copy the linklink copied!

2. The National Context


This chapter examines the national context within which state higher education systems operate. It notes key characteristics of the country’s labour market and the national higher education landscape, and surveys graduate labour market outcomes nation-wide. It focuses on four areas of federal policy important to state efforts to ensure successful labour market outcomes among graduates: student financial assistance, accreditation, workforce development, and the provision of information.


This chapter provides an overview of the national conditions in which state higher education systems operate in the United States. The first three sections of the chapter highlight the labour market context in the United States, key features of US higher education, and the links between the labour market and higher education, by briefly reviewing the labour market returns to higher education and skills-matching in the labour market.

The fourth section provides a brief presentation of the role of the federal government in higher education, with a focus on policy areas most relevant to the labour market relevance and outcomes of higher education, namely student financial assistance, accreditation, workforce development, and the provision of information on higher education students and outcomes. The chapter closes by briefly highlighting the space that US states have in a federal system of higher education to orient higher education provision towards labour market relevance.

copy the linklink copied!

2.1. Overview of the US labour market

The US labour market is more flexible and characterised by higher average wages and greater wage dispersion than in many OECD Member countries

The United States has one of the most flexible labour markets among OECD Member countries. It has the lowest minimum wage relative to median wages in the OECD, and among the lowest legislative protections for permanent workers and levels of regulation on temporary employment. Unemployment benefits are relatively limited: in 2016, only 10.5% of unemployed workers in the United States received unemployment benefits, compared to 23.3% on average across the OECD (OECD, 2019[1]). In addition, union membership and coverage of workers by collective bargaining agreements is low: for instance, collective bargaining covered 11.6% of US workers in 2017, below the OECD average of 32.4% (OECD, 2020[2]) .

The American labour market is characterised by higher wages and a wider wage dispersion than on average across OECD countries. Average wages in the United States were USD 63 093 in 2018, well above the OECD average of USD 46 686 (OECD, 2020[3]). However, wage growth has not recovered since the 2008-09 economic recession, as it averaged 1.9% per year in the United States from 1997 to 2007, but decreased to about 0.7% per year since 2008, a figure that is close to the OECD average.

At the same time, wage disparities in the United States are considerably larger than on average in the OECD. One-quarter (24.5%) of US workers earned less than two-thirds of median annual earnings in 2017, which was the highest figure in the OECD among countries for which data was collected, with an average of 15.4% (OECD, 2020[4]). Differences by race and ethnicity, as well as by gender, are significant. Black/African American workers’ median wages were 24.2% lower than those of White workers in 2018, a gap that has grown from 19.7% in 2006 and is largest among men (U.S. Bureau of Labor Statistics, 2020[5]). Hispanic or Latino workers earned even lower median wages on average in 2018 (74.2% of median White workers’ wages), though they have made important gains since 2000. The gender wage gap was also notably higher in the United States (18.2%) than on average across the OECD (13.5%).

However, barriers may limit labour market flexibility in the United States. Occupational licensing has soared from 5% of workers in the late 1950s to over 20% today, a rate similar to those found in the European Union and Japan. While occupational licensing has a critical public protection role by setting standards for initial training and on-going professional development, it can result in barriers to employment and job mobility (Hermansen, 2019[6]). There are also concerns that rising industry concentration is creating labour market monopsony. For example, recent research found that 20% of all workers in 2014 had non-compete clauses in their contracts, including 12% of workers with less than college education, as well as a high prevalence of “non-poaching” agreements in franchise contracts, prohibiting franchisees from hiring workers away from each other (Starr, Prescott and Bishara, 2019[7]; Krueger and Ashenfelter, 2018[8]). At the same time, research suggests that in most industries, national-level concentration is actually increasing local-level competition (Rossi-Hansberg, Sarte and Trachter, 2018[9]).

Low labour market participation is a continuing challenge in the United States, even in periods of strong economic growth. Before the COVID-19 pandemic hit in the first quarter of 2020, the country had experienced the longest period of sustained economic growth on record and seasonally adjusted unemployment had fallen to 3.6% in October 2019, its lowest point in 50 years and well below the OECD average (OECD, 2018[10]; U.S. Bureau of Labor Statistics, 2020[11]). Employer demand has been high in recent years: the Manpower Group (2018[12]) found that 46% of American employers reported difficulty finding the right people for jobs in 2018, either due to a lack of applications or to the insufficient experience or skills of applicants. This is the highest share of employers reporting difficulties finding talent since 2012.

