Jump to ContentJump to Main Navigation
Jacob MincerA Founding Father of Modern Labor Economics$

Pedro N. Teixeira

Print publication date: 2007

Print ISBN-13: 9780199211319

Published to Oxford Scholarship Online: May 2007

DOI: 10.1093/acprof:oso/9780199211319.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 14 December 2017

The Influence of Jacob Mincer on Modern Labor Economics

The Influence of Jacob Mincer on Modern Labor Economics

Chapter:
(p.132) 7 The Influence of Jacob Mincer on Modern Labor Economics
Source:
Jacob Mincer
Author(s):

Jacob Mincer

Publisher:
Oxford University Press
DOI:10.1093/acprof:oso/9780199211319.003.0007

Abstract and Keywords

This chapter appraises the role played by Jacob Mincer in labor economics during his long and prolific career. Mincer is certainly a good example of the capacity that many academics have to attract the attention of those around them and to interest them in pursuing similar lines of research. His ability to lure students and colleagues to human capital research was extremely relevant for the development of the human capital research program, especially in the early crucial years. Moreover, his persistent and methodic interest in the analysis of the implications of human capital for labor market analysis had a lasting influence on several generations of labor economists, and contributed to the strengthening of the position human capital came to occupy in labor economics during the last forty years. This chapter analyzes the impact Mincer had through his publications, as a teacher, and especially as a mentor of several emerging leading labor economists. The impact this relationship with his students had on the development of the human capital research agenda is explored.

Keywords:   human capital, inequality, Columbia, Chicago, Gary Becker, labor

Despite their intrinsic capacities, all prominent scientists, especially those pioneering in a field or a topic, know that in order for a certain topic of research to prosper and endure it needs to develop a community of researchers. Hence, researchers, especially those working in new areas, need to go through a process of identification and self‐awareness in order to create a sociologically identified community that shares certain tenets providing methodological and theoretical coherence to the group (cf. Backhouse 2000). The creation of this social circle (see Crane 1972: 13–18) that can intellectually engage in this scientific quest is necessary and expresses the need this community has to organize itself and to create instruments of communication.

Mincer is certainly a good example of the capacity that many academics have to attract the attention of those around them and to interest them in pursuing similar lines of research. His ability to lure students and colleagues to human capital research was extremely relevant for the development of the human capital research program, especially in the early crucial years. Moreover, his persistent and methodic interest in the analysis of the implications of human capital for labor market analysis had a lasting influence on several generations of labor economists and contributed to the strengthening of the position human capital came to occupy in labor economics during the last forty years.

The aim of this chapter is to appraise the role played by Jacob Mincer in labor economics during his long and prolific career. A general picture of the influence of Mincer's work is provided through (p.133) a brief citation analysis followed by a more detailed analysis of the role Mincer played as a teacher and especially as a mentor of several emerging leading labor economists, and the impact he had on the training of several generations of labor economists. Then, the impact that this relationship with his students had on the development of the human capital research agenda will be examined. Finally, the discussions promoted by some of Mincer's main contributions to economics and his provisional legacy emerging from those debates will be assessed.

7.1. Citation Analysis

One of the possible instruments to analyze the impact of a particular author is through the use of citation analysis. In applying this analysis to Mincer's economic research, namely in labor economics, it provides a general impression of the quantitative impact of Mincer's work. Data have been collected for the period between 1972, the year the Social Science Citation Index became available, and 1991, the year of Mincer's retirement (see Table 7.1). Although Mincer continued to work intensively after his retirement and his works continued to be cited in large numbers, the period selected covers the most significant time of his career and the years of consolidation of the human capital research program.

Table 7.1 indicates that Mincer's work has been very influential as measured by the number of citations. It is also noticeable that the number of citations has remained stable throughout the years, even when referring to research published many years earlier. This is normally regarded as a clear sign of influential research. During the period analyzed, Mincer's work received an average of more than 200 citations and each piece of research has been cited on average more than ten times per year. This is highly significant due to his prolific career. It is also noticeable that some of his work continues to be cited more than thirty years after its publication.

Mincer's most highly cited publication is by far his 1974 book Schooling, Experience, and Earnings. The number of citations for this book is remarkable, even more so as this is largely an empirical piece of economic research. It is also extremely significant that almost twenty years later the book continued to be cited, even more than (p.134)

