Canadian Journal of Sociology Online March-April 2004

Note on the Discipline//Note sociologique

Where Have All the Sociologists Gone?
Explaining Economic Inequality

John Myles
Canadian Journal of Sociology 28, 4 (Fall 2003): 551-559.

[T]he distribution of income [has] grown markedly more unequal over the past three decades….sociologists have been strangely and remarkably silent on these issues. (Morris and Western 1999: 623)

Martina Morris and Mark Western’s judgement of the contribution made by sociologists to the study of the changing structure of income inequality in recent years is both accurate and disturbing. Their point is not that sociologists have failed to take note of the changes in the distribution of worker’s earnings or family incomes. Rather, the point is that the contribution of sociology to accounting for these trends has been so modest. With a few notable exceptions (e.g., Bernhardt et al, 2001), we have left most of the heavy lifting — both theoretical and empirical — on this topic to the economists (see Text Box).

Until the late 80s, economists were not much interested in the study of trends in economic inequality in large measure because of the remarkable stability of the income distribution from the 50s to the 70s. As Henry Aaron once put it, until the 1980s, studying trends in income inequality was like watching the grass grow.[1] All this changed in the 1980s when the secular trend of rising inequality that began in the 1970s became apparent. Since then the study of inequality has become a major growth industry in economics. In contrast, sociologists have been mostly on the sidelines, keeping track of the score perhaps but not contributing much to the game in process. Sociologists continue to study point-in-time differences in earnings and income — between men and women or between races — but even here, the most important work has been done by economists. In recent years the economists have even begun, albeit tentatively, to make headway in an area that was once a sociological monopoly, the study of intergenerational class inheritance (see Bowles and Gintis 2002; Erikson and Goldthorpe, 2002).

There are two facts to be explained here — the economists’ advance and the sociologists’ retreat. My hypothesis for the economists is that growing methodological sophistication has led successive generations of younger economists to search out new topics to which to apply their skills and, on average, sociologists have not keep pace. Economic inequalities concern quantities and it is inconceivable that sociologists will contribute greatly to their understanding without a high level of sophistication in quantitative analysis.

But why the sociologists’ retreat? Sociologists were more or less in the game until the early 1980s. Segmented labour market theories were all the rage at the time and sociologists arguably made some of the most important contributions to this field of research (Hodson and Kaufman 1982). Thereafter, sociology began retreating from its traditional dialogue, and competition, with economics. Instead, as Morris and Western (1999:650) suggest, we ended up a quarter century later fighting disciplinary boundary battles with literature and the humanities.
Here I want to advance the hypothesis that it was a theoretical loss of faith, not methodological immaturity, which triggered the retreat. To put the argument simply, it became increasingly embarrassing to find that sociology’s “leading indicators” for trends in the “structure of inequality” suggested that things were getting better and better while in the real world growing inequality and polarization in both earnings and family income were the order of the day.

Sociology’s shared legacy, one inherited from 19th century political economy, is that the structure of inequality is built on the division of labour. Frank Parkin (1971) called this the “positional” approach to the study of inequality. Wealth, power or prestige are attached to positions, not to the individuals who occupy those positions. The trick to understanding inequalities is to understand how the structure of empty places — let’s call it the “class structure” — is organized. Then, and only then, can we begin to ask second order questions about how people are allocated to these positions.

Of course we have always had enormous disagreements over how to think about the empty places in the “class structure.” To use Erik Wright’s (1980) familiar distinction, some of us like to think about empty places defined by the technical division of labour — the occupational structure — and some of us in terms of the social relations of production (Marxian-like classes). But in the last instance, our intellectual legacy from 19th century political economy — whether Marxian, Weberian or Durkheimian — makes us think first about the structure of empty places — how it changes over time or differs across societies. So let’s reflect a bit on where that idea has taken us, and especially in regard to what we often think of as our forte — accounting for historical change.

There are two ways in which an “empty places” approach might explain long term trends in inequality. The first is through compositional shifts, changes in the class structure or the technical division of labour. Segmented labour market theory fit this mode of thinking as did the initial intuition that led to speculation that wage and earnings inequality were on the rise in the 1980s. In 1982, Barry Bluestone and Bennett Harrison (1982) made the case that deindustrialization was a likely culprit leading to greater wage polarization. Let’s call this the McJobs thesis.

