Written by: Oren M. Levin-Waldman, Political Consultant, Labor Economist and Data Analyst
In this polarized political world that we live in, especially with wokeness, it has been commonplace to claim that systemic racism is everywhere. According to the narrative it is obvious because whites have privilege that blacks don’t have and if their incomes are higher than blacks, that must be because of systemic racism. If one wanted to believe the narrative because one believes the orthodoxy, then clearly it must be true. Orthodoxies can be challenged, especially when confronted with data. This article argues that a deep dive into demographic data can greatly undermine a reigning narrative, and in so doing provide candidates for office a basis upon which to discuss issues that may well resonate with voters.
Instead of simply claiming that the labor market is not racist, candidates would do better to demonstrate how it is not. Data from the Current Population Survey (CPS) shows that between 2012 and 2020 earnings for white workers declined while earnings for blacks and Latinos increased. Although it is true that white workers overall are earning more than black workers, it is also the case that when controlling for educational attainment, skills, and the specific occupations one works in that picture gets muddied. If a disproportionate percentage of minorities go into social welfare provision and social work, then their earnings will be lower than those going into STEM occupations which pay more. STEM requires considerably more skills and also that one will be training in math and sciences. If one group is disproportionately going into lower paying occupations, one cannot easily make the claim that it is because of systemic racism. If the same group is going into these occupations because educational requirements are less rigorous than those required for say law, engineering, computer science, and medicine, then clearly other factors are at work. A key factor might just be general education and the quality of schools one hails from. If one grows up in a wealthy neighborhood, the odds are that person will have a higher quality of education than the person who grows up in a poorer neighborhood. But that becomes an issue mainly of economic and social class.
When we look at specific occupations we will see that the gap between whites, blacks, and Latinos narrows considerably, thereby undermining the reigning narrative of systemic racism. To the extent that this is true, the implications may be enormous. Instead of joining into the narrative that each party is peddling, candidates would do better to discuss the issues that are important. Discussing the issues based on data should enable those candidates that argue that systemic racism does not exist to avoid being labeled racist themselves. Moreover, candidates that can discuss issues on the basis of data are in a better position to offer, once elected, serious policy proposals that can serve the interests and needs of their constituents.
The Wage Story
The theory of competitive markets holds that workers are paid their worth according to laws of supply and demand. If we assume all workers to be the same in terms of skills, endowments, attributes, etc., and that employers are also the same in terms of their demands, then wage determination is rather simple: more workers will be hired the lower the wages are for each worker. If one worker earns more than another among the same group, it is perhaps because his/her marginal utility, or marginal revenue product from his/her labor is greater, thereby leading to greater productivity. Indeed, the basis for giving workers raises is predicated on the assumption that for each additional unit of labor there will be an increase in marginal revenue product. The wage story, however, is more complicated because not all workers are the same and not all employers’ skills requirements are the same. Some employers require more skills than others, and if those skills are in short supply, they will then be forced to offer higher wages. There are in fact multiple supply and demand curves in the marketplace.
With data, we can counteract a reigning narrative that there is systemic racism in the labor market. The data is drawn from the Integrated Public Use Micro-Series (IPUMS) Current Population Survey (CPS). The wages shown in Table 1 are for those workers specifically between the ages of 18-72 working full-time. They are shown for all workers and by race and ethnicity.
Table 1 Median Wages According to Race and Ethnicity in 2020 Dollars
Over the past several decades, wages in the U.S. and much of the global economy have stagnated. Still, between 2012 and 2020, individual median wages of all workers increased by 21.9. The increase in median wages was slightly higher between 2016 and 2020 than between 2012 and 2016. When looking at wages by race and ethnicity, we can see these gains have not been shared equally. Whites still earn more than blacks and Latinos, but the median wages of Latino workers increased more than they did for White workers. Although Latinos earn less than blacks and considerably less than whites, their median wages increased by 23.1 percent between 2012 and 2020 compared to 21 percent for whites and 18.5 percent for blacks.
