angle-up angle-right angle-down angle-left close user menu open menu closed search globe bars phone store

Choosing Our Future: A Story of Opportunity in America
Irwin Kirsch, Henry Braun, Mary Louise Lennon, and Anita Sands

The Increasing Importance of Skills

As technological change races forward, demands for skills – some new and some old – are altered. If the workforce can rapidly make the adjustment, then economic growth is enhanced without greatly exacerbating inequality of economic outcomes. If, on the other hand, the skills that are currently demanded are produced slowly and if the workforce is less flexible in its skill set, then growth is slowed and inequality widens. Those who can make the adjustments as well as those who gain the new skills are rewarded. Others are left behind.

Claudia Goldin and Lawrence F. Katz , The Race between Education and Technology 10

Not so long ago, a high school diploma and a commitment to hard work were enough to secure a middle class life in towns and cities across America. Gary, Indiana, was one such place. A city created by the U.S. Steel Corporation, named for its then-chairman of the board, and dubbed the "Magic City" during its heyday, Gary emerged from the Great Depression to thrive in the decades during and following World War II.11 High school graduates working in the steel mills benefited from negotiated job security, decent salaries that were linked to productivity, and guaranteed pensions. Fast forward to the 1970s and 1980s and Gary was beset by plant closings and layoffs. As other cities and other industries went through similar seismic shifts, high-school-level skills, or the human capital associated with a high school education, no longer provided a ticket to the middle class.12

video thumbnail

Earnings and Unemployment Rates by Educational Attainment

Note: Data are for persons age 25 and over. Earnings are for full-time wage and salary workers.

Source: Current Population Survey, U.S. Bureau of Labor Statistics, U.S. Department of Labor.

Line graph showing changes in real wage levels of full-time U.S. male workers by education from 1963 to 2012. Lines show wages at five levels of education: greater than a bachelor's degree, bachelor's degree, some college, high school graduate, and less than high school. The line for male workers with more than a bachelor's degree shows the largest increase in wages, followed by the line for men with a bachelor's degree. The lines for those with less than a high school education, high school graduates, and those with some college shows that their wages rose from 1964 though the mid-1970s, fell through the 1980s and early 1990s, and have remained relatively flat. Those with less than a high school education suffered the greatest loss in wages since the mid-1970s peak.

The relationship between education levels (commonly used as a proxy for skills) and earnings has changed in important ways over the past 40 to 50 years.13 As shown in Figure 1, there has been a clear – and growing – advantage to male workers with college and graduate degrees. These increasing wage differentials are not just due to the fact that those with higher levels of education are surging ahead but also because others are falling further behind. Men with a high school credential or less have seen their wages stagnate or fall since the mid-1970s. While the trends are slightly more positive for women with those same levels of education, like men, their gains in wages have not matched those of more highly educated individuals.

Figure 1: Changes in Real Wage Levels of Full-Time U.S. Male Workers by Education, 1963 – 2012

Source: David H. Autor, Science 2014; 344: 843-851.

Additional data show that by 2013, Americans with four-year college degrees earned 98 percent more per hour, on average, than those without a college degree, up from a 64 percent advantage in the early 1980s – despite evidence that wages for college graduates have been flattening since around 2000.14 The relative economic returns to postgraduate and professional degrees are even more substantial. A comparison of median weekly earnings in 2014 shows that those with a master's, doctoral, or professional degree earned between 200 and 245 percent more than those with a high school diploma and 120 to 149 percent more than those with a bachelor's degree.15

This is not to imply that all routes to the middle class require college degrees, or even higher levels of education. But, unlike earlier eras, many alternative career paths involve the use of new technologies and, as a result, require more than a high school education, as well as specialized skills.16

