1、THE ECONOMICS OF LABOR: AN ASSESSMENT OF RECENT RESEARCH The Challenge of Dual and Radical Theories of the Labor Market to Orthodox Theory By GLEN G. CAIN* Ideally, a paper that attempts to evalu- ate a challenge to established theory should accomplish the following objec- tives: 1. Present the theo
2、retical substance of the challenging theory, which is here called the dual and radical theories, and indicate how it differs from the existing theory, which I will refer to as neoclassical or orthodox theory; 2. Point out the empirical implications of the new theories and explain their dif- ferences
3、 with the existing theory; 3. Assess the empirical basis for dis- criminating between the competing hy- potheses or, if this is lacking, provide the theoretical or evidential counterarguments of the neoclassical response to the dual and radical challenge; and, finally, 4. Spell out the policy implic
4、ations of the competing theories. Meeting these objectives is an impos- sible task for a short paper. The space constraint is compounded by the fact that in my judgment the dual and radical theories are too varied, incomplete, and amorphous to present concisely. The strategy I adopt to describe thes
5、e theories in the longer paper consists of, first, dis- cussing their linkages to historical criti- cisms of classical and neoclassical theory; second, developing their empirical chal- lenges to orthodox theory. Only the second part is emphasized below. To establish the context of the chal- lenge, l
6、et us begin with the observation of Leo Rogin (p. 13) that: “new systems (of economic doctrines) first emerge in the guise of arguments in the context of social reform.“ The dual and radical (or D-R, for short) theories began to emerge in the 1960s when the movement for social re- form mainly involv
7、ed the war on poverty and the drive for full participation in the economy by minority groups, including women. Dissatisfaction with the pace and progress of reform in these areas and dis- satisfaction with the conventional analysis of the problems and their remedies have led to arguments within the
8、economics profession, especially on the part of the younger labor economists. Let me simply assert, without defending the proposition as I do in the longer paper, that the neoclassical school does dominate * Department of economics and Institute for Re- search on Poverty, University of Wisconsin. Th
9、is is an abbreviated version of a longer paper, which is available from the Institute for Research on Poverty, University of Wisconsin, Madison, Wisconsin 53706. The longer version contains a bibliography, which is omitted here. I am indebted to a number of people at the U.S. Depart- ment of Labor,
10、especially Fred Siskind, and at the Institute for Research on Poverty for their support. They are not responsible for errors, nor do they neces- sarily agree with the interpretations expressed in the report. I am grateful to Marc P. Freiman for his exten- sive assistance. 16 VOL. 65 NO. 2 RECENT LAB
11、OR ECONOMIC RESEARCH 17 the profession today, even within labor economics where, as I mention below, it was not dominant during the 1940s and 1950s. I will also assert that neoclassical theory means, in general: (1) the theory of the demand for labor contained in the marginal productivity theory and
12、 the as- sociated research on production functions; and (2) the theory of labor supply con- sisting mainly of models involving the labor/leisure choice and human capital investments. Theory, however, operates on at least two levels. As a framework for analysis, neoclassical theory implies a set of m
13、eth- ods and techniques-for example, margi- nal analysis of behavioral relations in which income and prices are key variables. Second, as a body of substantive be- havioral propositions, neoclassical theory implies a large number of qualitative pre- dictions and quantitative estimates of parameters
14、about behavioral relationships. The dual and radical critics attack both the methodology and the collection of pre- dictions and substantive hypotheses. T hose attacks inevitably spill over into policy disputes. I. A Classification of Issues Raised in the Dual and Radical Literature Out of this back
15、ground we can discern three types of issues raised ih the D-R literature: first, criticisms of the outcomes and processes taking place in the labor market; second, a rejection of the existing theories and methods that have purported to explain the operations of the labor market; and third, a call fo
16、r new and more radical policies. A. Empirical Generalizations It is useful to distinguish between two types of facts which contribute to con- troversy. T he first are those that indicate some sort of hardship or distress, such as high levels of unemployment and poverty. The facts about these problem
17、s may not be in dispute. The second type of descrip- tive empirical findings consists of unre- solved and inadequately treated problems, which may or may not concern matters of hardship. The empirical challenges raised by the D-R spokesmen usually concern the overlap empirical anomalies about pressi
