OCCUPATIONAL AND CAREER MOBILITY

Occupational and career mobility in adulthood is often referred to as intragenerational social mobility. It involves change in an individual's position in the labor market over the adult life course. Change is studied with respect to both type of work and the rewards derived from work. The term career refers to an individual's job history. Empirical regularity in the careers of individuals in the labor force defines what we call a "career line" or "job trajectory," since a work history common to a portion of the labor force reflects the existence of structurally determined linkages among jobs in the economy. Jobs are located in particular firms, whereas occupations and industries encompass jobs in many firms. An individual may remain in the same occupation or industry but change firms (and jobs within the same firm) any number of times. Since the process of job change does not necessarily involve a change of occupation or industry, but a change of occupation or industry always involves a job change, the process of job change provides a more detailed account of career movement. Changes in the rewards derived from work usually accompany job changes but can also occur during the course of tenure in a job.

Research on intragenerational mobility has focused on the labor force as a whole and on employees in particular occupations and firms. Research on the labor force as a whole has usually considered change in occupation and industry, as measured for detailed categories or more aggregated groupings that define broad occupational and industrial groups. It has also considered change in the rewards derived from work, focusing primarily on occupational prestige and earnings. Research on particular occupations and firms has usually focused on job status and authority changes within organizational hierarchies and on changes in work rewards.

Early work on intragenerational mobility involved the mathematical modeling of transition probabilities, usually among a few, highly aggregated categories (see Mayer 1972 for a review). This work used Markov models and semi-Markov models to analyze transition probabilities in a sequence under the assumption that the job category an individual will occupy in the future depends only on the job category occupied in the present and not on job categories occupied previously. Although there was empirical support for this assumption in some studies, it was not found to be broadly applicable. In semi-Markov models, transition probabilities are permitted to vary with time and for subgroups of the population. These models capture declines in mobility with age or duration of stay and allow for the fact that some individuals are more likely to move than others. There is evidence that mobility is an exponentially declining function of time (Mayer 1972) and that some individuals become "movers" while others become "stayers" (Blumen et al. 1955).

Later analyses have considered more refined models of job change, focusing on job shifts as elementary acts in the mobility process. Job shifts may be either voluntary or involuntary, and may occur at a decreasing rate with time in the labor market. They are also more likely to occur within a firm than between firms as labor market experience increases (Rosenfeld 1992; DiPrete and Nonnemaker 1997). As time spent in a firm increases, both rates of promotion and rates of leaving the firm also decline (Petersen and Spilerman 1990; DiPrete and Nonnemaker 1997). When the effect of labor market experience is considered simultaneously, firm tenure may have a positive effect on within-firm job mobility, but this effect indicates that the rate of within-firm job mobility declines with labor market experience more slowly for each year of tenure with an employer than for each year of pre-employer labor market experience. Job shifts (both voluntary and involuntary) are strongly and negatively related to job tenure. As the duration of a job increases, the probability of leaving it declines. Most of this decline occurs in the first year of a job, after which there is a leveling off (Topel and Ward 1992). Within firms, however, job tenure has a positive effect on the probability of promotion, in some instances increasing up to a point and then decreasing (Felmlee 1982; Althauser and Kalleberg 1990; Petersen and Spilerman 1990). The rate of mobility varies with characteristics of the individual, the job, the occupation, the employing organization, and the economic environment.

Change in the rewards derived from work usually accompanies a job shift (Sørensen 1974; Rosenbaum 1984; Topel and Ward 1992). Voluntary job shifts are associated with increases in prestige and wages; involuntary job shifts, with losses. Job shifts within firms and occupations are more often associated with gains in prestige and wages than those between firms and occupations (DiPrete and Krecker 1991; DiPrete and McManus 1996; Cheng and Kalleberg 1996). In general, the rewards of changing jobs decline with labor market experience and job tenure. Wages, as well as the opportunity for wage growth and promotion, bear a negative relationship to the probability of a job shift (Petersen and Spilerman 1990; Topel and Ward 1992). Change in wages also occurs during the course of a job (Topel 1991), declining with labor market experience and, at a decreasing rate, with job tenure (Topel and Ward 1992). For the labor force as a whole, occupational prestige and earnings increase over the adult working life. The shape of these trajectories tends to be concave downward—with a rise early in the career, a plateau during the middle years, and a slight decline as the end of the work career approaches. Status and earnings trajectories have the same form, but the former is flatter (Mincer 1974; Rosenfeld 1980). There is also variation in the shape of these trajectories for those in different career lines (Spilerman 1986).

Given the heterogeneity of career lines and the important role that age, or duration since career entry, plays in shaping career lines, there have been attempts to study the path of careers in recent years. Not only does mobility decline sharply with age, and therefore with the proximity of a job to the end of a ca reer line, but job changes that occur later in an individual's work life tend to involve jobs requiring skills that are more similar than those that occur earlier in the work life (Spenner et al. 1982; Althauser and Kalleberg 1990). There is also evidence that early career experiences have an important effect on later career outcomes. This evidence indicates not only that those entering different career lines, who receive different rewards at career entry, can expect different career outcomes (Marini 1980; Spenner et al. 1982), but also that early experiences within a career line can condition subsequent progression and the level of reward attained relative to others who enter the same career line (Stewman and Konda 1983; Rosenbaum 1984). The career lines most often studied have been trajectories within institutional structures, but there have also been attempts to describe career lines that cross institutional boundaries (Spilerman 1977; Spenner et al. 1982; Althauser and Kalleberg 1990).

LABOR MARKET STRUCTURE

The concept of "career line" or "job trajectory" derives from the view that the labor market is structured in a way that makes some types of job changes more likely than others. Early work on career mobility ignored this structural differentiation, estimating the overall (linear) relationship between the status and earnings of an individual's first job and the status and earnings of a job held later in the career (Blau and Duncan 1967; Coleman et al. 1972; Marini 1980). Jobs resembling each other in status, pay, and working conditions, however, are sometimes part of a career line and sometimes not, even if part of a career line can be attached to different career lines. Jobs providing similar current rewards may therefore not offer the same prospects for future mobility.

Since career lines are rooted in labor-market structure, their existence demonstrates that intragenerational mobility is influenced by the structure of the labor market and changes in that structure, as well as by the demography of the labor force and individual characteristics that affect movement within segments of the labor market. If jobs have entry requirements and confer rewards, the structure of jobs plays a critical role in establishing the link between the attributes of individuals and work rewards. Recognition that the labor market is structured in a way that produces segmentation among career lines has led to attempts not only to describe career lines but to identify the forces shaping them.