Table 7.1. Citations of the main publications of Jacob Mincer, 1972–91

Total pY

Avg pY

Thesis 1957

JPE 1958

JPE–S 1962

AspLabEc 1962

MeasEc 1963

ProspUne 1966

SchExpEr 1974

JPE 1974

JPES 1976

JPE 1978

JHR 1978

Total

2236

203.3

3

90

132

284

148

139

815

353

111

103

58

Average

111.8

11.2

0.2

4.5

6.6

14.2

7.4

7.0

45.3

20.8

6.9

7.9

4.8

1972

25

4.2

0

3

6

4

6

6

1973

55

9.2

1

3

11

20

12

8

1974

69

9.9

0

6

16

15

7

11

14

1975

77

9.6

0

5

14

18

9

6

19

6

1976

132

14.7

0

9

13

18

9

15

45

20

3

1977

128

14.2

1

6

7

20

14

10

46

18

6

1978

107

11.9

0

9

6

12

9

9

38

21

3

1979

134

13.4

0

1

4

19

9

11

52

19

17

2

1980

145

13.2

0

4

6

23

7

11

57

18

11

7

1

1981

147

13.4

0

12

11

20

8

5

47

29

5

5

5

1982

158

14.4

0

4

6

11

7

9

56

29

18

9

9

1983

131

11.9

0

7

5

14

10

5

46

25

8

7

4

1984

119

10.8

0

4

5

10

4

5

49

25

4

5

8

1985

135

12.3

0

2

5

22

5

7

55

22

6

8

3

1986

102

9.3

0

2

2

9

4

4

42

20

2

12

5

1987

124

11.3

1

2

2

11

6

5

51

25

8

8

5

1988

116

10.5

0

5

1

10

6

1

47

19

6

14

7

1989

115

10.5

0

4

3

11

6

4

50

15

3

14

5

1990

102

9.3

0

1

2

7

6

2

49

22

3

6

4

1991

115

10.5

0

1

7

10

4

5

52

20

8

6

2

Source: SSCI–ISI.

(p.135) in the years after its publication. These are clear indications of the enduring relevance of the book for labor economics.

However, the impact of Mincer's research cannot be fully appreciated through this simple quantitative analysis. A large part of his lasting influence has been through his role as a mentor for several generations of labor economists. We now turn our attention to that role in order to obtain a fuller picture of his impact in shaping contemporary labor economics.

7.2. Mincer's Influence as a Teacher

The obvious networks for attracting other researchers to work on the same topic are graduate teaching and research training. In the case of human capital, the teaching hubs were certainly Chicago and, later, Columbia. The early examples of teaching of human capital issues come arguably from T. W. Schultz's courses at Chicago in the second half of the fifties in growth topics and the economics of agriculture. Also in Chicago was Gregg Lewis (though teaching on labor topics), who had a close supervisory role in the doctoral work of people who played an important role in the development of human capital such as Gary Becker, Marvin Kosters, Morton Zeman, Walter Oi, Glen Cain, Sherwin Rosen, and Giora Hanoch. Becker moved to Columbia in 1957, making it a major center—with Chicago—of research on human capital at that time.

The visibility of human capital research at Columbia was intensified when Mincer went back to teach there in the early sixties. Throughout the sixties, Mincer and Becker joined forces to establish a community of human capital researchers based at the Labor Workshop and the NBER. During that period they attracted many of their students at Columbia to human capital topics, and guided them through the early phase of their research careers by supervising their Ph.D.s or making them their research assistants at the NBER which, during the early years of the human capital research program, had its offices located in mid‐town Manhattan. The impact of this community was emphasized by the fact that several of them eventually became leading researchers in human capital, and in economics.

The recollections of former students are quite unanimous about the impact that both Mincer and Becker had on them in their first (p.136) year of graduate education. The students normally had Becker for Microeconomic Theory and Mincer for Statistics (which included some econometrics) and, from the mid‐sixties onward, Labor Economics. By the late sixties, Mincer concentrated his teaching activity on labor economics, since he felt increasingly attached to that field. Moreover, he regarded econometrics as more relevant than statistics for empirical economics. Mincer was praised for his clear, articulate, informed teaching. He taught the students basic elements about statistics especially its usefulness for economic analysis.

Mincer's approach to teaching labor economics was both seductive and somehow elusive to students. His deep commitment to price theory led him to tell students who enquired about the course that labor economics was simple, and basically consisted of supply and demand analysis. Many students only later understood that this apparent simplicity was not trivial. Underlying his effort to identify the essential traits of the labor market, there was a number of fundamental methodological and theoretical issues needing to be understood.1 Mincer's aim at making labor analysis simple and concise was both sophisticated and demanding. The fact that so many of his students persisted in this effort is a tribute to his scientific and pedagogical skills.

