors of richer cities.


The human capital theory deepens our understanding of the theoretical debate whether a city's economic performance causes its leader's promotion. Empirical data shows that a mayor's promotion speed is a reward for his previous career advantages, supporting the logic of human capital theory and echoing the context of Chinese policy, such as the formal personnel regulations initiated by Deng. This strongly suggests that the odds of promotion depend on the cadre's personal characteristics. The personality attributes facilitate the promotion of a higher proportion of younger, well-educated and on-the-job trained officials than was possible before the reforms, but weakens the link between good governance and political rewards. In turn, the key finding that faster climbers are more likely to govern a richer city may not fully reject the causality between economic performance and political promotion constructed by the performance model. However, this postulation makes the explanation of the model more comprehensive, more logical and easily understandable. Compared to other competing candidates, those who are nominated as mayor are successful climbers of the Chinese political hierarchy. But the degree of their success depends on their human capital. The research question is, therefore "Why are some climbers more successful?" (rather than "Why are some climbers successful and others unsuccessful?" that compares those who were promoted with those who were not from the pool of potential candidates). This implies we look only at leaders who were already promoted as mayors for clues to understanding the causes. Climbers with better human capital arrive at prefecture-level positions faster than others; and the faster climbers are more likely to govern a city with a better economic record. According to the performance model, the city's better economic record will in turn increase the chance of mayor's further promotion. Hence, mayorship here is a "filter" which differentiates faster climbers from slower ones by placing them in cities with different economic records. Meanwhile, mayorship of a rich city is an "accelerator" for career advancement.

The fact that a mayor's further promotion depends heavily on his/her previous accumulated advantages implies that political elitism has emerged as an important factor in the Chinese polity. This means, to some extent, that the polity will be dominated by leaders with specific education or work experience. This is a form of "political path-dependence", implying that the leaders' chance of further promotion is limited by their promotion speed in the past; and the promotion speed in turn is determined by the human capital accumulated, even though their human capital may no longer be directly relevant. Put simply, a strong starting point somehow leads to every step that follows. However, in practice, "political path-dependence" will lead to a greater degree of elitism and political inequality in China.


The author would like to thank the anonymous reviewers for their valuable comments and suggestions.

(1) Susan L. Shirk, The Political Logic of Economic Reform in China (Berkeley: University of California Press, 1993); Jia Hao and Lin Zhimin, eds., Changing Central-Local Relations in China: Reform and State Capacity (Boulder, CO: Westview, 1994), p. 3; Peter T.Y. Cheung, Jae Ho Chung, and Zhimin Lin, eds., Provincial Strategies of Economic Reform in Post-Mao China: Leadership, Politics, and Implementation (Armonk, NY: M.E. Sharpe, 1998); and Tony Saich, Governance and Politics of China (New York: Palgrave Macmillan, 2004).

(2) Saich, Governance and Politics of China, p. xv.

(3) According to Article 8 of the Organic Law of the Local People's Congresses and Local People's Governments of the PRC, local people's congresses at and above the county level shall exercise the power to elect mayors and deputy mayors of cities, at <> [12 Sept. 2006].

(4) Such as Bo Zhiyue, Chinese Provincial Leaders: Economic Performance and Political Mobility since 1949 (Armonk, NY and London: M.E. Sharpe, 2002); Lin Tingjin, "Jingji jixiao yu zhengzhi shengqian" (Economic Performance and Political Mobility: A Case Study of L City in Zhejiang Province), Master's thesis, Peking University, 2003; Zang Xiaowei, Elite Dualism and Leadership Selection in China (London and New York: Routledge Curzon, 2004); and Li Hongbin, and Zhou Li'an, "Political Turnover and Economic Performance: The Incentive Role of Personnel Control in China", Journal of Public Economics 89 (2005): 1743-62.

(5) There are two exceptions: Cheng Li and David Bachman, "Localism, Elitism, and Immobilism: Elite Formation and Social Change in Post-Mao China", World Politics 42, no. 1 (Oct. 1989): 64-94; and Pierre F. Landry, "The Political Management of Mayors in Post-Deng China", The Copenhagen Journal of Asian Studies, no. 17 (2003): 31-58.