Differential consequences linked to variations in credentials and skills characterize the current economic landscape, one that has been shaped by the complex interplay of globalization and technology, as well as policy choices at the federal, state, and local levels. Many features of this new landscape have been positive. We have seen, for example, extraordinary changes in how we communicate and access information as a result of global digital networks. We are able to collaborate across borders, purchase products in an international marketplace, and instantaneously access information posted by people and organizations from around the world. An increasing variety of material goods, with many available at lower prices than ever before, has led to improved living standards in many parts of the world.17

thumbnail for Human Capital video

Human Capital

Source: Educational Testing Service

However, over the past four or five decades, these forces have also led to fundamental changes in our economy – changes that have devastated large segments of the American workforce. A dramatic decline in the number of manufacturing and production jobs and a concomitant increase in service-sector work have changed both the type and numbers of jobs available, as well as the pay and benefits associated with those jobs.18 As Frank Levy and Richard Murnane point out in The New Division of Labor, for decades, the economy expanded and contracted in predictable ways. Jobs were lost during the downturns and then, for the most part, replaced on the upturns. But, as they explain, this time is different. "That world is largely gone now. Many of the jobs lost in the post-2000 recession – clerical and factory jobs lost to automation, call center jobs lost to India, manufacturing jobs lost to China – will not be coming back."19 There are no signs that the combined trends of job losses and stagnant or decreasing wages will change any time soon, if at all, and our ability to leverage emerging technologies and markets to create new jobs to replace those that have been lost remains an open question. Understanding the present reality, and the factors that underlie it, are critical to developing appropriate policies and solutions.

For decades, the economy expanded and contracted in predictable ways. This time is different. Many of the jobs lost in the post-2000 recession will not be coming back.

Skills in the Workplace

Trends in Routine and Non-Routine Tasks in Occupations, United States, 1960 to 2009 video

Trends in Routine and Non-Routine Tasks in Occupations, United States, 1960 to 2009

Source: David H. Autor and Brendan Price, 2013, "The Changing Task Composition of the US Labor Market: An Update of Autor, Levy, and Murnane (2003)," MIT Mimeograph, June 2013,

The skills required of many workers a generation or two ago consisted primarily of those possessed by the typical high school graduate. But the demands of a technology-infused economy, the rapid pace of change, and global competition have interacted to alter the workplace and increase the demand for more broadly skilled employees. Analytical, technical, and problem-solving abilities, along with communication, collaboration, and teamwork skills, are increasingly valued. Employers also seek workers who can keep pace with rapidly changing technologies. As a result, they are looking for individuals who have skills that enable them to benefit from ongoing training programs and, perhaps most importantly, have the ability and initiative to learn on their own and continuously upgrade their skills.

The combined forces of rapid technological change and globalization have differentially impacted workers with varied levels and types of skills. We will look at each, defining them as follows: Low-skill workers are those with a high school education or less; middle-skill workers have more than a high school education but less than a bachelor's degree (e.g., an associate degree or postsecondary certificate); and high-skill workers have a bachelor's degree or higher.

Low-skill workers

Individuals with lower skills, the kinds of workers who once earned a secure living in the steel mills of Gary as well as in other industries and cities across America, have been hit hard. As a result of increasingly sophisticated technologies, millions of low-skill jobs have simply disappeared. For example, many assembly-line workers have been replaced by industrial robots, and scanners are doing the work of cashiers. Other jobs have been exported to take advantage of the low-cost labor and increasingly skilled workforces in developing countries. Production jobs associated with apparel manufacturing and the assembly of electronic components, as well as service jobs at help desks, are but a few examples.

The loss of low-skill production and manufacturing jobs has been somewhat offset by an increase in low-skill service jobs such as food services, call centers, and retail sales. This has meant employment for millions of low-skill workers, but the trade-off has been far from even, as wages for service-sector jobs are well beneath what such workers had been able to earn in the past. For men without a high school diploma, median real earnings decreased by 20 percent between 1990 and 2013. Similarly skilled women have fared somewhat better, but their earnings have still dropped by 12 percent. Analyses by the Brookings Institution's Hamilton Project reveal that the move to lower-paying service occupations accounts for one-third of the drop in earnings for men over this time period and two-fifths of the drop for women.20

These low-skill jobs tend to share characteristics that do little to promote the economic and social stability of employees and their families. Lower wages mean that having two income earners has become an economic necessity for many families, and millions of individuals are working two or three jobs just to get by. These jobs tend to be part time and most often do not include benefits such as sick days or personal time, putting workers in the precarious position where missing a day due to illness or family responsibilities not only reduces their pay but also may put them at risk of losing their jobs. Many of these jobs also operate using just-in-time, or flexible, scheduling. Software makes it possible for managers to predict staffing needs in real time and adjust workers' shifts, sometimes just minutes before they are scheduled to begin. Although this saves employers from paying workers who are not needed on a particular day or time, employees cannot plan on a set income week to week, and must juggle childcare and transportation plans at the last minute.21