However, labour force participation among those aged 25-64 has not recovered in the United States since the financial crisis of 2008-09. As Figure 2.1 demonstrates, labour force participation has been in decline since 1997, falling from 80.2% to a low of 76.7% in 2015. Whereas the United States used to have much higher labour market participation than on average in the OECD and Western Europe, this is no longer the case. Gains since 2015 have been modest, amounting to a 1.1 percentage-point increase.

copy the linklink copied!
Figure 2.1. Labour force participation rate among individuals aged 25-64, 1970-2018
United States, OECD average and European Union-22 average
Figure 2.1. Labour force participation rate among individuals aged 25-64, 1970-2018

Note: European Union - 22 refers to the 22 European Union members of the OECD in 2018: Austria, Belgium, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Luxembourg, Netherlands, Poland, Portugal, Slovak Republic, Slovenia, Spain, Sweden and United Kingdom.

Source: OECD (2018[10]), Labour Force Survey Sex and Age Indicators,


The earnings advantage for high levels of skills has risen and highly skilled workers are increasingly concentrated in urban areas

Technological change and globalisation have contributed to a shift in the levels, types, and combinations of skills demanded and rewarded by employers. Two main phenomena have taken place in parallel: an increasing share of displaced workers due to automation, and a change in job tasks and the skills required to perform these tasks.

The US labour force has undergone significant transformation since the 1980s, with employment shifting from the middle level of skills towards either high- or low-skill jobs (Autor, Dorn and Hanson, 2013[13]). Much of the decline in middle-skilled employment has occurred in the manufacturing industry, which experienced a loss of approximately seven million jobs from the 1980s to 2016, even as output doubled (Carnevale, 2016[14]). While automation has displaced many routine tasks, it has begun to affect non-routine tasks as well (Acemoglu and Restrepo, 2017[15]; Deming, 2017[16]). OECD estimates using the Survey of Adult Skills (PIAAC) suggest that about 10% of US jobs are at high (greater than 70%) risk of automation, while approximately 28% are at significant risk (50%-70%) of automation (OECD, 2019[1]). This compares to OECD averages of 16.6% and 30.2% respectively, suggesting there are fewer jobs facing a high risk of automation in the United States than in other OECD countries.

The skills demanded of workers have changed significantly in the past decades. For instance, there was a strong shift in the skills profiles of workers in manufacturing, with a decline in jobs for high school graduates and a rise in jobs for post-secondary graduates. Skills requirements have risen significantly among publicly traded firms with large increases in capital stock, pointing to the complementarity between high levels of skills (Hershbein and Kahn, 2018[17]). Research also suggests that the share of US jobs “requiring high levels of social interaction” increased by 11.8 percentage points between 1980 and 2012, and that the greatest employment and wage growth was found in jobs that required both high social and mathematics skills, followed by jobs with high social skills and low mathematics skills (Deming, 2017[16]). This pattern has applied even at the bottom of the earnings distribution, in positions such as continuing care assistants, food service workers and security guards (Autor, Dorn and Hanson, 2013[13]). There is also evidence that skill requirements within occupations have been increasing in the United States since the 2008-09 recession (Hershbein and Kahn, 2018[17]; Atalay et al., 2018[18]).

Skill-biased technological change and rising housing costs are key contributors to shifts in worker mobility in the United States. While Americans with lower education levels previously had higher rates of inter-state migration, the inverse is now true: high-skilled workers are increasingly migrating towards already skill-abundant cities, while less-educated workers are moving away from higher-income cities. Leading metropolitan areas in turn benefit from dynamic “knowledge economy” clusters that concentrate innovation activity and rely on advanced skills, causing wages for educated workers to rise, even as supply increases (Giannone et al., 2017[19]; Moretti, 2012[20]). International immigrants, at all skill levels, also tend to concentrate in more prosperous regions and cities (Moretti, 2012[20]; Card, 2009[21]).