Pioneering work by White (1970) directed attention to the importance of social structure in shaping careers by modeling the way in which vacancies in a structured labor market trigger career movement. More recent models of vacancybased movement have further delineated the ways in which vacancy chains, job distributions, and managerial staffing and hiring practice structure the relative career chances of individuals (Sørensen 1977; Skvoretz 1984; Stewman and Konda 1983; Stewman 1986). In these models the availability of job openings determines career advancement possibilities, and the shape of organizational hierarchies affects the probability of advancement. Since most organizational hierarchies are pyramidal, with many more low-level than high-level positions, the average employee's advancement must slow down over time. Aggregate age-promotion curves appear to be described empirically by an exponential-decline function where the highest promotion chances occur at the outset, declines are a fixed proportion of an individual's current chances, and promotion chances become increasingly rare but not impossible. In short, promotion favors youth, declines gradually, but does not disappear for older workers. There may also be chances for an increase in promotion during the initial career years when most on-the-job training occurs.

Although career advancement often occurs as a result of job change to fill a vacant position, not all career movement depends on or is affected by a job opening. New jobs in organizations are sometimes created for particular individuals, and jobs that are vacated may be eliminated rather than filled. New jobs can also be created by individual moves to become self-employed, and the termination of self-employment does not necessarily create a vacancy. Within organizations, career advancement occurs through upgrading as well as vacancy filling. An employee receives an upgrade promotion when a certain level of seniority or a certain performance marker is reached. Such promotions involve reclassification to a higher rank and do not depend on a vacancy's being present (Stewman and Yeh 1991; Barnett and Miner 1992). Both upgrade promotions and vacancy promotions may occur during the course of an individual's career. These alternative advancement mechanisms, however, are not necessarily found equally throughout an organizational hierarchy. In several occupational categories studied in one organization, most lower-level promotions were vacancy based, and a higher proportion of upper-level promotions occurred through upgrading (Barnett and Miner 1992). The mechanisms available for promotion affect an individual's promotion chances. It has been found that if promotion occurs through upgrading based on performance rather than vacancy filling, salary grade level is not necessarily associated with a lower probability of promotion (Petersen and Spilerman 1990). Similarly, the effect of hiring temporary workers on the promotion chances of permanent workers depends on the extent to which promotion occurs through vacancy filling rather than upgrading (Barnett and Miner 1992).

Because the structure of the labor market affects career movement, there have been attempts to understand the structural bases of segmentation among career lines. During the 1940s and 1950s, institutional economists called for an understanding of well-defined systems of jobs and firms, drawing a distinction between internal and external labor markets. For example, Dunlop (1957) argued that within a firm there are groups of jobs, or "job clusters," each of which is linked together by technology, the administrative organization of the production process, and the social customs of the work community. A job cluster usually contains one or more key jobs and a group of associated jobs, and the wage rates for the key jobs mediate the effects of labor market influences, including union and government wage policies, and forces in the market for products on the wage structure of the firm. The distinction between internal and external labor markets was later reintroduced by Doeringer and Piore (1971), who discussed what they called a "mobility cluster." The central idea was that administrative rules and procedures tend to set up separate markets for those already hired (an internal labor market) and those seeking employment (an external labor market). A firm hires workers from the outside labor market into "entry jobs," and other jobs are filled internally as workers progress on well-defined career ladders by acquiring job-related skills, many of which are firm-specific. Thus, firms make investments in individuals, and these investments segment the workforce with respect to advancement opportunity.

The concept of the internal labor market was developed further in what was first called dual and then segmented labor market theory (Gordon 1972; Kalleberg and Sørensen 1979). This theory draws a distinction between primary and secondary jobs, arguing that the internal labor market is only one kind of work setting. Primary jobs emphasize long-term attachment between workers and firms and offer built-in career ladders and promotion opportunities, whereas secondary jobs do not offer these advantages. The distinction between primary and secondary jobs may occur within the same firm or between firms, since primary jobs are considered more likely to be found in oligopolistic, unionized industries, and secondary jobs in competitive industries.

Dual and other segmented labor-market theories have come under attack as being too crude to meaningfully characterize the multiple dimensions on which labor markets vary. Attempts to measure labor-market segmentation by crude topologies, including industrial and occupational categories, have also been criticized. Because there is extensive heterogeneity within industrial and occupational categories, and because there is no way to link them to the kinds of job clusters hypothesized to exist, it has been argued that more disaggregated analyses of jobs and firms are needed. Many subsequent analyses have focused on particular bureaucracies or firms, and those examining larger segments of the labor market have sought to measure the characteristics on which employing organizations and occupations vary.

Attempting to develop a systematic conceptual scheme for studying labor market segmentation, Althauser and Kalleberg argued that "the concept of an internal labor market should include any cluster of jobs, regardless of occupational titles or employing organizations, that have three basic structural features: (a) a job ladder, with (b) entry only at the bottom, and (c) movement up this ladder, which is associated with a progressive development of knowledge and skill" (1981, p. 130). Based on four possible pairings of type of control and prospects for advancement, they differentiated four types of labor market structures: (1) firm internal labor markets, which are internal labor markets controlled by firms; (2) occupational internal labor markets, which are internal labor markets controlled by occupational incumbents; (3) firm labor markets, which provide firm-specific security without advancement prospects; and (4) occupational labor markets, which provide occupational security without advancement prospects.

Within a firm or an occupation, there may be multiple job ladders that vary in length and shape, and there are structured relationships among job ladders. Job mobility occurs not only along formal organizational and occupational ladders but across them (Baron et al. 1986; Stewman 1986; DiPrete 1989). For example, job ladders often cross detailed and even aggregated occupational boundaries. DiPrete (1987) described the permeability of these boundaries as depending on several types of contingencies: (1) skill-based contingencies, (2) information-based contingencies, (3) contingencies due to the configuration of positions in an organization, and (4) contingencies that emerge from the institutionalization of formal structure. Internal labor markets limit competition from outsiders by offering a combination of closed and restricted mobility contests that involve movement on and across job ladders (DiPrete and Krecker 1991; Stewman and Yeh 1991).