The appeal that Mincer had for so many students is even more striking due to his well‐known perfectionism and rigor which he applied firstly to his own work than to that of those who worked closely with him. As one of his former students recalled, ‘He was reluctant to let a student finish until he was convinced no stone was left unturned to verify a thesis’ assertions’ (Polachek 2006b: 30). The students’ task was even more complicated by the fact that Mincer did not seem to be terribly impressed by the use of sophisticated mathematical and statistical techniques. Instead, he expected them ‘to apply a sound specification using a number of data‐sets so one could assess robustness’ (30). Thus, his sense of rigor required from students a solid grasp of theory and a rigorous testing through robust empirical work.2

The influence that Becker and Mincer had on their students meant that many of those who had gone to Columbia aiming to work with other professors in the department would change their minds and turn to them at the time of choosing their dissertation topic. Despite cooperating closely Becker and Mincer always had very different styles of research, which in this case proved to be effectively (p.137) complementary. Whereas Becker explored matters on a theoretical basis, Mincer took a more empirically oriented approach, showing his students how human capital research could be applied in order to explain aspects of the actual labor market.

The success of human capital in creating a network of researchers was certainly linked to its ability in attracting graduate trainees and providing the area with new and able researchers, notably in the concerns of labor research. Several studies confirm that the years of doctoral training, and the period immediately after, are crucial in setting the pattern of research productivity of new (successful) researchers (see for instance Clemente 1973). The same studies also emphasize the importance of learning‐by‐doing in terms of scientific activity, both in graduate training and in early career research (postdoctoral fellowships, initial appointments, etc.).

In the postwar decades, doctoral training in economics was still very much concentrated in a handful of departments, which contributed to their dominance in research and publication activity. In the specific case of labor economists this pattern of concentration in a handful of departments was also very noticeable (Table 7.2). The field was dominated by four departments (Wisconsin, Columbia, Harvard, and Chicago), which remained the backbone of the field, though part of this dominance was eroded as the number of departments producing labor economists increased significantly during the postwar period. These four departments were joined by the increasing vitality of the MIT (since the fifties), the University of California (especially Berkeley), New York, and Cornell (during the fifties). Some departments, such as the MIT, Penn State, Stanford, Princeton, and Yale, that had traditionally overlooked labor research started to change their attitude from the fifties onward, producing in some cases an important number of graduates.

The content of the dissertations was also changing rapidly. Much more attention was given to training and manpower issues, supply and migration of labor, wage differentials, and employment/unemployment matters. Other aspects that used to be prominent, such as unionism, collective bargaining, industrial relations, and analysis of labor legislation retained some importance but lost ground in their relative importance for new generations of labor researchers. The approach was also different, with the increasing adoption of econometric techniques and the shift away from the (p.138)

Table 7.2.

Ph.D.s awarded by American universities in labor economics, 1940–68

Ph.D.s

Total

CHG

COL

HVD

WIS

MIT

CAL

CRN

NY

PRIN

PEN

NC

YAL

MCH

STA

Other

Total

917

67

86

72

102

51

55

43

36

21

28

16

12

24

10

280

1940–9

222

24

41

23

32

3

10

6

5

4

3

4

0

13

1

48

1950–9

319

30

26

25

41

19

19

27

14

10

11

5

8

5

0

77

1960–8

376

13

19

24

29

29

26

10

17

7

14

7

5

6

10

155

Source: American Economic Association (various years).

(p.139) institutional study of specific markets (industry, region, and firm studies).

In terms of Ph.D.s related to human capital issues, until the early sixties human capital research was almost exclusively confined to Chicago (the exception was Mincer's 1957 Columbia dissertation). These Ph.D. students did their research work on occupational differentials (Morton Zeman and Becker in 1955, Mincer in 1957, Robert Polkinghorn in 1958, Paul Keat in 1959, and Henry Sanborn in 1960) and on the role of education in economic growth (Zvi Griliches in 1957). By the mid‐sixties the situation started to change with the first Ph.D.s coming from other departments (Harvard and Washington in 1962; Yale and Virginia in 1963). Columbia became more evident with the work of students supervised by Becker and Mincer.

One of the major vehicles of Mincer's influence on labor research has been through the supervision of many doctoral dissertations. Although we do not have complete data on all the doctoral students whose dissertations Mincer acted as first or second sponsor, Table 7.3 provides an approximate idea of that role.

Accordingly, among his former students there are several subsequent prominent human capital and labor researchers. These include people such as Barry Chiswick, Dave O'Neill, Reuben Gronau, Robert Michael, Michael Grossman, June O'Neill, Arleen Leibowitz, George Borjas, Masanori Hahimoto, and Solomon Polachek. In their dissertations they developed several building blocks for the human capital research program and for contemporary labor economics. Many of Mincer's former students went on to become prominent academics and leading experts in their fields of research, and several of them confirmed the influence that Mincer's work and personality had not only on their research interests but also on their method of working.3

Mincer's role as a mentor to new generations of labor economists contributed to the fact that by the late sixties, doctoral research on human capital presented a significant expansion in terms of the number of graduates and departments from where they graduated. However, in order for this research potential to become effective and provide vitality to the area, it was essential that those newly trained professionals joined the academic profession and became successful academics. In the case of human capital there was not only high retention of these young researchers in the academic (p.140)

Table 7.3.

