The Credential Society: An Historical Sociology Of Education and Stratification

What effects has credentialism had on the overall shape of stratification?

In some areas, it appears to have had no effects at all. Rates of intergenerational social mobility, for example, have been constant from decade to decade as far back as the 1920s (Rogoff, 1953; Jackson and Crockett, 1964; Blau and Duncan, 1967: 110). Indeed, the rates have been virtually constant back to the early nineteenth century as far as we can tell from more limited information on economic and political elites (Taussig and Joslyn, 1932: 97; Warner and Abegglen, 1955: 37–68; Newcomer, 1955: 53; Bendix, 1956: 198–253; Mills, 1963: 110–139) and on particular communities (Thernstrom, 1964). The nineteenth century, in other words, was no golden age of opportunity, but neither was it an inferior situation to the modern one. The implication of this for the rise of educational stratification has gone unnoticed. The enormous expansion of education since the mid-nineteenth century has had no effects at all for increasing opportunities for social mobility. There is no alleged shift from “ascription” to “achievement.” On the face of it, there has been the same level of correlation between fathers’ and sons’ occupations with a large educational system, a moderate-sized one, or virtually no educational system at all.

In one respect, this is not surprising. If education does not provide occupational skills, one should not expect that making education more widely available should have any effects on occupational careers. Yet educational credentials have become the currency for employment, so that a greater abundance of even this artificial good might be expected to make a difference. But perhaps precisely because of this artificiality, the order of groups’ access to credentials has not changed their relative stratification. As education has become more available, the children of the higher social classes have increased their schooling in the same proportions as children of the lower social classes have increased theirs; hence the ratios of relative educational attainment by social classes has remained constant throughout the last 50 years (Spady, 1967) and probably before.1

A comparative finding from central Africa shows the basic pattern even more sharply: Kelley and Perlman (1971) calculated mobility rates for an older generation whose careers had started before the introduction of a formalized school system and the rates for a younger generation whose careers began afterwards. The mobility rate declined, and in approximately the proportion that education now played in careers. Cultural goods, it appears, are even more readily passed on from parents to children than are economic and political resources.2

THE EARLY TWENTIETH-CENTURY INCOME REVOLUTION

Nevertheless, there has been one major shift in the socioeconomic structure. The distribution of income and of property did change between approximately 1930 and 1950, and in a more egalitarian direction. We find this shift in Gini coefficients, describing the overall shape of income distribution: averaging .433 during 1920–1929 and .421 during 1949–1958 (Stack, 1976: Table 3.2);³ and in the declining share of the upper 1% of the property owners, from 36% of the wealth in 1929 to 23% in 1945, and fluctuating between 20% and 26% since then (Statistical Abstract, 1976: Table 693).

This shift coincides with the New Deal period of government and with a period of significant expansion of the educational system. Here, at least, one might find one egalitarian promise of educational liberalism paying off.

If we examine the twentieth-century income revolution, however, it appears that the redistribution has been to specific groups rather than across the board. Summary measures like Gini coefficients hide which portion of the income curve has shifted. Examination of the shares going to each decile of the income-receiving population shows that the bottom tenths have not improved their positions: the lowest 10%, in fact, was best off in 1910, when it received 3.4% of total income; but this group declined to 1.0% in 1937 and has fluctuated very slightly around that figure ever since. The ninth decile has fluctuated at 2–3% for 60 years, the eighth decile at 4–5%, the seventh decile at 6%, the sixth decile at 7–8%, the fifth decile at 8–9%, and the fourth decile at 10–11%. The major shift has been from the top decile to the second and third deciles. The top decile fell from its height of 39% of the income in 1929 to 29% in 1945 and has fluctuated around that figure ever since. The second decile was the major gainer, rising from 12% in 1929 to 16% in 1945, while the third decile rose from 10% in 1929 to 13% in 1945, both groups holding constant since then.4 And within the top decile, it appears that the holdings of the top few percent have declined in favor of those just below them.

The pattern looks like a shift from the upper class to the upper middle class. But is it correct to describe the second and third deciles or even the portion of the top decile below the top few percentages as “upper middle class” in terms of occupation? Pilcher (1976: 216–244), in analyzing British data, found that this was not so, and the American data presented in Table 7.1 show a similar pattern.

Table 7.1 Occupational Composition of Decile Shares, 1972^a

Source: Calculated from U.S. Bureau of the Census (1972: 136–138). Percentage totals vary from 100% due to rounding.

^a Full-time employed income earners, civilian labor force.

^b Total numbers in parentheses in thousands; other figures in this table represent percentages of each decile share.

The top 10% of American income earners in 1972 breaks down into the following: 34% managers and administrators, most of them salaried; 30% professional and technical, again most of them salaried; and 9% salesmen. I say salesmen explicitly because all the groups cited here are male; female workers in all categories make up only 3.4% of the top income decile. The managers, administrators, and professionals are readily enough described as upper middle class; given their high proportions of salaried employees, it appears that most of them work for the large corporations and government bureaucracies. Salesmen fit the label less obviously. Although most of these are not in retail sales, and hence are likely to be employed in these same corporate structures, they represent a more entrepreneurial component of it. The most striking finding, though, is that 18% of the top decile is working class or upper working class: 11% craftsmen, 3% foremen, and 4% operatives.