However, these shifts are taking place in a context of lower job and geographic mobility than in the past. While the US labour market has traditionally experienced high levels of job turnover, the frequency of labour market transitions has been in steady decline. In 2002, 12% of the workforce reported a change of employer within the preceding 12 months, and 6% a change of industry; in 2012, these shares were respectively 9% and 4% (Molloy, Smith and Wozniak, 2017[22]).

Globalisation has also played a role in the changing demand for skills and growing differences between geographic areas. Foreign competition has greatly affected industries concentrated in regions with lower educational attainment, and favoured skill-intensive economic sectors where the United States has a comparative advantage (Autor, Dorn and Hanson, 2013[13]; Moretti, 2012[20]). Population ageing and gradual secular shifts in demand from goods towards services are also reshaping the labour market (Abraham and Kearney, 2019[23]; Lawrence, 2018[24]).

copy the linklink copied!

2.2. The scope and resourcing of higher education

Higher education attainment is rising, though more slowly than in some other OECD systems

In 2018, 47% of Americans aged 25-64 had a tertiary credential, which is the fifth-highest rate of attainment in the OECD and nine percentage points above the OECD average, as shown in Figure 2.2. Tertiary education in the context of international comparisons includes the associate’s degree level (ISCED 5), the bachelor’s degree level (ISCED 6), and the doctoral level (ISCED 8). The United States stands out especially in terms of the highest level of degrees, ranking fourth in the attainment of doctoral degrees, behind Slovenia, Luxembourg and Switzerland.

copy the linklink copied!
Figure 2.2. Tertiary attainment of 25-64 year-olds, by higher education level, 2018
As compared to overall higher education attainment for 25-34 year-olds
Figure 2.2. Tertiary attainment of 25-64 year-olds, by higher education level, 2018

Source: OECD (2019[25]), Education at a Glance 2019: OECD Indicators, Tables A1.1 and A1.2,


Yet, America’s skills advantage is driven in part by the generation of American workers currently nearing retirement (aged 55-64), who have a tertiary education attainment rate close to 43% compared to an OECD average of 27% (OECD, 2020[26]). While the higher education attainment rate for younger Americans (aged 25-34) remains above the OECD average, the United States ranks tenth for that age group, behind Canada, Australia, the United Kingdom, South Korea and Japan, among others.

Furthermore, fewer Americans are participating in higher education than a few years ago. Overall, enrolment peaked at almost 21.6 million students in 2010, and remained almost 1.3 million students below that level in 2016 (NCES, 2017[27]). Part of the decline in enrolment is driven by demographic change, namely that the youth cohort is smaller than in past generations. However, lower enrolment also reflects that the share of young high school graduates pursuing higher education has stagnated, even as an increasing number of students graduate from high school. In October 2018, the share of young high school graduates aged 16 to 24 who were enrolled in some form of higher education was 69.1%, a participation rate that is only half a percentage point higher than in 2008 (U.S. Bureau of Labour Statistics, 2019[28]).

Important differences exist in higher education attainment in the United States by socio-economic status and ethnic and racial group. For example, research suggests that among students scoring at the top of the scale in standardised exams, the rate of bachelor’s degree attainment is 41% for students from low-income families compared to 74% for students from high-income families (Page and Scott-Clayton, 2016[29]). As shown in Figure 2.3, among 25-34 year-olds, in 2018, the group with highest attainment (Asians) had an attainment rate close to 2.5 times that of those with the lowest attainment (Hispanics). Between 2010 and 2018 however, gains in higher education attainment were largest for Hispanics and Black/African Americans, indicating modest progress in closing racial and ethnic attainment gaps. On the other hand, differences in attainment by socio-economic status have increased over the past few decades, according to several studies (Page and Scott-Clayton, 2016[29]).

copy the linklink copied!
Figure 2.3. Tertiary education attainment in the United States, by race and ethnicity, 2010 and 2018
25-34 and 25-64 year-olds
Figure 2.3. Tertiary education attainment in the United States, by race and ethnicity, 2010 and 2018

Sources: Adapted from U.S. Census Bureau (2019[30]), Current Population Survey, 2018 Annual Social and Economic Supplement, and U.S. Census Bureau (2019[31]), Current Population Survey, 2010 Annual Social and Economic Supplement,


On average, US higher education is well resourced, with extensive private spending and a diminishing role for public spending

Spending on higher education in the United States is high. When factoring in both individual and government spending, the percentage of Gross Domestic Product (GDP) spent on higher education in the United States is second only to Chile, and per capita student expenditure is second only to Luxembourg (OECD, 2019[25]).