A selective list of doctoral students in economics supervised by Jacob Mincer at Columbia University, 1964–81

1964

Morris Silver—Birth, Marriages, and Business Cycles in the United States

Gonan Smith—Occupational Pay Differentials for Military Technicians

1966

Dave O'Neill—Occupational Incidence of Unemployment

Robert Rice—An Analysis of Private Wage Supplements

William Landes—The Effect of State Fair Employment Legislation on the Economic Position of Nonwhite Males

1967

Barry Chiswick—Human Capital and Regional Inequality

Reuben Gronau—The Effect of Traveling Time on the Demand for Transportation

1969

Robert Michael—The Effect of Education on Efficiency in Consumption

1970

Michael Grossman—The Demand for Health: A Theoretical and Empirical Investigation

June O'Neill—The Effect of Income and Education on Interregional Migration

Albert Zucker—Some Aspects of the Economic Effects of Minimum Wage Legislation: 1947–66

Beth Niemi—Sex Differentials in Unemployment in the US and Canada 1947–66

1971

Linda Edwards—Investment in Human Capital: A Study of the Teenage Demand for Schooling in the USA

Masanori Hashimoto—Factors Affecting State Differences in Unemployment

Elizabeth F. Durbin—Family Instability, Labour Supply, and the Incidence of Aid to Families With Dependent Children

1972

Carmel Chiswick—Income Inequality in LDC's

Arleen Leibowitz—Women's Allocation of Time to Market and Nonmarket Activities: Differences by Education

Cynthia B. Lloyd—The Effects of Child Subsidies on Fertility: An International Study

Fredericka P. Santos—Some Economic Determinants of Marital Status

1973

Solomon Polachek—Work Experience and the Difference Between Male and Female Wages

1975

George Borjas—Job Investment, Labor Mobility and the Earnings of Older Men

1976

Cordelia Reimers—Factors Affecting the Retirement Decisions of American Men

1978

Linda Leighton—Unemployment over the Work History: Structure, Determinants, and Consequences

1980

William Alpert—The Economic Determinants of Private Wage Supplements

Margaret L. Hashimoto—The Effect of Industrial Composition Changes of the Relative Demand for Skilled Labor

Gregory de Freitas—The Earnings of Immigrants in the American Labor Market

1981

Nancy A. Garvey—Job Investment, Actual and Expected Labor Supply, and the Earnings of Young Women

(p.141) profession, but also high productivity. Moreover, many in this generation managed to get positions in some of the top departments in economics in the USA and abroad, as well as in major research institutions, helping to enhance the dissemination of research in this field.

The development of a field of research also requires institutional support. This is very important for the support and training of new research cohorts. In the case of human capital, the NBER was the research hub notably for large projects. In fact, the attraction of working at the NBER was one of the main reasons for Becker's decision to leave Chicago and move to Columbia in late 1957. This growing engagement of the NBER with human capital research was also a major factor in bringing Mincer to the institution in the early sixties (1961). He came primarily to develop his work on the role of on‐the‐job training in labor market earnings, and eventually in 1966 took charge of labor research, especially on labor participation and unemployment and its relation with skills differentials and investment in human capital. Many other young researchers came at that time to the NBER to start projects related to human capital. These included people like Albert Fishlow (economic history), Victor Fuchs (differentials in hourly earnings and eventually health economics), Paul Taubman and Terence Wales (returns to higher education and the potential screening effects), and Thomas Juster (who coordinated a series of studies on the relation between the stock of human capital and certain types of economic and social behavior).

The contribution of the NBER to the development of human capital research was also very important as a springboard for many of these new generation researchers. Many of the research assistants were Becker's and Mincer's doctoral students coming from Columbia and included people such as Barry Chiswick (regional income inequality), Dave O'Neill (black–white employment rate differentials), Robert Michael (consumption efficiency), Michael Grossman (health demand), and Gilbert Ghez (lifetime allocation of time and earnings within a human capital framework). The NBER played a very important role by providing an opportunity for these young researchers to interact and develop their work, mostly under the mentoring of Mincer and Becker. Several of these researchers continued to develop their research on human capital topics at the NBER after they finished their dissertations.

(p.142) 7.3. Creating Discipleship and Developing the Human Capital Research Program

The success in attracting, training, and supporting new groups of researchers had a major impact on the potential of human capital research. It allowed the exploration, empirically and theoretically, of many of the aspects identified in the initial contribution of the three pioneers. The fact that these young researchers worked in most cases closely with Schultz, Becker, and Mincer meant that the first installments of the human capital research agenda were very much shaped by their interests. The interaction with these new researchers also meant that human capital research explored other aspects that enriched its applicability.