Middle-skill workers

Middle-skill workers have been impacted by the changing workplace as much as, if not more than, low-skill workers. Technologies that excel at repetitive tasks involving the storage, processing, and retrieval of information continue to replace workers who perform routine work, including clerical workers, meter readers, telemarketers, and travel agents, to name a few. Although such routine jobs made up 58 percent of employment in 1981, they declined to 44 percent by 2011.22

However, the picture is not entirely bleak, as not all categories of middle-skill jobs are declining. Instead there is what economist Harry Holzer calls a "tale of two middles."23 Employment opportunities still exist, and are projected to increase, for middle-skill jobs that involve non-routine tasks. Examples include technical jobs in the health care field such as respiratory and radiation therapists; jobs involving the installation, maintenance, and repair of mechanical systems such as heating and air conditioning systems; and heavy-vehicle maintenance work. Examples can be found on the farm as well, where technology such as the equipment used in automated milking stations must be monitored and repaired. These jobs differ from more traditional middle-skill jobs in that they involve the use of new technologies and, consequently, require additional education and technical training beyond high school. But for individuals who can successfully complete the required programs, good employment opportunities still exist.

Economist Robert Lerman believes that a robust apprenticeship system is one pathway that could help workers develop the occupational and employability skills needed to obtain such jobs.24 He argues that the singular focus of policy makers and funding programs on educational attainment and the development of academic skills ignores the critical role of occupational skills. Sociologist and demographer Andrew Cherlin concurs, noting that, "Decent-paying professions exist for the non-college-educated . . . . However, we need to better train young adults for the skills needed for jobs such as these. Many experts urge that we provide more work-based teaching in career-oriented high schools, in apprenticeships and in community college partnership with local firms."25

Unlike other advanced economies, the U.S. has a relatively underdeveloped apprenticeship system. However, private-sector workforce development efforts are underway across the country. Companies in the Houston area, for example, expect some 60,000 jobs in the construction and petrochemical industries to be generated over the next three years. A local staffing agency reports that it disqualifies 60 percent of job seekers, many for a lack of skills.26 One plant manager believes that the skilled-worker shortage is partly because local high schools channel students toward four-year university degrees instead of technical careers.27 Without denying the returns to four-year college degrees, Lerman and others argue that we need alternate paths to the middle class. Not everyone needs, or may want, to complete four years of college to get a good job and support a family. Variations among individuals and differences across regions of the country are too great to take a "one size fits all" approach to expanding opportunity. Instead, there must be a focus on identifying and supporting multiple pathways to developing the skills and knowledge needed for the jobs of today and tomorrow.

High-skill workers

Finally, high-skilled professionals are not immune to job insecurity, as some analysts predict that the range of jobs displaced by technology will continue to expand.28 The combination of increased computing power and more sophisticated software will make it possible for a growing number of complex cognitive tasks to be automated, meaning that technology can take on an ever-broadening set of jobs including medical diagnostics, certain types of research, tutoring, accounting, and translating.

Whatever the skill level, an individual's ability to find and keep a good job is related to the skills he or she can bring to that job and continue to develop over time. On average, better skills result in better economic outcomes. However, the challenge of improving the prospects for America's workers is more complex than a single-minded focus on developing and expanding their skills, as important as that is. In fact, as economists such as Jared Bernstein warn, that focus could lead to segments of the population being "all dressed up with no place to go" if appropriate job opportunities don't exist for them. Policies to stimulate job creation as well as improve wages and benefits, while beyond the purview of this initiative, are equally essential to any strategy to meet this challenge.29

The challenge of improving the prospects for America's workers is even more complex than developing and expanding their skills, as important as that is. Policies to stimulate job creation as well as improve wages and benefits are equally essential to any strategy to meet this challenge.

Skills in Everyday Life

Discussions in the media and popular press focus almost exclusively on the greater demand for skills in the workplace. Largely ignored is the simultaneous increase in the skills required for everyday life as a result of both new technologies and evolving business practices. For example, more and more everyday tasks require the ability to navigate, critically analyze, and problem solve in data-intensive, complex digital environments.