As indicated in Figure 2.4, in 2016, spending per student for all services in the United States was almost double the OECD average (USD 30 165 vs USD 15 556). The United States spends more than twice the OECD average on teaching (“core services”), six times the OECD average on other services to students including campus facilities (“ancillary services”), and only about 80% of the OECD average on research and development (OECD, 2019[25]).

copy the linklink copied!
Figure 2.4. Total expenditure on educational institutions per full-time equivalent student, by category of spending, 2016
In equivalent USD converted using PPPs
Figure 2.4. Total expenditure on educational institutions per full-time equivalent student, by category of spending, 2016

Note: Year of reference is 2017 for Chile and Colombia.

Source: OECD (2019[25]), Education at a Glance 2019: OECD Indicators, Table C1.2,


Compared to other OECD countries, the United States is characterised by a large share of private funding for higher education. Private tertiary spending relative to GDP was over three times the OECD average (1.6% versus 0.5%) in 2016, a figure equivalent to that of Chile, and well above countries such as the United Kingdom (1.2%), Canada (1.1%) or Japan (1.0%) (OECD, 2019[32]). Household expenditures account for almost half (46.2%) of tertiary funding in the US, followed by public funding (34.6%) and expenditures by other private entities (19.3%). Private not-for-profit institutions are an important feature of US higher education, as discussed in the next section.

As charitable organisations, private not-for-profit institutions supplement tuition fees with tax-favoured philanthropic gifts from individuals and business. At the end of the 2016 fiscal year, 84 universities had endowments of USD 1 billion or more (NCES, 2018[33]). These funds are more prevalent and larger in private not-for-profit institutions, but they are becoming more common in public institutions. For instance, the University of Texas system, Texas A&M University and the University of Michigan, Ann Arbor each have endowments worth at least USD 9.8 billion.

While the share of public expenditure dedicated to higher education is 15% higher in the United States than in other OECD countries, public expenditure in higher education, in aggregate, has not increased since 2008. By contrast, it has increased by 17% on average across other OECD countries (OECD, 2019[34]). This lack of increase reflects declines in state funding for higher education, driven in part by states seeking to balance their budgets in the wake of the 2008-09 economic crisis (Mitchell et al., 2018[35]).

In response to reduced state spending, public higher education institutions have resorted to increasing tuition fees and, in some cases, seeking additional international students or out-of-state students (both of whom pay higher tuition fees) to make up budget shortfalls (Ripley, 2018[36]). Between the 2007/08 and the 2015/16 academic school years, tuition and other fees at public institutions increased on average by 27%, outpacing earnings growth and inflation (NCES, 2018[37]).

This has accelerated a long-term pattern in the United States, whereby households pay a larger share of the cost of higher education. As shown in Figure 2.5, only the United Kingdom has higher tuition fees at public institutions than the United States.

copy the linklink copied!
Figure 2.5. Tuition fees in bachelor’s or equivalent programmes, 2017/18
Annual fees charged by institutions for full-time national students, in equivalent USD using PPP
Figure 2.5. Tuition fees in bachelor’s or equivalent programmes, 2017/18

Source: OECD (2019[25]), Education at a Glance 2019: OECD Indicators, Table C5.2,


Increased private spending has led to rising student debt levels

To help offset the costs of attending post-secondary education, considerable public spending takes the form of financial aid. As indicated in Figure 2.6, in 2017/18, 89% of US bachelor’s degree students received some form of public or government-guaranteed financial support, which is higher than most OECD countries and equivalent to the rate found in Australia, New Zealand, and Norway, but slightly below Sweden and England (94%) (OECD, 2019[25]). In the United States, most students (53%) receive both public/government-guaranteed private loans and public grants or scholarships, while 26% receive public grants or scholarships only, and another 10% receive public/government-guaranteed private loans only. Compared to the other countries with high student aid, the United States has a larger share of students receiving only public grants or scholarships.

At the bachelor’s degree level, student loans are common. Over 70% of first-time full-time students in private for-profit four-year colleges received loan aid in 2015/16, compared to 60% of students in private not-for-profit institutions and under 50% in public institutions (McFarland et al., 2018[38]).