In the case of Mincer, his influence was clearly felt in the analysis of the role of human capital for income distribution and labor market analysis. Several of the students and younger researchers with whom he interacted developed their careers, at least initially, around two large issues. On the one hand, a large group focused on analyzing the explanatory role of investments in human capital for income inequality, either in the overall population or between specific population groups, with particular attention paid to gender differences and migrant workers. On the other hand, another large group focused on exploring the contribution of human capital to understanding the interactions between market and nonmarket behavior, including issues such as labor force supply, nonmarket effects of human capital, and home investment in human capital. The analysis that follows is not an exhaustive list of those influenced by Mincer's work on human capital and labor economics, but rather a set of examples that aims at illustrating the breadth and depth of Mincer's legacy through his various collaborations with younger generations.

7.3.1. Income Inequalities

One of the areas that initially attracted most interest in human capital, following the pioneering work of Mincer's doctoral research, was income inequality. Barry Chiswick, one of Becker's and Mincer's earliest students at Columbia, analyzed in his dissertation the effect of education and training on (regional) inequality (finished in 1966, but only published in an extended version in 1974). Chiswick attempted to show that distribution of income was related to (p.143) investments in human capital and that schooling and postschool training were important determinants not only of individual differences in income, but also of regional differences (1974). The preliminary results of his work with Gary Becker (on the role of human capital for a theory of income distribution) presented at the 1965 meeting of the AEA (Becker and Chiswick 1966) suggested that education had a reasonable explanatory power in terms of the distribution of earnings between and within US regions. Chiswick pursued this area further with Mincer some years later when they analyzed the role of human capital in US personal income distribution (Chiswick and Mincer 1972).

Important research developed on the role played by education in explaining differences in the economic performance of several specific population groups (especially in the USA). This attempted to explore the variability of the return to education between groups identified by Becker's pioneering work (Becker 1964).

One of the aspects that received significant attention was the possible role of human capital in explaining gender differences. The human capital explanation developed by Mincer and Solomon Polachek (1974) focused on the effect of human capital accumulation on market earnings and on market activities, but also took into consideration the elements of discontinuity in labor force participation and in human capital investments that characterized much of women's work behavior. This led them to adapt the so‐called Mincerian earnings function in order to take account of discontinuous processes of human capital accumulation. The issue of discontinuity and its obsolescent effects was further developed in Mincer's collaboration with Haim Ofek (Mincer and Ofek 1982). Accordingly, they considered that the issue of depreciation and possible restoration of human capital became a central aspect of labor market analysis: the longer the interruption the greater the impact.

This view of the gender gap was strengthened throughout the eighties by several studies that frequently adopted the human capital framework as a major explanatory principle. Among these are the studies by June O'Neill (1985), a former student of Mincer's at Columbia, and James Smith (see Smith and Ward 1989), one of the first NBER postdoctoral fellows in human capital in the late sixties. These studies enhanced the emphasis on women's market skills, notably human capital, as the primary shaper of their (changing) economic status.

(p.144) Another group that attracted particular attention was that of ethnic minorities. Both Mincer (1958) and Becker (1964) noted that although education was largely beneficial for all population groups, there were important variations between specific groups. Initial interest was concentrated on the Afro‐American population; Dave O'Neill, one of Mincer's and Becker's first students at Columbia, used military test scores to analyze how far the differences in terms of earnings were due to discrimination. His results suggested that although some of the differences in earnings could be ascribed to current discrimination in the labor market, a significant, potentially bigger difference was the result of a poorer provision of schooling to certain groups. This idea was further developed by Finis Welch (1967), another of the earliest postdoctoral human capital researchers at the NBER, when studying income differences between white and black populations in the US rural south. Welch considered that the lower income was due not simply to labor market discrimination, but to the fact that the black community had less financial capacity to invest in education, which reduced their income potential. By the late seventies human capital researchers were suggesting that the vintage hypothesis seemed to get some support from this population group, and that human capital and experience seemed to be the most important characteristics in explaining the improving status of black Americans (see work by James Smith 1978). This work continued in the late eighties, suggesting that the evolution of the black–white wage gap was mainly explained via two types of human capital: education and migration opportunities (Smith and Welch 1987).

The analysis of ethnic groups in the USA was also expanded, notably to the Asian and Jewish communities. This allowed a better picture of the role of human capital in ethnic and racial minorities, enhancing differences in level of schooling and returns to schooling. Although it was recognized that discrimination played a role in access to schooling and in the labor markets, this seemed to be less relevant than expected. Based on the work of Barry Chiswick (since the late seventies), and later by James Heckman, it was suggested that the differences in the rate of return to education by ethnic group could be linked to complementarities between schooling and other family investments in human capital, notably home human capital, and had links with fertility behavior and labor force participation (Chiswick 1988; Heckman and Walker 1990).