The entire upper half of the income distribution is heavily male, urban, and (surmising from other data, e.g., in Thurow, 1975: 6; and Pilcher, 1976), full-time employed. The lower half consists of female workers, agricultural workers, and—extrapolating from Pilcher’s data (1976: 216–221, 44)—those who are part-time employed, unemployed and underemployed, and those living on welfare payments. Thus the most crucial dividing lines turn out not to be occupational class divisions per se, but: (a) the segregation between higher-paying male occupations; and (b) admission into full-time and relatively secure jobs within the urban labor force. The big distinctions are essentially ones of “positional property.” Getting into and holding onto a male-designated position, especially within the wealthy corporations and within governmental bureaucracies, by itself is enough to ensure an above-average income.

More specific occupations show the same pattern. The highest paying of all occupations is that of the physician, especially self-employed ones (of which 68% of the latter group were in the top 3.5% of all income earners), and this is the occupation with the strongest monopolistic controls over their own numbers. Similarly, other self-employed professionals such as lawyers are also very likely to be in the most favored income levels (41% in the top 3.5% of income earners), while salaried professionals such as engineers (5.9%) and teachers (1.9%), with much less control over their own positions, are much less likely to be in this group (all figures from Thurow, 1975: Table 6). And at a lower occupational level, it is the craftsmen rather than the foremen who make up the significant working-class representation in the top income decile; it is the former, with their strong unions, amounting to guildlike control over the numbers in their own ranks, who achieve the high incomes rather than the latter, whose positions are subject to an external organizational control. And equally significant is the shaping of positions in the artificial distinction between male and female jobs, resulting in a form of positional property with major economic consequences.

What do these data imply about the twentieth-century income revolution? Lacking a comparable occupational breakdown for an earlier point in the century, we cannot be sure. But we do know that personal shares of the total incomes became less concentrated in the top 1–2% of the population, and have shifted from the top decile generally to the next two deciles. In terms of the current occupational composition of those deciles, the groups that have lost out relatively have been the big business executives, entrepreneurs, elite (especially self-employed) professionals, and the leading salesmen; those who have gained have been the salaried professionals, the craftsmen, and to a smaller degree, bureaucratic managers and administrators below the highest level, and further down, male clerical and manual workers generally in the urban and bureaucratic sectors of employment.5

The height of the income shift came during the New Deal period when government employment substantially took off. This was a period of liberal reform as well as of consciously Keynesian employment policies. Substantial bureaucratization occurred in that period, but not only within government: This is the period of the massive bureaucratization of the private corporations as well. This bureaucratization, I would suggest, was responsible for the income revolution. But notice: It was not a shift from the top to the bottom, despite the liberal and egalitarian rhetoric within the government at the time. The welfare system developed at that point has done nothing to improve the position of the lowest decile groups, unless one assumes that transfer payments keep them out of even direr poverty. Rather, the beneficiaries of the income revolution have been the bureaucratic middle class themselves and particularly favored craft-union enclaves of the upper working class.

At the same time, one notable bureaucratic group did not benefit from the expansion of their employment sector: women workers. For bureaucratic position shaping meant an improvement in the male-defined managerial and professional positions, but an actual decline in the relative income position of secretaries, typists, and other positions segregated into a sex-defined, subordinate enclave (Pilcher, 1976: 90).

How has the expansion of educational credentials figured in this income revolution? It has been by the use of educational credentials that the lucrative professions have closed their ranks and upgraded their salaries, and it has been in imitation of their methods that other occupations have “professionalized.” The expansion of bureaucracy, especially in the tertiary sector, has been accompanied by the most prominent emphasis on educational requirements of anywhere in the economy (as Table 2.2 indicates). Educational credentials in such organizations go along with spending large proportions of time—and money—on administration and control rather than on production; credentials have been a principal vehicle by which the sinecure sector has been built. The rhetoric of professionalization, of liberal reform, and of educational expansion has been all of a piece; these calls for reforms have gone together as a single mode of argumentation that underlay the transfer of some of the income flow from the upper to the middle class.⁶

Not all of the income shift may be directly explained in this manner. The upper working class—especially the craftsmen who make up a sizable fraction of the top income decile, as well as some of the operatives and foremen—has made gains especially by means of strong craft unions and by advantageous seniority pay systems within the wealthy corporations. But even these mechanisms bear a relation to the shift to a highly credentialized society. That shift has been building up since the late nineteenth century in response to the pressures of a multiethnic version of the economic class struggle and in response to the problems of overproduction and insufficient aggregate demand. Both issues came to a head simultaneously in the New Deal period when ethnic minorities achieved substantial concessions in the national political arena, including governmental legitimation of labor unions, and when the huge credentialized bureaucracies made their largest strides. The same underlying conditions that produced the credential system, with its most prominent effects on the middle class, also produced the powerful forms of labor organization that have benefited the upper working class. The upper working class does not rely particularly upon formal school credentials to control admission to its ranks (though many craft unions now use high school degree requirements); they use instead the mechanisms of lengthy apprenticeship programs, state licensing procedures, and in many unions, outright nepotism (Greer, 1959). But these (except for the last) are the types of occupational monopolization pioneered by the formally credentialed professions. The same background conditions have operated in each case, and the atmosphere of credentialism has both advertised the path for the upper working class to follow and made it easy to secure acceptance from employers and from the state for restrictive practices of their own educational (apprenticeship) and licensing requirements.

