Part IV - Generating Stratification

Social Mobility

Number of theories have been proposed as an explanation of the trend of social mobility under the process of modernization and industrialization, for whether we are indeed moving toward a society in which we are freer to move and achieve based on what we can do rather than who we

are is one of the most central concern social scientists studying modernity are concerned with. Then, it follows that one theory that has been central in this debate is the most optimistic one - what might be broadly labeled the liberal theory - and it would be helpful to first categorize existing theories classified according to the relationship to this central conception and followingly, review some empirical researches that have attempted to answer the question.

Theories

1. The Liberal Theory

This theory is the most intuitive one associated with the industrialization and modernization, and the most optimistic as well. It posits that the rate of social mobility increases - further, the rate of the increase should increase over time as well - as society is freed from the constraint of highly stratified feudal structure based on ascriptive stratification. Because of the expansion of bureaucratic and managerial jobs, it is also postulated that upward mobility should be more common than the downward mobility.

2. Lipset and Zetterberg Theory

Modification of the above theory, altering the former in some key respects. It posits that the overall pattern of social mobility is much the same in industrial countries of the West. However, unlike the former theory it does not maintain that the rate of economic expansion and mobility rates are correlated. Rather, it posits the much more restricted "threshold effect", that once certain level of industrialization is achieved then the social mobility rate tends to be relative high.

3. FJH modification of the Lipset and Zetterberg theory

Paying heed to the empirical discovery that the pattern of social mobility is not equal across various industrial nations, it modifies the former theory further by maintaining that what is changed by industrialization is not the pattern of exchange among occupations, but rather the composition of the industries and occupations. This alternation in composition is then to lead to the changing mobility pattern.

4. Sorokin

In direct contrast to the Liberal theory, Sorokin maintains that the trend of social mobility in human history is one of trendless fluctuation. The seeming increase in mobility for him is merely a rather limited and specific historical phenomena applying to only some nations of the West at certain time. However, Sorokin does state that during the times of great social change, upheaval or transformation the mobility rate does tend to increase, and we may be living in such situations as of now.

5. Turner

Though this one is in entirely different vein, it has some unique claim that is worth mentioning briefly. The interesting insight in this theory on the relationship between education and mobility is that whereas other theories tend to focus on the effects of the former on the latter, Turner looks at the mode of the mobility pattern pre-existing in the society as system of values and norms operating to shape the system of education.

Empirical Researches

1. Both Featherman/Hauser study and Grusky/Hauser study makes use of the loglinear analysis, the former focusing on the mobility pattern within the U.S. while the latter focuses on those of 16 selected nations. The merit of the loglinear analysis is that it can uncover the effects of variables independent of the effects of marginal distribution, thus making it a particularly effective method to test for such hypothesis as the FJH revision that calls for the control of the industry composition. Briefly, the main findings are that: There tends to be great deal of immobility at the top and bottom of occupational hierarchy - people do not move out of farming or the nonmanual jobs. In contrast, the midddle of the hierarchy - namely the blue collar -is characterized by great amount of mobility, and both upward and downward mobility seems to be equally common generally. The latter study rejects the Lipset/Zetterberg theory in favor of the FJH revision. It extends the FJH revision in two respects: uniformity in mobility is not restricted to industrialized nations but may extend to all nations, and that there is a symmetry of exchanges between occupational strata in nearly all of the nations.

2. The Erikson and Goldthorpe study provides the serious challenge to the liberal theory, by demonstrating that there has not been a constant tendency to increase in mobility in the 20th century by using data 16 European nations. Erikson and Goldthorpe distinguishes between absolute and relative rate of mobility: in terms of the former, the trend has been one of trendless fluctuation, whereas in terms of the latter the mobility trend has been remarkably stable over time. Both of these findings, naturally, challenge the liberal theory seriously. This finding regarding the relative rate, then, does not rule out the FJH revision hypothesis however, as it may be that the basic similarity in relative rates once factors as industrial composition are controlled for, assumed by the FJH revision, is the source of the temporal stability in mobility trend.

Status and Income Attainment

Theories regarding this issue of status and income attainment - just exactly what factors determine the occupational status, and particularly the income, of people - can broadly be classified according to differential stress they place on two of the key aspects: characteristics of individuals themselves, or the structural factors in the economy and society generally beyond the control of individual workers. In practice, most of the theories have focused almost exclusively on either one or the other of these two crucial factors of income and occupational status determinant, a fact deeply lamented by Granovetter who makes the subsequent assertion that coherent theoretical integration of these two factors along with the even more neglected factor - how the matching processes between employers and employees occur - is needed. But, anyway, let's look at and classify each of the selections from the perspective of which factors they give precedence to as determinants of one's income and status attainment.

The classic article by Blau and Duncan, "The Process of Stratification", makes use of the then newly introduced technique of the path analysis to determine which factors contribute exactly to the occupational status attainment process quantitatively. Four variables are identified as particularly crucial in determining the occupational status: Father's educational attainment, father's occupational status, respondent's educational attainment, status of respondent's first job - and the dependent variable is the status of respondent's current occupational status. While Blau and Duncan makes also a potentially helpful assertion such as the warning for the misinterpretation of the multicollinearity as standing for the strong association of variables, the fairly simple theoretical implication of this model is immediately clear. It assumes occupational status attainment process is almost exclusively a matter determined by one's individual characteristics. In this sense, as Granovetter has pointed out, the sociological account provided by Blau and Duncan is strangely similar to the human capital theory in the field of economics, also focusing almost exclusively on the variables of individuals' characteristics as the determinant of the income.