There is a growing expectation that everyone has access to a computer, smartphone, and the Internet and is capable of navigating the online world. Doctors increasingly provide results from medical tests to their patients via online portals. Employers share information about office closings or schedule changes via text or email. Information about obtaining a driver's license or photo ID is provided on (often complex) government websites. A growing number of employers now accept online applications only. Each of these tasks requires not only access to a computer and the Internet but information technology skills.30

Although it is true that access to technology has dramatically increased as the digital divide continues to shrink, many do not possess the skills required to operate in that domain. Those without the necessary skills find themselves unable to adequately cope with these demands and are at a comparative disadvantage with those who do. Evidence of this skills gap is found in results from the Programme for the International Assessment of Adult Competencies (PIAAC), a large-scale international assessment of adults between the ages of 16 and 65 developed under the auspices of the Organisation for Economic Co-operation and Development (OECD). PIAAC includes an assessment of how well respondents can access and use the technology-based information found in websites, email, and spreadsheets to solve presented tasks. Over 60 percent of adults in the U.S. scored in the lowest levels, demonstrating only limited skills needed to acquire, evaluate, organize, and utilize information found in digital environments.31

Another example where a lack of skills can be detrimental to individuals is in their management of employee benefits. For jobs that offer benefits as part of the employment package, there has been a significant shift in the responsibility for managing health care and retirement accounts. Whereas workers once were able to rely on defined benefit pension plans, the number of employers offering such plans has declined sharply. In 1979, 28 percent of all workers were enrolled in such plans. By 2012, that figure had dropped to 3 percent.32 The increasing use of what are called defined contribution plans, such as a 401(k), allows employees to take their accounts with them when they move to new jobs – a positive development given our increasingly mobile workforce. However, the concern is that such plans shift responsibility onto workers, making them responsible for deciding whether or not to participate, how much to save, and where to invest their contributions.

The percentage of those under the age of 65 with employer-sponsored health insurance also has been on the decline, from 69 percent of all workers in 2000 to 59 percent in 2010. Data show that coverage varies by education, with 80 percent of college graduates receiving employment-based coverage in 2010 in comparison to 62 percent of those with a high school diploma. From 2000 to 2010, the decline in medical coverage rates for high school graduates was more than twice that for college graduates.33 The gap in coverage between full-time and part-time workers also grew substantially during this period. The share of full-time workers who were uninsured increased 3.2 percentage points, compared with a rise of 9.3 percentage points for part-time workers.34 This leaves individuals, most often those with the lowest skills, to find appropriate coverage. The advent of the Affordable Care Act, while making health coverage accessible to many previously uninsured individuals, requires the use of state or federal exchanges to select health care plans, meaning that individuals are responsible for understanding plan features and weighing costs and benefits relative to their needs – another instance where cognitive skills have become more important.35

The connection between skills and health outcomes is not limited to insurance issues. Large disparities in health behaviors and outcomes by level of education also have been well documented.36 The Centers for Disease Control and Prevention reports that, "Educational attainment and income provide psychosocial and material resources that protect against exposure to health risks in early and adult life. Persons with low levels of education and income generally experience increased rates of mortality, morbidity, and risk-taking behaviors and decreased access to and quality of health care."37 For example, there is a strong association between rates of smoking and educational attainment. In 2013, smoking among adults ranged from 24 percent of those with no high school diploma, to 22 percent of high school graduates, 9 percent of those with a college degree, and just 6 percent of those with a graduate degree.38 Although obesity rates in U.S. adults have increased across all levels of income and education, for women in particular, obesity rates are higher among those with lower levels of education: Among females over the age of 20, some 42 percent of those with no high school diploma in 2005-2008 data were obese compared with just over 23 percent of those with a college degree.39

Such behaviors and conditions impact overall life expectancy. In 2005, 25-year-old men who had not graduated from high school could be expected to live about 16 years less than those with graduate degrees. For women, that difference was close to 12 years.40 Although the relationships between levels of education (as a proxy for skills) and health behaviors are clear and persistent, such behaviors are complex and interact with other factors such as income, access to health care and information, and the social environment.