Those who argue that there is systemic racism in the labor market will focus on what percentage of white wages blacks and Latinos are earning. In 2012, the median wages of black workers was 83.3 percent of the median wages of white workers, and the median wages of Latinos was 69.4 percent of the median wages of white workers. In 2020, the median wages of black workers decreased to 81.6 percent of white workers, but the median wages of Latino workers increased to 71.4 percent of white workers. And yet, none of this really says anything about the occupations and industries that one group of workers is employed in relative to others, or that educational attainment of one group compared to another.
Table 2 Median and Wages by Educational Attainment in 2020 Dollars
It certainly becomes clear that median wages increase with education, which can also be used as a proxy for skills levels. Although the biggest increases in wages between 2012 and 2020 were among those with less than a twelfth grade education, those with at least a BA degree are still earning more, and those with an advanced degree that much more. Consider that in 2012 the median wages of a College graduate were 32.1 percent greater than a High School graduate and the median wages of those with advanced degrees were 48.6 percent higher than the median wages of those with only a BA degree. The median wages of those with an advanced degree were 96.4 percent higher than those with a high school degree. By 2020, the median wages of a college graduate were only 28.6 percent higher than the median wages of High School graduates, but the median wages of those with advanced degrees were now 51.1 percent greater than the median wages of those with BA degrees. Those with advanced degrees were only earning 94.3 percent more than those with High School degrees, principally because the percentage increase in the median wages of High School graduates were greater than for those with higher levels of educational attainment. Goldin and Katz (2008) have noted that as the economy has changed towards one that places a greater emphasis on technical skills, the high school premium has declined. But it would also appear that the BA premium has also been declining as well. A BA alone appears not to have as much value as it once did. We can see some of that alone just between 2012 and 2020. The point is, as it has been for some time now, that there are returns to education, and as much as that is still the case the greater return is from advanced degrees, which include MAs, Ph.D.s, and professional degrees in fields of medicine, dentistry, and law.
At the same time, if we look at each group according to educational attainment, we may be seeing some interesting things.
Table 3 Median Wages by Education
In 2012, those with less than a 12th grade education were earning $20,254 and in 2020, they were earning $25,000. The median wages of white workers with less than a 12th grade education was $20,524, which was 1.3 percent above the median wages for all workers with less than a 12th-grade education. The median wages for black workers and Latino workers with less than a 12th grade education was $15,498 and $20,479 respectively in 2012, which was 76.5 percent and 101.1 percent respectively of all workers’ median wages with less than a 12th grade education. In 2020, the median wages of white workers with less than a 12th grade education was $25,725, compared to $24, 440 and $25,000 for black and Latino workers respectively with less than a 12th grade education. Now white workers were earning 102.9 percent of all workers with less than a 12th grade education, while black workers with less than a 12th grade education were earning 97.8 percent of all workers with less than a 12th grade education. Latino workers with less than a 12th grade education were earning 100 percent of all workers with less than a 12th grade education.
There are a couple of different ways to look at this. The first and obvious way is to say that white workers with no skills, are still earning more than black and Latino workers with no skills. That is clearly the most objective way to view it, but it doesn’t say why some of those differences may exist. Higher wages among say low-wage workers may be due to age differences, in that some have more seniority. Educational attainment may be one attribute, but there may be others that are perhaps equally as important. In 2012, black workers with no skills are earning 75.5 percent of white workers with no skills and Latino workers are earning 99.8 percent of white workers with no skills. In 2020, black workers were earning 95 percent of white workers, and Latino workers were earning 97.2 percent of white workers. And yet, another view of this tells us that black workers with less than a 12th grade education saw considerable improvement with regards to their ratio of the median wage of white workers, while Latino workers saw a slight drop. Median wages for black workers in this educational group increased by 25. 9 percent from 2012 to 2020 compared to an increase of 10.9 percent for all workers, 11.0 percent for white workers, and 14.3 percent for Latino workers. In other words, there were significant wage gains for black workers with less than a 12th grade education, even if they were earning less than whites and Latinos.