Because employing organizations vary in the kinds of jobs they offer, the occupational groupings in which those jobs fall, and the wages they pay, there has been some attempt to examine the effect of organizational characteristics on job mobility. Large organizations offer better pay, have more job ladders offering opportunities for promotion and wage growth, and offer better employee benefits. Organization size is therefore positively related to mobility within an organization and negatively related to voluntary movement out of an organization (Hachen 1992; DiPrete 1993; Cheng and Kalleberg 1996). Organizations in concentrated industries and industries dominated by conglomerates experience both lower rates of within-firm mobility and lower rates of voluntary exit because they have higher average wage levels, and there is a negative relationship between the average wage level of an industry and the probability of both within-firm mobility and voluntary firm exit. High-wage industries have lower rates of within-firm mobility than low-wage industries because, net of organizational size, internal labor markets are less prevalent in firms in high-wage industries. Capital-intensive industries have higher rates of both within-firm mobility and voluntary firm exit but lower rates of involuntary exit. These differences are linked to the greater prevalence of internal labor markets in firms in capital-intensive industries (Hachen 1992).

Although there is general agreement that the technical character of work influences organizational development, existing variation in institutional and personnel structures, especially cross-nationally, indicates that the technical character of work alone does not determine job ladders and career lines. These are affected by the historical circumstances surrounding an organization's founding, including the gender, racial, and ethnic composition of the workforce; by market and other social conditions that affect organizational change and employment growth; and by the negotiating strength of various bargaining units.

A number of explanations have been advanced for the emergence of internal labor markets. One focuses on the development of specialized knowledge and skill relevant to employment in particular occupations, jobs, and firms. Becker (1964) noted that when a worker receives training specific to the needs of an employer, the worker's value to the employer increases since a new employee would have to receive similar training before being able to perform at the same level. It is therefore in the interest of the employer to retain workers with such specific training by providing them with opportunities for promotion, salary growth, and employment security. When firm-specific skills are rewarded, workers also have an interest in remaining with the employer because the wages and other benefits they could obtain from another employer are lower. Based on this reasoning, Williamson (1975) argued that internal labor markets emerged because they were preferable to long-term contracts for maintaining the employment relationship. Thus, internal labor markets are seen as resulting from employers' needs for renewable supplies of otherwise scarce, highly skilled workers. Substantial empirical evidence is consistent with this view. Another explanation, offered by Bulow and Summers (1986), is that job hierarchies offering promotion prospects and wage increases are an important means of motivating workers when individual performance is not easily monitored.

The emergence and spread of internal labor markets is also seen as related to the development of personnel departments, formalized rules, and the rise of bureaucratic control. Although there is some empirical support for this view, at least some internal labor markets preceded the emergence of bureaucratic control systems and rules (Althauser 1989; Stovel et al. 1996). Spilerman (1986) has noted that workers have an interest in barring lateral entry and in having high-level positions filled through promotion. Workers may also wish to limit employer discretion by having decisions about promotion and layoff tied to seniority. These worker interests cause labor unions to work to create more widespread job hierarchies and promotion. Although internal labor markets have been argued to be a result of unionization, evidence suggests that they are to some degree an alternative to unionization (Pfeffer and Cohen 1984). In manufacturing establishments, the technology of production helps to manage the workforce, binding workers to a specific organization with specific machines and production technology, and providing necessary skills and training. Unionization reduces turnover by providing workers with a voice, skewing compensation toward deferred benefits over current wages and bonuses, and paying a larger proportion of total compensation as fringe benefits valued particularly by senior workers. In the presence of a governance system under unionization, the internal labor market and other forms of bureaucratic control are redundant if not in competition with union mechanisms.

Because the prevalence of internal labor markets varies across organizations, there has been some attempt to explain why they are found more often in some organizations than others. It was initially argued that firms characterized by high profit levels, oligopolistic pricing, and large organization size were more likely to create internal labor markets because they could better afford to. However, there is evidence that internal labor markets are not merely a derivative feature of core-economy organization (Althauser 1989). Hachen (1992) has suggested that they are affected by three characteristics of employer personnel policy: the need to retain workers, the availability of alternative mechanisms for retention, and the ability to use specific mechanisms. In labor-intensive industries where labor costs are substantial and worker replacement is used as a strategy to maintain or even lower wage levels, retaining workers is less important. As a result, within-firm mobility rates are low, and involuntary exit rates are high. When an organization desires to retain workers, offering higher wages and increasing internal opportunities are two means of retention, but organizations vary in their ability to use these mechanisms. A large organization will be better able to offer internal opportunities. An organization that can offer high wages may have less need to offer advancement opportunities.

Career lines not only involve substantial movement between job ladders within firms but often cross firm and industry boundaries rather than remaining within them. In the U.S. economy, most job change involves a change of employer (DiPrete and Krecker 1991). It has been estimated that only somewhat over a quarter of U.S. workers are continuously employed by the same employer for twenty years or more (Hall 1982). In some careers, such as the salaried professions, crafts, and "secondary" labor-market positions, firm and industry are not a locus of career line structure. Nevertheless, moves between firms and industries often occur between related positions, so that labor-market segmentation emerges even without institutional barriers.

Aspects of the structure of labor markets, other than internal labor markets, that have implications for occupational and career mobility have been documented in cross-national research. One type of cross-national variation is in the relationship between the educational system and the labor market, which varies according to the structure of the educational system. In countries with more stratified and standardized educational systems, such as Germany, Austria, France, and Norway, there is a tighter linkage between educational preparation and labor-market position than in the United States. In these countries, reliance on educational degrees and other credentials as screening devices is greater, and rates of occupational and job mobility are lower (Haller et al. 1985; Allmendinger 1989; DiPrete and McManus 1996). In the United States, where schooling is more open and comprehensive, the number of years of education attained matters more, and career mobility is higher because a larger proportion of employees are allocated to low-level entry positions in organizations. In Great Britain, which has a stratified educational system, similar to those of the other European countries, occupational mobility is higher than in the United States because the linkage between education and training prior to labor-market entry and job placement is weaker. Workers are also more likely to attain part-time education after labor-market entry that leads to occupational change (Winfield et al. 1989).

Another aspect of labor market structure that varies cross-nationally is the degree of institutional separation between manual and white-collar work, which affects mobility across this boundary. Existing throughout Europe, but strongest in the German-speaking middle-European countries, is the institution of apprenticeship for manual work, which restricts access to jobs for skilled workers to those who complete an apprenticeship after elementary schooling. In countries where the institution of apprenticeship is less strong, as in France and the United States, access to a job as a skilled worker is more dependent on affiliation with a specific enterprise and is open to those moving up from unskilled jobs. The greater stratification of the educational system in countries where the institution of apprenticeship is strong also restricts upward mobility from manual to nonmanual jobs.

