President Lyndon B. Johnson signs the Civil Rights Act of 1964 (Source: Wikimedia Commons)
by Hadas Mandel and Moshe Semyonov
Following the implementation of the 1964 Civil Rights Act and the enactment of affirmative action policies in the United States, both the educational level and the relative income of black men and women rose. Consequently, earnings disparities between blacks and whites declined; this trend continued until the end of the 20th Century, for both genders.
But from the turn of the new millennium, the trend reversed for women and men for the first time since 1970. In light of the continual convergence in black and white pay and the gender differences in the size and sources of the pay gaps, this uniform reversal of the trend is intriguing.
To study these trends, we examined trends in earnings inequality from the period following the Civil Right Act (1970) until 2013, using the IPUMS data. The findings support several expectations, while leaving others open to debate.
First, as shown in Figure 1 below, our expectation of much smaller racial gaps between women than between men was firmly supported. In addition, the unexplained portion of the pay gap – the indicator we used as a proxy for racial discrimination – was also found to be much smaller among women than among men. In fact, from 1980-2000, there were virtually no racial pay gaps among women – a finding which contradicts the idea of a double disadvantage from being a member of both disadvantaged gender and racial groups, and supports the argument that black men are the prime target of economic discrimination. This is because black men (but not black women) pose a threat to whites; a threat that results in negative stereotypes of black men as violent, criminal, hostile, unreliable, and lazy.
Figure 1: The Gendered nature of racial inequality
Figure 1 also shows that earnings inequality in the US is more gendered than racialized; women of both races share a considerable earnings disadvantage, relative to men. This finding is in line with the argument that women of different races have much more in common than men do; an argument based on the universal tension between family obligations and paid work that black and white women share. Thus, the “racial advantage” of black women (compared to black men) should be understood, first and foremost, within the context of their “gender disadvantage.”
Despite the differences between men and women, the findings underscore a striking similarity across women and men in the over-time trend of the racial earnings gap. Moreover, the over-time trend in the racial earnings gap is closely related to over-time changes in government reforms. Our findings underscore not only the reversal of the trend of declining racial disparities during the new millennium, but also the significant enlargement of the unexplained portion of the gap, our proxy for racial discrimination.
Despite the differences between men and women, the findings underscore a striking similarity across the two gender groups in the over-time trend of the racial earnings gap. Moreover, the over-time trend in the racial earnings gap is closely related to over-time changes in government reforms. Figure 2 displays a disaggregation of the racial gaps into explained and unexplained components from 1970, and also the trend during the first decade of the new millennium. The figure underscores not only the reversal of the trend of declining racial disparities during the new millennium, but also the significant enlargement of the unexplained portion of the gap, our proxy for racial discrimination.
Figure 2: Similarities in the over-time trend of the racial pay gap
Intrigued by these findings, we examined the trend and offer several alternative explanations for its reversal, summarized in Table 1 below. First, we examined whether this reversal was a post-recession effect, but found no evidence to support this. The “Great Recession” officially began at the end of 2007 and ended in June 2009. Although its consequences may have begun earlier and may still persist today, we found no evidence for a dramatic change in the years before and after the recession, but rather of a continuing trend of widening pay gaps over the years, one that finally peaked in 2013 for both gender groups.
Second, we examined the possible effect of the rise in income inequality in the US. We studied this because blacks are overrepresented at the bottom, and whites overrepresented at the top of the wage hierarchy. Thus, a rise in overall income inequality (increasing the distance between top and bottom wages) could result in an increase in the gap between blacks and whites. This rising income inequality thesis is supported in the case of men, but not women, suggesting that income inequality is more detrimental to the earnings of black men than to others.
Third, we examined whether educational differences contributed to the enlargement of the pay gaps between black and whites. We found no support for this expectation with both genders. However, we did find that the increasing economic rewards for working hours contributed to the enlargement of the gaps between black and white men, because the latter group tended to work fewer hours than the former group. The process of occupational segregation – one of the major sources of pay differentials – might also have contributed to widening the gap. This is so because the role of occupations in explaining the gap has widened since the beginning of the new millennium, for both gender groups. Likewise, the continual racial occupational integration that characterized the period after the 1970 reforms has come to a halt in the recent decade.
Fourth, we entertained the possibility that an increase in market discrimination took place. Although an increase in the unexplained portion of the racial gaps, and stagnation in occupational desegregation, are only implicit indicators of discrimination, simultaneous changes to both alongside changes in government reforms over the course of time may be an indication of growing discrimination in the American labor market.
We are not the only ones to link pay discrimination and slowed desegregation to government reforms. Sociologists Kevin Stainback and Don Tomaskovic-Devey showed that between 1980 and 2005 – when the regulatory environment switched to advancing diversity programs rather than affirmative action policy – the pace of improvement in blacks’ access to good jobs, and the decrease in racial occupational segregation, largely declined. Such a switch was also described by sociologist Silvia Cancio and colleagues, who attributed the rise in economic discrimination against blacks after 1980 to the retreat from the enforcement of antidiscrimination policies during the Reagan and Bush Sr. presidential administrations. Sociologist George Wilson and colleagues have shown that after public sector privatization of employment practices, racial pay gaps widened. The findings presented by our research lend indirect support for this argument by revealing the affinity between government reforms and fluctuations in racial gaps, following the Civil Rights Act and in the new millennium.
We have no direct way of examining whether the reversal of the trend is a one-time deviation from the general trend of declining racial earnings disparities, or a real and consistent change. It is our hope that future research will provide further direct tests of the explanations entertained in this article, and will provide answers to the questions that remain.
Hadas Mandel and Moshe Semyonov are Professors at Tel-Aviv University. This article is based on their paper ‘Going Back in Time? Gender Differences in Trends and Sources of the Racial Pay Gap, 1970 to 2010,’ published in the American Sociological Review.