White Working-Class Men in a Changing American Workforce

 What's NEXT?

How can it be that white working-class men earn $10,000 more than non-college-educated women and minorities, yet feel aggrieved? Commentators noted last November that the key constituencies in President Trump’s win were white, working-class men frustrated by their economic and sociocultural status. A common interpretation is that these workers feel left behind as the new economy charges ahead. Can it really be that white men, historically the most privileged demographic in America, are now disadvantaged?

A new paper by economist Stephen Rose digs into the heart of the matter. By analyzing Census data on wages, education, and demographics from 1960 to 2014, he demonstrates that the plight of working-class white men is all about relative change. Half a century ago, simply being a white male high-school graduate substantially improved one’s chances of out-earning women and people of color. As a result, white men without a college degree held most high-paying jobs in the country.

Today, that world is a distant memory. White men constitute a smaller portion of the population; higher educational attainment has increased across age and gender groups; the achievement and wage gaps between both race and gender have narrowed. Further lengthening the playing field, the wealthy and upper-middle class went from receiving 30% of all income in 1979 to 61% in 2014. The result: a high-school-educated white male still earns more than his 1960 counterpart, as well as women and people of color today, but these improvements pale in comparison to the economic gains made by these same women, people of color, and skilled professionals.

White Working-Class Men in a Changing American Workforce is the latest in a series of ahead-of-the-curve, groundbreaking pieces published through Third Way’s NEXT initiative. NEXT is made up of in-depth, commissioned academic research papers that look at trends that will shape policy over the coming decades. In particular, we are aiming to unpack some of the prevailing assumptions that routinely define, and often constrain, Democratic and progressive economic and social policy debates.

In this series, we seek to answer the central domestic policy challenge of the 21st century: how to ensure American middle-class prosperity and individual success in an era of ever-intensifying globalization and technological upheaval. It’s the defining question of our time, and one that, as a country, we’re far from answering.

Each paper dives into one aspect of middle-class prosperity—such as the workplace, education, retirement, or the safety net. And by doing that, we’ll be able to help push the conversation toward a new, more modern understanding of America’s middle-class challenges—and spur fresh ideas for a new era.

Jonathan Cowan
President, Third Way

***

In 2015, the annual wage gap between the typical white worker and black worker was $10,000, and the gap between the typical male and female worker was $10,000.¹ So how is it that in the same country, a populist political movement to the right is blamed on the misfortunes of those who are seemingly doing better than their minority or female counterparts? In this paper, I describe that what is happening to the white working class that may explain this phenomenon. Some in this group are suffering from what sociologist Samuel Stouffer described in the 1940s as "relative deprivation." This is the notion that legitimate life expectations seem blocked for societal or economic reasons.

Many commentators have noted that members of the “white working class,” especially men, are angry at the current state of the country and are demanding changes. Their discontent is supposedly rooted in the decline in good-paying manufacturing jobs, which has led to layoffs, inability to get comparable jobs, and large income declines. Politically, they are cited as the ones who propelled the Tea Party movement to prominence and provided the votes for Donald Trump to win the presidency.

This paper examines the magnitude of decline for this group of Americans. In 1960, 55% of the top-paying managerial and professional jobs in the U.S. economy were filled by white men without college degrees. Fast forward to 2014, and that group held only 14% of top-paying jobs.

Arlie Hochschild in her book Strangers in their Own Land ran into many people complaining about being left behind. She summarized their arguments as saying that others had “cut into the line” ahead of them unfairly. Have these traditional breadwinners, non-college-educated white men, genuinely been cheated by the government in favor of policies that help women and minorities at their expense? The answer is no. But in the second section of this paper I detail how their relative standing has fallen.

Indeed, the white male working class has exhibited a dramatic fall in status from 1960 to today. But the reasons have to do with a changing economy and the opening of opportunity to large segments of the population who were denied the chance for much of the country’s history. Specifically, college-educated women and people of color have increasingly earned these top managerial and professional jobs because they began earning bachelor’s and graduate degrees.

Source: Author’s computations from 1960 Census and 2015 and 2014 March Supplements of the Current Population Survey (CPS)

The dramatic increases of white women and people of color were driven by the gradual opening up of these jobs to people with a college degree, regardless of race or gender: an extra 18 percentage points for white women, an extra 14 percentage points for African-Americans and Hispanics, and an extra 8 percentage points for Asians and other races. At the same time that the opportunities for non-college white men were contracting for the best jobs in the economy, their bread and butter blue-collar jobs were also drying up. In addition to fewer jobs, the relative pay of these jobs versus white-collar jobs was declining.