The income revolution of the first half of the twentieth century is all of a piece. It has been carried out under the rhetoric of liberal reform that has simultaneously built up the school system, the highly protected trade unions, massive government employment, and bureaucracy generally. None of these institutions has actually done what it officially claims to be doing. There has been no wholesale revamping of the class structure, nor a shift to widespread equality of opportunity or of distribution. The bureaucratic middle class has gained at the expense of the upper class, but not all bureaucratic workers have gained as a whole: Only the male-designated jobs have benefited, whereas female-designated jobs, although expanding equally rapidly, have not. The upper segment of the working class has achieved a comfortable and well-protected position, while the less-protected mass of the manual labor force has made only modest gains (in the case of urban males), and the lower section—female workers, rural labor, the chronically unemployed, and others living on transfer payments—have gained nothing at all (Pilcher, 1976: 57–59).

The income revolution represents the outcomes of class conflicts. It illustrates once again that these conflicts are not between two large, broadly defined classes, but between a much larger number of contenders. Within the white-collar and blue-collar sectors alike, some classes have won and others have lost.

This is in keeping with the hypothesized effects of a multiethnic and decentralized society. Indeed, the ethnic nature of these multiple class divisions has not disappeared. Particular enclaves continue to be heavily populated by particular ethnic groups. White Protestants tend to be disproportionately in the professions and in entrepreneurial businesses, Catholics heavily in government and bureaucracy generally.7 Particular craft unions have their own ethnic stamp and set their boundaries against ethnic outsiders (Greer, 1959). Blacks are disproportionately in lower-working-class positions and in specialized sectors of the white-collar world, especially government employment.⁸ Thus the lines of class struggle are doubly fragmented, with ethnic bonds continuing to subdivide the interest groups contending over economic spoils.

Given this fragmentation, there are severe restraints upon large-scale, Marxian-style confrontations between class blocs. The multiclass structure and the related credential system shape conflict into other channels. But it is still possible for general, society-wide crises to appear. The focus of such crises is likely to be the market for credentials.

THE LATE TWENTIETH-CENTURY CREDENTIAL CRISIS

As of the 1960s, the credential system went into a state of explicit crisis. The rising credential price of jobs had been going on for many decades, but at this point the change began to be consciously seen as inflationary. With near-universal high school completion and one-half the youth cohort attending college, these formerly valued goals lost much of their appeal. They no longer guaranteed a respectable job, at the high school level, or an elite one, at the college level. At the same time, there was tremendous pressure from subordinated ethnic groups, especially blacks and Latin Americans, for integration into the dominant educational and occupational institutions. The result has been a multifaceted crisis in confidence in the system and a variety of reactions and criticisms.

Initially, in the context of a militant civil rights movement for minority integration, mass student rebellions broke out within the universities. The rebellions took advantage of the state of growing delegitimation of credentials to demand revision of traditional curricular requirements. Such demands were usually put in the form of a shift to greater “relevance,” or toward the cultures of the ethnic minorities themselves. But in fact, the alternatives lacked substance; their principal appeal was negative, a reaction against the traditional requirements that were now recognized as purely procedural formalities of the process of gaining a credential. More recently, the idealistic rhetoric of curricular alternatives has been replaced by a manipulative cynicism. Students electing to remain within the system have adopted the goal of high grades, irrespective of content and by any means whatsoever, producing an inflation in college grades, while at the same time achievement levels have been steadily dropping.

Most of the “deschooling” talk was another version of Progressive education. It contained the same ideals, many of the same slogans, and it arose among school teachers at times in which schools have undergone crises of credential devaluation that destroy belief in their old functions. “Deschooling” is not so much a way of abolishing schools because they are proven useless, but of revising them internally in order to retain students. The influence on this of teachers’ interests in maintaining jobs for themselves is obvious; less obvious is the way in which the proposed curricular and authority changes—like those of Progressive education half a century earlier—help reestablish rapport between teachers and students, giving them a common rhetoric and de-emphasizing the authority relations that have been the focus of so much rebellion. In all this, “deschooling” simply carries the Progressive innovations one step further. Indeed, there are many proposals—generally made by those who consider themselves educational “liberals,” but perhaps even advocated by adherents of “deschooling” plans—for extending compulsory education requirements to age 18 or beyond.

Outside the schools, there has been a heightened criticism of education as insufficiently relevant to jobs and a revival of the technical–utilitarian rhetoric that was so prominent in the mid-nineteenth-century crisis of American education. As in the previous crisis, we see the reemergence of competition among different types of educational institutions. The conventional secondary and higher educational sequence has been challenged by a resurgence of commercial trade schools, business-operated training programs, and the crowding of professional and paraprofessional schools such as business schools.