If workers do not contribute to retirement accounts and invest wisely, or do not find a health plan and select the coverage that best matches their medical needs and those of their family members, the consequences can severely impact their health and financial security, as well as increase costs to public-health systems. This general shift of responsibility, sometimes referred to as the privatization of risk,41 reflects, in part, current economic realities faced by employers. But the result is that employees must be able to understand and prepare for a variety of risks to their welfare, making it more important than ever that they have the skills, knowledge, and requisite support to do so wisely.

Unfortunately, at a time when skills are increasingly consequential – both in the workplace and in everyday life – evidence indicates that many Americans do not have what it takes to succeed in this environment.

America's Skills Gap

The growing importance of educational attainment has not gone unnoticed. Accordingly, there has been a national focus on increasing attainment rates – making sure that more students graduate from high school, and more go on to college. In fact, there have been improvements on both fronts. The high school graduation rate reached 81 percent in the 2012-2013 school year.42 And the percentage of students enrolling in college increased by 46 percent from 1990 to 2014.43

While such statistics can justly be celebrated, they do not tell the full story. Although college enrollment rates are up, the number of students who go on to graduate in six years has not shown a similarly dramatic increase. Between 2008 and 2013, the overall six-year college graduation rate at four-year degree-granting institutions increased from 57 to just 59 percent.44 Similarly, a report from the William T. Grant Foundation shows that while the gap in college enrollment rates for Black and White students has narrowed since 2000, the gap in college completion rates has increased.45

One important factor in understanding such gaps is the confounding of educational attainment and skills. When looking at preparedness for higher education and work, most studies use educational attainment as a proxy for skills. For example, the data comparing employment rates and wages for high school graduates against those for college graduates shown in Figure 1 is a way of displaying the relationship between skills and income. This is a reasonable strategy given that educational attainment and credentials are indeed important, particularly for opening doors and expanding access to additional educational opportunities and jobs. But ultimately it is the skills individuals actually possess that determine if they will succeed in those environments and if doors to additional opportunities will be open.

Educational attainment and credentials are indeed important, but ultimately it is the skills individuals actually possess that determine if they will succeed.

We do have data from a number of large-scale national and international assessments that provide more direct evidence regarding the actual skill levels of our student and adult populations. One such assessment is the National Assessment of Educational Progress, or NAEP, which is a nationally representative and continuing assessment of students in grades 4, 8, and 12. NAEP results show that the reading and mathematics skills of our in-school 12th graders have remained essentially unchanged since 1970.46 In 2013, NAEP reported that 74 percent of the nation's 12th graders, students headed to the workforce or higher education, were below proficient in mathematics, and 62 percent were below proficient in reading.47

When we look at various subpopulations of American students, we find persistent gaps in academic achievement across racial and socioeconomic groups.48 Although achievement gaps among racial groups have not disappeared and should continue to be cause for concern, results from standardized assessments show that they have narrowed since the 1970s. In contrast, as shown in Figure 2, income-related achievement gaps have grown substantially. The gap between children from high- and low-income families is about 30 to 40 percent larger for children born in 2001 than it was for those born in the 1970s – and is now more than twice as large as the Black-White achievement gap.49

Figure 2: Trends in Race and Income Achievement Gaps, 1943-2001 Cohorts

Line graph showing trends in achievement gaps by race and income from 1943 to 2001.

Source: Whither Opportunity?: Rising Inequality, Schools, and Children's Life Chances by Duncan, Greg J.; Murnane, Richard J. Reproduced with permission of Russell Sage Foundation in the format Republish in other published product via Copyright Clearance Center.

Although achievement gaps among racial groups have not disappeared and should continue to be cause for concern, they have narrowed since the 1970s. In contrast, income-related achievement gaps have grown substantially.

As if these national data were not concerning enough, recent studies show that our 15-year-olds, students near the midpoint of their high school careers, are outperformed by many of their international peers. The OECD's Programme for International Student Assessment (PISA) administers an assessment of reading, mathematics, and science skills every three years in all 34 OECD member countries along with an additional group of participating countries. In 2012, U.S. students performed below the OECD average in mathematics, with 21 OECD member countries scoring higher. Not only is our "average" mediocre, but even our best performers, those who scored at the 90th percentile, are relatively weak when compared with their peers in OECD member countries, with those in 19 countries scoring higher in mathematics.50

Data about how well our high school graduates perform as they move forward in the educational system should raise concerns as well. Although readiness standards vary across states and institutions, well over half of students in community college take one or more remedial classes at some point during their enrollment.51 This lack of readiness for college-level coursework is alarming, and data show these students are more likely to drop out than their more prepared peers. Fewer than 1 in 10 students who begin their community college careers with remedial classes graduate within three years.52 Similar skill gaps are evident in four-year colleges and universities as well, where some 20 percent of students must take remedial classes in their freshman year.

