Discrimination, human capital, and life expectancy in a model of economic development

• DOI: http://doi.org/10.1111/ijet.12155
• Author: Norman Sedgley,Bruce Elmslie
• Published date: 01 September 2018
• Date: 01 September 2018

Discrimination, human capital, and life expectancy

in a model of economic development

Norman Sedgley



and Bruce Elmslie

y

Evidence suggests that race- and gender-based discrimination are prevalent. If discrimination

misallocates resources then it is likely to generate social costs. This paper provides a general

equilibrium model of the impacts of discrimination. We analyze effects of labor market

discrimination in a model where agents make human capital decisions based on comparing the

marginal benefits and marginal costs of additional human capital accumulation. Life expectancy

is a consideration in making human capital decisions, and is allowed to be endogenous. We find

that the impact of discrimination on equilibrium depends on the nature of any skills bias in

discrimination.

Key words discrimination, economic development, life expectancy, human capital, welfare

economics

JEL classification E10, J71, D61

Accepted 5 May 2017

1 Introduction

Most economists agree that labor market discrimination, based on race or gender, is a prevalent

phenomenon. If discrimination misallocates resources then it is likely to have significant social costs.

Macroeconomic models of these costs are rare, but the studies that exist suggest that they may be

large. This paper attempts to fill a gap in the literature by providing a macroeconomic model of

the general equilibrium impacts of discrimination on economic development. We analyze the

general equilibrium effects of labor market discrimination in a model where agents make human

capital decisions based on comparing the marginal benefits and marginal costs of additional human

capital accumulation. The cost of additional schooling is modeled as the value of an agent’s time

measured in terms of forgone earnings. Life expectancy is a consideration in making human capital

decisions. We allow life expectancy to be endogenous and dependent on the level of skills or human

capital accumulated. A major result of the model is that the impact of discrimination on

macroeconomic equilibrium depends on the nature of any skills bias in discrimination.

Economists have highlighted many fundamental reasons for differences across countries in the

degree of economic development. These reasons include factors as diverse as international trade

openness, multinational enterprise investment, equipment investment, the distribution of income,

inflation, and socialism (Easterly 1993; Fischer 1993). Perhaps the variable that has received the most

attention is educational attainment and training. In the case of lagging countries, Verspagen (1991)



Department of Economics, Sellinger School of Business and Management, Loyola University in Maryland, Baltimore,

Maryland, USA. Email: nsedgley@loyola.edu

y

Department of Economics, University of New Hampshire, McConnell Hall, Durham, New Hampshire, USA.

doi: 10.1111/ijet.12155

International Journal of Economic Theory 14 (2018) 211–231 ©IAET 211

International Journal of Economic Theory

models what he calls a country’s ‘‘intrinsic learning capability’’ as a determinant of knowledge

spillovers from advanced countries, while Hikino and Amsden (1994) include educational

attainment (especially of civil servants) as a key determinant of convergence. Moreover, Barro (1991,

1997), Levine and Renelt (1992), Mankiw et al. (1992), and Jones (2002) find educational attainment

to be an important determinant of growth for both leading and lagging countries. Hanushek and

Kinko (2000) use international math and science scores as an output measure of educational

attainment. They find a strong correlation between test scores and growth rates. Most empirical

research on growth finds that human capital accumulation is an important key to economic growth

and/or levels of per-capita gross domestic product.

This paper contributes to the discussion by modeling the role labor market discrimination plays

in determining capital to labor ratios and output per capita. It is important to recognize that there are

likely to be cumulative effects of discrimination on macroeconomic performance, with

consequences for the whole of the population. The research to date, however, has not recognized

the potential importance of the market distortions created by discrimination, endogenous human

capital choices and life expectancy that are highlighted in the current work.

There are three questions in the literature that directly impact the contribution of this paper.

First, what is the extent of discrimination in labor markets? Second, given that evidence of

discrimination can be found, what is the nature of the bias in discrimination with regard to skills?

Finally, what are the impacts from discrimination and the nature of its skills bias on macroeconomic

equilibrium outcomes such as growth rates and/or per-capita income?

We begin with the extent of discrimination which evidence suggests is a prevalent phenomenon.

A majority of work on discrimination uses data from the US labor market, though the results are

representative of a large cross-section of nations with significant minority populations and ethnic

diversity. Studies based on the decomposition described by Oaxacca (1973) document robust

evidence of gender- and race-based discrimination (Cotton 1988; Neumark 1988; Boston 1990; Main

1991; Gyimah-Brempong et al. 1992; Weinberger 1998; Coleman 2003; Bertrand and Mullainathan

2004; Jarrell and Stanley 2004; Carneiro et al. 2005). These papers suggest that race and gender wage

gaps have ranged between 30% and 60%, and 35–70% of the gap can be explained by factors other

than labor market discrimination, such as individual human capital choices.

Given that evidence of significant wage discrimination exists for various groups, is the

discrimination biased by skill level? Couch and Daly (2002) report that the black wage gap fell from

50% in 1967 to 30% by the late 1980s. The gap then fell, on average, by about 1% per year throughout

the 1990s, leaving approximately a 20% wage gap in 2000. Also of significance is their estimate that

the wage gap does not decrease at the same rate across skill levels. They report the most rapid

convergence for young workers with little work experience. This implies that relative levels of

discrimination for blacks in the USA are increasing with education. This seems reasonable since

minimum-wage laws create an artificial wage floor only for low-skill workers. Many other papers

document a potential skills bias in wage gaps (Cotton 1988; Duncan 1992; Gyimah-Brempong et al.

1992; Rogers and Armentrout 1996; Garc

ıa et al. 2001; Bertrand and Mullainathan 2004;

Munasinghe et al. 2008). Generally the evidence suggests that women and minorities gain lower

marginal wages from additional education, experience, and on-the-job training.

Much of the literature finds evidence that discrimination and its skills bias are measurable aspects

of real labor markets. As people respond to these differential returns to human capital accumulation,

discrimination might lead to a misallocation of resources. If resources are not efficiently allocated

significant costs to society can arise. Studies that investigate the potential impacts of discrimination

on an aggregate level are surprisingly rare.

Discrimination, human capital, and life expectancy Norman Sedgley and Bruce Elmslie

212 International Journal of Economic Theory 14 (2018) 211–232 ©IAET