The term “social inequality” encompasses the many ways in which members of society have unequal access to resources, opportunities, status, and protection. Income inequality is one dimension of social inequality. Over the past two decades, economists have made substantial progress in characterizing the nature of income inequality in the United States, how it has changed over time, and the underlying forces that are responsible for it.¹⁾See, for example, T. Piketty and E. Saez, “Income Inequality in the United States, 1913–1998,” The Quarterly Journal of Economics 118, no. 1 (February 1, 2003): 1–41, doi:10.1162/00335530360535135; Claudia Dale Goldin and Lawrence F. Katz, The Race Between Education and Technology, (Cambridge, Massachusetts: The Belknap Press of Harvard University Press, 2009); D. H. Autor, “Skills, Education, and the Rise of Earnings Inequality among The ‘Other 99 Percent,’” Science 344, no. 6186 (May 23, 2014): 843–51, doi:10.1126/science.1251868; David H. Autor, “Why Are There Still So Many Jobs? The History and Future of Workplace Automation,” Journal of Economic Perspectives 29, no. 3 (August 2015): 3–30, doi:10.1257/jep.29.3.3; Thomas Piketty, Capital in the Twenty-First Century, trans. Arthur Goldhammer (Cambridge Massachusetts: The Belknap Press of Harvard University Press, 2014). In this essay, I will bring these findings into dialogue with the conceptual framework for understanding social inequality developed by john a. powell and Stephen Menendian in the inaugural issue of this journal.³⁾john a. powell and Stephen Menendian, “The Problem of Othering: Towards Inclusiveness and Belonging,” Othering and Belonging: Expanding the Circle of Human Concern 1, no. 1 (2016). Their framework is built on the concept of “othering.” I will present an approach to understanding income inequality that is complementary to, yet distinct from, othering. I hope that examining how these approaches are related will improve our overall understanding of social inequality and spark further debate.

The concept of othering, as developed by powell and Menendian, is grounded in group position theory, which comes from sociology. This theory posits that humans have a universal tendency to assign themselves and others to social categories and to judge members of one’s own category or group as superior to others. This innate bias engenders beliefs and narratives about the inferiority of other groups that are deployed particularly when there is a conflict over symbolic or real resources. These beliefs and narratives justify the priority of the claims of one group over another. Conflict of this sort, in turn, reinforces beliefs in group differences and produces new narratives. Social scientists have long referred to this process of stereotyping other groups as “othering.” powell and Menendian adopt this term.

powell and Menendian point to a wide variety of contemporary conflicts around the world in which the dynamics of othering appear to be important, including the persecution of the Rohingya people by the government of Myanmar, white supremacist attacks on African Americans, and renewed action by the Turkish state against its Kurdish minority. One need look no further than the writings of Dylann Roof, the white supremacist who murdered eight parishioners of the Emanuel African Methodist Episcopal Church in Charleston, South Carolina, in 2015, to see how conflict over resources and prejudiced narratives are direct antecedents to racial violence.²⁾Dylann Roof, “Dylann Roof’s Manifesto,” 2016, https://www.nytimes.com/interactive/2016/12/13/universal/document-Dylann-Roof-manifesto.html?_r=1. Othering has real explanatory power.

Within economics, ideas closely related to othering have been part of the study of group-based discrimination for many years, and economists have produced evidence consistent with othering being responsible for part of income inequality in the United States.⁴⁾Much of the empirical literature concerns distinguishing animus from other motives for unequal treatment. See, for example, Jonathan Guryan and Kerwin Kofi Charles, “Taste-Based or Statistical Discrimination: The Economics of Discrimination Returns to Its Roots,” The Economic Journal 123, no. 572 (November 2013): F417–32, doi:10.1111/ecoj.12080. However, the study of income inequality also shows that forces not related to othering are also important. These forces relate to how the wages of workers with different levels of education are determined in the labor market.

Before we explore this link between inequality and education, I would like to ask the reader to engage in a thought experiment. Suppose that we lived in a world just like our own except that the human tendency to engage in othering was completely removed. What would the nature of social inequality in such a world be like? Would there be inequality of income, and if so, what would determine it? While powell and Menendian present an alternative social vision in their article, they do not address this issue directly. A plausible answer to such questions in contemporary America is that such a world would enact the ideal version of the American system, which, at least for the last forty years or so, has been thought of as a meritocracy. Many view othering as a distortion of this ideal vision of America.

The economics of income inequality puts this interpretation of an othering-free America in doubt in an interesting way. During the past fifty years, the per-capita output of the US economy more than doubled. This increase results from ongoing efforts to improve the efficiency by which labor effort and machines turn raw materials into finished goods and services. Economists call these improvements “technical change” because they often result from the adopting of new technologies. The widespread incorporation of computers into production processes is an example of technical change.

Suppose that we lived in a world just like our own except that the human tendency to engage in othering was completely removed. What would the nature of social inequality in such a world be like?

For at least the past century, technical change has increased the demand for more highly educated workers relative to less educated ones.⁵⁾Goldin and Katz, The Race Between Education and Technology. At the same time and partially in response to this increase in demand, the education levels of workers have risen. Elementary economics tell us that when the demand for highly educated workers increases relative to the less educated, upward pressure will be put on the wages of the highly educated relative to those of the less educated. To be concrete, college graduates earned a 32-percent premium over high school graduates in 1950. Technical change increased demand for college graduates more than high school graduates, putting pressure on the premium to rise.