(6) Some previous research has found different promotion patterns, named as "elite dualism", between administrative leaders and Party cadres. While it would be interesting to compare Party secretaries with mayors, this study focuses solely on mayors. The comparison between Party secretaries and mayors deserves a new study.

(7) Organic Law of the Local People's Congresses and Local People's Governments of the PRC, at <> [12 Sept. 2006].

(8) Currently, there are 15 sub-provincial cities: Changchun, Jilin; Chengdu, Sichuan; Dalian and Shenyang, Liaoning; Guangzhou and Shenzhen, Guangdong; Hangzhou and Ningbo, Zhejiang; Harbin, Heilongjiang; Jinan and Qingdao, Shandong; Nanjing, Jiangsu; Wuhan, Hubei; Xiamen, Fujian; and Xi'an, Shaanxi.

(9) It is necessary to distinguish a prefecture-level city from a prefecture administrative unit. The former does represent a formal level of government which is included in the "city divided into districts" category by the Constitution of the People's Republic of China (PRC). But the latter does not represent a formal a level of government as defined by the Constitution and just exercises executive authorities delegated by the provincial government.

(10) Examples of sub-prefecture-level cities include Jiyuan, Henan; Xiantao, Qianjiang, and Tianmen, Hubei; Golmed, Qinghai; Manzhouli, Inner Mongolia; and Shihezi, Tumushuk, Alar, and Wujiaqu, Xinjiang.

(11) Li and Bachman, "Localism, Elitism, and Immobilism", World Politics 42, no. 1 (1989): 64-94; and Landry, "The Political Management of Mayors in Post-Deng China", pp. 31-58.

(12) Jurgen Domes, "Intra-Elite Group Formation and Conflict in the PRC", in Groups and Politics in the People's Republic of China, ed. David S.G. Goodman (Armonk, NY: M.E. Sharpe, 1984), pp. 26-39; Andrew J. Nathan, "A Factionalism Model for CCP Politics", The China Quarterly, no. 53 (Jan.-Mar. 1973): 34-66; and Lowell Dittmer and Yu-Shan Wu, "The Modernization of Factionalism in Chinese Politics", World Politics 47, no. 4 (July 1995): 467-94.

(13) Li Cheng, China's Leaders: The New Generation (Lanham, MD: Rowman & Littlefield, 2001), p. 28.

(14) Ibid., pp. 128, 149.

(15) Bo, Chinese Provincial Leaders: Economic Performance and Political Mobility since 1949, pp. 2-3.

(16) Li and Zhou, "Political Turnover and Economic Performance", pp. 1743-62.

(17) Li Cheng and Lynn White, "Elite Transformation and Modern Change in Mainland China and Taiwan: Empirical Data and the Theory of Technocracy", The China Quarterly, no. 121 (Mar. 1990): 22.

(18) See Landry, "The Political Management of Mayors in Post-Deng China", pp. 54-6. Note the measurement of economic performance is different from Bo's study of provincial leaders.

(19) W. Arthur Lewis, Development Economics: An Outline (Morristown, NJ: General Learning Press, 1974).

(20) John P. Burns, "Civil Service Reform in China", in Governance in China, OECD (Paris: OECD, 2005), pp. 61-2.

(21) Jacob Mincer, "Investment in Human Capital and Personal Income Distribution", Journal of Political Economy 66, no. 4 (Aug. 1958): 281-302; Gary S. Becker, Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education, 3rd ed. (Chicago, IL: University of Chicago Press, 1993); and Jacob Mincer, Schooling, Experience, and Earnings (New York: Columbia University Press, 1974).

(22) Becker, Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education, p. 16.

(23) James S. Coleman, Foundations of Social Theory (Cambridge, MA: Belknap Harvard, 1992), p. 304.

(24) Deng Xiaoping, Deng Xiaoping wenxuan (Selected Works of Deng Xiaoping), vol. 2 (Beijing: Renmin chubanshe, 1983), pp. 326, 361.