From the point of view of non-college white men, these shifts meant that they lost many of their places at the top of the pecking order as compared with their fathers and grandfathers. Some of them failed to realize that the world around them had changed dramatically, as more people were college educated, and a declining share was without a high school diploma or GED. What they saw was that they had as much or more education as their fathers, but that they were missing out on the better life of those in the upper middle class.

What also was setting back non-college white men was that more than 25% of children of non-college-educated parents got four-year degrees. Of parents who didn’t have a high school diploma, 80% of those children at least completed high school, many had some postsecondary education, and a modest portion got a college degree. The people today who have finished high school but didn’t attend any college are those who didn’t participate very much in this huge educational attainment leap.

What many non-college white men may not appreciate is that their earnings still are significantly higher today than those of white women and people of color with the same level of education. Instead of seeing that they are performing better than others with the same education, some are using a reference based on a nostalgic vision of how well their fathers and grandfathers did in the distant past.

This paper will track the major changes that occurred from 1960 to 2014. The first section will present data on huge demographic and educational shifts that distinguish the modern economy from that of the past. One of the oddities in the modern discussion of the white working class is that it is mainly based on education—not having a college degree—rather than the more traditional definition based on blue collar occupations. In the second section, I will argue that the traditional definition of working class as being blue-collar manual workers is too narrow, and I will offer an alternative, expanded definition based on occupations.

The third section will look more closely at the huge increases in earnings and income inequality. I will show that economic growth has indeed lifted all boats, but that the divide between the upper middle class—composed mainly of college-educated people working in managerial and professional jobs—and the rest of society has widened. So this analysis will look at both absolute and relative changes, taking into account that sociologists have found that relative status decline can lead to strong feelings of loss even if it isn’t accompanied by declining living standards.

Part I: Demographic and Economic Changes

Few people realize how white a nation we were in 1960, as nearly 8 out of every 9 workers were non-Hispanic whites (Figure 2). Another 9.4% were African-Americans, leaving just 2.4% for Hispanics of all races, Asians, and other races. By 2014, the white share of workers had dipped below two-thirds (65%), with only a small increase in the African-American share of the work force (11%). The biggest gainers were Hispanics, whose share of the labor force jumped more than 14 percentage points to nearly 16% in 2014. 

Source: 1960 Census and 2013-2015 Current Population Surveys.

The small numbers of Hispanics and Asians in 1960 mean that it is not statistically helpful to compare their statuses in 1960 and 2014. Therefore, the changing experiences of people of color will be presented as the different trajectories of whites and African-Americans.

As education has become more central to preparing for better employment success, it is sometimes used as a proxy for defining the working class as those who don’t have a four-year college degree. In particular, this approach is used in opinion polls that have information about education and incomes but not about occupations. Figure 3 shows the huge change in the educational attainment of American workers from 1960 to 2014. Remarkably, half of the labor force (51%) hadn’t finished high school in 1960, while only 10% were college educated. This meant that the 27% with a high school diploma were in the third quartile (four ordered groups from least educated to most educated) of the educational distribution.

Source: 1960 Census and 2013-2015 Current Population Surveys

As all of the workers who were born before World War II aged out of the labor force, the share of those without a high school diploma plummeted from 51% of all workers in 1960 to 9% in 2014. By contrast, the share of those with a bachelor’s or graduate degree grew from 10% to 35%, and the share with some postsecondary education but no four-year degree grew from 12% to 29%. Thus, if the lack of a college degree is a defining characteristic for being in the working class, then the size of the working class shrank from 90% of workers in 1960 to 65% in 2014.

As the amount of education grew, the place of those with just a high school diploma shifted from spanning the 52nd to 78th percentiles in 1960 to running from the 10th to 36th percentile of the education distribution in 2014.² Or, put in a different way: the high-school-only educated in 2014 were basically in the same place on the educational ladder as those who finished 8th to 11th grade in 1960. Similarly, the change in the relative position of the “some college” group was also large: going from the bottom half of the top quartile to straddling the median. Since many of these people were the first generation in their families to go to college, their aspirations for what college attendance would get them was high. And as will be shown, some college education in 1960 provided more opportunity for top managerial and professional jobs than it did in 2014.