Nevertheless, given the evidence that job skills of all sorts are actually acquired in the work situation rather than in a formal training institution, it is apparent that the technical training rhetoric is a response to the crisis of the credential market rather than a substantively significant change in educational content. New types of credentials are proposed because the public has lost confidence in the value of the old types. Hence the inflationary struggle for credentials seems to be building up in new directions. We now hear of the previously unprecedented use of Ph.D.s in accounting as credentials to acquire business jobs; we also hear of a massive expansion of internal and external credentialing in the business school sector. Skilled trades, contractors, and realtors continually establish more restrictive licensing programs, usually based on their own formal training requirements. Business corporations have established their own training programs.

The effect of these sorts of shifts is not to open up vocational career channels, but to increasingly monopolize and control technical jobs. For example, major automobile companies and chain stores are now monopolizing auto repair jobs for graduates of their own training programs. This not only replaces the current—and technically effective—pattern of mechanics acquiring skills through their own experience, but also results in company control of procedures (so-called “preventive maintenance” and automatic replacement rather than repair of parts) that tighten their financial grip over consumers. In general, the emerging pattern is to build up restrictive credentialing in new sectors where the traditional school credentials had not penetrated.

The shift to private credentialing is the counterpart of the existing crisis of the public, formal school credentialing system. Apart from conflicting rhetorical claims, the crisis has a material aspect. The public schools and the universities are facing difficulties from two directions. In the context of a general inflation affecting all prices in the material economy, schools have become expensive luxuries. In the public sector, taxpayers’ reactions are cutting the level of financial support, and seem likely to cut it still further in the future. In the private sector (especially in higher education), rising operating costs and declining enrollments are likely to force the closing of a considerable number of institutions.

The underlying source of pressure is the condition of credential inflation. Education is both more costly and promises less of a payoff for given levels of credentials than previously; hence students and those who pay their bills are relatively less willing to make the investment (Freeman, 1976: 9–50). Thus the historical peak in the proportion of the youth group attending both high school and college was reached in the early 1970s. Since 1972, high school completion rates actually fell (Statistical Abstract, 1976: Tables 230 and 231). Attendance rates at colleges and universities fell off for male students from 54% of the 18- to 19-year-old group, and 29% of the 20- to 24-year-old group in 1970, to 50% and 26% respectively for those age groups in 1975. Only an increased attendance rate for females—from 42% of the 18- to 19-year-olds and 15% of the 20- to 24-year-olds in 1950, to 44% and 19% respectively in 1975—offset this trend (Statistical Abstract, 1976: Table 191). The survival of the traditional educational credential, then, seems to be increasingly tied to the efforts of women to break out of their subordinate occupational position, while male employment may well be shifting into a separate set of trades- and business-controlled credentialing institutions.

The credential system, severely challenged at one point, seems to be reshaping itself in new directions. From a long-term viewpoint, it may be the case that the crisis of the 1960s and 1970s is a temporary one, and that credentialing sequences may extend indefinitely into the future. Once temporary periods of imbalance are past and the numbers of students return to a level at which inflation in the credential price of jobs is kept at an acceptably gradual rate, it may be that the finances of schools will stabilize. Possibly even at some point in the future, the system could begin to grow again.

Of course, the growth of the credential system has not occurred simply by its own dynamics, but in interaction with the struggle for economic position and with the level of economic productivity. The accumulation of highly efficient capital has steadily decreased the labor needs of the economy. The American industrial plant operates at only about half of full capacity, and a continual problem has been unemployment and underemployment and the related problem of maintaining aggregate consumer demand. It is here that we can see the economic importance of our educational system—not because of the technical skills that it might provide, but rather as a counterbalance to excess industrial capacity. This works in two ways: Both because education is a major area of government spending (7.5% of the GNP in 1975; Statistical Abstract, 1976: Table 183) and because it absorbs a considerable portion of the labor force as students. The 9.7 million college and university students alone would add 10% to the labor force. Adding any sizable portion of the 15.7 million high school students would, of course, increase the problem proportionately (Statistical Abstract, 1976: Tables 185 and 571). Cutting back on education to any considerable degree would tend to have disastrous economic consequences.

Thus the credential system of occupational placement is caught between opposing forces. On one side the system has become central to propping up an economy of excess productive capacity and surplus labor. On the other side the system has become extremely expensive and relatively unrewarding for many individual investors and for its political supporters. Although a rough balance continues to be possible, the potential for crisis exists in either direction. Too rapid growth in the credential market brings disillusionment and withdrawal of material investment in it; too little investment, on the other hand, feeds economic depression.

VARIETIES OF SINECURE POLITICS

A number of different political positions have been taken regarding the credential market. Few of them explicitly recognize the arbitrary currency nature of the educational system; most operate within some more familiar legitimating rhetoric. Nevertheless, they might be named as follows:

Credential capitalism is the traditional laissez faire attitude toward individual competition in the credential marketplace. It naively assumes that one should get as much education as possible in order to cash in on as much career advancement as possible. As individualistic advice, it ignores aggregate effects on the value of the credential currency, or simply proposes to override them by outcompeting others, at whatever level it takes.

Credential socialism is the program of government intervention to equalize the distribution of educational opportunities. Like traditional political socialism, it has failed to have much effect on the underlying system producing the inequalities and instead has added on to it a superstructure that has mainly been successful in redistributing some of the wealth to itself. Like political socialism (including that which has gone under such American labels as “liberalism”), credential socialism is a popular position among employees of the alleged redistribution system itself: in the one case government bureaucrats, in the other case, teachers and administrators.