(25) Frederick C. Teiwes, Provincial Party Personnel in Mainland China, 1956-1966 (New York: East Asian Institute, Columbia University, 1967), p. 57.

(26) Li and Bachman, "Localism, Elitism, and Immobilism", pp. 64-9.

(27) Gary S. Becker and Nigel Tomes, "Human Capital and the Rise and Fall of Families", in Approaches to Social Theory, ed. Siegwart Lindenberg, James S. Coleman, and Stefan Nowak (New York: Russell Sage, 1986), pp. 201-39; and Becker, Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education, p. 16.

(28) James S. Coleman, Adolescents and the Schools (New York: Basic Books, 1965); and Coleman, Foundations of Social Theory, p. 304.

(29) For example, Andrew G. Walder, "Career Mobility and the Communist Political Order", American Sociological Review 60, no. 3 ( June 1995): 309-28; and Andrew G. Walder, Bobai Li, and Donald J. Treiman, "Politics and Life Chances in a State Socialist Regime: Dual Career Paths into the Urban Chinese Elite, 1949 to 1996", American Sociological Review 65, no. 2 (Apr. 2000): 191-209. Chih-Jou Jay Chen, "Elite Mobility in Post-Reform Rural China", Issues & Studies 42, no. 2 (June 2006): 53-83. Also see Zang, Elite Dualism and Leadership Selection in China.

(30) "The Political Management of Mayors in Post-Deng China", pp. 31-58.

(31) See <>. Once a mayor is identified in official sources, his name will be checked in the Chinese Military and Political Elites Online website <> and cross-checked in various biographical sources through official websites, such as Xinhuanet <>, <> and local government webpages. A surprising amount of career data has been compiled in this fashion but it suffers from a number of gaps since not all of the 268 mayors' personal data are available online. The greatest deficiencies concern Party-age, educational background and work experience. Effort was also made to collect the data of mayors who were promoted after 1 January 2006. However, success was limited because the accessibility of the website Chinese Military and Political Elites Online was disabled intermittently due to political sensitivity. The information released on governmental websites is too basic to be included in statistical analyses.

(32) See Landry, "The Political Management of Mayors in Post-Deng China", Tables 3 and 4.

(33) Melanie Manion, Retirement of Revolutionaries in China (Princeton, NJ: Princeton University Press, 1993), pp. 1-3.

(34) See <> [19 Sept. 2006].

(35) Teiwes, Provincial Party Personnel in Mainland China, 1956-1966, p. 31.

(36) According to "Decisions on Establishing a Retirement System for Elder Cadres" (Zhonggong zhongyang guanyu Jianli laoganbu tuixiu zhidu de jueding) issued by the CPC Central Committee on 20 February 1982 on the rule of retirement, officials who occupy prefecture-level positions in the Party and governmental units should not be older than 60 years of age, at <> [18 Sept. 2006].

(37) Zang, Elite Dualism and Leadership Selection in China, p. 5.

(38) Li and White, "Elite Transformation and Modern Change in Mainland China and Taiwan", p. 21.

(39) Li, China's Leaders: The New Generation, 105; and Li and White, "Elite Transformation and Modern Change in Mainland China and Taiwan", pp. 19-21.

(40) Zang, Elite Dualism and Leadership Selection in China, p. 58.

(41) Li, China's Leaders: The New Generation, p. 149; and Bo Zhiyue, Seminar, Department of Politics and Public Administration, University of Hong Kong, Apr. 2006.

(42) Li, China's Leaders: The New Generation, p. 149; Li and White, "Elite Transformation and Modern Change in Mainland China and Taiwan", p. 21; and Teiwes, Provincial Party Personnel in Mainland China, 1956-1966, p. 28.

(43) Li and Bachman, "Localism, Elitism, and Immobilism", p. 80; and Teiwes, Provincial Party Personnel in Mainland China, 1956-1966, p. 15.

(44) A. Doak Barnett, Cadres, Bureaucracy, and Political Power in Communist China (New York: Columbia University Press, 1967), p. 51.