18、ng social problems. 1 will list eight problem areas, some of which have sub- topics. 1. As noted above, the most important problem motivating the D-R economists was the persistence of poverty in conjunc- tion with the affluence of society and the scope of governmental efforts to eradicate poverty. I
19、his problem is too broad, how- ever, and we must turn to its more par- ticular aspects, listed below as items 2 through 7. 2. One such problem is the failure of education and training programs which, according to the D-R critics, demonstrate a failure of the human capital models and call into questi
20、on the conventional ac- ceptance of a causal connection between formal education and labor productivity. I comment on the alleged failure of social action programs below. Concerning the general education-productivity relation, my judgment is that its empirical support is shakiest regarding the inten
21、sive margin of educational investment-that is, the effect of more educational resources, hold- ing years of schooling constant. There is, I believe, ample empirical support for a payoff at the extensive margin that is, positive returns to more years of school- ing. The orthodox interpretation of thi
22、s empirical relation is, however, called into question by the next point. 3. Employers used educational and training indicators merely as screening devices either to discriminate on grounds other than productivity, or to serve as a proxy for potential productivity that is not directly related to wha
23、t was actually 18 AMERICAN ECONOMIC ASSOCIATION MAY 1975 learned in the educational institution. The latter purpose, it should be noted, redefines the investment nature of education as “information“; it does not deny that any investment takes place. However, a denial of either the real productivity
24、of education or of the potency of the informational con- tent of educational attainments does chal- lenge the orthodox assumptions about ra- tional behavior and/or about the degree of competitiveness in the economy. 4. T he income distribution remains as unequal now as it did twenty years ago, in sp
25、ite of the decrease in the variance of educational attainments in the adult population. 5. A collection of problems concern the large and persistent differentials in earn- ings between white and black males and between males and females. These chal- lenge neoclassical models of discrimination which
26、predict a withering away of the dif- ferential under conditions of competition. Several more specific empirical puzzles in connection with labor market discrimina- tion may be listed: the decline in black male labor-force participation rates rela- tive to white males in recent years; the near-consta
27、nt ratio of black-to-white earn- ings from 1950 to 1966 or so; the long- standing empirical relation showing a de- cline in the ratio of black-to-white male income with higher educational attain- ment; the flat wage-earnings profile of black males and all females relative to white males as revealed
28、by cross-section surveys; and, finally, the stagnant trend in earnings and occupational attainments of women relative to men during recent years. Associated with this trend is the slower growth in female educational at- tainment compared with men from 1940 to 1973, despite the sharp increase in labo
29、r-force participation of women during this period. 6. The roles of monopolies, unions, and other sources of protected labor markets have been a long-standing problem area. The D-R spokesmen imply that these are so pervasive as to make untenable the em- pirical work of orthodox labor economists, whic
30、h usually assumes competition. 7. Another area is the role of psychologi- cal variables measuring the attitudes and motivations of workers. Conventional economists have customarily viewed tastes for work as exogenous and as one of the (unexplored) causal variables explaining employment, earnings, an
31、d occupational achievements. The challenge of the D-R theories lies in their claim that tastes are endogenous and a result of ones labor market achievements. 8. Alienation among American workers is the final listed problem. If we assume as true the disputed contention that aliena- tion and dissatisf
32、action among U.S. work- ers are widespread and increasing, the challenge to orthodox theory may be stated as follows: Why has the market not responded to the workers tastes and pref- erences either by a redesign of jobs and up- grading of working conditions or simply by appropriate wage rate compens
33、ations? The foregoing list of empirical general- izations about outcomes and operations of the labor market covers, I believe, the main bill of particulars in the D-R in- dictment. Let us now comment briefly on their theories of how the labor market operates. B. Alternative Theories In the absence o