These have been the traditionally dominant ideologies about American education. More recently, there has been an upsurge of demands by particular ethnic groups for more opportunities for themselves to acquire credentials. Sometimes this has been demanded with socialist–egalitarian rhetoric, sometimes in the name of ethnic cultural nationalism. In either case, its actual material goals are entrée on easier terms into the credential system, a demand entirely analogous to the traditional pattern in American politics of ethnic groups demanding a share of political patronage. This might simply be called ethnic-patrimonial credentialism or patronage-credentialism.

In reaction to such pressures from subordinated minorities, some members of dominant ethnic groups have created what might be called credential fascism: the effort to exclude particular minorities on principle. Earlier versions of racial ideologies have been updated for this purpose in the form of genetic arguments for inherited IQ differences. That this is a highly ideological stance is apparent from the evidence (cited in Chapters 1 and 2) that IQ predicts success only within the school system, and that the link between school success and occupational success is an entirely artificial one. Thus credential fascism attempts to shore up an arbitrary system of domination as the exclusive possession of its own ethnic group.

In a different direction, we find an upsurge of credential radicalism: the advocates of “free schools” or “deschooling” mentioned earlier. Their politics of liberating the schools by giving control to a communal group of students (and sometimes teachers) is rather like an extreme version of “communism in one country.” For local communal control of the credential-producing institution does nothing to affect the larger credential market within which they exist. If, as such schools usually attempt, their policy is to make credentials easy for students to attain, perhaps even automatically awarded (whether giving all “A” grades or automatically conferred degrees), they simply contribute to galloping credential inflation. In the larger context, such regimes tend to destroy their own economic base.

In my opinion, there are only two honest and realistic positions in regard to credentialism. One might be called credential Keynesianism. This would explicitly recognize that education creates an artificial credential currency and that this is economically useful to offset deficiencies of aggregate demand. Thus both investment in the school system and the credentialing of occupations would be encouraged, not in order to promote work efficiency or equal opportunity, but simply in the interest of keeping the economy running. The danger of this kind of policy, like that of ordinary economic Keynesianism, is inflation, but this can be taken as a side effect of the system to be accepted and managed by quantitative manipulation of whatever variables are under government control. In other words, one could decide to work openly within the sinecure system, to recognize the migration of leisure into the interior of jobs, and to deliberately set out to enhance such sinecures.

Such a policy, of course, has implicitly been in force for some time. Schools have often been supported for reasons of employment policy, and there has been a long-standing liberalization within the system that has affected the ethos of many jobs, making them more casual and less subject to ritual acknowledgments that nothing but productive work is going on. An open, unhypocritical recognition of the sinecure component in our society would certainly constitute a cultural revolution in public honesty and would make it easier to assess and control the effects of the factors affecting sinecures.

My own preference, however, is for an opposite policy: credential abolitionism.9 The prospects of continuing to expand the credential system indefinitely, to let job requirements inflate to the point where 4 years of college is needed for a manual laborer or 20 years of postdoctoral study is required for a technical profession, would be exceedingly alienating to all concerned. Moreover, it would not affect the rate of mobility, if past precedent is any indication, nor change the order of stratification among ethnic groups; it would simply reproduce their order at higher and higher levels of education. The alternative, to freeze the credential system at some point by allowing only given numbers of students to reach given levels, would be to freeze the existing system of stratification and to keep credential barriers in place that segregate the labor force into noncompeting sectors. Existing advantages of monopolization of lucrative job sectors would be maintained. Either way, letting the credential system expand, or holding it at its present level, would maintain existing stratification and would have culturally debilitating effects as well.

A serious change would depend upon abolishing the credential system. This does not mean abolishing the schools, but it does mean returning them to a situation where they must support themselves by their own intrinsic products rather than by the currency value of their degrees. Legally, this would mean abolishing compulsory school requirements and making formal credential requirements for employment illegal. Within the framework of civil rights legislation, such legal challenge to credential requirements has already established precedents (White and Francis, 1976). Since the evidence strongly shows that credentials do not provide work skills that cannot be acquired on the job, and that access to credentials is inherently biased toward particular groups, the case for discrimination is easy to make.

The main advantages of legal decredentialing would be two. It would improve the level of culture within those schools that continue to exist, and it would provide the opening wedge of a serious effort to overcome economic inequality.

The expansion of the credential-producing school system to mammoth proportions has made intellectual culture into a short-term obstacle for students to pass through on their way to their credentials. Thus it is hardly surprising that high school achievement test scores have fallen precipitously in the period since 1963 (Advisory Panel on the Scholastic Aptitude Test Score Decline, 1977), when school promotion became not only perfunctory but near universal. The pressures to get high college grades in order to enter graduate schools has had a similar effect upon substantive intellectual concerns. Humanistic culture has become very nearly the exclusive province of professional teachers of the humanities, and they themselves have turned their subjects from cultural ends to be created and enjoyed in themselves into the mere basis of a currency of numbers of publications on a professional vitae by which their careers in academic bureaucracies are made.