(45) In Chinese, it is "genzhe zuzhibu, niannian youjinb". A Chinese novel named The Head of Reception Section, vol. 4: 23. See a blog of a township cadre, at <> [19 Sept. 2006]. The author mentioned that his personal network improved in the process of evaluating other officials or other institutions. Such a network, built up while working in an OD, is invaluable.

(46) Teiwes, Provincial Party Personnel in Mainland China, 1956-1966, p. 55; and Barnett, Cadres, Bureaucracy, and Political Power in Communist China, p. 13.

(47) Teiwes, Provincial Party Personnel in Mainland China, 1956-1966, p. 58.

(48) See Bo, Chinese Provincial Leaders: Economic Performance and Political Mobility since 1949, pp. 4-5; and Li and White, "Elite Transformation and Modern Change in Mainland China and Taiwan", p. 22. This statement is not necessarily contradictory with the previous discussion that leaders contribute little to the local economy. The real logic behind this is that leaders need good economic performances in order to stay in power and attain further promotion.

(49) Liu Changyu and Ren Feilin, "Lan Jun ren Songyuan shiwei shuji" (Lan Jun is Appointed Party Secretary of Songyuan City), Xin wenhua bao (New Culture Daily), 12 Apr. 2006, at <> [10 Sept. 2006].

(50) Kenneth C. Farmer, The Soviet Administrative Elite (New York: Praeger, 1992), p. 211.

(51) Thomas A. DiPrete and Whitman T. Soule, "Gender and Promotion in Segmented Job Ladder Systems", American Sociological Review 53, no. 1 (Feb. 1988): 28.

(52) Farmer, The Soviet Administrative Elite, pp. 215-7.

(53) Li and Bachman, "Localism, Elitism, and Immobilism", p. 80; Xiaowei Zang, "Provincial Elite in Post-Mao China", Asian Survey 31, no. 6 (June 1991): 513-4; and Xiaowei Zang, "Elite Formation and the Bureaucratic-Technocracy in Post-Mao China", Studies in Comparative Communism 24, no. 1 (Mar. 1991): 116, 120.

(54) Li and Bachman, "Localism, Elitism, and Immobilism", pp. 80-1.

(55) Gil Eyal and Eleanor Townsley, "The Social Composition of the Communist Nomenklatura: A Comparison of Russia, Poland, and Hungary", Theory and Society 24, no. 5 (Oct. 1995): 733.

(56) Zang, Elite Dualism and Leadership Selection in China, p. 188.

(57) Ibid., p. 118.

(58) Li and Bachman, "Localism, Elitism, and Immobilism", p. 82.

(59) Stuart Elaine MacDonald, "Political Ambition and Attainment: A Dynamic Analysis of Parliamentary Careers", PhD diss., University of Michigan, 1987, pp. 113-4.

(60) Ibid., p. 122; and Li and Bachman, "Localism, Elitism, and Immobilism", p. 82.

(61) Li and White, "Elite Transformation and Modern Change in Mainland China and Taiwan", pp. 16, 21.

(62) Robert D. Putnam, "Elite Transformation in Advanced Industrial Societies: An Empirical Assessment of the Theory of Technocracy", Comparative Political Studies 10, no. 3 (Oct. 1977): 404.

(63) Teiwes, Provincial Party Personnel in Mainland China, 1956-1966, p. 30.

(64) Li and Bachman, "Localism, Elitism, and Immobilism", p. 82.

(65) Since mayors at the prefecture level are chosen by the provincial Party standing committees, logically it is necessary to examine the personal networks between these mayors and their provincial leaders. In addition to shared work experiences, school ties and family ties are also important sources of personal networks. However, this article left these informal ties untouched because information of this nature is unobservable to outsiders.

Lin Tingjin ( is Professor of Public Administration and the founding Director of the Institute of Urban Development Studies, Nanjing University of Finance and Economics, China. He obtained his PhD in Politics and Public Administration from the University of Hong Kong. His research interests include public finance, public service provision and local political elites in China.

                        Minimum   Maximum    Mean       Std.

