Our measure of economic security is based on white economic security. This way of thinking is baked into how many of us were taught about the economy: focus on the aggregate picture, on high-level combinations of individual behaviors and experiences. Take, for example, the latest aggregate unemployment numbers for the United States. They indicate that the economy may be growing again. But dig a little deeper, and they also show that Black workers are having a very different experience than white workers are.

The economy gained 2.5 million jobs in May, bringing the net losses since February to 15.2 million. However, while overall unemployment fell slightly, it ticked up for Black people. The Black employment to population ratio went up slightly in May, but it is below 50 percent (about 4 percentage points lower than the current figure for white workers). And when a true recovery does begin, the evidence indicates that Black workers will be some of the last workers rehired. The aggregate does not fully capture the experience of everyone.

Neoclassical assumptions—while actively contested by more and more scholars and activists—still permeate public consciousness. They are taught in introductory economic courses, and they are pushed into law by political leaders whose class and race interests lie in the continued dominance of neoliberal ideology in business and policymaking. As Darrick Hamilton has put it, neoliberalism thrives on the idea that “a rising tide ‘lifts all boats,’” and yet inequities of employment, income, and wealth persist for Black Americans no matter our educational attainment. Even if economic growth lifts all baseline living standards, persistent inequity, even at a higher standard, is still unjust.

Racism is a system that manifests in norms, institutions, and policies. When economic analyses fail to account for these institutions, they perpetuate racist outcomes. Therefore, we must interrogate the assumptions often made in economic research and pedagogy that prevent economics from promoting anti-racist public agendas.

There are many such assumptions, including the first discussed above: that the aggregate picture is a good measure of the whole picture. Here, I’ll focus on two others: the assumption that value equals price, and the assumption that behavior represents the independent and rational preferences of individuals.

The price we are asked to pay for a product is assumed to reflect both how much we are willing to pay for it and a rational assessment of that product’s value. Wages are the labor market’s version of prices. In a perfectly competitive market (which does not exist), wages are equal to the marginal value a worker adds to a firm’s ability to produce. They are derived from the level of competition between workers, firm and sector productivity, each worker’s skill level, and so on.

Recently, we have seen the limits of this model, especially when it comes to the treatment of workers in the services deemed “essential” during the pandemic. The Center for Economic Policy and Research analyzed the demographics of frontline workers and found that people of color are overrepresented in these industries. Essential workers are more likely to be paid less than $20 per hour than non-essential workers. Some argue that this difference can be explained solely by differences in education and skill—with lower-skilled and less-educated workers happening to be predominately people of color who are therefore sorted into lower-skilled, lower-paid jobs. This explanation fails to account for the ways in which rules and the lack of institutional investment in work done by Black people leads to lower-paid, lower-quality work.

We often classify work as low-skill that is actually high-skill, with a particular tendency to undervalue the work done by women and people of color. Care work is a prime example. Black workers are overrepresented in the child care and social service professions, some of the fastest growing sectors in the country. Median hourly wages for home health and personal care aides was $11.37 in 2018. Yet these are jobs that require a high level of emotional intelligence, hours of training, and, in some states, an occupational license or certificate.

It is the market power of employers in many of these industries that affects wages and job quality. The median wage of food and agricultural essential workers with a high school degree was $13.07 in 2019, while the median wage of essential manufacturing workers with the same education level was $18.89. Food and agricultural essential workers are 50 percent non-white while critical manufacturing is 37 percent non-white. Market concentration in agriculture is reaching alarming levels, but with low levels of countervailing union power—the unionization rate in food and agricultural services is 8 percent, compared to 16 percent in critical manufacturing jobs. And declining union density is correlated with higher income inequality.

Persistent labor organizing brought about union power in manufacturing work. Meanwhile, our system of outdated labor standards reflects underinvestment in industries with large concentrations of people of color. The evidence is clear: the presence of strong labor institutions and public investment in standards enforcement makes a difference in outcomes for workers.

Why isn’t this investment happening in occupations with high numbers of Black workers and other marginalized communities of color? Stratification economics—which centers race in economic analysis and our understanding of inequality—argues that social group identity plays a major role in how institutions are wielded to shape economic outcomes. The intentional stratification of racial groups—in order to protect white advantages and gains—create intergenerational disparities in wealth and, therefore, in economic and political power. This, in turn, shapes both individual behavior and the design of our public institutions. In fact, the wealth gap between Black and white families is as wide now as it was in 1968. And extreme wealth, which is overwhelmingly white, has increasingly more control over political agendas as inequality increases.

This leads into the final assumption: that individual behavior is just a revealed set of preferences of a rational actor. In reality, individual behaviors reflect group identities, norms, and ideas that help uphold racist policies and institutions. This is why explanations that assume discrimination is irrational and will be competed away, like Gary Becker’s theory of racial discrimination, are insufficient. The “free market” does not inherently deliver equitable outcomes. Racism is rational when it upholds institutional arrangements that preserve white wealth and economic power—a point that applies just as much to public health and police violence as it does to the labor market. And economists who want to challenge racism must recognize the role of history, power, and institutions in shaping behavior.

Joelle Gamble is an economist and organizer, currently serving as a principal for the Reimagining Capitalism initiative at Omidyar Network. She sits on the Board of Directors of the Roosevelt Institute.