In science as well, it is likely that smaller would be better. Despite the self-serving rhetoric of university lobbyists, it is not at all necessary to have an extremely large component of university research to support the national economy or national security. Most practical inventions, in fact, are made in applied settings, and the basic science that they may draw upon tends to come from relatively small sectors of research carried out decades earlier (Sherwin and Isenson, 1967; Price, 1969). Moreover, the massive size of current American science is not proportionately efficient; the much smaller but proportionately more creative and better integrated organization of British science shows the superiority of a more elitist structure (see figures presented in Collins, 1975: 578). In the very large system, on the contrary, a kind of bureaucratization of scientific ideas takes place so that specialties are minute, mutually remote, and hard to integrate; information retrieval problems are serious; and the huge numbers of researchers enforces a lingua franca of the most rigidly operational and quantifiable concepts, to the denigration of more powerful theoretical ideas. The latter sort of problem especially afflicts the American social sciences, making them prematurely specialized around quantitative techniques while ignoring theoretical issues that give meaning to research. One might well claim that American social science—and natural science as well—has been living off of theoretical ideas that it has itself been unable to produce, either by importing the ideas from Europe or actually importing European theorists (as in the case of the refugees of the 1930s and 1940s, who have led American sciences in the last generation).

In sum, existing levels of mass credential production are unfavorable to American science and especially to humanistic culture. Further expansion of the credential system would be even more debilitating. Probably the initial building up of the educational system in the early twentieth century encouraged high-level scientific production, but the scale of operations has long passed the level of diminishing returns.

The other major argument for decredentialing is that only in this way can we move toward overcoming income inequality. Educational requirements have become a major basis of separating work into distinct positions and career lines, and hence in keeping labor markets fragmented. The gap between blue-collar and white-collar jobs constitutes a barrier to direct promotion in almost all organizations and is upheld by the disparity in educational requirements for each. These requirements are not necessary for the learning of work skills of these different sorts, but operate precisely to prevent members of one group from having the on-job opportunities to learn the skills of the other. Similarly, specialized “professional” and “technical” activities are reserved for separate labor pools by the same means. Thus it is not only ethnic and sexual segregation that produces “dual labor markets” (Edwards et al., 1975), but above all educational requirements that have become built into the definitions of “positions” themselves. Moreover, as direct ethnic and sexual discrimination becomes increasingly illegitimate and subject to legal challenge, educational discrimination becomes increasingly relied upon as a surrogate means of group domination.10

Hypothetically, income equality would occur if there were no barriers to movement among occupations. Hence positions that paid less than others, or whose work is especially dirty or unpleasant, would have to raise their wages to attract labor; positions that paid more than the average, or that had especially attractive work (such as planning and giving orders), would attract a surplus of applicants and could lower their wages. It has been proposed that educational barriers are the principal impediment to this income-equalizing tendency of a free labor market (Thomas, 1956). This is correct as far as it goes. But eliminating credential requirements would be only the opening wedge of the necessary restructuring. Under existing conditions, a large surplus of qualified applicants for managerial positions would not likely lower their salaries (Thurow, 1975); the number of positions would remain limited, and those few who did hold such positions would act to appropriate high salaries in any case.

This is clear from a realistic model of organizational behavior that sees power (i.e., “political labor”) rather than productivity as the key to income and advancement. Hence organizations themselves are the obstacle to a freely operating labor market. Similarly, although the elimination of credential requirements might increase competition within the professions, this would not necessarily eliminate stratification within any particular profession. Lawyers and engineers with links to wealthy organizational clients would still get large fees, and only the most decentralized and small-scale aspects of legal services would be subject to competitive price reductions.

Nevertheless, elimination of educational requirements for jobs would be a necessary step in any overall restructuring of the occupational world to produce greater income equality. The key would be to break down current forms of positional property. Managerial work could be brought back into responsiveness to labor market pressures if it did not constitute a distinctive, long-term “position” but was one activity to be shared among workers who carried out immediate production as well. By job rotation across the existing lines of authority and specialization, all types of work would become subject to a common labor pool and respond to the same wage conditions. This would mean that opportunities for learning various kinds of work on the job, including technical and managerial work, would have to be rotated or otherwise widely shared, possibly by rotating apprenticeship of “assistant-to” assignments. To do so would require eliminating current definitions of jobs as allegedly based upon prior, external preparation by specialized education.

Educational credentials, then, are not the only basis of barriers to a free labor market, but they are a crucial component of the system of barriers that would have to be removed. Secretaries, for example, are in a perfect situation for on-job learning of managerial skills. At present, the sex-caste barrier defines their positions as a separate enclave, however, so that virtually no secretary is ever promoted to take her boss’s position. Nevertheless, this is not only technically feasible, but once was the standard promotion line, before the late nineteenth century, when secretaries were males acting as apprentices for later administrative responsibility.

The current feminist movement has by and large ignored this form of positional discrimination; its emphasis has been on getting into the elite professions and managerial positions by following existing career channels—hence female college enrollments have gone up rapidly even as male enrollments have tended to decline. Desirable as this may be, it is an elitist reform that will have little effect on the economic prospects of the majority of women, especially in the vast clerical sector of bureaucracies. Indeed, applicants bringing more educational credentials to the professional and managerial labor markets will further raise, and in turn specialize, the credential requirements of those positions, making them even less accessible to promotion from within from the secretarial ghetto.