(p.145) The analysis of income inequality between different ethnic communities was also closely linked to the study of the labor market performance of immigrant workers. Most of the contributors to this latter topic in recent decades were Mincer's former students, such as Barry Chiswick, or research assistants, such as George Borjas. Their work and that of others on immigrants have contributed to a much better understanding of the determinants and impact of immigration emphasizing the role of human capital investments (Borjas 1999a and 1999b). This research indicated that the economic status of immigrants improved with duration of residence, although refugees experienced greater difficulty (Chiswick 1986). Although the same amount of years of schooling and labor market experience had a smaller effect on the economic status, if accumulated prior to immigration, the strong progression of earnings of economic migrants at the country of destination (Duleep and Regets 1997) could, ceteris paribus, eventually make them level with those native born workers. In contrast, research has downplayed both the importance of chance as a determinant of earnings and the possibility of selective return migration, that is, the return to the country of origin of the less successful immigrants. These results were found to be robust in both cross‐sectional and longitudinal data.

The work on the economic effects of migration has provided empirical support to the human capital approach, namely to the idea that migration propensities are clearly associated with educational level (see Greenwood 1997). In many of these studies, nurtured by human capital analysis, the influence of Mincer's work can be found. First, human capital analysis has been used in the issue of transferability of skills between different labor markets and different jobs, especially concerning the now classic distinction between general and specific human capital. Second, the possible complementarities between different types of skills and their impact on the economic return to human capital have been pointed out, notably concerning linguistic skills in the case of migration. Third, Mincer's work with Polachek and Ofek on interrupted work careers and skills’ obsolescence has been applied extensively in the case of migrant workers. Fourth, the analysis of migrant workers has highlighted the need for a long‐term perspective of the identification of their income patterns, something that has been stimulated by Mincer's focus on a long‐term view of the labor market. Finally, the study of migrant workers has enhanced the importance of the family as an essential unit of analysis for many (p.146) important aspects of the labor market, a view to which Mincer has contributed significantly.

Mincer's approach to inequality was also influential in the development of the analysis of the role of human capital in a life‐cycle framework. Some initial work had been developed by Yoram Ben‐Porath in his doctoral dissertation at Harvard (1967) in which he attempted to integrate human capital into a lifetime framework, and analyze its interaction with earnings, notably by taking into account the role of foregone earnings.4 Ben‐Porath's model used human capital theory to explain the shape of the wage profile, its upward slope, deceleration, and eventual decline, by offering a productivity‐based explanation of the growth of earnings with working age.5

At that time Becker was also increasingly interested in this view that broadened human capital's horizons for intergenerational aspects of education, as shown by his Woytinsky lecture (1967), where he developed a model of wealth maximization in order to explain the distribution of human capital investments, notably their concentration at earlier ages. This approach aimed as well to include the interaction between changes in wage rates over the life cycle resulting from the accumulation of human capital, the allocation of time between market and nonmarket sectors, and the impact of human capital on the productivity of household behavior (see Ghez and Becker 1975).

Becker's model was significantly strengthened empirically by Mincer's 1974 book, notably by emphasizing the role played by on‐the‐job training in the model's explanatory power. For Mincer, this wage growth was certainly related to firm training. He analyzed the issues by comparing indirect estimates of total worker investment costs derived from observed wage profiles with directly observed costs of job training investments. His initial estimates were based on the former method since at that time there were no direct estimates of training available (this only became possible in the mid‐seventies for the US case). The direct estimates of job training investment costs require data on the time spent in training per period and the period opportunity cost of that training. The initial estimates were the ones initiated by Mincer's 1962 study and then became much more tractable empirically through his innovative use of the parametric wage function.

One important issue for the analysis of lifetime patterns of income was that of allocation of time. Ben‐Porath's model assumed a two‐way (p.147) allocation of time between learning and earning, not taking into consideration the time spent in consumption or leisure or assuming it as fixed. Ghez and Becker (1975) had also considered a two‐way option but between labor and leisure. This was due to the fact that considering three‐way choices (between work, leisure, and human capital) was far more complicated. This was further developed in the mid‐seventies by people such as James Heckman, Alan Blinder, and Andrew Weiss. Their results added complexity to the analysis of life‐cycle patterns, notably in terms of the effects of production‐function specifications and of consequences of differences in initial conditions, though they did not question Ben‐Porath's main outcomes.