According to the argument, the wage distribution generated by the empty places characteristic of mature industrial societies was relatively benign. In the age of high industrialism, following the Second World War, there were lots of jobs in the middle of the wage distribution and a smaller number of both low and high paying jobs. The result was Goldthorpe’s “affluent” worker or what came to be known in postwar North America as the “middle class”, a growing number of workers able to achieve a middle class lifestyle.

Deindustrialization was changing all this. The shift from goods to service employment meant the elimination of a large number of manufacturing jobs in the middle of the distribution and an increase in the number of both low and high paying service jobs. At the end of the day we would be left with an economy made up of well paid software engineers, film producers, and lawyers, on the one hand, and the people who made their lunch and emptied their wastepaper baskets on the other. This was a thesis most sociologists could feel comfortable with: inequality was on the rise because the structure of empty places was changing.

As it turned out, the prediction was correct — wages did grow more unequal from the seventies to the eighties — but not for the reasons expected. Changes in the industrial and occupational composition of the work force — the structure of empty places — had a little to do with the change but not very much. The surge in wage inequality in Canada occurred in the early 1980s and most of the change was a result of changes in the wage and earnings distribution within occupations and industries not because more workers were employed in high and low end services (Myles, Picot, and Wannell 1988). U.S. results lead to identical conclusions (Levy 1998). Something other than changes in the structure of empty places — what we often like to call “structural” change — had to account for the rise in inequality.

More embarrassingly, to the extent that we were able to describe changes in the structure of empty places, things seemed to be moving in the opposite direction. Managerial, professional and technical occupations were growing and the share of routine working class jobs was shrinking, despite the fast food industry. The most rigorous efforts to detect growing “proletarianization” of the class structure in rich capitalist democracies by researchers theoretically disposed to find such trends inevitably concluded that the opposite was true (Esping-Andersen, Assimakopoulou, and Kersbergen 1993; Myles 1988; Wright and Martin 1987). This is not to say there had not been a quantitative expansion of “lousy” jobs but their rate of growth was simply lower than that of not-so-lousy jobs, at least as sociologists define them.

A second way that a positional or empty places account might explain trends in inequality is through change in the relative incomes of different classes or occupational groups. Marx’s thesis of the immiseration of the working class relative to capital was of this sort. It is true that the relative incomes of CEOs have grown dramatically since the 1970s. CEOs, however, account for a very small share of the population and their incomes, however large, do not have much of an impact on the overall trends. Braverman’s (1974) deskilling thesis, if true, could account for the change. Implicitly, at least, if the polarization of job skills that Braverman predicts is at work, we should observe a parallel polarization in the distribution of wages and earnings — between the “direct producers” on the shop floor and the managerial and professional employees who have appropriated their knowledge. The reality, however, is that most of the action has been taking place inside classes and occupational groups, not between them. In short, changes in the structure of income inequality appear to be largely unrelated to any observable changes in class structure, however defined.

Grusky and Sorensen (1998) catalogue a number of different responses to this development. The revisionist impulse, shared by the present author (Clement and Myles 1994) among many others, was to refurbish and revamp the traditional approach with new and presumably improved occupational or class aggregations. The result has been an “embarrassment of choice” among boundary-mapping schemes with no clear standard for evaluating among them. The postmodern deconstructionists who announced the “death of class” were at the other end of the continuum (Pakulski and Waters 1996). Their claim was that all types of “structural” divisions are eroding and that new social cleavages are cultural in origin, hence the emergent dialogue with the humanities.

The average sociologist probably falls into neither camp. Like the average Anglican, s/he can sing the familiar hymns when called upon but doesn’t go to church very often since neither approach has helped much in explaining emergent patterns of economic inequality. As an inveterate revisionist, I still hold out hope for the positional approach since we have yet to push it to the limits suggested by Grusky and Sorensen (1998), an approach that has met with some success in explaining in black-white earnings differentials in the United States (Grodsky and Padger 2001). Analysis of the growth of market-mediated employment relations — “contingent workers” — also holds promise, a topic that we may be able to study adequately with the arrival of a new generation of firm-level studies as in Statistics Canada’s Work and Employment Survey (WES).