Another variation observed cross-nationally is in the degree of separation between hierarchically ranked white-collar positions. In Austria and France, there is greater separation among hierarchical layers of white-collar work than in the United States, where the higher levels of white-collar work are more open to entry from below (Haller et al. 1985). In the United States, both the highest and lowest categories in the occupational structure are less separated from the middle layers than they are in Austria and France. As a result, mobility over relatively long distances is more common in the United States. This difference may be due in part to the less stratified educational system in the United States.

Another type of variation in labor market structure found across countries is in the links of the labor market to other national institutions such as the state and centralized employer and labor associations. In some countries, such as China, the state remains the major employer (Zhou et al. 1997); in others, such as Austria and France, large enterprises and whole industrial sectors are nationalized (Haller at al. 1985). Labor markets in Europe have also been described as less flexible than those in the United States because "corporatist" institutional arrangements of various types coordinate and regulate the labor markets. Countries described as corporatist have relatively centralized wage-setting institutions that result in less volatility in wages and smaller wage differences (DiPrete and McManus 1996; Gottschalk and Smeeding 1997). They may also have greater employment security that reduces involuntary job mobility, but restrictions on wages may result in relatively high unemployment as well.

Because the availability of vacancies in organizational hierarchies affects career advancement, and the supply of labor affects both the probability of advancement and wages, career progress in all countries is affected by demographic factors such as the size and distribution of education and skills in various age cohorts and the rate of exit from positions (Stewman and Konda 1983). It is also affected by organizational growth and contraction that result in the creation and termination of jobs. Change in the actual structure of jobs occurs in response to social and economic influences, including technological development. At the level of the firm, expansion increases the rate of promotion, and contraction decreases it (Stewman and Konda 1983; Barnett and Miner 1992). Expansion of an industry, as measured by growth in the average size of firms or the establishment of new firms, increases the rate of job shifts within and between firms in the industry and decreases the rate of involuntary exit (Hachen 1992; DiPrete 1993; Haveman and Cohen 1994; DiPrete and Nonnemaker 1997). Contraction of an industry, as measured by decline in the average size of firms or the closing of firms, decreases the rate of job shifts between firms in the industry and increases the rate of movement out of the industry, including movement to unemployment. Contraction of an industry may either decrease or increase the rate of within-firm mobility, since contraction may lead to the internal redeployment of labor as a substitute for new hiring. Internal mobility that reflects reorganization occurs primarily among lower white-collar, service, and blue-collar workers. Movement out of the industry and to unemployment occurs primarily among service and blue-collar workers (DiPrete 1993). Change that involves merger and acquisition within an industry has fewer effects, but they tend to be similar to the effects of industry contraction (DiPrete 1993; Haveman and Cohen 1994). The effects of occupational expansion and contraction differ somewhat from those of industrial expansion and contraction. Expansion of an occupation increases the rate of mobility between firms in an industry but not within firms. Contraction of an occupation increases the rate of mobility within firms (due to the redeployment of labor) and increases the rate of movement to unemployment but does not increase the rate of movement out of the industry (DiPrete and Nonnemaker 1997). The adverse effects of economic turbulence in either an industry or occupation are dampened by employer tenure.

Since the structure of the labor market affects both job mobility and wage growth, aspects of labor market structure account for the aggregate declines in job mobility and wage gain observed over the course of workers' careers. One aspect of labor-market structure that plays a role in explaining the declining change in career rewards is the pyramidal shape of organizational and occupational hierarchies, which contain many more lower-level than higher-level positions. Given this structure and the role of vacancy filling in promotion, the chances of promotion and wage growth linked to promotion decline with labor-market experience. To the extent that upward mobility is possible, workers entering a career line are in competition with others at the same level for advancement to the next highest level. Because jobs at the next level are filled from those occupying positions in the level below, entry to a job ladder and performance at each step on the ladder affect ultimate career attainments. Rosenbaum (1984) has described the process by which employees in a cohort are progressively differentiated throughout their careers in a series of implicit competitions as being like a tournament. Selections among the members of a cohort occur continually as careers unfold, and each selection affects the opportunity to advance further. As a result, individuals are distributed more and more to "winner "and "loser" paths, and interpersonal variance in wages rises with labor-market experience and firm tenure. Those who are not promoted are more likely to leave the firm, but their gain from doing so is less than the gain of those who are promoted internally.

According to human capital theory, the shape of wage trajectories over the career course is a function of the changing productivity of workers. Wage growth through labor-market experience results from learning on the job that increases a worker's productivity and value to the employer. On-the-job training occurs in formal training programs provided by employers or as a result of informal instruction by supervisors and coworkers and simply by doing the job. Time away from the job, in contrast, can lead to skill depreciation. Because investments in training are costly, these are concentrated in the early part of careers, leading to greater wage growth at early rather than later career stages. Formal and informal training in a particular firm and on particular jobs can provide knowledge and skills that are general or specific. General knowledge and skills are applicable in other firms and other jobs, whereas specific knowledge and skills are applicable only in the firm or job where they are acquired. As workers develop firm- and job-specific human capital, they are less likely to leave that firm or job because their productivity and resulting wages are higher than they would be in another firm or job. Although productivity may be greater in the firm or job where specific human capital is acquired, increments in productivity decline over time. As a result of this decline and biological aging, productivity may be lower at the end of a career than it was earlier.

One problem with human capital theory is that the relationship between labor market experience and worker productivity has not been adequately tested. Labor-market experience may affect wages because it is a proxy for seniority if wages rise with seniority and job tenure regardless of productivity. For women, labor-market experience may be an outcome rather than a determinant of career mobility if entry into career lines offering little opportunity for advancement affects labor-force participation (Marini 1980; Marini and Fan 1995). Evidence on the relationship between labor market experience and productivity within occupations suggests that, although labor market experience has some bearing on productivity, its effect on career advancement is largely independent of productivity (Horowitz and Sherman 1980; Medoff and Abraham 1980, 1981; Maranto and Rodgers 1984). Much variability is also seen across occupations and work contexts in the extent to which work experience affects either productivity or earnings (Horowitz and Sherman 1980; Spilerman 1986). Evidence also contradicts the "tradeoff hypothesis"—that individuals sacrifice earnings at the beginning of their careers for better long-term career prospects. Individuals with lower earnings early in their careers actually have lower rather than higher job status and earnings later.