Additional evidence, using PIAAC data, demonstrates that gaps in skills continue into adulthood. This assessment of literacy, numeracy, and problem-solving skills was first administered to some 160,000 adults in 24 countries in 2011-2012. At a time when more skills are needed, findings from this survey show that the skills of U.S. adults have, in fact, decreased when compared with results from previous adult surveys.

Analyses of the PIAAC results for millennials, young adults between the ages of 16 and 34, showed that U.S. millennials performed well below the OECD average on all domains.53 They ranked last in numeracy (along with Italy and Spain) and in problem solving in technology-rich environments (along with the Slovak Republic, Ireland, and Poland). Some will argue that such international comparisons are not relevant to policy because America is exceptional; our diverse population along with the structure of our educational system makes us different from other countries. But looking at subgroups, defined by characteristics such as educational attainment or immigration status, does not change the picture. Furthermore, PIAAC demonstrates that large numbers of even our best performing and most educated millennials possess relatively weak skills. Even our highest-scoring millennials, those at the 90th percentile, scored lower than their top-scoring counterparts in 15 of the 22 participating countries. Such results strongly suggest that too many students are graduating from high school and even completing postsecondary education without developing the skills they need.54

Such evidence, as well as the life experiences of millions of Americans, attests to the fact that skills matter. But focusing solely on skills, even the broad range of human capital that is needed in today's society, is too narrow a lens. Social capital is an equally important component of the opportunity story. Social capital rises out of, and develops through, a variety of social relationships. It is these relationships that help determine how human capital is developed and used.


10 Claudia Goldin and Lawrence F. Katz, The Race between Education and Technology (Cambridge, MA: Harvard University Press, 2008), 352.

11 Electronic Encyclopedia of Chicago, s.v. "Gary, IN,"   (Chicago: Chicago Historical Society, 2005),

12 Throughout this narrative, the terms "human capital" and "skills" will be used interchangeably. While the term "skills" is often associated solely with cognitive abilities, it is used in its broadest sense here. So "skills" and "human capital" are defined as the knowledge, competencies, and attributes needed for work and personal life. These include reading, writing, and mathematics skills; general and specific knowledge; analytical, problem solving, and technical skills; the ability to learn independently and continuously upgrade skills; interpersonal skills such as collaboration, communication, and teamwork; and character traits such as motivation, persistence, conscientiousness, reliability, self-discipline, and curiosity, often referred to as noncognitive skills.

13 David H. Autor, "Skills, Education, and the Rise of Earnings Inequality among the ‘Other 99 Percent,' " Science 344, no. 6186 (2014): 843-51. Data from Current Population Survey. As we explain in the subsection entitled "America's Skills Gap," although datasets commonly use educational attainment as a proxy for skills, it is important to distinguish between them.  The two are clearly related, but data show that some students with educational credentials do not have the skills we might expect from their level of schooling. 

14 David Leonhardt, "Is College Worth It?  Clearly, New Data Say," The Upshot, New York Times, May 27, 2014,

15 Authors' calculations based on Bureau of Labor Statistics data on earnings and unemployment rates by educational attainment, last modified April 2, 2015,

16 Harry Holzer, "Job Market Polarization and U.S. Worker Skills:  A Tale of Two Middles," Economic Studies at Brookings, Brookings Institution, April 2015,

17 While this narrative focuses on opportunity in the United States, another positive outcome is that there has also been a dramatic reduction in global poverty as documented by The World Bank ( and the United Nations (

18 Derek Thompson, "A World without Work," Atlantic, July/August, 2015, Since 2000, the number of manufacturing jobs has fallen by almost 5 million, or about 30 percent; The Low-Wage Recovery: Industry Employment and Wages Four Years into the Recovery, National Employment Law Project Data Brief, April 2014,