In such circumstances, young people will be attracted by the prospect of higher wages and seek out more education. In other words, the supply of highly educated workers changes in response to the increased demand brought about by technical change. The supply response has been very large. At the beginning of twentieth century, only around 10 percent of American thirty-year-olds had graduated from high school and only 5 percent had graduated from college. By the end of the century, those levels had risen to 86 percent and 30 percent.

Differences in wages by education level is thus the result of an interplay between the rising demand for more educated workers due to technical change and the response of the population to acquire more education. During the first half of the twentieth century, income inequality by education level fell as rapid increases in education outpaced the increases in demand for the more highly educated. As noted above, in 1950, a college graduate earned about 32 percent more than a high school graduate. During the latter half of the twentieth century and up to the present, the relative earnings of college graduates has increased markedly. A college graduate now earns more than 100 percent more than a high school graduate.

Acquiring a college education requires a sacrifice of time and expenditure on living expenses. It is reasonable to expect that college-educated workers will be paid a premium to offset this sacrifice and expenditure. However, the 100-percent premium currently earned by college graduates is much larger than needed to compensate them for their sacrifices.

Further, note that othering is a less-compelling general explanation for an employer’s decision to hire a college graduate to fill a position rather than a high school graduate, than the explanation that the college graduate has abilities needed for the position that the high school graduate does not. The notion of considering education level in hiring as not primarily a question of animus is especially clear when we consider occupations that require specific training, such as plumbers, crane operators, nurses, and engineers.

If compensation for sacrifice and othering do not explain why college graduates went from earning 32 percent more than high school graduates to 200 percent more, what does? Is it that college graduates today are more creative, smarter, and more capable than their predecessors in the 1950s? No. Has college increased markedly in difficulty? No.

The explanation is that the increase in college graduates stopped outpacing the steady march of technical change, particularly after 1970 and particularly among men. And like any item for which a shortage is experienced, the cost of employing the college educated increased. Some reflection will show that increased inequality as a result of shortage is consistent with the idea of meritocracy. It is not ability or effort that college graduates have been rewarded for, but scarcity. From the point of view of an individual, the change in their fortune is mainly due to the actions of others.

A more concrete example may better illustrate the inherently arbitrary nature of the relationship between technical change and economic inequality. In the nineteenth century, professional musicians were employed entirely in live performances. The best musicians in the country could perform for, at most, a few thousand people per day. Since the demand for musical performance was much larger, many thousands of good musicians could earn a living practicing their art. The advent of recording and radio markedly increased the size of the audience the best musicians could reach. Earnings inequality among musicians rose as a small number captured a large share of the total audience. However, nothing about the talent of the musicians involved had changed. Again, the change in inequality had nothing to do with othering but with the impersonal and anonymous forces of technical change.

powell and Menendian leave the moral argument against the practice of othering implicit, perhaps because it is widely accepted among educated people that discrimination against others on the basis of arbitrary differences is wrong. What they perhaps failed to anticipate is that there are sources of arbitrary social inequality other than othering, such as the typical operation of the labor market. They address the moral question more directly in the solution they propose to the problem of othering, which is a society based on “inclusion and belongingness.” Creating such a society requires us to ensure that the “circle of human concern” is widened to encompass all members of society and to engage in the project of “humanizing the other, where negative representations and stereotypes are challenged and rejected.”⁶⁾powell and Menendian, “The Problem of Othering: Towards Inclusiveness and Belonging.”

If we were to be fortunate enough to create a society that fulfills powell and Menendian’s vision, we would still be left with the problem of distribution. As we have seen, wages are determined in an important way by relative scarcity, and it is difficult to see how one can reasonably use the term “meritocracy” to describe the inequalities produced by the labor market. If the reason for being opposed to othering is that othering disadvantages people for arbitrary reasons, then it is difficult to see why the same logic does not apply to market-driven income inequality. We ought to call a society that would result from combating othering an improvement, but it would still leave important questions about social inequality unanswered.

Further, contrary to powell and Menendian’s suggestion that “the right to belong is prior to all other distributive decisions since it is members who make those decisions,” it is not clear prima facie why we ought to give priority to solving problems of depravation that result from othering over those that result through the operation of the labor market.⁷⁾Ibid. A more comprehensive and philosophically grounded theory of social inequality and justice would help here. For example, the capability approach popularized by Amartya Sen and Martha Nussbaum would argue that we consider the roles played by income, belonging, social esteem, and so on in providing all members of society the capacity to achieve well-being.⁸⁾Ingrid Robeyns, “The Capability Approach,” ed. Edward N. Zalta, The Stanford Encyclopedia of Philosophy, 2016, https://plato.stanford.edu/archives/win2016/entries/capability-approach/.

If we were to be fortunate enough to create a society that fulfills powell and Menendian’s vision, we would still be left with the problem of distribution.

I believe that this exploration of sources of arbitrary social inequality in acts of othering and in the operation of the labor market serves to reorient us toward the underlying philosophical and practical questions about how we ought to conceive of a society in which institutions and resources are oriented toward the fulfillment of all. In considering arbitrary inequality, we might reflect on the first article in the Declaration of the Rights of Man and of the Citizen, the key document of the French Revolution, which holds that “social distinctions may be based only on considerations of the common good.”⁹⁾France, “Declaration of the Rights of Man and of the Citizen,” 1789, http://www.refworld.org/docid/3ae6b52410.html. Holding to this principle enables powerful arguments not only for the amelioration of othering but also for distribution of social resources in a way that takes the needs of all into full account.