There are also important private sources of financial aid in the United States, financed in part by institutional endowments. The College Board (2019[39]) estimates that in 2018/19, institutional grants and private and employer grants accounted for 32% of the USD 260 billion in total financial aid provided to students that year. Additionally, private lenders also offer loans to students, which account for approximately 10% of student borrowing (Scott-Clayton, 2017[40]).

copy the linklink copied!
Figure 2.6. Percentage of bachelor’s and master’s long first degree (or equivalent) students receiving financial support, 2017/18
Figure 2.6. Percentage of bachelor’s and master’s long first degree (or equivalent) students receiving financial support, 2017/18

Note: Where there are no markers, data is unavailable.

Source: OECD (2019[25]), Education at a Glance 2019: OECD Indicators, Table C5.2,


Because of increased reliance on tuition fees to finance higher education, student debt has risen over the past decades. Nation-wide data indicate that for the academic year 2015/16, 61.8% of undergraduate degree and certificate completers received loans over the course of their programmes, for an average cumulative loan amount of USD 24 480 (NCES, 2018[33]). This represents a significant increase from 1999/2000, where the share was 52.5% of students for an average amount of USD 14 260 in current dollars. Spread across 43 million Americans, household student loan debt reached USD 1.5 trillion in 2018, making it the second most important form of household debt after mortgages (FRBNY, 2019[41]).

Debt loads and default rates vary substantially according to students’ degree types, institution attended, and demographics. Analysis by the Urban Institute suggests that individuals in the highest income quartile hold about one-third of all student debt, as more students from higher-income background participate and complete higher education (Urban Institute, 2017[42]). Many of them also carry larger debt loads as a result of completing graduate studies (master’s or doctoral programmes). While individuals in the lowest income quartile hold a smaller share of total student debt, they are more likely to carry debt than their higher-income peers: 75% of students from the lowest income quartile have debt, compared to 57% of individuals in the highest income quartile. Only 14% of Black/African American students carry no debt, and about one-third of them carry debt loads of more than USD 40 000. In contrast, 30% of Hispanic and White students carry no debt.

Students at public two-year institutions have the highest default rates (18.3% of the cohort entering repayment in 2013/14), even though they have lowest debt levels, followed by students at for-profit two-year institutions (17.5%). This is in comparison to default rates of 7.0% at private, not-for-profit four-year institutions and 7.5% at public four-year institutions (Urban Institute, 2017[42]). For-profit institutions have the highest debt levels, particularly at the four-year level. Research points to several challenges affecting graduates of these institutions including poor repayment conditions, earnings that are lower several years after attendance compared to earnings before starting the programme of study, and evidence that employers value credentials from these institutions less than credentials from public institutions (Deming et al., 2016[43]; Cellini and Turner, 2018[44]; Dynarski, 2015[45]).

copy the linklink copied!

2.3. Linking labour markets and higher education: Returns to higher education and skills matching

The labour market returns to higher education are high, but highly variable

Higher education provides a broad range of tangible and intangible benefits to those who complete it, ranging from better employment prospects to better health and higher levels of life satisfaction. Measuring the labour market returns on higher education attainment is of special concern to individuals and governments as they make decisions on investment in higher education. For reasons of data availability and comparability, this review focuses on the employment and earnings of graduates, as well as average debt levels (see Tables 3.1. and 3.2 in Chapter 3).

Across OECD countries, higher education graduates are more often employed and enjoy higher earnings than their peers with upper secondary education. Those who started higher education but did not complete their programme generally experience poorer outcomes, and are at higher risk of not being able to repay their student loans, compared to their peers who have obtained a qualification (Itzkowitz, 2018[46]).

According to 2017 true cohort data, the completion rate of full-time students at a bachelor’s or equivalent degree programme in the United States was similar to the OECD average, at around 40% by the theoretical duration of the programme and close to 70% by the theoretical duration of the programme plus three years. However, this rate was notably below that of countries like Ireland, Israel or the United Kingdom, which ranged from 60% to 70% on the first measure, and exceeded 80% in the second (OECD, 2019[25]). The measures, however, do not take into account the much lower completion rates of students enrolled in certificates or associate’s degrees programmes at two-year institutions. As discussed in the next chapters, low completion rates disproportionately affect disadvantaged students.