In my view, however, most of the recent efforts to “salvage” class analysis is flawed because its starting point is flawed. It begins by theorizing the independent variable — classes — rather than with the outcome to be explained — trends in economic inequality in advanced capitalist societies. The aim of the exercise is still to explain who gets what and why. Marx’s labour theory of value may have been wrong but it did start in the right place. Sorensen’s (2000) effort to revive class analysis based on a theory of “rents” is also in the right spirit. In particular, he attempts to account for the contemporary pattern of growing inequality within classes and occupational groups (Sorensen, 2000: 1552), including university professors (Sorensen 2000:1548). But to date, neither he nor any other sociologist has shown how one might actually analyse contemporary changes in the structure of earnings and wage inequality from a class-based perspective.

Class-based accounts have fared much better at the macro-institutional level. Erik Wright’s (2000) analysis of the varieties of “class compromise” among nations provides a useful introduction for sociologists and the theoretically inclined to this literature. Cross-national studies of trends in wage inequality have demonstrated the importance of the institutions that regulate relations between capital and labour. The role of unions and the presence or absence of corporatist-like bargaining systems play a prominent role in contemporary cross-national studies of wage structures and how they change. But as the economists have also taught us, the demonstration of these institutional effects, hinges on the prior capacity to estimate appropriate micro-level models of wage determination and change, a task that precious few sociologists are intellectually equipped to do.

Once upon a time, one could argue with some reason that economists’ views on these issues were short-sighted since their faith in markets blinded them to the role of power and institutions in shaping the structure of inequality. But this is no longer true or at least much less true than in the past. At least since the early nineties, cross-national research has become a major part of the field and variations in the institutional arrangements that regulate relations between capital and labour, gender and family, have been moving to the core of the discipline (Blau and Kahn 2002; Freeman 1993; Freeman and Katz 1995).

So what’s the answer? I have no solution to these big questions but I would propose a strategy. The young Marx began by debating with Hegel, the philosopher, but ended up debating Ricardo, the economist. In the past quarter century much of sociology seems to have moved in the opposite direction. We began in the 1960s and 1970s by debating the economists but ended up a quarter century later fighting disciplinary boundary battles with literature and the humanities. There are of course other things worth explaining besides basic forms of economic inequality but economic inequalities do remain pretty basic. To go beyond what economists have to say on the issue, we must first master what they have to say and the theoretical and empirical tools required to do it better. If our aim is to answer the question, who gets what and why?, Marx remains a fine model to emulate. Where could we begin?

When translated into the language of contemporary economics, any theory that relies mainly on changes in the “structure of empty places” to account for labour market inequality is likely to fail since it puts all of its eggs in one analytical basket, namely labour demand. But if we believe that markets exist, and most of us do, this cannot be true. Labour supply, both its quantity and quality, matter as well. Feminist scholars were quick to highlight this point: part of the reason women do less well than men is the constraint on their labour supply produced by an unequal household division of labour (Seecombe 1986). Why did wage inequality grow so much more in the U.S. than in Canada during the 1980s and 1990s? Part of the reason is that the supply of post-secondary graduates expanded more quickly in Canada so that the sharp increase in the wage gap between the more and the less educated experienced south of the border was suppressed (Freeman and Needels 1993).

Excoriating markets is not enough; we must also understand how they work. As Roemer (1982: 196) puts it: “The neoclassical model of the competitive economy is not a bad place for Marxists to start their study of idealized capitalism.” But markets are too central and important an institution in our societies to abandon their study to the economists and the neoclassical model. To surpass the model, however, our theories must subsume it, not ignore it.

Notes
1 Recent U.S. research on the “great wage compression” of the postwar years (Goldin 1992) has changed this assessment. Long term historical shifts in the distribution of income and wealth — who gets what and why? — had of course been a topic long studied by economists and sociologists including Gerhardt Lenski’s magisterial Power and Privilege (1966). Back to text

References

Bernhardt, Annette, Martina Morris, Mark Handcock, and Marc Scott
2001	Divergent Paths: Economic Mobility in the New American Labor Market. New York: Russell Sage.
Blau, Francine and Lawrence Kahn
2002	At Home and Abroad: U.S. Labor Market Performance in International Perspective. New York: Russell Sage.
Bluestone, Barry and Bennett Harrison
1982	The Deindustrialization of America. New York: Basic Books.
Bowles, Samuel and Herbert Gintis
2002	“The inheritance of social inequality.” The Journal of Economic Perspectives 16:3–30.
Braverman, Harry
1974	Labour and Monopoly Capital: The Degradation of Work in the Twentieth Century. New York: Monthly Review Press.
Clement, Wallace and John Myles
1994	Relations of Ruling: Class and Gender in Postindustrial Societies. Kingston and Montreal: McGill-Queens University Press
Erikson, Robert and John Goldthorpe
2002	“Intergenerational inequality: a sociological perspective.” The Journal of Economic Perspectives 16:31–44.
Esping-Andersen, Gosta, Zina Assimakopoulou, and Kees Van Kersbergen
1993	“Trends in contemporary class structuration: a six-nation comparison.” Pp. 32–57 in Changing Classes, edited by G. Esping-Andersen. London: Sage.
Freeman, Richard
1993	Working Under Different Rules. New York: Russell Sage.
Freeman, Richard and Lawrence Katz
1995	Differences and Changes in Wage Structures. Chicago: University of Chicago Press.
Freeman, Richard and Karen Needels
1993	“Skill differentials in Canada in an era of rising labor market inequality.” in Small Differences That Matter: Labor Markets and Income Maintenance in Canada and the United States, edited by D. Card and R. Freeman. Chicago: University of Chicago Press.
Goldin, Claudia and Robert Margo
1992	“The great compression: the wage structure of the United States at mid-century.” The Quarterly Journal of Economics, CVII:1–34.
Grodsky, Eric and Devah Padger
2001	“The structure of disadvantage: individual and occupational determinants of the black-white wage gap.” American Sociological Review 66:542–567.
Grusky, David B. and Jesper B. Sorensen
1998	“Can class analysis be salvaged?” American Journal of Sociology 103:1187–1234.
Hodson, Randy and Robert Kaufman
1982	“Economic dualism: a critical review.” American Sociological Review 47:727–39.
Lenski, Gerhardt
1966	Power and Privilege: A Theory of Social Stratification. New York: McGraw Hill.
Levy, Frank
1998	The New Dollars and Dreams: American Incomes and Economic Change. New York: Russell Sage.
Morris, Martina and Bruce Western
1999	“Inequality in earnings at the close of the twentieth century.” Annual Review of Sociology 25:623–657.
Myles, John
1988	“The expanding middle: some Canadian evidence on the deskilling debate.” Canadian Review of Sociology and Anthropology 25:335–364.
Myles, John, Garnett Picot, and Ted Wannell
1988	“The changing wage distribution of jobs, 1981–1986.” The Canadian Economic Observer November:1–33.
Pakulski, Jan and Malcom Waters
1996	The Death of Class. London: Sage.
Parkin, Frank
1971	Class Inequality and Political Order. New York: Praeger.
Seecombe, Wally
1986	“Reflections on the domestic labour debate and prospects for Marxist-Feminist synthesis.” Pp. 201–235 in The Politics of Diversity: Feminism, Marxism and Nationalism, edited by R. Hamilton and M. Barrett. Montreal: Book Centre.
Wright, Erik Olin
1980	“Class and occupation.” Theory and Society 9:117–214.
2000	“Working-class power, capitalist class interests and class compromise.” American Journal of Sociology 105:957–1002.
Wright, Erik Olin and Bill Martin
1987	“The transformation of the American class structure, 1960–1980.” American Journal of Sociology 93:1–29.

http://www.arts.ualberta.ca/cjscopy/articles/myles2003.html
March 2004
© CJS Online