In this sense, a better long-term strategy for overcoming sexual stratification in employment would be to press for job reshaping instead of educational credentials, to explicitly substitute on-job apprenticeship as a means of managerial recruitment.

How might the credential restructuring of a strong profession such as medicine take place? As it stands, American medical training is attached at the end of a very long and expensive education that keeps the supply of physicians low and their incomes and social backgrounds very high. This formal education appears to have little real practical relevance; most actual training is done on the job in the most informal circumstances, through the few years of internship and residency. The existing medical structure is not only highly expensive, inefficient, and inegalitarian in terms of career access; but it is also tied to a system of job segregation in which the menial tasks are shunted off onto a separate medical hierarchy of women with the assistance of low-paid ethnic minorities in service jobs with no career possibilities.

It is likely that far greater quality and efficiency could be attained by eliminating the distinction between nurses and doctors and combining their career sequences with that of hospital orderlies. (No doubt this would offend the status concerns of doctors, but it would at least challenge them to take their altruistic claims seriously.) All medical careers would begin with a position as orderly, which would be transformed into the first stage of a possible apprenticeship for physicians. After a given number of years, successful candidates could leave for a few years of medical school (2 years seems sufficient background for most practitioners, and this could be done equally well at an undergraduate or postgraduate level, with the option being left open) and then return to the hospital for advanced apprenticeship training of the sort now given in internship and residency programs. The motivation of orderlies would be enhanced, and the implicit opportunities for apprenticeship-type training could simply be brought into the foreground. Advanced specialties could continue to be taught as they now are—through further on-job training; only medical researchers would be involved in lengthy schooling. The overall effect would certainly be less expensive and would provide better medical care from all personnel; there is no evidence to make one believe that the technical quality of treatment would suffer.11

Fundamental changes in the structure of inequality, then, and in the quality of modern culture, imply the abolition of credentialing. A thoroughgoing program of this sort would eliminate approximately half of the existing inequality. It would not directly touch the other source of inequality, the distribution of physical and financial capital. But socialist programs for overcoming this inequality attack only half the problem; Gini coefficients in socialist countries are approximately one-half of those in capitalist countries, averaging around .240 in the former and .440 in the latter (Stack, 1976). But even these lower Gini coefficients indicate a structure of occupational inequality than remains when capital is socialized. The socialist countries as well as the capitalist ones still need a second revolution.

I have argued that decredentialing of this order would have momentous, even revolutionary consequences. It could not be carried out without a thorough restructuring of organizational forms. This would be especially necessary because the currently existing credential system helps counteract the problem of excess capacity, and some other means of keeping up employment would have to be found. In the context of a thoroughgoing reform, such measures would not necessarily be difficult, by institutionalizing a shorter work week and/or by giving longer vacations. Under current conditions, of course, such measures are much more difficult to implement because of the income redistribution they imply; instead, spending on education has been a politically cheap way of practicing Keynesian economics in America.

The issue of credential stratification points us at the central feature of occupational stratification today: property in positions. To restructure these would be a more fundamental economic revolution than any we have yet seen. For that very reason, we cannot expect this reform to be easily made. It is possible that organizational and professional career hierarchies could be restructured piecemeal by local action. But local resistance would be hard to overcome without a widespread atmosphere of reform and a highly mobilized movement in this direction. It is far easier for allegedly liberal or even radical movements to continue the long-standing tradition of expanding access to the credential system. These efforts will only extend the inflationary nature of that system. In that direction, one can foresee that current issues around educational costs, discrimination, and integration will go on unresolved into the indefinite future.

To be realistic, one should bet on an expansion of current credentialism, even though this brings reformers no closer to their avowed goals than a donkey chasing a carrot held over its nose. This means, though, that crises of the class struggle continually threaten, not only within the material economy, but within the cultural economy as well. And using the educational system as a basis for an arbitrary currency of domination means that it suffers a continually increasing internal contradiction in the consciousness of its inhabitants. For all its claims to be raising the level of rationality of its students, education itself operates as part of a larger system that denigrates its own contents and ignores any insights it might provide into the nature of that system.

Hence although it would be unrealistic to expect a decredentialing revolution in the short run, it would be equally unrealistic to rule it out in the long run. Given the trend of credential expansion toward potentially absurd levels, it seems more than likely that credential abolitionism will come to the fore whenever any very sharp imbalance occurs between the size of the school population and the distributional processes of the material economy.

In effect, we are very much more like a tribal society than we like to admit. Despite our self-image of rational control, our institutions are no more reflectively chosen than the tribal initiation rites, secret societies, and implacable gods that our educational and occupational procedures resemble so much. Or to shift the analogy to more large-scale societies, we are subject to the same forces that transformed India over the centuries into a series of closed occupational castes, or that made medieval Europe a network of monopolistic guilds. Such societies undergo convulsions from forces beyond their control, as in the Reformation, which destroyed the religious currency upon which the medieval monopolies were legitimated. In the long run, unless we raise our own level of rational control over our institutions, we can expect that such forces will be waiting for us.

