While the contest-sponsored typology poses the normative constraints on organizational selection procedures, social science theories provide more detailed models and causal mechanisms. Human capital theory in economics and structural theories in sociology and economics (i.e., internal labor market theory) offer counterparts of the contest and sponsored models, respectively, in their emphasis on opportunity or structure. But these theories also go beyond these norms; they elaborate the underlying processes and provide specific hypotheses for empirical study of the issues. These theories provide the basis for a great deal of current research on labor markets and social mobility generally, and they are the basis for much of the research reported in this study.
Human Capital Theory
Human capital theory in economics proceeds from the same assumptions as the contest mobility norm. It posits that labor markets offer open opportunity and that employees' attainments are largely a function of how hard they work and the ability, education, and training they possess. This theory conceptualizes the latter factors-ability, education, training-as human capital. Like physical capital, human capital is increased by investments, and education and training are the primary investments individuals can make in their human capital (Becker 1964; Mincer 1974; Blaug 1976).
This theory suggests an explanation for the shape of career attainments. Investments in education or training are assumed to be costly. Consequently, individuals will concentrate their investments in the early part of their careers in order to maximize the amount of time they have to receive the benefits (amortize their investments). This explains the rapid increase in earnings which occurs in the early part of careers and the leveling off of earnings later in their careers.
Like other neoclassical economic theories, human capital theory assumes that labor markets are perfect markets; thus employers must pay employees what they are worth (their human capital) or else the employees will go to work for another employer. Economic theories do not consider the influence of job structures; structures merely represent labor market imperfections which, according to the theory, will tend to be eliminated by market processes.
Individuals are the active forces in the model. Individuals determine their productivity, and this is the basis for compensation. Indeed, human capital theory contends that individuals create their own productive capacity by the investments they make in themselves.
But investments in education or training do involve a trade-off. While individuals are attending school or receiving training at work, they generally earn less than those who are not receiving formal education or training. Individuals must choose whether they wish to receive greater earnings immediately or to sacrifice immediate earnings to invest in their training. Employees who choose to sacrifice immediate earnings for the benefits of greater training will generally attain much higher earnings in their later careers than will comparable employees who do not make these early sacrifices. In other words, human capital theory contends that great advancement opportunity exists for those who are willing to make the sacrifice.
In this model, the job world is assumed to act like a perfect market. It is not structured, and it does not impose restrictions on individuals' careers. Individuals alone are the main actors in this model. They choose whether and how much they will invest in themselves: how much effort they will expend and how much immediate earnings they will sacrifice to get better training. No barriers exist for able individuals if they choose to make the sacrifices and efforts to acquire the skills necessary for advancement.
Structural Theories
In contrast, structural theories posit a situation comparable to that described by the sponsored mobility norm. The most common version of structural theory, internal labor market theory, posits that organizations make important investments in individuals and these investments segment the work force into separate opportunity circumstances (Doeringer and Piore 1971; Spilerman 1977). Individuals in the primary labor market receive investments from their employing firms, and, as a result, receive advancement opportunities. Individuals in the secondary labor market do not receive such investments; and, even if they invest in themselves, the firm will not respond to these investments. These individuals have no advancement opportunities. In effect, employees in the primary labor market are sponsored by the firm, and those in secondary labor markets are deprived of sponsorship.
Like Caplow's (1954) description of the bureaucratic labor market, internal labor markets are characterized by fixed ports of entry, internal promotions as the main way of filling vacancies, and normatively approved, established procedures for hiring, firing, and promoting. Promotions, the central concern of this study, are primarily determined by on-the-job training. Doeringer and Piore (1971) describe what they call the "promotion unit" in which "each job in the progression line develops skills requisite for the succeeding job and draws upon the skills required in the job below it" (p. 21). On-the-job training is often more economical than formal instruction, for it takes advantage of the physical proximity of workers, the use of free time and idle equipment during a production break, the filling in by subordinates during a temporary absence of superiors, and the virtual absence of excess training since training is tailored to the learning capabilities and needs of each trainee.
Since internal labor market theory makes on-the-job training the primary mechanism determining career advancement, entry jobs have great effects on careers. After individuals are assigned to entry jobs, their subsequent careers are largely determined since entry jobs lead into predetermined progression systems. This view differs radically from human capital theory, for it assumes that individual attributes have no influence beyond determination of entry job assignments. En-try jobs lead to distinctive career lines within which future mobility is constrained (Doeringer and Piore 1971). After an individual enters an organization through a particular entry job, "the key decisions about his career follow. The occupational decisions that he made earlier will be of relatively little significance" (Ginsberg 1971, p. 17).
Compared with status attainment research, the real strength of internal labor market theory is that it describes a fine-grained structure of job mobility which status attainment research must dismiss as "random exogenous effects" (Spilerman 1977, p. 522). "What is sacrificed in such a (status attainment) formulation is a comprehension of the mechanism by which background variables operate, because the mechanisms are well defined only at the disaggregated level of the career line" (p. 586).
However, Spilerman suggests that internal labor market theory may not go far enough: "Despite the dualists' sensitivity to institutional factors, theirs is a bare-bones description, in which the considerable variety of labor market structures is reduced to the distinctions between primary and secondary jobs" (p. 582). Refinements of internal labor market theory to discuss multiple segments have not yet led to an adequate description of these segments. What is most lacking is an analysis of the multiple internal labor markets in the primary sector, that is, the multiple career lines of employees who stay with a single employer. This is a focus of the present research.
One other feature of internal labor market theory and other structural theories must also be noted: their assumption that career history is irrelevant. Internal labor market theory is generally interpreted to suggest that career history is irrelevant because present jobs embody the entire effect of past job history. If jobs have clear prerequisites for the kinds of previous training required to do them, or if previous training is less specialized or less valuable than the training one acquires on one's current job, then earlier training will have no independent effects on future career attainment. This would be particularly the case for very early jobs, which, though important for defining one's area of specialization, are probably not as highly specialized as the jobs that follow them.