As we move up the educational latter, serious differences in educational attainment begin to emerge. In 2012, the median wage of all workers with a high school diploma was $31,506 and the mean was $36,506. In 2020 their median wage was $35,000, an increase of 10.7. In 2012, the median wages for whites, blacks and Latinos with a high school diploma were $38,355, $25,880, and $27,005 respectively. With regards to median wages, whites were earning 121.7 percent of all workers with a high school diploma, while blacks and Latinos were earning 82.1 and 85.7 percent respectively. Blacks with high school diplomas were only earning 67.5 percent of whites with a high school diploma while Latinos were earning 70.4 percent of whites with a high school diploma. Latino workers in this educational cohort were actually earning more than black workers in this cohort.
By 2020, the median wages for those with a high school diploma were $36,200 for whites, $30,000 for blacks, and $31,200 for Latinos. Now, blacks with a high school diploma were earning 83.3 percent of whites and Latinos were earning 86.7 percent of whites. Whereas the percentage increase in median wages was 11.1 percent for all workers with a high school diploma, the percentage increase for both black and Latino workers with a high school degree was 15.9 and 15.5 percent respectively. The median wages of whites with a high school diploma actually decreased by 5.6 percent. Both black and Latino workers with high school diplomas actually made gains relative to whites, in that their percentage share of white workers with high school diplomas was much greater than it was in 2012.
For those workers with a four-year college degree, we can see that gains in median wages were greater between 2012 and 2020 for blacks and Latinos than for whites. In 2012, the median wages for workers with a BA degree were $41,633, and in 2020, they were $45,000, which was an increase of 8.1 percent. In 2012, the median wages for whites with a BA degree were $43,208 compared to $37,025 for blacks and $37, 650 for Latinos with a BA degree. White workers’ median wages were 103.8 percent of all workers with a BA degree while black and Latino workers’ median wages were 88.9 and 90.4 percent respectively of all workers’ median wages. Blacks with a BA degree, then, had median wages that were 85.7 percent of the median wages of white workers with a BA degree while Latino workers with a BA degree had median wages that were 87.1 percent of white workers in the same educational cohort. In 2020, the median wages of whites with a BA degree were $46,000 compared to $40,000 for both blacks and Latinos. Whites with a BA degree had median wages of 102.2 percent of the median wages of all workers, while blacks and Latinos had median wages of 88.9 percent of all workers and 87.0 percent of white workers with a BA degree. Whereas median wages of whites with a BA degree increased by 6.5 percent, they increased by 8.0 percent for blacks and 6.2 percent for Latinos. Although the median wages of white workers with a BA degree continued to be higher than those of blacks and Latinos, the median wage gains of blacks with a BA degree were greater. And yet, these comparisons tell us little about the career choices made by those in each respective educational cohort. Social workers and teachers aren’t going to be earning as much as accountants and engineers. Therefore, more needs to be probed on this question.
In 2012, the median wages of all workers with an advanced degree were $61,886, and in 2020, they were $68,000, an increase of 9.0 percent. The median wages of white workers with an advanced degree were $61,887, compared to $55,135 for blacks and $50,635 for Latinos. The median wages of whites with advanced degree were the same as that of all workers with an advanced degree. The median wages of blacks with an advanced degree were 89.1 percent of the median wages of whites with advanced degrees and the median wages of Latinos with advanced degrees were 81.8 percent. In 2020, the median wages of whites was $69,000 which were now 101.5 percent of the median wage of all such workers. The median wages of blacks with an advanced degree were $56,940 compared to $55,000 for Latinos. In 2020, then, the median wages of blacks with advanced degrees were 82.5 percent of the median wage for whites with advanced degrees and the median wages of Latinos with advanced degrees were 79.7 percent of the median wages of whites in the same cohort. Therefore, it would appear that while white workers saw median wage gains, black and Latino workers did not. Median wages in this group increased by 11.5 percent for whites, but they only increased by 3.3 percent for blacks and 8.6 percent for Latinos. Again, the operative question here is just what professions are whites in that their median wages increased while median wages of blacks and Latinos, perhaps in different professions, did not.