People with more education have two advantages over those with less education. First, every advanced industrial country spends huge amounts of money on education because they believe that these years of training lead to skills that translate into higher economic productivity. Approximately one-half of college graduates major in fields that are related to their occupations, and all students benefit from the practice of task completion—taking harder and harder courses and knowing how to get a reasonable grade.³ Others describe this as gaining the crucial skill of learning how to learn.

Second, because higher education now plays such a crucial role in qualifying for the highest-paying jobs, the most talented people will be drawn to staying in school. Talent has many dimensions, and doing well at school is just one of them. But staying in school develops more discipline, more awareness of different types of thinking, and more self-confidence. Consequently, children of non-college-educated parents, on average, face fewer intellectual challenges and develop fewer intellectual skills.  

Another crucial element in the increasing educational attainment of young adults is the big gains by African-Americans. As Figure 4 shows, more than 75% of African-Americans had not finished high school in 1960 versus 53% of whites. In fact, a majority of African-Americans (55%) had not made it to 9th grade (versus 30% of whites). By 2014, just 5% of whites and 10% of African-Americans had not finished high school.

At the top end of the education ladder, college completion was relatively rare in 1960 throughout the population—just 9% for whites and 3% for African-Americans. By 2014, these numbers had changed to 38% for whites and 24% for African-Americans. While this is a sizable gap, it marks a significant contrast with 1960. As will be shown below, the population of African-American college graduates now competes for a sizable share of the best jobs in the economy.

The final demographic shift affects the relative labor market participation of men and women: in 1960, women only constituted 33% of the labor force; by 2014, they had almost reached parity, making up 47% of the labor force. This was accomplished by big changes in the employment-to-population ratios of each gender. In 1960, 89% of males aged 23 to 69 were in the labor force; the comparable figure for women was 39%. By 2014, while women were massively increasing their attachment to working, men were retiring earlier or dropping out of the labor force. Consequently the employment-to-population ratio for 23-69-year-old women rose to 64%, and for men it fell to 73%.

Many commentators have been alarmed by the rise of “missing men” in the labor force and argue that our relatively low unemployment is deceptive because of the large number of prime-age men who have dropped out. As Appendix Figure A1 shows, this problem hits less-educated men more significantly and hits African-Americans more severely than whites. A major 2016 study from President Obama’s Council of Economic Advisors shows that the decline has been fairly steady since 1965, with much of the action occurring during economic downturns.⁴ They don’t find that government assistance programs (e.g., disability insurance and welfare) explain the decline but argue that it is a changing economy with a lower demand for unskilled labor that is the major culprit.

Part II: Occupations

Demographic and educational shifts are not the only changes affecting the employment status of white working class men. So have changes in occupations, which this section will explain. First, some background on a framework for looking at occupational changes:

Government socioeconomic surveys that are used in this study have between 350 and 450 separate occupational designations. From work over the last 20 years, I have developed procedures to combine these detailed occupations into 21 hierarchical groups.⁵ In this study, I further combine these occupations into eight subcategories and into three major tiers, which are described in the box below.

The top-tier managers and professionals are divided into two subcategories: The first sub-group in this tier consists of a variety of high-paying jobs, from mid-level business managers (but not including “managers” in retail and food outlets) to those in the corner offices of large corporations. The middle tier includes an array of business professionals in accounting, finance, real estate, and sales (but not sales clerks). And finally, there are many different types of technical professionals—architects, engineers, scientists of various types that don’t teach in universities—in the bottom tier.

Paradoxically, some occupations with the highest concentration of workers with a graduate degree have relatively modest earnings. These “intellectual and caring professionals” are largely motivated by the nature of the jobs and less by the pay. Members of clergy, other religious workers, K-12 teachers, social workers, and other similar occupations take these jobs on the basis of making a difference and don’t receive pay anywhere near what people with their levels of education get in the top tier business and technical managers. Finally, artists of various sorts—writers, actors, painters, musicians, etc.—pursue their passion knowing that making a decent living is difficult to accomplish and that only a precious few become big stars with huge salaries. 