Although the analysis was focused on analyzing intertemporal differences in human capital investments over the life cycle, Mincer considered that it provided important insights into interpersonal differences. Accordingly, at any moment of the life cycle the marginal cost of producing human capital was lower for people with greater learning abilities. On the other hand, the marginal revenue was expected to be higher the easier the access to financing or the lower the interest rate. Altogether, these two conditions suggested that persons with greater ability to learn, lower funding costs, and lower time preference for the present would invest more in human capital in all periods. Since individuals with more schooling were more likely to be fast learners and to face lower discount rates, they were also more likely to invest more in job training. From this analysis, Mincer drew three major empirical implications in terms of lifetime patterns of investment in human capital (1997a: S41). First, persons with higher levels of schooling were expected to invest more in job training. Second, those who invested more in job training at earlier stages were also more likely to continue to do so at later stages of life. Finally, persons with greater learning ability or better schooling engaged in more job training activities, even when they had the same schooling attainment in quantitative terms.

7.3.2. Linking Market and Nonmarket Behavior

One of the most important developments brought about by human capital research was in drawing economists’ attention to the interactions between market and nonmarket choices. Mincer's influence (p.148) in this respect was very important due to his work on human capital and labor supply.

The impact of Mincer's work on labor supply has been acknowledged in various contexts. His influence was particularly felt in work dealing with family supply models and life‐cycle profiles, though arguably the major effect was to greatly stimulate economists’ interest in the topic (Gronau 2006). This interest has promoted important advances in the study of labor supply during the past decades, namely by recognizing and interpreting empirically the different labor supply functions (Heckman 1993). Other advances refer to the distinction between choices at the extensive margin (participation and employment) and at the intensive margin (about hours and weeks of work), and to the distinction between descriptive and structural labor supply functions. Although some of these advances questioned aspects of Mincer's pioneering work, it is hard to find a researcher working on this who has not been strongly influenced by his work (namely his 1962 paper). Moreover, several of these critical advances came from former students and close colleagues, illustrating the fact that intellectual admiration did not get in the way of analytical discernment.6

Some of Mincer's influence in linking market and nonmarket behavior was also felt through the analysis of the social and nonpecuniary benefits of education that was carried out in the late sixties and seventies at the NBER, namely by some of Becker's and Mincer's students. Particularly important at the time was Robert Michael's doctoral work which analyzed the impact of education on consumption behavior, notably on consumer efficiency. This would pursue in a more systematic and elaborate way one of the insights contained in Mincer's paper on opportunity costs and time (1963). According to Michael (1972) one should analyze the efficiency effect of education in terms of nonproductive activities, such as consumption, by using the new developments in consumer theory, developed by Becker and others. Michael's results suggested that expenditures changed with education (in particular, more educated individuals consumed more services) and that education enhanced people's capacity to produce useful commodities from a given level of factor inputs in the nonmarket sector (Michael 1972: 88).

Michael's work also followed in the footsteps of Becker who had been focused in reformulating consumer theory by adjusting it to household behavior (1965). Becker wanted to focus on the (p.149) increasingly important nonworking time, especially given the scarce attention this part of human activity had received. This built on the contribution of people attached to the Columbia Labor Workshop, in which Mincer played a leading role, and on some significant developments in the economic analysis of education, in order to integrate production and consumption into the household. Hence, the traditional framework of choice between work and leisure would be adjusted to encompass the allocation of time and goods within the household.7 Using this reformulation of consumer theory, which focused on household behavior, Becker moved increasingly into the analysis of individual behavior in a socially interactive context (1974). Becker has recently publicly acknowledged his profound (mutual) intellectual indebtedness to Mincer in what concerned their research activity during the sixties, namely by referring to his difficulty in disentangling their individual contributions from each other's.8

Another area stimulated by Mincer's work was research focusing on home human capital. This work was very much initiated by Arleen Leibowitz and Jacob Mincer at Columbia and the NBER at the beginning of the seventies. In her work, Leibowitz suggested that what was frequently regarded as the effect of natural ability could be in fact the result of preschooling human capital. Moreover, she challenged the view that women with higher levels of human capital and more attractive careers would invest less in their children's education (1974). In fact, these women were apparently reducing the time spent in other house activities (home production), but not the time spent with children and in particular doing activities that could have a significant positive effect on their cultural capital. Hence, there seemed to be a positive correlation between mothers’ education and children's education, due to higher amounts invested in the children's (home) human capital.

The attention to home human capital led other researchers to explore better the factors underlying children's attainment, especially in terms of schooling, and the direct and indirect effects of family background on income, notably via educational achievement. Children's economic success was analyzed in terms of the impact of governmental policies (setting the environment), parents’ behavior (work and earnings choices), and children's choices in using their talents and the resources made available to them (especially in terms of education, work, and family behavior) (cf. Haveman and (p.150) Wolfe 1995). The more persistent attention to the role of the family, education, and socioeconomic background led to several important empirical developments. Notably among these were the development of sibling studies, the improvement in the measurement of the role of education, and a better analysis of home investment in children. Altogether, these led to a more complex picture, in which factors such as the genetic endowment of ability, the family cultural background, and the family's endowment of physical and human capital converged. Moreover, it helped to consolidate the role of human capital both directly ascribed to the children and in inter‐generational terms, especially through the mother's education (Hill and O'Neill 1994).