We could hypothesize that whites perhaps earn more than blacks because larger percentages may have higher levels of educational attainment and be in higher paying occupations and industries. The real issue is skills levels and there is no way to measure that from this type of census data. Rather levels of educational attainment are serving as a proxy for skills, but it is by no means a perfect one. One with an advanced degree is social work is not likely to be earning as much as one with an advanced degree in engineering and/or finance. With advanced degrees we are lumping together highly paid attorneys and physicians with not as well paid teachers and/or social welfare professionals. And yet, when looking at those who apparently don’t have skills, as measured by those with less than a 12th grade education, the wage differences between whites, blacks, and Latinos don’t appear as large as they were among the other educational cohorts.
Narrowing the Gap
When we look at some selected occupations, which may require varying degrees of skills and levels of education, the gap between whites, blacks, and Latinos narrows. In some cases, the median wages of blacks and Latinos is even higher than whites. And in those instances where whites are earning more than blacks and Latinos, the share of white wages earned by blacks and Latinos increased between 2012 and 2020.
Table 4 Comparative Earnings of Selected Occupations in 2020 Dollars
In 2012, the median wages of black Accountants are higher than the median wages of whites. Black pharmacists and nurses also appear to have higher wages than white pharmacists and nurses. The median wages of black lawyers and judges are higher than white lawyers and judges, although the wages of white lawyers and judges are higher at the 90th percentile. The median wages of black dentists also appears to be higher. The next table presents the share of white workers’ wages earned by black and Latino workers
Table 5 Share of White Workers’ Wages in Selected Occupations
Although the share of black and Latino wages of white workers’ wages decreased in some of these occupations from 2012 to 2020, there were some significant increases as well. The black workers’ share of white workers median wages in engineering decreased by 7.7 percent. We see similar increases among black physicians and pharmacists. Given the gains in black and Latino wages, it is very difficult to make the claim of systemic racism, or at least to make it stick.
To a certain extent, these selected occupations can be divided into two categories: higher paying and lower paying. Even if we assume that the lower paying occupations require a minimum of a four year college degree, those at the higher end require advanced degrees, many of which are in STEM. That is, a masters of social work, which may be the requirement for social and welfare workers, is not the same as a masters in mathematics, physics, or even finance. Not all degree institutions are equal in terms of their rigor. Although white workers still earn more than do black and Latino workers, the gap between them appears to narrow as they move up the educational latter. In some cases, when controlling for occupation and educational attainment, black and Latino workers in some cases do earn more than their counterparts. Moreover, it becomes clear that both black and Latino workers have made tremendous strides. Given that, it might appear somewhat spacious to continue trotting out the claim that there is systemic racism in the labor market. Actually, the claim is that there is systemic racism in society at large, which includes the labor market, but that doesn’t appear to be entirely supported from the data. If it turns out that blacks for the most part gravitate towards certain occupations which happen to pay less, is that proof that they are being discriminated against? Why wouldn’t they be discriminated against in say social work, nursing and teaching?
Nevertheless, there are implications for where black and Latino workers stand in relation to whites when it comes to wealth creation. A path to wealth accumulation is employment in high paying occupations. Many are found in the fields of engineering, mathematics, science, and medicine. All require a high degree of proficiency in mathematics, and/or English language skills. Blacks account for approximately 14.6% of the U.S. population but are substantially underrepresented as accountants (7.8%), chemists (6.8%), dentists (1.7%), engineers (4.2%), biological scientists (3.1%), mathematics (7.6%), physicist (1.4%), optometrist (3.3%), pharmacist (7.0%) and physicians (6.0%) - all high paying occupations. The occupations in which blacks represent at least 10% of workers - such as dancers and dance teachers (21.1%), and social and welfare workers (20.3%) musicians and music teachers (14.1%), recreation and group workers (11.7%) and personnel and labor administrators (10.8), nurses (10,3%) are all relatively low paying.