The bottom tier combines all occupations that have few, if any, higher educational requirements. The skilled blue-collar workers and blue-collar supervisors combine relatively high levels of manual skill and leadership roles and are paid much higher than the other occupational groups in this tier. In many ways, these are middle-skill jobs but are included in this tier in order to keep all manual workers in the same group. While supervisors usually are prohibited from being members of labor unions, skilled blue-collar workers were most of the first workers to be unionized in the U.S. In fact, it wasn’t until 1955 that the labor movement became united when the mostly craft unions of the American Federation of Labor joined with the industrial unions of the Congress of Industrial Organizations (CIO), which had only been created in 1935.

Most of the CIO unions were populated by members of the sixth category above—workers in steel, autos, auto parts, and other manufacturing industries. Unionization also was prevalent in other large conglomerations of manual laborers in trucking, utility workers, construction, and the dock workers who unloaded and loaded cargo. These successful worker organizations were able to win compensation packages that were much higher than the compensation of other manual laborers who weren’t in large work sites. However, the combination of new technologies (containerization on the docks, mechanization in the factories, strip mining, and efficient long-haul transportation that led to global supply chains), lower economic growth after 1971, and less government support for collective bargaining eroded the power of unions and decreased the number of manual workers.  

Middle-tier jobs consist of a variety of non-professional, non-manual, and non-low-skill jobs. The vast majority of these workers today have at least some postsecondary education. The first group of jobs consists of non-professional self-employed and white-collar supervisors.

Historically, the working class was first defined as all those who had nothing to sell but their labor. This distinguished them from peasants, who were tied to the land, and from business owners. But as industrial economies grew, a growing share of the workforce became professionals and business managers. While many of these workers relied solely on wages and salaries, their working conditions and allegiances were much different from most other workers. Consequently, a new definition of working class developed that focused mainly on manual laborers, often in large factories. By contrast, small shopkeepers and the managers and professionals were then categorized as part of a “middle class” because they were based in cities and were neither capitalists nor members of the working class.

But manual labor jobs today constitute only 27% of all employment, so the modern definition of working class should probably be expanded. At a minimum, low-skill service and retail workers should be included. But, even with this addition, the working class includes less than 40% of American workers in 2014.

Thus, it is reasonable to expand the modern definition of working class to combine the bottom two tiers into a white collar/blue collar working class group to get to 65% of the workforce. This works especially well for men, because men in the second tier only earned 13% more than men in the bottom tier in 2014. But it doesn’t work as well for women workers, because middle-tier female workers earned 71% more than their lower-tier counterparts. Nonetheless, the earnings of all sectors of the expanded working class grew much more slowly than pay of managers and professionals.

As Figure 5 shows, the bottom tier—the group with the clearest justification to be classified as working class—declined by nearly 25% of the entire labor force (24 percentage points). The top tier—those with the lowest connection to being considered working class—doubled in size by adding 18 percentage points of the entire labor force. This left the middle tier to grow by 6 percentage points. The combination of the bottom and middle tiers—the broad definition of the working class—went from 83% of workers to 65%.

These shifts were driven by separate movement of each of the eight occupational clusters described above. In the bottom tier, the four groups that are part of the narrow definition of working class all lost significant shares: factory operatives lost more than 12 percentage points and ended with a share half its 1960 share; farmers and farm workers also declined by more than half; even the high-end manual jobs of skilled workers and supervisors lost more than a third of its share. Only laborers and low-skilled service and retail workers had a small decline (see Appendix Figure A2). As a consequence of these changes, the share of the lowest-paying subgroup of laborers and low-skill sales and service workers increased from 35% in 1960 to 47% in 2014.

Big employment gains can be seen in some mid-tier jobs, while the share of clerical workers declined because of the huge gains in information technology—professionals now are expected to work on their computers and not rely on secretaries. The second biggest share increase occurred among the combination of jobs that includes technicians, technologists, licensed practical nurses, non-professional self-employed, and white-collar supervisors, including managers at retail and food outlets.

This leaves top-tier jobs as the main source of new jobs. The share of the intellectual and caring professions started from a small base (3% of employment) but nearly doubled to 7%. Few have failed to appreciate the tremendous growth in the number of high-end managerial and professional jobs. The professional leadership and high-end tasks in private businesses and government grew by more than 12 percentage points, leading to top managerial and professional jobs comprising more than one in four jobs.

The bottom line is that the share of the working class, using several occupational definitions, declined substantially. Now that the top tier has grown to more than one-third of all workers, those workers’ economic clout and their ability to mainly socialize with each other has grown, meaning that the social divide between these groups has grown substantially. And unsurprisingly, the winners think they earned their elevated place and their right to determine social mores.