Then, the social psychological model of Sewell, Haller and Portes is basically the extension of this Blau and Duncan model. The main insight here is that since the income and occupational status attainment process is a matter of behavior, the outcome of behavioral actions is not so much determined by the variables as father's status and educational characteristics directly but rather is mediated by psychological factors as motivation, cognition and the significant other's influence. These psychological variables are what really determine the behavior of individuals directly, and the authors make an additional claim that this approach has the added merit of specifying points in the process of overall status and income attainment that can be altered effectively for the people to achieve the outcome they most desire. So then, their model takes the form of having added mediating variables to the simpler Blau and Duncan model. The results of the quantitative analysis reveals the particularly important effects the significant other's influences have on income and status attainment process. But, of course, the basic framework of almost exclusive focus on the individual level variables is unchanged.

In direct contrast, many social scientists, typically of more liberal disposition, have emphasized the role of the structural aspects at the societal level beyond the control of individuals as largely determining the income and occupational status attainment processes. For instance, the writing by Jencks et al. is one of the article pointing out the weakness of the individual level approach by pointing out empirically contradictory evidence in the U.S. For this, they first note that policies of "fighting poverty" in the U.S. have traditionally been based on the assumptions of individual level approaches - that is, these policies tended to posit that eliminating poverty is largely a matter of helping disadvantaged children attain needed individual characteristics (i.e., education) to succeed in occupational attainment process, and that the primary reason that poverty exists is the lack of cognitive skills needed in the labor market. Consequently, the result is the policy output that the most effective means for fighting poverty is the educational reform. The authors maintain, drawing upon empirical evidence, that all three of these propositions the U.S. policies are based on are false. They assert that factors as "lack of cognitive skills" cannot account for the existence of all poverty, and that educational reforms have not been that instrumental in helping people rise out of poverty at all. Consequently, their assertion is that in order to eliminate poverty what we need is not the policy to help people improve their individual assets that may or may not improve their occupational prospects but rather a policy to reduce the impact of income inequality directly.

While the previous article did not identify what are the exact structural factors that maintain the income and status inequality, it at least did set the tone by negating the idea that such factors can largely be reduced to individual characteristics of each workers. Other articles in this group have tried to identify specific structural mechanisms within the capitalistic market system that maintains the inequality. Piore's "The Dual Labor Market" theory is one of the pioneer in this type of work. The basic insight of this theory is rather simple: there are good jobs and bad jobs, for the market itself is divided into two sections so that the "primary" section has a structural advantage over the "secondary" section to provide for all the goodies as higher income and better security. Further important claim Piore makes is that this division of labor market is perpetuated by the fact that the employers and those working in the primary sector have economic interests in perpetuating this division. On the other hand, as Granovetter has pointed out, the problem with this formulation is also quite simple - it is unclear as to who gets positioned into which section in the first place. The article of "An Outline of a Theory of the Matching of Persons to Jobs" by Sorensen and Kalleberg is in similar vein as that of Piore's, but this latter one provides for somewhat more sophisticated level of theorizing as it is clearer on the internal mechanism of income and status attainment processes within two different segements of the market. In this formulation, "wage competition" system is roughly comparable to the secondary sector as identified by Piore, while the "vacancy competition" system is comparable to that of the primary sector. Briefly, labor market operating under the wage competition principle is the one in which most of the neoclassical arguments of individual assets determining income level holds true. The needed assumption for this system to work is that employers have perfect control over employees as to who and when to hire and fire personnel, and over the wages as well. However, when employees have the control over the terms of employment, then the system tends to operate under the vacancy competition principle. Now, under this system, the employers cannot link marginal productivity to the wage rate, as they lack complete control over the terms of employment and wages. This is supposed to have important ramifications regarding who should be hired, and the process of income and status distribution within firms. In terms of who to hire, since employees cannot be easily fired and so employers become more concerned about the personal attributes of the employees, various ascriptive characteristics tend to come into focus in making a decision as to who should be hired. Further, as employers lack good control over the wages as they relate to production, the wage determination process tends to be dependent on various institutionalized factors as unionization and collective bargaining. Wage structure then comes to be used as the means for the control of internal promotion system, so that wages come to be increasingly determined by jobs, not by individual incumbents who fill those jobs. It is by now clear that the conditions of "vacancy competition" is largely the characteristics that tendto be assoiciated with the "primary sector" as identifed by Piore. In this type of system, access to jobs is the primary barrier in occupational status and income attainment, not the lack of needed individual assets. Thus, when this system is the primary mechanism governing labor market, no wonder worker training programs have rather limited success as training programs themselves cannot create jobs - and that ascriptive characteristics tend to influence hiring decision. So, overall, this article has some merit of sophistication over the previous two article, in the sense that it identifies specific mechanisms within the labor market operating to render educational reform and training programs ineffective, and how method of hiring and promotion occurs differently in different markets.