It is true that there may be occupational segregation in the labor market, but is that necessarily the same as systemic racism? The data also makes clear that many blacks with advanced degrees tend to be more concentrated in lower paying occupations like nurses, social welfare workers, and teachers. Higher paying occupations can generate more wealth in terms of property purchased and retirement savings.
When we ask the question of who is more likely to be earning at the 90th percentile, race becomes less of a factor, rather educational attainment and the specific types of occupations are. Through a logistical regression analysis we can establish correlations between variables and the probabilities that individuals who are earning wages at the 90th percentile will have certain attributes over others. We can see this in the next table with a logistical regression analysis. The dependent variable is earning a wage at or higher than the 90th percentile.
Table 6 Logistical Regressions: Dependent Variable of Having earned equal to or greater than the 90th Percentile
The coefficients from the above table make it clear that those with the strongest probabilities for earning wages at the 90th percentile or above are those with graduate and/or professional degrees, thereby reinforcing the returns to education data that has long been out there. Although the coefficient is weaker in 2016 than in either 2012 or 2020, they still effectively say that those who have graduate and/or professional degrees are more likely to be earning at that percentile or above. Also, those who are professional/managers are more likely to be earning at the 90th percentile or above. Consistent with what we were saying earlier, higher paying industries like manufacturing have a strong positive coefficient which is statistically significant in 2020 while lower paying professions like social and welfare workers and teachers have strong negative coefficients which are statistically significant (except for 2016) for earning at that percentile. On the question of whether whites are more likely than blacks to earn wages at that percentile, the only positive coefficient which is statistically significant is for whites in 2012. It weakens over time, however, and it is not statistically significant in 2020. Blacks have negative coefficients for earning at that percentile, but they aren’t statistically significant until 2020, whereby the coefficient is more strongly negative. And yet, there are positive coefficients for black professional/managers, although it isn’t statistically significant in 2016. Blacks with graduate and/or professional degrees has a strong positive coefficient for earning wages at this percentile in 2016, but it is not statistically significant. It is negative and statistically significant in 2012.
As the strongest coefficients are connected to educational attainment, industry and occupational categories, it actually becomes more difficult to make the claim that the labor market is racist. Racial type is not on the face of it a good predictor of who is likely to be earning wages at the 90th percentile or above.
Conclusion
Even though whites still earn more than blacks and Latinos, they don’t in all occupations. As the data shows, wage gains of blacks and Latinos have in fact been greater. And yet, as the data demonstrate that when controlling for educational attainment and occupation, the gap between different racial and/or ethnic groups does narrow considerably. The larger point of this analysis was to demonstrate why data is essential for challenging a narrative and why it could assist certain candidates attempting to run a campaign on the issues and specifically against woke orthodoxy. The data here does not fit the prevailing narrative that has come to dominate woke culture, and which has served to maintain elite control. On the contrary, it challenges the reigning narrative, and serves as an example of the power of data that should be a staple of any campaign for public office. Candidates armed with a demographic analyses based on sound data can find themselves in a better position to counter orthodoxies that merely divert attention from real issues. As such, they are in a better positions to offer serious policy solutions to pressing problems.
Written by: Oren M. Levin-Waldman, Political Consultant, Labor Economist and Data Analyst
Oren M. Levin-Waldman is a VIP Specialist at PoliticalVIP. As a Labor Economist and Former Professor, Levin-Waldman has served at prestigious universities like Rutgers University, Newark and New School University in New York.
Levin-Waldman’s expertise includes demographic and economic data analytics, labor economics, voting behavior statistics and researching. Just a few publications authored by Levin-Waldman are listed here:
References
Goldin, Claudia and Lawrence F. Katz. 2008. The Race Between Education and Technology. Cambridge, MA/London: Belknap Press of Harvard University Press.