More recent economic explanations of the relationship between labor-market experience and wage growth have focused on new arguments about the role of productivity in shaping the desire of employers, especially those who make large investments in screening and training employees, to retain workers and motivate high performance over time (Lazear 1981; Lazear and Rosen 1981; Malcomson 1984; Bulow and Summers 1986). Promotions and wage increases are seen as a means of eliciting effort from workers when the monitoring of their efforts and outputs is prohibitively expensive. These explanations deviate from the view that labor is paid its marginal product in each short period, assuming that workers are paid their marginal product over the life cycle or in some cases in excess of their marginal product. So far, none of these explanations has an empirical basis, and, as single-factor explanations, they are unlikely to account for the diverse compensation schemes observed across occupations and work settings (Talbert and Bose 1977; Spilerman 1986). A further problem is that they ignore the role of nonmonetary incentives in retaining workers and motivating high performance.

Other economic explanations of career mobility have combined human capital theory and information theory to consider the role of imperfect information and the acquiring of additional information as influences on career decision making (e.g., Jovanovic 1979). As in human capital theory, voluntary job mobility is assumed to result from a comparison of the benefits derived from one's current job with those that could be obtained from another job. The decision to search for or accept another job is argued to be affected by new information about either the current job or a possible alternative. Because certain aspects of a job cannot be assessed prior to employment, new information is argued to be acquired through actually working on the job, and this information affects assessment of the quality of the person-job match. Jobs that survive are those evaluated as a good match, or at least as a better match than could be obtained elsewhere. With increases in labor-market experience, workers acquire more information about the labor market and are better able to demonstrate their performance, which in turn provides more information to employers. As a result, matches improve over the course of a career, and, because it becomes increasingly less likely that a better match can be found, job mobility declines. Although the role of information is implicit, a related argument is linked to the vacancy-based model of career advancement (Sørensen 1977). According to this argument, as time in the labor market increases, people who enter the labor market in a position below their optimum level because a better vacancy is unavailable have more time to take advantage of vacancies and to close the gap between their current and potential rewards. What these models do not consider is the path dependency of careers, whereby there are long-term effects of early job placements on later career outcomes. Job placement early in a career not only shapes worker experience and training but is used as a basis for screening by employers that affects later career opportunities.

Information about the labor market and specific alternative jobs may also change with labor-market experience in ways that affect assessments of the likelihood that a better alternative is available. At labor-market entry and during the early career, both occupational aspirations and the importance attached to a variety of job attributes decline. These changes suggest a diminished view of the labor market, which makes it less likely that a better alternative to one's current job will be perceived to be available. As firm and job tenure increase, it is also likely that knowledge of specific alternative job opportunities declines, since the time since the worker's last involvement in job search is longer and contacts with others who would be sources of information are fewer.

If opportunities for career advancement and wage growth are used by employers to retain workers who have firm- and job-specific knowledge and skill, career advancement and wage growth may also decline over the course of individual careers because they become less necessary means of retaining workers. They become less necessary because increases in firm-and job-specific knowledge and skills increase the extent to which a worker will be more valuable to the current employer than to an alternative employer and because the value of accrued employee benefits, including retirement benefits, makes it increasingly costly for a worker to change employers. As firm and job tenure increase, personal relationships and familiarity with the work environment also increase, reducing the likelihood that a worker will change employers or jobs. There is also less time left as a career advances to recover the costs of a change of employer or job and achieve the same level of functioning as was attained with the current employer or a higher level (Groot and Verberne 1997).

WORKER CHARACTERISTICS

Given the structure of the labor market, particularly the local labor market where most job search occurs, entry into a career line is affected by worker characteristics, such as job-related credentials (e.g., education, intelligence, physical attributes), job preferences, and access to resources (e.g., information, material support, sponsorship by influential others). In some career lines, worker characteristics at entry may also influence career progression, whereas in others such characteristics may have little effect after access to a career line is obtained. If worker characteristics at entry continue to have an effect, they may do so in part because they influence subsequent on-the-job training and performance.

Most theoretical attempts to explain individual differences in career mobility have focused on individual differences in worker performance. In human capital theory, workers are seen as rational actors who make investments in their productive capacities to maximize lifetime income (Becker 1964; Mincer 1974). The investments usually studied are education and on-the-job training, although the theory applies to other investments, such as effort, job search, geographic mobility, and health. It is assumed that labor markets offer open opportunity and that earnings growth is a function of change in productivity due to how hard individuals work and to the ability, education, and training they possess. Individuals can increase their productivity not only through formal education but by learning on the job. Because investments in education and training are costly, workers concentrate their investments in the early part of their careers, sacrificing immediate earnings for better long-term career prospects. Although worker qualifications that may affect productivity have been found to be associated with advancement, these associations are consistent with explanations other than that afforded by human capital theory. In the theory of vacancy-based competition, workers' performance-related resources are also assumed to affect career advancement, but the mechanism by which they affect advancement is left unspecified (e.g., Sørensen 1977).

Sociologists generally view the relationships between education and experience, on the one hand, and career advancement on the other, as influenced primarily by the organizational, and even the broader societal, context in which the administrative arrangements that govern promotions and salary advancement arise. Spilerman (1986) has suggested that education and experience bear weaker relationships to promotion and earnings when organizational rules rigidly prescribe the temporal paths of occupational and earnings advancement. These rigid schedules are usually found in workplaces where the majority of workers are engaged in a very few career lines, or where multiple career lines exist, but there is little opportunity for transferring among them. Such schedules often result from unionization, since labor unions seek to standardize work arrangements. Even the effects of education and experience that exist under this type of personnel system vary across cities in ways suggesting an influence of general societal beliefs that educated and experienced workers should be paid more. Because such beliefs do not specify how much more, compensation schedules vary widely. Education and experience appear to bear stronger relationships to earnings in large nonunionized organizations that encompass many occupational specialties. However, even in these organizations, societal notions of equity and custom may affect wage structures. In Japan, for example, both seniority and family size are major determinants of salary in large companies (Dore 1973).

Sociological accounts differ from economic explanations in recognizing that the rewards of work derive from job occupancy and that a relatively enduring structure of jobs determines the relationships of individual effort, ability, and performance to work rewards. In the sociological view, jobs are assigned wage rates via processes operating at the societal and organizational levels; and the mechanisms that match individuals to jobs produce associations between individual effort, ability, and training, on the one hand, and work rewards, on the other. Jobs differ in the routes by which they are entered and in the extent to which performance can affect work rewards.

What influences advancement is not the attributes of individuals per se, but their attributes in relation to organizational positions. Organizations define the criteria by which ability is identified. To the extent that ability is not tied to those criteria, it will not be recognized and rewarded. If individuals of limited ability are able by chance or other more calculated means to meet the criteria by which ability is identified, they will be assumed to have ability by an inference process in which the direction of causality is reversed. An important consequence of this system is that factors affecting access to positions, of which ability is only one, and factors affecting performance in accordance with organizationally recognized criteria, including the willingness to conform to organizational goals and practices, have an important influence on long-term career outcomes.