Among students who have graduated, employment and earnings vary significantly. The selectivity of the institution, the type (level and length of the programme), and the field of study, have all been shown to have an important effect on outcomes, after controlling for student characteristics and the particular geographic and economic context where institutions are located (Chakrabarti and Jiang, 2018[47]; Webber, 2014[48]; Andrews, Li and Lovenheim, Michael, 2016[49]).

The employment advantage of having attained a higher education degree in the United States is similar to that observed in other OECD countries. In 2018, 85% of Americans with a post-secondary qualification (ISCED 4-8) were employed, which is equivalent to the OECD average (84%). However, more young Americans with post-secondary credentials were inactive than on average in the OECD, at 13% versus 11% (OECD, 2019[50]). The earnings premium for higher education is larger in the United States than in most OECD countries. However, they are disproportionately higher for graduates with a bachelor’s degree or above (ISCED 6, 7 or 8), whereas it is moderate for graduates with a two-year degree (ISCED 5). As shown in Figure 2.7, holders of master’s and doctoral degrees (ISCED 7 and 8) in 2017 had an earnings advantage of 131% compared to workers with an upper secondary degree, which was 40 percentage points above the OECD average and the third highest among OECD countries, behind Chile and Mexico. In contrast, workers with associate’s degrees (ISCED 5) had an earnings premium of 13% compared to upper-secondary graduates, well below the OECD average (20%).

copy the linklink copied!
Figure 2.7. Earnings of tertiary graduates relative to upper secondary graduates, 2017
Pre-tax annual earnings – upper secondary and post-secondary, non-tertiary graduates = 100
Figure 2.7. Earnings of tertiary graduates relative to upper secondary graduates, 2017

Note: Full- and part-time workers are included.

Source: OECD (2019[25]), Education at a Glance 2019: OECD Indicators, Table A4.1,


The particularly high return on investment of bachelor’s and graduate-level degrees in the United States results from many factors. Research suggests that one-quarter of pay premia for higher degrees are the result of between-firm pay differences, with higher degrees associated with the best-paid firms (Engbom and Moser, 2017[51]). There is also growing evidence that differences in productivity growth are rising between American firms as some concentrate unique managerial and technological capabilities, especially in high productivity growth sectors (Autor et al., 2017[52]). The rise of “intangible capital” such as productivity-enhancing technologies, as well as branding and patent protections, is likely an important contributing factor (Crouzet and Eberly, 2019[53]). The resulting gains of the highest-performing firms are being passed along (at least in part) to their employees, creating greater firm-based differences in earnings.

Comparable data on the employment and earnings of graduates with post-secondary credentials below the associate’s level, such as certificates, is scarce at the national and international level. However, there is evidence to suggest that these credentials hold value in the labour market, although this depends greatly on the field of study pursued, and whether graduates work in the field in which they obtained their certificate. Research also suggests greater benefits on average of these credentials for men (Carnevale, Rose and Hanson, 2012[54]; Strada; Gallup; Lumina Foundation, 2019[55]). States often track the employment and earnings of certificate graduates, as will be discussed in the state chapters. While these are typically lower than the returns of associate’s degree, there is evidence to suggest that the returns can exceed those of longer programmes in certain high-demand occupations, and when the graduate works in the field in which they completed their certificate (Carnevale, Rose and Cheah, 2013[56]; Carruthers and Sanford, 2018[57]; Schneider, 2015[58]).

Differences in earnings by field of study are particularly large in the United States. The wide spread in earnings by field is similar to that observed in countries such as Chile, Estonia, Latvia or Germany; whereas countries such as Australia, Nordic countries and the United Kingdom combine both lower median earnings across fields and less variation in graduate earnings between fields of study. It is worth noting that the earnings premium of graduates in arts, humanities, social science, journalism and information are comparatively higher in the United States than in many OECD countries, which may result from a large proportion of these graduates pursuing graduate-level education (Figure 2.8).

copy the linklink copied!
Figure 2.8. Relative earnings of tertiary-educated adults by field of study, 2017
Pre-tax annual earnings – upper secondary graduates = 100
Figure 2.8. Relative earnings of tertiary-educated adults by field of study, 2017

Note: Full-time and part-time workers are included.