19 Levy and Murnane, New Division of Labor, 157.

20 Up Front, "Profiles of Change: Employment, Earnings, and Occupations from 1990-2013," blog entry by Melissa S. Kearney, Brad Hershbein, and Elisa Jácome, Brookings Institution, April 21, 2015,

21 In These Times, "How Flexible Scheduling Is Making American Workers' Lives Miserable," blog entry by Robert Reich, April 24, 2015, Note that some companies, such as Starbucks, have made efforts to establish more regular scheduling for employees.  See Jodi Kantor, "Starbucks to Revise Policies to End Irregular Schedules for Its 130,000 Baristas," New York Times, August 14, 2014,

22 Anton Cheremukhin, "Middle-Skill Jobs Lost in U.S. Labor Market Polarization," Economic Letter, Federal Reserve Bank of Dallas, May 2014,

23 Holzer, "Job Market Polarization."

24 Robert Lerman, "Restoring Opportunity by Expanding Apprenticeship," in The Dynamics of Opportunity in America: Evidence and Perspectives, eds. Irwin Kirsch and Henry Braun (New York: Springer, 2016).

25 Andrew Cherlin, "The Missing Working Class," Washington Post, February 13, 2015,

26 Ana Campoy, "Match Game: Companies Push Training to Close Skills Gap, Wall Street Journal, April 23, 2015,

27 Ibid.

28 Erik Brynjolfsson and Andrew McAfee, The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies (New York: W. W. Norton & Company, 2014).

29 Jared Bernstein, "Wages in the United States: Trends, Explanations, and Solutions," in The Dynamics of Opportunity in America: Evidence and Perspectives, eds. Irwin Kirsch and Henry Braun (New York: Springer, 2016); On the Economy; "The Reconnection Agenda: Reuniting Growth and Prosperity," blog entry by Bernstein, April 2015, See Ishwar Khatiwada and Andrew M. Sum, "The Widening Socioeconomic Divergence in the U.S. Labor Market," in Dynamics of Opportunity, for a detailed discussion of the uneven distribution of labor market outcomes across workers based on differences in household income and educational attainment.

30 Gerry Smith, "Without Internet, Urban Poor Fear Being Left Behind in Digital Age," Huffington Post, March 1, 2012,

31 In fact, the figure of 60 percent is an underestimate of the proportion of adults at the lowest levels as no data was collected for just over 10 percent of adults who either did not demonstrate the minimal computer skills required to complete the assessment (including the ability to use a mouse to move, point and click, and type on a keyboard) or refused to participate in the computer-based portion of the survey.

32 "Pensions Decline as 401(k) Plans Multiply," Bankrate, July 24, 2014, Statistics cited are from the Employee Benefit Research Institute.

33 Elise Gould, A Decade of Declines in Employer-Sponsored Health Insurance Coverage, Briefing Paper 337 (Washington, DC: Economic Policy Institute, February 28, 2012),

34 Ibid.

35 Ellen Peters, Louise Meilleur, and Mary Kate Tomkins, eds. "Appendix A: Numeracy and the Affordable Care Act: Opportunities and Challenges," Health Literacy and Numeracy: Workshop Summary (Washington, DC: National Academies Press, 2014), In addition, appropriate support must be available for those individuals who have not been able to develop requisite skills.  Such support can be provided through social networks in the community or, in some cases, through technological solutions.  One example of the latter is tax preparation software that greatly reduced the cognitive skills required to file taxes. 

36 Paula A. Braveman, Catherine Cubbin, Susan Egerter, David R. Williams, and Elsie Pamuk, "Socioeconomic Disparities in Health in the United States: What the Patterns Tell Us," American Journal of Public Health 100, no. S1 (April 2010), S186-S196, doi:10.2105/AJPH.2009.166082.

37 Pamela A. Meyer, Paula W. Yoon, and Rachel B. Kaufman, CDC Health Disparities and Inequalities Report – United States, 2013, Centers for Disease Control and Prevention Morbidity and Mortality Weekly Report, Supplement 62, no. 3, November 22, 2013,

38 Current Cigarette Smoking among Adults in the United States,  Centers for Disease Control and Prevention, accessed January 2015,

39 Obesity and Socioeconomic Status in Adults: United States, 2005-2008, Centers for Disease Control, NCHS Data Brief 50, December 2010,

40 Brian L. Rostron, John L. Boies, and Elizabeth Arias, "Education Reporting and Classification on Death Certificates in the United States," Vital and Health Statistics, Series 2, no. 151 (2010): 1-16.