____________________

1. This consistency of mobility rates does not mean that particular ethnic groups have not changed their average occupational levels, or even, in some cases, their levels relative to each other. As of the 1960s, Jews held the highest occupational level of any ethnic group, and Catholics as a whole had passed white Protestants as a whole. The latter result, though, was due primarily to the fact that the white Protestant population is much more rural than the Catholic population, whereas occupational opportunities are better in the cities. The white Protestant population, in fact, seems to be split into two segments: an urban group that dominates the upper class and much of the upper middle class, and a rural-small town group of below-average occupational level. See Glenn and Hyland (1967); Duncan and Duncan (1968); Jackson et al. (1970); Featherman (1971).

2. Given this pattern, in fact, we should have expected that mobility rates would decline in the twentieth-century United States as education expanded. That rates have stayed constant may indicate some offsetting force in the noneducational realm, or it may indicate only that mobility rates are higher in a mixed horticultural society than in an industrial one.

3. Another source of data (Pilcher, 1976: 297) gives Gini coefficients from 1949 to 1969 as fluctuating from .383 to .406. Such differences in Gini values can be produced by calculations based on different categorizations of the same data (deciles, quintiles, etc.) or by different data sources.

4. Data from various sources for 1910 through 1969 are presented in Pilcher (1976: 57–59).

5. What if the occupational composition of the deciles has shifted since before the income revolution? Whatever their composition, we know that the relative proportions of money going to each decile (and within the top decile) have shifted, and we know what occupations are now the beneficiaries of this shift. It is conceivable that some occupations have both lost relative preponderance of incomes and have also moved out of the top income groups; the latter is not necessary for the known shift in decile shares (and hence in Gini coefficients) to have occurred. Conversely, some occupations may have both gained relative income and even moved up into a higher decile of the population; again, the latter is not necessary, unless the shift in relative absolute incomes is sufficiently great. Lacking direct information on the occupational composition of income deciles in the 1920s or earlier, we can only estimate that the amount of income shifting that has occurred seems compatible with either considerable stability in occupational composition of the decile shares or some change. I would estimate that the formerly highest-paying occupations are still the highest paying, but that their distance from the next highest-paying occupations has narrowed; and that the presence of the upper working class in the top two deciles is a new development.

6. Why hasn’t the creation of these new middle-class jobs brought about increased upward mobility? It has, in a sense. But our measures of mobility are constructed to capture “pure mobility,” independent of shifts in the occupational structure. “Pure mobility” rates have not changed because the credential system and bureaucratic employment are no more meritocratic with respect to family background than the organizational forms they replaced.

7. See sources in footnote 1 to this chapter.

8. In 1975, 56.5% of black workers were operatives, laborers, or service workers, as compared to 32.8% of whites (Statistical Abstract, 1976: Table 601). In the same year, 21% of government employees were minorities, as opposed to 16% of private employees (U.S. Civil Service Commission, 1976; U.S. Equal Employment Opportunity Commission, 1977). It seems especially likely that minority white-collar employees are disproportionately in government jobs.

9. The various forms of “credential politics” might be labeled more generally as factions in “sinecure politics.” This generates some interesting historical parallels. “Sinecure capitalism,” in fact, existed in the late Middle Ages in Europe, and in many other agrarian societies, where sinecures (prebends) could actually be bought and sold. The very mention of this is taboo for our society, though, and “sinecure capitalists” are doomed to operate under a misleading ideology. “Sinecure socialism,” on the other hand, is rather close to the goal of Marxists, who could share out equally the fruits of superproductive technology if they would propose to do this by distributing occupational positions instead of incomes. “Sinecure ethnic-patrimonialism” or “sinecure patronage” is familiar already: It is what ethnic patronage politics always was. “Sinecure fascism” would be the equivalent to returning to the medieval principle reserving nonwork positions for a hereditary aristocracy. “Sinecure radicalism” would be making work relations highly egalitarian; current movements for participatory democracy are quite close to this, but without yet recognizing that most of what is being shared is not work responsibilities but on-the-job leisure and on-the-job politicking. “Sinecure Keynesianism” means a radical extension of WPA-style make-work hiring; some short-lived revolutions such as the Paris Commune of 1871 carried this out fairly widely. “Sinecure abolitionism” has its religious precedents, notably the Protestant Reformation, which led to the elimination of monasteries and many other church prebends. In that case, the result was government confiscation of these properties, greatly enriching the new absolutist state—the real economic significance of the Reformation. Whether there would be analogous consequences of a modern-day “sinecure Reformation” is worth careful consideration.

10. Educational requirements were highest in those organizations making the strongest efforts to racially integrate employment, according to the 1967 San Francisco Bay Area Employer survey reported in Chapter 2.

11. To repeat the relevant points cited in Chapters 1 and 6: Medical school requirements are essentially arbitrary screening devices, as virtually no one ever flunks out of medical school; subsequent performance bears no relation to school grades; and the actual practical training of doctors is of the most casual sort—orderlies probably could acquire as much over the course of their work experience, especially if they are at all motivated (contrary to the enforced expectations of their current roles) to acquire it. Moreover, a reformed medical system ought to be more technically efficacious than the one that exists now. American medical care, despite the haze of national glorification in its own pronouncements and in the mass media, is a good deal less effective than that provided in the less professionally autonomous medical systems of Europe. This is illustrated by the higher infant mortality rate and lower longevity in the United States than in almost every European country, despite the superiority of the United States in GNP per capita of from 50% to 300% (Taylor and Hudson 1972: 253, 314).