Given the importance of the structure of jobs in mediating the relationship between individual attributes such as education and experience and job rewards, it is not surprising that unchanging attributes have little direct effect on rewards within job status categories. However, they have an important effect on careers via their influence on access to positions in the structure of jobs. Education is a major determinant of access to job ladders and career lines, and movement within these produces relationships between education and experience and work rewards. The organizational hierarchies to which college graduates, especially those from preferred colleges, have access increase the effects of college on career attainments over time. These hierarchies are moved through via experience. Within at least some career lines, education increases rates of promotion and wage gain (Rosenbaum 1984; Petersen and Spilerman 1990; Sicherman and Galor 1990). Education bears a positive relationship to within-firm and within-occupation mobility and a negative relationship to mobility between occupations, firms, and industries, as well as a negative relationship to moves to unemployment (Sicherman and Galor 1990; Cheng and Kalleberg 1996; DiPrete and Nonnemaker 1997). When mobility does occur between occupations and firms, education increases the likelihood that these moves are voluntary and in an upward direction. Education also helps to buffer workers from economic turbulence. Young, educated workers are both helped less by organizational expansion and hurt less by organizational contraction (Rosenbaum 1984; DiPrete 1993; DiPrete and Nonnemaker 1997).

Over time, career histories become differentiated in the timing and occurrence of advancement. Early recognition and achievement relative to members of the same entry cohort have long-term effects on career mobility. Those receiving early promotions have more rapidly advancing subsequent careers and ultimately attain higher-level positions and higher earnings than those not receiving early recognition (Stewman and Konda 1983; Rosenbaum 1984). The age when a person enters a position bears a negative relationship at career advancement (Petersen and Spilerman 1990; Rosenfeld 1992). The early recognition of "stars" and its cumulative effect on subsequent career movement has been argued to occur as a result of labeling processes that identify such individuals as especially able—as high achievers who are likely to achieve more.

One problem in research on career mobility has been the limited availability of actual measures of worker ability and performance. Most research has studied the effects of education; formal on-thejob training; and informal training acquired through experience in the labor market, in a firm, or on a job. This set of influences is not only limited, but measures of these influences have tended to be crude and indirect. Informal training through experience has usually been inferred from the durations of time spent in the labor market, in a firm, or on a job. Since direct measures of ability and performance have rarely been obtained, knowledge of the effects of ability and performance on career mobility remains limited.

In both theory and research, there has been relatively little attention paid to the influence of worker characteristics other than ability and performance. There is evidence that individual differences in work and job values affect entry into career lines and that these change over the career course, in part as a result of experiences in the labor market (Lindsay and Knox 1984; Judge and Bretz 1992). However, the effect of workers' values, normative beliefs, preferences, and personality characteristics on career mobility has not been studied, and individual variation in these characteristics has often been assumed to be either unimportant or nonexistent. Similarly, little attention has been paid to the influence of noncareer events and activities on career advancement. There has been some limited study of the effects of family roles, which indicates that marriage and children have different effects on the career advancement of women and men, but relatively little is known about the effects of family and other noncareer events and activities on career mobility.

Apart from ability and performance, the worker characteristics receiving the most attention in research on career mobility have been ascriptive characteristics such as gender, race, and ethnicity. The structure of the labor market mediates the relationship of these characteristics to job rewards, since these characteristics become a basis for the differentiation of job ladders and career lines, as well as a basis for access to the jobs within them (Spilerman 1977; DiPrete 1989; Marini and Fan 1995; Yamagata et al. 1997). Because other measures of worker characteristics and the process of employer evaluation and selection have received little research attention, however, understanding of the multiple factors that produce differences in career mobility among groups defined by ascriptive characteristics remains limited.

Another type of worker characteristic that has received some research attention is the social ties of workers, particularly as used in the process of job change. It has been argued that social networks provide a kind of capital—social capital—that is useful to workers in career advancement. Social ties of high status are seen as furthering career advancement more than those of low status, and it has also been argued that weak social ties are more helpful than strong social ties because weak ties provide a less redundant source of information and influence (Granovetter 1974; Lin 1990). There is evidence that workers do rely on others whom they know in the process of job change, but the extent to which social ties are a source of information and influence varies with the structure of the labor market and a worker's place in it. In China's largely state-controlled economy, a worker's social ties are not used as a source of information but are a source of influence on state authorities with decision-making power over jobs. In market economies, social ties are used as a source of both information and influence, although the extent to which they are used varies across countries and for different types of career lines (Wegener 1991; Bian and Ang 1997). In at least some countries, workers in lower-status positions are more likely to use social ties when changing jobs than workers in higher-status positions. Evidence has been inconsistent on whether the use of social ties in obtaining jobs produces better outcomes, but when social ties are used and have an effect, obtaining help from those of high status rather than low status increases job rewards. Empirical evidence on the influence of the strength of social ties has been inconsistent, in part because the strength of social ties has been defined and operationalized in a variety of ways. In some countries strong social ties have been shown to be more helpful than weak ones. In other countries weak social ties are more helpful for workers in jobs of high status, whereas strong social ties are more helpful for workers in jobs of lower status (Wegener 1991; Bian and Ang 1997).

SELECTION BY EMPLOYERS

Movement within a career line is affected not only by the characteristics of workers but by the characteristics and actions of those empowered by employers to make hiring and promotion decisions. Evaluation of performance and ability is usually based on incomplete information. In many jobs, performance is difficult to assess. Ability is even harder to assess, because it is inferred from performance. In addition, human perceptive capabilities are limited and variable. As a result, employers tend to rely on readily available information, or "signals," such as the amount of education attained, where it was attained, the amount and types of prior job experience, observable personal attributes, and evidence of past performance such as the rate of career advancement and prior job status and earnings (Spence 1974). The use of such signals is particularly likely when job shifts are made between organizations and between job ladders within an organization rather than within an organization or a job ladder (Rosenfeld 1992). The criteria on which individuals are evaluated are therefore often superficial, and having the resources (i.e., money, knowledge, and skill) to identify the way the evaluation process works and acquire an appropriate set of signals plays an important role in career advancement. Because of the difficulty of obtaining information, employers are susceptible to employees' attempts to supply and manipulate information about themselves as well as others.

Another influence on the evaluation of ability and performance is attitudes and beliefs previously acquired by those with decision-making authority. Information is unconsciously filtered and interpreted through that cognitive lens. The effect of prior attitudes and beliefs is evident in the prejudice and stereotyping triggered by ascriptive characteristics such as gender, race, and ethnicity (Hamilton 1981; Marini 1989), which have been a focus of theories of discrimination in the labor market (see, e.g., Blau 1984).