Age (as of 1 January     39.00     59.42    49.7827    4.16403
Age (when he/she was     35.08     56.50    47.5273    3.81873
  nominated as mayor)
Years in current           .00      9.58     2.2555    1.53039
  position (as mayor)
Valid N                    264


                         Number   Percentage   Number   Percentage
                         1989 *     1989 *      2006      2006

Sex           Male         243       98.4       254        96.2
              Female         4        1.6        10         3.8
              Total        247      100.0       264       100.0

Nationality   Han          226       91.5       229        86.7
              Minority      19        7.7        16         6.1
              Unknown        2        0.8        19         7.2
              Total        247      100.0       264       100.0

Age group     35-39          8        3.2         1          .4
              40-44         57       23.0        39        14.8
              45-49         67       27.1        96        36.4
              50-54         71       28.7        97        36.7
              55-59         35       14.1        31        11.7
              60-64          7        2.8         0         0
              Unknown        2        0.8         0         0
              Total        247       99.7 **    264       100.0

* Source: Cheng Li and David Bachman, "Localism, Elitism, and
Immobilism: Elite Formation and Social Change in Post-Mao China",
World Politics 42, no. 1 (Oct. 1989): 69.

** Percentages do not add up to 100 due to rounding-off


                        N    Minimum    Maximum    Mean       Std.

Party Age              264     9.08      40.58    26.6133    5.51446
  (as of 1
  January 2006)
Party Age              264     9.08      36.67    24.3661    5.16964
  (when he/she was
  nominated as mayor)
Length of Service      264    14.42      43.17    30.3073    5.38893
  (1 January 2006)
Length of Service      264    13.67      39.58    28.0518    5.09215
  (when he/she was
  nominated as
  a mayor)
Length of Service      264    11.33      38.25    26.8538    5.24170
  (when he/she


                       Number      Percentage   Number   Percentage
                       Jan. 2006   Jan. 2006    1989 *     1989 *

Middle School              0           0           9        3.6
Senior Middle School       1            .4        22        8.9
Part-time College         13           5.3        39       15.7
Full-time College         18           7.3
Part-time University      36          14.6       147       59.5
Full-time university      69          28.0
Part-time Master's        85          34.6
Full-time Master's        10           4.1         5        2.0
PhD                       14           5.7
Foreign Study             --          --           2        0.8
Unknown                   --          --          23        9.3
Total                    246         100.0       247       99.9 **

* Source: Cheng Li and David Bachman, "Localism, Elitism, and
Immobilism: Elite Formation and Social Change in Post-Mao China",
World Politics 42, no. 1 (Oct. 1989): 73. The data includes both
part-time and full-time study.

** Percentages do not add up to 100 due to rounding-off.


               Full-time    Part-time     Total          (% of

Arts           43 (23.8%)   138 (76.2%)   181 (100.0%)   (79.7%)
Science        41 (89.1%)     5 (10.9%)    46 (100.0%)   (20.3%)
Total          84           143           227            (100.0%)

               Value        df            Asymp. sig.

Pearson        67.244       1             .000
N of valid     227


                                Frequency   Percentage

No Work Experience in CYL         156          65.0
Work Experience in CYL             84          35.0
Total                             240         100.0

Mishu                              83          65.3
Non-mishu                         156          34.7
Total                             239         100.0

Local                             185          75.5
Non-local                          60          24.5
Total                             245         100.0

No Work Experience in             192          81.0
  Organisational Department
Work Experience in                 45          19.0
  Organisational Department
Total                             237         100.0

No Work Experience in             134          56.1
  Economic Sector
Work Experience in Economic       105          43.9
Total                             239         100.0

Mayor as the Non-first             93          37.2
  Prefecture-level Position
Mayor as the First                157          62.8
  Prefecture-level Position
Total                             250         100.0

Promoted from Sub-provincial      136          54.2
Promoted from Provincial          115          45.8
Total                             251         100.0


                              Promoted from   Promoted from    Total
                               Provincial     Sub-provincial
                               Government       Government

Mayor as the   Count(%)          70 (75.3)       23 (24.7)     93
Mayor as the   Count(%)          44 (28.0)      113 (72.0)     157
Total          Count (%)        114 (45.6)      136 (54.4)     250