34、f any single, well-articu- lated theory in the dual and radical litera- ture, I will mention three versions that veer progressively further from orthodox theory. Perhaps closest to the orthodox is Lester C. Thurows job competition the- ory, which is contrasted with the orthodox wage competition theo
35、ry (Thurow and Thurow and R. E. B. Lucas). Its main ele- ments are: (1) the number and type of job VOL. 65 NO. 2 RECENT LABOR ECONOMIC RESEARCH 19 slots are technologically determined; (2) the workers skills (i.e., human capital) and wage offers are nearly irrelevant in deter- mining the number and
36、type of job posi- tions actually filled; (3) queues of workers for jobs offered at fixed wages constitute the supply of labor, and the employers as- sessment of the workers trainability and adaptability determines which workers are hired; and (4) fluctuations in macro policies will lead to changes i
37、n the demand for la- bor and thus to changes in the lengths of the queues. The theory emphasizes the within-firm (or internal) labor market as the locus of decisions about allocations, promotions, and on-the-job training all of which are relatively insulated from the external labor market. In many r
38、espects this theory is similar to the dual market theory. The dual labor market theory of Peter B. Doeringer and Michael J. Piore depicts a labor market divided into primary and secondary markets. The former contains the jobs in large or profitable firms and the unionized jobs. These jobs are higher
39、 paying, offer promotion possibilities, better working conditions, and greater stability. The secondary labor market contains the low-paying jobs which are held by workers who are discriminated against and who have unstable working patterns. The dis- cussion of this duality tends to be merely descri
40、ptive, and perhaps descriptive only of polar cases. Many of the theoretical ideas are similar to the job competition theory; namely, the demand-determined allocation of jobs and the downgrading of human capital characteristics as determi- nants of wage levels. In addition, con- siderable attention i
41、s given to the role of employer discrimination. Also, the work- ers attitudes, motivations, and work habits are viewed as mutually reinforcing determinants of a workers assignment (and confinement) to the primary or sec- ondary sector of the labor market. The third version of a labor market theory i
42、s referred to variously as radical, segmented, and stratified theory (see T)avid M. Gordon). It expresses a more explicit critique of capitalism, acknowl- edges its ties to Marxian dialectical analy- sis, and emphasizes class conflicts. The dual labor market idea is sometimes ex- pressed in terms of
43、 an analogy with an underdeveloped economy or even with a colony which is exploited by the primary economy. Radical theories are also similar to dual market theories in drawing upon sociological analysis of institutional change and power relations and upon psychologi- cal analyses of the determinati
44、on of work- ers and employers attitudes, preferences, and motivations. The expositions of these theories are stronger in their criticisms of neoclassical theory than they are in advancing a co- herent replacement. Criticism of classical and neoclassical theory has a long and, in many instances, a di
45、stinguished history, so casting the D-R writings in this mold can be complimentary. To the extent that the issues raised have appeared before (even though in different terms) and re- main unresolved, the challenge is all the more compelling. On the other hand, if the issues have been satisfactorily
46、answered before, the challenge is less compelling. In either case an historical perspective can be informative. Space limitations prohibit this develop- ment, but mention should be made of the similarities of the dual and segmented market theories to the Mill-Cairnes theory of noncompeting groups an
47、d to several ideas of the neoinstitutionalist group of labor economists who dominated their field in the 1940s and 1950s. Of this latter group, John Dunlop had written of inter- nal labor markets, and Clark Kerr had advanced the idea of labor market seg- mentation in his famous article about the “ba
48、lkanization“ of labor markets. 20 AMERICAN ECONOMIC ASSOCIATION MAY 1975 C. Policy Implications of the D-R Theories One set of policies advocated in the D-R literature concentrates on the labor market; another deals with the larger issues of power relationships and nonlabor market institutions in so
49、ciety. The first set of policies is most clearly distinguished by a focus on the demand side of the labor market. Specifically, public employment, wage subsidy, and antidiscrimination pro- grams are stressed. Intervention on the supply side of the market, particularly the human capital investment programs of education, training, and job search assis- tance, is deemphasized if not rejected. A second set of policies advocated by the radical economists is less specific. T hese include prelabor market condition- ing of the consciousness of people, perhaps calling for a reorganizatio