Part III: Changing Economic Relationships

There are two separate earnings trends. On the one hand, rising inequality means that earnings differences based on education levels and occupations have grown. On the other hand, earnings gaps have narrowed when it comes to white women versus white men and African-Americans versus whites. Narrowing these gaps does not mean that they have closed—white women and African-Americans still have earnings significantly lower than comparably educated white men. Most of the narrowing of the gap between African-Americans and whites occurred before 1980, while most of the rising inequality on the basis of education and occupation occurred after 1980. Only the narrowing of the female-male gap was consistent over the 54 years between 1960 and 2014.

Table 3 presents data on the median earnings of the eight occupation groups for men in both 1960 and 2014. The biggest winners were those business and technical managers and professionals who started out with the highest earnings in 1960. In second place are the intellectual and caring professionals who benefited from the high demand for college-educated workers and had their real cash earnings (not including employer benefits) grow by 47%. The earnings of middle-skill white-collar workers and the skilled blue-collar workers grew modest amounts—33% and 28% respectively. While factory and related blue-collar workers had rather anemic earnings’ growth of 18%, male clerical workers and low-skill blue collar, sales, and retail workers had the earnings decline by 2% and 11% respectively. The steep decline of the latter category is one of the indicators of backward movements of non-college-educated men.

Table 3: Male earnings gains by 8 occupational groups, 1960-2014 (2014 dollars)

Source: 1960 Census and 2013-2015 Current Population Surveys.

Strikingly, the economic reward structure changed over these years to the benefit of business and technical managers. Compared with factory operatives, high-paying male technical business managerial and professional jobs earned 62% more than factory workers in 1960 and 132% more in 2014. For the highest-paying manual workers, the comparable gap grew from 40% in 1960 to 85% in 2014. And for the lower paid—laborers and low-skill sales and services workers—the gap went from 99% in 1960 to 180% in 2014.

This movement in jobs underlies that change in relative earnings of workers with different levels of education. As Figure 6 shows, all the B.A. premiums were higher in 2014 than they were in 1960. The B.A. advantage over high school in 2014 was 69% for both men and women. The advantage over some college (including a two-year degree) in 2014 was 58% among women and 50% among men.⁶

Source: 1960 Census and 2013-2015 Current Population Surveys.

Turning now toward gaps that were narrowing, African-Americans and women made progress from their dismal earnings in 1960. Figure 7 shows a startling finding: In 1960, white men with a high school diploma and no postsecondary education working as factory workers earned more than both white women and African-American men with a college degree working in a top managerial and professional job. By 2014, this odd relationship was reversed as the higher education and better pay in top managerial and professional jobs won out over lower education and employment in factory work.⁷

Source: 1960 Census and 2013-2015 Current Population Surveys.

Table 4 shows the earnings of African-Americans and whites by educational level and gender in 1960 and 2014. There are several interesting relationships in these results. First, with the exception of women without a high school diploma and women with a graduate degree in 2014, whites earned more than African-Americans with the same education. Second, the earnings gap declined in 2014 for every gender/educational level combination. Third, with the exception of women without a high school diploma in 1960, the earnings gap was smaller for African-American women than for African-American men.  Fourth, the notion that African-Americans have jumped in the line ahead of non-college-educated white men is not borne out by these figures. While the earnings gap is small for the 5% of white men without a high school diploma versus the 10% of African-American men without a diploma, African-Americans at all other levels of education trail their white counterparts by 17-29%. The biggest gap—29%—is for those with a high school diploma or no postsecondary education. Because white workers are more likely to have completed higher education degrees than African-Americans, the overall male earnings gap is 31%.

Table 4: Earnings of african-americans relative to whites with same education

Source: 1960 Census and 2013-2015 Current Population Surveys.

African-American women have always had a greater attachment to the work force than white women and therefore tended to work more hours. Again, because of an educational deficit, the overall African-American/white women gender gap was 12%.

The final comparison involves all women versus all male workers. As Figure 8 shows, women trailed men with the same education by 44-47% in 1960 and by 32-36% in 2014. The consistency across the different educational levels demonstrates that the gender penalty persisted across a range of occupations.

Source: 1960 Census and 2013-2015 Current Population Surveys.