                Value              df          Asymp. sig.
                                                (2 sided)

Pearson         52.548              1              .000
N of valid     250


Dependent Variable                 Model 1        Model 2
                                 Pro. speed A   Pro. speed A

Independent Variable              45.150 ***     45.008 ***
(Constant)                       (10.135)       (10.128)

Age (1 Jan. 2006)                 -1.546 ***     -1.539 ***
                                 (-8.605)       (-8.592)

Age Square                          .0141 ***      .0141 ***
                                  (7.774)        (7.787)

Party Age (as of 1 Jan. 2006)       .0079         --

Years of Education                  .057 **        .0552 **
                                  (2.441)        (2.386)

Academic Discipline                 .314 ***       .305 ***
                                  (2.948)        (2.888)

Mayor as the First                 -.375 ***      -.367 ***
Prefecture-level Position        (-3.877)       (-3.828)

Provincial/sub-provincial           .180 *         .182 *
                                  (1.853)        (1.876)

Organisational/                    -.0703         -.0652
non-organisational                (-.648)        (-.603)

Economic/non-economic              -.339 ***      -.347 ***
                                 (-3.609)       (-3.724)

CYL/non-CYL                        -.193 **       -.180 **
                                 (-2.110)       (-2.018)

Mishu/non-mishu                    -.085          -.0874
                                  (-.950)        (-.980)

Local/non-local                    -.176 *        -.175 *
                                 (-1.836)       (-1.825)

R Square                            .688           .687

F                                 38.403 ***     41.965 ***

No. of cases                     222            222

Dependent Variable                 Model 3        Model 4
                                 Pro. speed B   Pro. speed B

Independent Variable              37.829 ***     37.682 ***
(Constant)                       (10.659)       (10.636)

Age (1 Jan. 2006)                 -1.281 ***     -1.273 ***
                                 (-8.945)       (-8.914)

Age Square                          .0115 ***      .0115 ***
                                  (7.960)        (7.968)

Party Age (as of 1 Jan. 2006)       .0082         --

Years of Education                  .0460 **       .0443 **
                                  (2.479)        (2.401)

Academic Discipline                 .287 ***       .278 ***
                                  (3.385)        (3.300)

Mayor as the First                  .116           .125 *
Prefecture-level Position         (1.500)        (1.630)

Provincial/sub-provincial           .106           .108 *
                                  (1.369)        (1.396)

Organisational/                    -.0342         -.0289
non-organisational                (-.396)        (-.336)

Economic/non-economic              -.139 *        -.147 **
                                 (-1.855)       (-1.978)

CYL/non-CYL                        -.173 **       -.159 **
                                 (-2.369)       (-2.238)

Mishu/non-mishu                    -.0324         -.0349
                                  (-.455)        (-.491)

Local/non-local                    -.0053         -.0039
                                  (-.069)        (-.051)

R Square                            .698           .697

F                                 40.324 ***     43.974 ***

No. of cases                     222            222

Note: * p < .1; ** p < .05; ***[micro]p < .01; figures in
parentheses are t-statistics.


Independent Variable            B          t.     Sig.

(Constant)                  -26449.234   -2.575   .0107
Population (100 times         3450.321    3.900   .0001
  inversed population)
Promotion speed A             2353.055    3.512   .0005
Age (as of 31 Dec. 2005)       415.337    2.561   .0111
Economic/non-economic         2573.950    2.688   .0077
Provincial/sub-provincial    -2187.371   -2.085   .0382
Growth Rate                    409.924    3.293   .0012
Local/non-local               -315.897    -.286   .7748

R square                          .389
F                                6.910            .0000
No. of cases                   229

Dependent variable: GDP per capita, the year before
mayoral nomination.
COPYRIGHT 2012 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 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Tingjin, Lin
Publication:China: An International Journal
Article Type:Report
Geographic Code:9CHIN
Date:Dec 1, 2012
Previous Article:Competitive or complementary? China's role in trade cooperation with ASEAN in the 1990s and 2000s.
Next Article:China's currency devaluation and WTO issues.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters