Labor markets, gender, and macroeconomics
James Heintz
Informal write-up of presentation (SOAS workshop & consultation on women's economic empowerment, Jan 26-27, 2012)
6 Feb 2012
The functioning of labor markets shapes the performance of the economy as a whole along a number of dimensions: macroeconomic adjustments, long-term structural change, the process of human development, and the realization of collective well-being. However, the models and theories frequently used to explain labor market dynamics, and from these explanations draw policy conclusions, fail to reflect the actual functioning of labor markets. This failure is particularly acute in developing countries, in which a wide variety of institutions, market transactions, and power relationships mediate the exchange of labor, yet the theories and standard empirical approaches used to characterize these dynamics bear a poor resemblance to what is taking place in reality. There is a need to consider 'real labor markets' - the actual ways in which labor is exchanged, including how choices regarding allocation of labor are facilitated or constrained and the full set of conditions under which returns to labor are realized.
These issues are of critical importance. The vast majority of the world's population - in developed and developing economies - earn the income they need to support themselves and their familiesthrough their labor. Choices around the allocation of this labor, including between market and non-market activities, and the factors determining the returns to labor are therefore of paramount importance. These issues are particularly critical for vulnerable populations for whom their own labor frequently represents their most abundant, and most valuable, resource. Those individuals who are constrained in their ability to exchange labor - due to age (either very young or very old), restricted physical capacity, or disability - depend on the labor of others, particularly when the social policies that exist to protect those unable to participate in labor markets are not enough.For many women, specialization in unpaid work affects current and future labor market choicesand control over time and money in the household. For these reasons, issues of labor markets and employment therefore must feature as core elements of any discussion of 'inclusive growth.'
But the functioning of labor markets not only makes growth more or less inclusive, it also directly affects the process of growth itself. Labor markets determine the allocation of human resources in an economy, and thereby affect growth and long-run economic performance. A reallocation of labor from low-productivity to high-productivity activities will raise aggregate productive performance - which is necessary, but not sufficient, for improving the average quality of employment. Similarly, the ways in which labor is allocated to unpaid care work has significant implications for the level of investment in human beings and, hence, long-run growth. Such investments improve the well-being of individuals, but they also have implications for the economy as a whole.
Approaches within the discipline of economics to explain and analyze labor markets are overly narrow.They focus on the productive characteristics of individuals or isolated firms. They frame labor supply decisions as individual trade-offs between work and leisure. Theyprimarily approach labor demand through the lens of labor costs and the optimizing behavior of employers. These factors are important and need to be taken into account. However, by only focusing on a relatively narrow set of variables to describe labor market dynamics, much of what makes labor markets tick gets lost. Moreover, those parts of the larger picture which are jettisoned may turn out to be more important for understanding how labor is exchanged than the core variables and techniques that tend to define standard labor economics.
One tension exists in the bifurcation of labor markets between enterprises (self-employment) and wage employment. For example, enterprise-based approaches emphasize the endowments of factors of production and the productive relationships between these endowments for determining the returns to labor and capital. However, this approach misses important aspects of how very small household enterprises function in low-income countries. The endowments of fixed assets in these household enterprises are typically extremely small and are particularly small for enterprises operated by women. Those individuals who are self-employed in household enterprises are primarily dependent on the returns to their labor, not their assets, and are effectively selling their labor to generate income. The realization of the returns to their labor are mediated by other markets and institutional dynamics, not the typical interaction between employers and employees characteristic of the standard labor market models. In additions, individuals may be engaged in multiple employment activities simultaneously (including both self- and wage employment) and, from what little longitudinal evidence we have available, it appears that individuals move between different employment activities - including between formal and informal employment.
Therefore, separating self-employment from wage employment and using different models and analytical techniques to understand what is going with regard to employment, livelihoods, and labor markets provides a fractured framework for understanding how real labor markets operate. What is needed is a broader, more inclusive conceptualization of labor markets - one that sees 'labor markets' as the collection of all institutions through which labor is allocated and exchanged. While this alternative does not spurn the usefulness of standard wage labor or enterprise based methods for specific research questions, it does recognize that such approaches are incomplete and, at times, inappropriate. A broader notion of what labor markets really are requires new levels of analysis. Labor supply is not simply the outcome of atomistic decisions made by optimizing individuals, but rather determined in the context of households, with multiple interdependent agents motivated variously by self-interest, commitment, reciprocity, and love. The supply-chain is often a more important unit of analysis than the enterprise for understanding market structures and power dynamics. The relationship between a street trader and her supplier (or a waste collector and her buyer) may have a more significant impact on her returns to labor than her productive endowments.
Within the enterprises of the self-employed, the role of assets (capital) needs to be interrogated more thoroughly in the context of understanding labor markets and the factors affecting returns to labor. For many very small enterprises, working capital (e.g. cash on hand) is frequently as critical as fixed capital (e.g. equipment) in day-to-day operations. Women processing rice in northern Ghana benefit from being able to buy rice when the prices are lowest, but to do so requires access to money at the right time. This represents a distinct role for particular types of capital assets - not as a productive factor, but rather affecting the terms under which labor is ultimately exchanged.
Because of the focus on individual characteristics, analysis of labor markets tends to favor the supply-side rather than the demand-side, explaining earnings using age, education, experience, sex, and other demographic variables. The supply-side is important. The invariably strong estimates of the effect of education on labor market outcomes testifies to this. However, there is a risk to reducing the explanation of labor market outcomes to a collection of core individual attributes with gender dynamics being whittled down to a significant coefficient on a dummy variable. Equally critically, the demand-side may receive scant attention. This is problematic. A lack of sufficient labor demand contributes to high levels of part-time and contingent work, unemployment and underemployment of various kinds, and growing informality as formal opportunities are rationed. Due to segregation of activities, a lack of labor demand has different consequences for women and men, producing more demand pressures for women's market labor relative to men's. Women are frequently crowded into activities in which labor demand may be 'saturated', putting downward pressure on the returns to women’s remunerative labor.
The standard wage labor model does address labor demand, but often by focusing on labor costs and sidestepping other factors which may influence the demand for labor. Labor demand curves are downward sloping - weak demand is explained by high costs. Of course, labor costs are important and cannot be ignored. However, other factors, for which there may be greater policy leverage, affect labor demand. Within the metaphor of the standard model of demand for wage labor, there is a need to pay attention to factors which shift the labor demand curve, instead simply trying to slide down it. Constrained access to markets, insufficient accumulation of productive capital, limited complementary public investment (e.g. infrastructure), and lack of reliable information can all affect labor demand. The recent global economic crisisprovided a clear and devastating reminder that the macroeconomic environment has a far-reaching impact on the demand for labor. Moreover, the standard model has nothing to say about the demand for the labor of the self-employed. It is critical to keep in mind that labor demand is frequently derived demand, flowing from the dynamics of markets other than a narrowly defined labor market.
When looking at how labor markets function, it becomes clear that the exchange of labor deviates significantly from the textbook ideal - labor markets are rife with market imperfections. The extent and nature of these imperfections have been theorized and documented at length: information asymmetries, the existence of transactions costs to labor force participation, segmentation and barriers to perfect mobility, market power, gender dynamics, and the role of social norms (e.g. norms of fairness), among others. It has been noted that labor is the one commodity which can contest the conditions under which it is exchanged, embedding the exchange of labor in social relationships which are far more complex than relatively anonymous one-shot market transactions. Despite these deviations from the abstract market clearing relationships of standard labor economics, labor market policy is often formulated to try to get labor markets to behave as much like the perfectly competitive ideal as possible. Such a task is sisyphean, unless at some point it becomes possible to take the human beings out of labor markets. Labor markets, broadly defined, are inherently messy, complex, and indeed imperfect. The challenge is to make labor markets work better, not to try to force them to conform to an imagined ideal.
Since Keynes' General Theory, it has been recognized that the functioning of labor markets has important implications for macroeconomic outcomes. Robert Solow, whose name is closely linked to the development of the neoclassical model of economic growth, recognized this fact in The Labor Market as a Social Institution(1990), in which he argues that the social character of labor markets affects macroeconomic adjustment and equilibrium. Akerlof and Yellen (1985a, 1985b) show how small deviations from rational behavior and perfect market outcomes, including in labor markets, can have large effects on macroeconomic outcomes;the macro consequences outweigh the costs at the micro level. The lesson is clear: routine imperfections in labor markets not only affect interactions at the individual level, but also have economy-wide implications and affect general social welfare. In addition, we have already stressed that the allocation of labor alters overall economic performance and influences the path of growth and structural change.
Perhaps the most significant channel through which labor affects the macro-economy is also the one which is most frequently brushed aside: the recognition that labor is a produced factor of production.Labor is not simply exchanged; it is also produced. Yet, labor often appears as an exogenous variable in growth models, with the growth rate of the population and the workforce taken as given. There are exceptions - for example, in the article by Robert Barro and Gary Becker (1989) “Fertility choice in a model of economic growth”. In Barro and Becker's model, parents make costly investments in children - represented equivalently by either less disposable income after paying for social reproduction or fewer hours dedicated to market work. Such investments ensure that labor is available for productive activities in future periods. Altruism at the household level provides the motivating force for making investments in human beings. Indeed, altruism plays a critical role in sustaining growth in Barro and Becker's model - set altruism equal to zero and future growth collapses.
The assumptions of Barro and Becker's model can be criticized on a number of fronts - e.g. households do not have a single utility function, altruism does not characterize all intra-household or inter-generational dynamics, gender relations shape how investments in children are made, etc. However, putting these general critiques of Becker's approach to household economics aside, it is interesting that behavior other than individualistic self-interested optimizing decisions is central to guaranteeing future growth. Non-market institutions, various interactions and a wider range of human motivations become central to growth when we consider labor as a produced factor of production. Markets alone are not enough to secure long-run growth, let alone inclusive growth. As has been long recognized by feminist economists, how labor and other resources are allocated to produce the human resources on which an economy depends is of the upmost importance, and yet this central part of the growth story is frequently overlooked.
These issues have important implications for understanding how labor markets really work and how they affect growth, development, and social welfare. It is not sufficient to consider how labor markets, broadly defined, allocate labor between market activities (e.g. various forms of wage employment and self-employment), how this allocation affects economic growth and structural change, and how the allocation ultimately shapes social well-being. We also need to consider, in the analysis of real labor markets, how labor is allocated to the production of labor itself - in the household, through the state, and in markets. Gender dynamics feature prominently. The question of women's economic empowerment cannot be boiled down to making markets work for women, since there is a risk of side-stepping the importance of non-market spaces and institutions. Markets in general, and labor markets in particular, depend on non-market institutions in order to function. It is this constellation of institutional forms and dynamics which needs to be considered if we are to understand how real labor markets work.
Refs.
Akerlof, George and Yellen, Janet.1985a. "Can small deviations from rationality make significant differences to economic equilibria?"American Economic Review 75(4): 708-20.
Akerlof, George and Yellen, Janet. 1985b. "A near-rational model of the business cycle with wage and price inertia" Quarterly Journal of Economics 100: 823-38.
Barro, Robert and Becker, Gary. 1989. “Fertility choice in a model of economic growth”Econometrica 57(2): 481-501.
Folbre, Nancy. 2008. Valuing Children: Rethinking the Economics of the Family. Cambridge, MA: Harvard University Press.
Keynes, John Maynard. 1964. The General Theory of Employment, Interest, and Money. Harcourt Brace.
Solow, Robert. 1990. The Labor Market as a Social Institution. Cambridge, MA: Blackwell.