In addition to the difficulties that arise in assessing performance and ability, personal relationships and political coalitions influence mobility. As noted, there is growing evidence that personal contacts and relationships can constitute important sources of information and influence in gaining access to jobs. For workers seeking advancement, they can provide information about available positions and how to apply or present oneself favorably as a candidate. They can also result in preferential treatment based on personal liking and trust, shared interests, or expected gain on the part of those making hiring and promotion decisions. For employees responsible for hiring and promotion, personal contacts and relationships can provide information about candidates that reduces the risk of error in the selection process and increases the likelihood that individuals with similar interests and a sense of loyalty or indebtedness to them will be advanced. Personal contacts and relationships can also affect decisions if they lead to a desire to gain favor with others supportive of a candidate. Because those making hiring and promotion decisions can act to advance their own interests and the interests of others, a candidate's position in workplace political coalitions can affect advancement prospects.

references

Allmendinger, Jutta 1989 Career Mobility Dynamics: AComparative Analysis of the United States, Norway, andWest Germany. Berlin: Max-Planck-Institut für Bilungsforschung.

Althauser, Robert P. 1989 "Internal Labor Markets." Annual Review of Sociology 15:143–161.

——, and Arne L. Kalleberg 1981 "Firms, Occupations and the Structure of Labor Markets: A Conceptual Analysis." In I. Berg, ed., Sociological Perspectiveson Labor Markets. New York: Academic.

——1990 "Identifying Career Lines and Internal Labor Markets within Firms: A Study in the Interrelationships of Theory and Methods." In R. L. Breiger, ed., Social Mobility and Social Structure. New York: Cambridge University Press.

Barnett, William P., and Ann S. Miner 1992 "Standing on the Shoulders of Others: Career Interdependence in Job Mobility." Administrative Science Quarterly 37:262–281.

Baron, James N., Alison Davis-Blake, and William T. Biebly 1986 "The Structure of Opportunity: How Promotion Ladders Vary within and among Organizations." Administrative Science Quarterly 31:248–273.

Becker, Gary S. 1964 Human Capital. New York: National Bureau of Economic Research.

Bian, Yanjie, and Soon Ang 1997 "Guanxi Networks and Job Mobility in China and Singapore." SocialForces 75:981–1005.

Blau, Francine D. 1984 "Discrimination against Women: Theory and Evidence." In W. Darity, Jr., ed., Labor Economics: Modern Views. Boston: Kluwer-Nijhoff.

Blau, Peter M., and Otis Dudly Duncan 1967 The American Occupational Structure. New York: John Wiley.

Blumen, Isadore, M. Kogan, and P. J. McCarthy 1955 The Industrial Mobility of Labor as a Probability Process.Cornell Studies in Industrial and Labor Relations, vol. 6. Ithaca: New York State School of Industrial and Labor Relations, Cornell University.

Bulow, J .I., and Larry H. Summers 1986 "A Theory of Dual Labor Markets with Application to Industrial Policy, Discrimination, and Keynesian Unemployment." Journal of Labor Economics 4:376–414.

Cheng, Mariah Mantsun, and Arne L. Kalleberg 1996 "Labor Market Structures in Japan: An Analysis of Organizational and Occupational Mobility." SocialForces 74:1235–1260.

Coleman, James S., Zahava D. Blum, Aage B. Sørensen, and Peter H. Rossi 1972 "White and Black Careers during the First Decade of Labor Force Experience. I. Occupational Status." Social Science Research 1:243–270.

DiPrete, Thomas A. 1987 "Horizontal and Vertical Mobility in Organizations." Administrative Science Quarterly 32:422–444.

——1989 The Bureaucratic Labor Market. New York: Plenum.

——1993 "Industrial Restructuring and the Mobility Response of American Workers in the 1980s." American Sociological Review 58:74–96.

——, and Margaret L. Krecker 1991 "Occupational Linkages and Job Mobility within and across Organizations." Research in Social Stratification and Mobility 10:91–131.

DiPrete, Thomas A., and Patricia A. McManus 1996 "Institutions, Technical Change, and Diverging Life Chances: Earnings Mobility in the United States and Germany." American Journal of Sociology 102:34–79.

DiPrete, Thomas A., and K. Lynn Nonnemaker 1997 "Structural Change, Labor Market Turbulence, and Labor Market Outcomes." American Sociological Review 62:386–404.

Doeringer, Peter B., and M. J. Piore 1971 Internal LaborMarkets and Manpower Analysis. Lexington, Mass.: Heath.

Dore, Ronald P. 1973 British Factory—Japanese Factory. Berkeley: University of California Press.

Dunlop, John 1957 "The Task of Contemporary Wage Theory." In G. W. Taylor and F. C. Pierson, eds., NewConcepts in Wage Determination. New York: McGraw-Hill.

Felmlee, Diane H. 1982 "Women's Job Mobility Processes within and between Employers." American Journal of Sociology 80:44–57.

Gordon, David M. 1972 Theories of Poverty andUnderemployment. Lexington, Mass.: Heath.

Gottschalk, Peter, and Timothy M. Smeeding 1997 "Cross-National Comparisons of Earnings and Earnings Inequality." Journal of Economic Literature 35:633–687.

Granovetter, Mark 1974 Getting a Job: A Study of Contactsand Careers. Cambridge, Mass.: Harvard University Press.

Groot, Wim, and Maartje Verberne 1997 "Aging, Job Mobility, and Compensation." Oxford Economic Papers 49:380–403.

Hachen, David S., Jr. 1992 "Industrial Characteristics and Job Mobility." American Sociological Review 57:39–55.

Hall, Robert L. 1982 "The Importance of Lifetime Jobs in the U.S. Economy." American Economic Review 72:716–724.

Haller, Max, Wolfgang König, Peter Krause, and Karin Kurz 1985 "Patterns of Career Mobility and Structural Positions in Advanced Capitalist Societies: A Comparison of Men in Austria, France, and the United States." American Sociological Review 50:579–603.

Hamilton, David L. 1981 Cognitive Processes in Stereotyping and Intergroup Behavior. Hillsdale, N.J.: Erlbaum.

Haveman, Heather A., and Lisa E. Cohen 1994 "The Ecological Dynamics of Careers: The Impact of Organizational Founding, Dissolution, and Merger on Job Mobility." American Journal of Sociology 100:104–152.