Finally, what happens in the economy is reflected in people’s living standards just as it is reflected in family incomes. In previous research, I showed that incomes were rising from 1979 to 2014 for all people except those in the bottom six percentiles, and that median income rose by 30%.⁸ I defined the upper middle class as those with incomes of $100,000 to $350,000, in today’s dollars, and showed that the share of the population in the upper middle class grew from 13% in 1979 to 29% in 2014.

In addition to growing in size, the relative incomes of the upper middle class and rich have grown faster than the middle class and lower middle class. Consequently, the rich and upper middle class went from receiving 30% of all income in 1979 to receiving 61% of all income in 2014 (see Appendix Figure A3). The discretionary consumption options of these high-income households are ever present in ads and human interest stories throughout the media—TV, magazine, and newspapers—on expensive restaurants, high-end national and international travel, and super expensive remodeling projects. So, even though the bottom 70% know that they have many things that their parents didn’t have—bigger houses, better-equipped cars and trucks, and gadgets galore, they also know that they don’t have what the upper middle class and rich have.

Conclusion

Many commentators have talked about the disastrous conditions of the white working class. But when you look at the change over time since 1960, it becomes clear just how much the story is one of relative change. In 1960, simply being a white man was all a high-school-educated factory worker needed to out-earn a black or female worker with a college degree. Today, that white male privilege still exists, but it has substantially eroded. White men occupy a smaller share of the population. Higher education attainment has increased across age and gender groups. The achievement gap between races has narrowed, as has the wage gap. And at the same time as these demographic shifts, our economy has evolved to favor the professional and skilled professions over the unskilled ones—to a greater degree than ever before. Now, that high-school-educated white male earns a little more in inflation-adjusted terms than his 1960 counterpart did, but his gains seem non-existent when compared to those in skilled or professional jobs, which employ many more women and racial minorities than they did a half century ago.

While the majority of non-college-educated white men have found adequate economic niches and do significantly better than white women and people of color with the same educational attainment, many aren’t doing that well. As noted above, there has been a shift away from good-paying manufacturing jobs, and the relative pay of male manual laborers to top business and technical managers and professionals has declined. Further, a higher proportion of prime-age white males are not in the labor force than in the past, and an extra four percentage points of those in the bottom occupation tier are in the lower-paying retail and low-skill service jobs. But these movements have been responses to the structural changes in the economy. What these people need, as is true of those without a college degree of every race and gender, is more training and education to perform better in today’s economy.

In terms of work, less-educated and rural people are “left behind” because the modern economy is very much an urban-based economy that relies on the interaction between different players for planning, financing, promoting, coordinating, and selling. It is the legion of business managers and professionals that have been the big winners in this economy, and they have developed their own means of communicating and interacting based on conventions learned at universities and graduate schools. The question for policymakers is how we get more opportunity in the hands of those left behind.

ABOUT THE AUTHOR

Stephen Rose is a nonresident fellow in the Income and Benefits Policy Center at the Urban Institute. He is a nationally recognized labor economist and has spent the last 35 years researching and writing about the interactions between formal education, training, career movements, incomes, and earnings. His book Social Stratification in the United States was originally published in 1978, and the seventh edition was released in 2014. His book Rebound: Why America Will Emerge Stronger from the Financial Crisis addresses the causes of the financial crisis and the evolving structure of the US economy over the last three decades.

Rose has worked with large longitudinal and cross-sectional data sets to develop unique approaches to understanding long-term income and earnings movements. He recently coauthored the report “The Economy Goes to College” showing that the high-end service economy of work in offices, health care, and education was the main driver of the US postindustrial economy, responsible for 64 percent of employment, 74 percent of earnings, and over 80 percent of workers with a bachelor’s or advanced degree.

Before coming to Urban, Rose held senior positions at the Georgetown University Center on Education and the Workforce, Educational Testing Service, the US Department of Labor, Joint Economic Committee of Congress, the National Commission for Employment Policy, and the Washington State Senate. His commentaries have appeared in The New York Times, The Washington Post, The Wall Street Journal, and other print and broadcast media. He has a BA from Princeton University and an MA and PhD in economics from the City University of New York.

Appendix 1: Data Sources and Methods

Source: 1960 Census and 2013-2015 Current Population Surveys.

Source: 1960 Census and 2013-2015 Current Population Surveys.

Source: 1960 Census and 2013-2015 Current Population Surveys.