This work on the intergenerational effects of investment in human capital linked with developments in the economics of household production. Some of the pioneering work was developed by Reuben Gronau, who interacted closely with Mincer at the beginning of his career. The late seventies and early eighties were characterized by a loss of momentum in this area of research (Gronau 1980), which was ascribed to the dissolution of much of the network associated with the Columbia Workshop that followed the move of Becker from Columbia (Grossbard‐Schechtman 2001). This area would acquire a renewed vitality at the start of the nineties (Gronau 1997).

7.4. Concluding Remarks

The development of human capital theory owes very much to the collective and articulated research efforts of a group of authors, of whom the pioneers were Theodore Schultz, Jacob Mincer, and Gary Becker. The importance of the three pioneers was not restricted to their individual contributions, since they played a crucial role in establishing a community of scholars that could extend, discuss, and substantiate their initial contribution. Mincer played no minor role in this respect, especially in what concerned labor economics.

Social science does not develop research groups as easily as most of the laboratory sciences do, and this makes even more prominent the need for any researcher, and especially those exploring new topics, to find colleagues with whom to discuss and improve their arguments. This capacity of Mincer, alongside the efforts of Schultz, Becker, and others, to convince their students that human capital analysis was an exciting project explained the rapid development and success of (p.151) human capital theory. Mincer stimulated several important developments related to human capital research. Among these, his influence is clear in such areas as income distribution, the comparison of the economic returns between different population groups, home human capital, and the effects of human capital on long‐term labor market performance.

Overall, Mincer was a mentor to many younger labor economists, attracting many of them to topics related to his research, notably topics associated with human capital research. The influence Mincer had on the early research careers of many of those labor economists was felt in the choice of topic for their dissertation, in the choice of their area of specialization, and in their methodological approach to labor research. However, it should be very clear that this mentoring did not mean that the work of these so‐called disciples was a mere imitation. Although influenced by Mincer, each of these researchers developed their own career and made individual contributions to human capital and labor economics. Their autonomy is clearly illustrated by the fact that several of these former students and research assistants introduced new topics of research, in some cases quite apart from human capital analysis. The point is that Mincer's important contribution to the development of human capital research was not limited to his own work but to the work that fructified through his interaction with his students and other labor economists.

Notes:

(1) Mincer's kind but extremely rigorous approach toward students is corroborated by one of his closest friends and colleagues, Gary Becker (2006).

(2) Polachek also recalls that this sense of rigor and perfectionism was deeply embedded in Mincer's way of doing economics. In their joint work, Polachek experienced how Mincer took it seriously, even to the point of completely rewriting an entire draft and rerunning the entire set of regressions. Thus, differently from many others, Mincer actually practiced what he preached.

(3) It is also interesting to note that someone who devoted so much interest to the labor supply of women has had such an important role in training so many women labor economists (Bloom and Siow 1993).

(4) According to Mincer (1997a: S27):

The lifetime accumulation of human capital is the process on which Ben‐Porath concentrated by modeling it as an optimal path of human capital investments over the individual's life‐cycle … his approach points out that human capital is a double‐edged response to puzzling findings in growth accounting and in income distribution statistics, i.e., whereas aggregate accumulation of human capital is a major factor in generating aggregate economic growth, individual accumulation is the process that generates individual economic growth. The latter has become a fundamental building block of modern labor economics.

Although the ideas expressed in the model were shared by others, this model provided possibly ‘the most succinct, rigorous and fruitful formulation’ (Mincer 1997a: S44).

(5) Accordingly, investments in human capital were done rationally, and therefore most investments occurred at younger ages since later investments faced a shorter payoff period. Moreover, if investments in human capital were profitable, their postponement meant a reduction in their net present value. Postponement of these investments could also be costlier since as earnings increased with lifetime, their opportunity cost would increase. Ben‐Porath's model assumed neutrality for matters of simplification. Mincer's empirical testing during the seventies suggested that the decline of investments in human capital was faster than predicted, and with stronger investments in human capital at early ages.

(6) As in the case of Gregg Lewis (1967), Ben‐Porath (1973), and work by James Heckman.

(7) This model was extended in 1967 (published later as an addendum to the second edition of Human Capital) to a framework of decisions over time and to investment in human capital. Along these lines was also Becker's work (p.177) with one of their graduate students, Gilbert Ghez (1975), on the allocation of resources by families over the lifetime of their members, where they analyzed the acquisition of skills, and why the investment tended to fall with age.

(8) This was done at Mincer's 80th anniversary conference that took place at Columbia in July 2002.