41 Jacob Hacker, "The Privatization of Risk and the Growing Economic Insecurity of Americans," in Privatization of Risk book series, eds. Craig Calhoun and Jacob S. Hacker (New York: Social Science Research Council and Columbia University Press, June 7, 2006),

42 U.S. Department of Education, "U.S. High School Graduation Rate Hits New Record High, U.S. Department of Education," press release, February 12, 2015, Graduation rates based on new uniform way of calculation adopted in 2010.

43 Undergraduate Enrollment, The Condition of Education (Washington, DC: National Center for Education Statistics, last modified May 2015),

44 Institutional Retention and Graduation Rates for Undergraduate Students, The Condition of Education (Washington, DC: National Center for Education Statistics, last modified May 2015). See Harry J. Holzer, "Improving Opportunity through Better Human Capital Investments for the Labor Marker," in The Dynamics of Opportunity in America: Evidence and Perspectives, eds. Irwin Kirsch and Henry Braun (New York: Springer, 2016) for a discussion of a policy agenda that includes improving completion rates at two- and four-year colleges.

45 Adam Gamoran, The Future of Educational Inequality in the United States: What Went Wrong, and How Can We Fix It?, 2014 Annual Report, William T. Grant Foundation,

46 Marc Tucker, "Are We Just Fooling Ourselves? Is American Education a Colossal Failure?" Opinion, Education Week, April 16, 2015, Because NAEP is a low-stakes assessment, there are long-standing questions about the level of engagement and effort of the 12th graders who participate in the assessment and, consequently, about the validity of the reported results. A randomized control study of the impact of monetary incentives on performance found that NAEP may both underestimate the reading abilities of students enrolled in 12th grade and yield biased estimates of certain achievement gaps. See Henry Braun, Irwin Kirsch, and Kentaro Yamamoto, "An Experimental Study of the Effects of Monetary Incentives on Performance on the 12th-Grade NAEP Reading Assessment," Teachers College Record 13, no. 111 (2011), 2309-44.

47 Information about how achievement levels are determined for NAEP can be found on the National Center for Education Statistics website at

48 For more discussion, see Gamoran, Future of Educational Inequality.

49 Sean F. Reardon, "The Widening Academic-Achievement Gap between the Rich and the Poor: New Evidence and Possible Explanations," in Whither Opportunity?: Rising Inequality, Schools, and Children's Life Chances, eds. Greg Duncan and Richard M. Murnane (New York: Russell Sage Foundation, 2011).

50 These results focus on mathematics as that was the major domain in 2012.  For information about performance in reading and science, see OECD, PISA 2012 Results: What Students Know and Can Do (Volume I). Student Performance in Mathematics, Reading and Science (Paris: OECD Publishing, 2013),

51 Community College FAQs, Community College Research Center (New York: Teachers College, Columbia University, September 10, 2015),

52 Remediation: Higher Education's Bridge to Nowhere, Complete College America, April 2012,

53 Madeline J. Goodman, Anita M. Sands, and Richard J. Coley, America's Skills Challenge: Millennials and the Future (Princeton, NJ: Educational Testing Service, 2015),

54 One important point to be noted is that the large-scale data cited here focus only on the cognitive components of human capital.  Important social, behavioral, and personality characteristics, often labeled noncognitive skills, are not yet targeted in these surveys.  See detailed discussions about noncognitive skills and their role in adult outcomes in Henry M. Levin, "The Utility and Need for Incorporating Noncognitive Skills into Large-Scale Educational Assessments," in The Role of International Large-Scale Assessments: Perspectives from Technology, Economy, and Educational Research, eds. Matthias von Davier, Eugenio Gonzalez, Irwin Kirsch, and Kentaro Yamamoto (New York:  Springer, 2013); and Patrick Kyllonen, "Soft Skills for the Workplace," Change 45 (no. 6): 16-23, November-December 2013,