Horowitz, Stanley A., and Allen Sherman 1980 "A Direct Measure of the Relationship between Human Capital and Productivity." Journal of Human Resources 15:67–76.

Jovanovic, Boyan 1979 "Job Matching and the Theory of Turnover." Journal of Political Economy 87:972–990.

Judge, Timothy A., and Robert D. Bretz, Jr. 1992 "Effects of Work Values on Job Choice Decisions." Journal of Applied Psychology 77:261–271.

Kalleberg, Arne, and Aage B. Sørensen 1979 "The Sociology of Labor Markets." Annual Review of Sociology 5:351–379.

Lazear, Edward P. 1981 "Agency, Earnings Profiles, Productivity, and Hours Restrictions." American Economic Review 71:606–620.

——, and Sherwin Rosen 1981 "Rank-Order Tournaments as Optimum Labor Contracts." Journal of Political Economy 89:841–864.

Lin, Nan 1990 "Social Resources and Social Mobility: A Structural Theory of Status Attainment." In R. L. Breiger, ed., Social Mobility and Social Structure. Cambridge: Cambridge University Press.

Lindsay, Paul, and William E. Knox 1984 "Continuity and Change in Work Values Among Young Adults." American Journal of Sociology 89:918–931.

Malcomson, James M. 1984 "Work Incentives, Hierarchy, and Internal Labor Markets." Journal of PoliticalEconomy 92:486–507.

Maranto, Cheryl L., and Robert C. Rodgers 1984 "Does Work Experience Increase Productivity? A Test of the On-the-Job Training Hypothesis." Journal of Human Resources 19:341–357.

Marini, Margaret Mooney 1980 "Sex Differences in the Process of Occupational Attainment: A Closer Look." Social Science Research 9:307–361.

——1989 "Sex Differences in Earnings in the United States." Annual Review of Sociology 15:343–380.

——, and Pi-Ling Fan 1995 "The Gender Gap in Earnings at Career Entry." American Sociological Review 62:588–604.

Mayer, Thomas 1972 "Models in Intragenerational Mobility." In J. Berger, M. Zelditch, Jr., and B. Anderson, eds., Sociological Theories in Progress, vol. 2. Boston: Houghton Mifflin.

Medoff, James L., and Katherine G. Abraham 1980 "Experience, Performance, and Earnings." QuarterlyJournal of Economy 95:703–736.

——1981 "Are Those Paid More Really More Productive? The Case of Experience." Journal of HumanResources 16:186–216.

Mincer, Jacob 1974 Schooling, Experience, and Earnings. New York: National Bureau of Economic Research.

Petersen, Trond, and Seymour Spilerman 1990 "Job Quits from an Internal Labor Market." In K. U. Mayer and N. B. Tuma, eds., Event History Analysis inLife Course Research. Madison: University of Wisconsin.

Pfeffer, Jeffrey, and Yinon Cohen 1984 "Determinants of Internal Labor Markets in Organizations." Administrative Science Quarterly 29:550–572.

Rosenbaum, James L. 1984 Career Mobility in a CorporateHierarchy. New York: Academic.

Rosenfeld, Rachel A. 1980 "Race and Sex Differences in Career Dynamics." American Sociological Review 45:583–609.

——1992 "Job Mobility and Career Processes." Annual Review of Sociology 18:39–61.

Sicherman, Nachum, and Oded Galor 1990 "A Theory of Career Mobility." Journal of Political Economy 98:169–192.

Skvoretz, John 1984 "The Logic of Opportunity and Mobility." Social Forces 63:72–97.

Sørensen, Aage B. 1974 "A Model for Occupational Careers." American Journal of Sociology 80:44–57.

——1977 "The Structure of Inequality and the Process of Attainment." American Sociological Review 42:965–978.

Spence, Michael A. 1974 Market Signaling. Cambridge, Mass.: Harvard University Press.

Spenner, Kenneth I., Luther B. Otto, and Vaughn R. A. Call 1982 Career Lines and Careers. Lexington, Mass.: Lexington Books.

Spilerman, Seymour 1977 "Careers, Labor Market Structure, and Socioeconomic Achievement." AmericanJournal of Sociology 83:551–593.

——1986 "Organizational Rules and the Features of Work Careers." Research in Social Stratification andMobility 5:41–102.

Stewman, Shelby 1986 "Demographic Models of Internal Labor Markets." Administrative Science Quarterly 31:212–247.

——, and S. L. Konda 1983 "Careers and Organizational Labor Markets: Demographic Models of Organizational Behavior." American Journal of Sociology 88:637–685.

Stewman, Shelby, and Kuang S. Yeh 1991 "Structural Pathways and Switching Mechanisms for Individual Careers." Research in Social Stratification and Mobility 10:133–168.

Stiglitz, Joseph E. 1973 "Approaches to the Economics of Discrimination." American Economic Review 63:287–295.

Stovel, Katherine, Michael Savage, and Peter Bearmen 1996 "Ascription and Achievement: Models of Career Systems at Lloyds Bank, 1890–1970." AmericanJournal of Sociology 102:358–399.

Talbert, Joan, and Christine E. Bose 1977 "Wage Attainment Processes: The Retail Clerk Case." AmericanJournal of Sociology 83:403–424.

Topel, Robert H. 1991 "Specific Capital, Mobility, and Wages: Wages Rise with Job Seniority." Journal ofPolitical Economy 99:145–176.

——, and Michael P. Ward 1992 "Job Mobility and the Careers of Young Men." Quarterly Journal of Economics 107:439–479.

Wegener, Bernd 1991 "Job Mobility and Social Ties: Social Resources, Prior Job, and Status Attainment." American Sociological Review 56:60–71.

White, Harrison C. 1970 Chains of Opportunity. Cambridge, Mass.: Harvard University Press.

Williamson, Oliver E. 1975 Markets and Hierarchies: Analysis and Antitrust Implications. New York: Free Press.

Winfield, Idee, Richard T. Campbell, Alan C. Kerckhoff, Diane D. Everett, and Jerry M. Trott 1989 "Career Processes in Great Britain and the United States." Social Forces 68:284–308.

Yamagata, Hisashi, Kuang S. Yeh, Shelby Stewman, and Hiroko Dodge 1997 "Sex Segregation and Glass Ceilings: A Comparative Statics Model of Women's Career Opportunities in the Federal Government over a Quarter Century." American Journal of Sociology 103:566–632.

Zhou, Xueguang, Nancy Brandon Tuma, and Phyllis Moen 1997 "Institutional Change and Job Shift Patterns in Urban China, 1949 to 1994." American Sociological Review 62:339–365.

Margaret Mooney Marini