Source: OECD (2019[25]) Education at a Glance 2019: OECD Indicators, figure A4.4.


Disparities in outcomes by institution are also important. Graduates from selective institutions have significantly higher returns than those from non-selective institutions, although there is much greater variation in returns between different non-selective institutions (Hoxby, 2018[59]). Research on public higher education institutions in Texas indicates that returns to some four-year colleges can be comparable to those from two-year colleges in some cases (Andrews, Li and Lovenheim, Michael, 2016[49]). Research by Chetty and colleagues (2017[60]) indicates that institutions vary greatly in the extent to which they facilitate students from the bottom income quintile moving up to the highest income quintile. Mid-tier public universities (e.g. City University of New York, California State University) perform the best, but the share of low-income students at these institutions fell significantly from 2000-11.

Attendance at selective institutions is uneven across demographic groups. As shown in Figure 2.9, 80.2% of post-secondary students whose family was in the highest income quintile attended a public or private not-for-profit four-year college as their first post-secondary institution. This compares to only 35.4% of students from low socio-economic backgrounds, who disproportionately attend two-year colleges, as well as private for-profit four-year colleges. Black/African American students were almost three times more likely to study at private for-profit four-year colleges than White students, while Hispanic students were 1.2 times more likely than White students to attend a public two-year institution. Black/African American and Hispanic students are underrepresented at more selective institutions, and the participation of Black/African American students in selective public institutions has fallen since the early 2000s (Carnevale et al., 2018[61]).

copy the linklink copied!
Figure 2.9. Enrolment in different types of higher education institutions by socio-economic status, race and ethnicity, 2018
Figure 2.9. Enrolment in different types of higher education institutions by socio-economic status, race and ethnicity, 2018

Notes: Two-year institution figures correspond to two-year or less institutions for figures by income quintile. Data by race and ethnicity are for 2018 and include all students in two-year and four-year institutions. Data by income quintile present the share of 2009 ninth graders who ever attended post-secondary by control and level of first post-secondary institution, as of 2016.

Source: Adapted from USDOE – NCES (2019[62]), Digest of Education Statistics, Tables 306.40 for 2019 and 302.44 for 2018


Research suggests that traditionally disadvantaged students frequently under-match, choosing institutions that are less challenging than those they would be qualified to attend, and often choose fields of study with lower completion rates and lower earnings (Blagg et al., 2017[63]; Backes, Holzer and Velez, 2015[64]). Disadvantaged students may be less responsive to wage signals because they may have less access to relevant information. Even when they graduate from high-earning fields of study, graduates from underrepresented groups face wage penalties due to the occupations in which they work, which tend to be lower-paying than those of their peers (Carnevale et al., 2017[65]).

While higher education can be a highly beneficial investment for underrepresented students (Zimmerman, 2014[66]), the lower returns they often face may dampen their incentives to participate. This can pose major challenges from an equity perspective and as public authorities aim to raise the attainment rates of their populations, as will be discussed in the next chapters.

Skills mismatches are modestly lower in the United States than in other OECD economies

The OECD Survey of Adult Skills provides information on the three different types of mismatches that can arise between workers’ qualifications and skills and the requirements of the job they hold. Qualifications or skills mismatches arise when workers’ educational attainment levels (or skills levels) are higher or lower than required for their jobs. Field-of-study mismatches rise when workers are employed in a different field than the field in which they have studied and are specialised (see Box 5.1 in (OECD, 2019[67]) for details on how these mismatches are defined and measured).

In the United States, skills mismatches are less prevalent than across the OECD, at 12% versus 15% (OECD, 2019, p. 118[67]). Qualification mismatches concerned around one-third of workers in the United States, in line with the OECD average. Field-of-study mismatches are the most common form of mismatch across the OECD, concerning about 40% of workers, compared to 45% of workers in the United States (Montt, 2015[68]).

While field-of-study mismatches can be a sign of saturated employment in some sectors, it can also signal high skills transferability and an ability to find better jobs in other occupations or sectors. Field-of-study mismatches alone carry a limited wage penalty: mismatched workers earn about 3% less on average than their well-matched peers. By comparison, skills mismatches carry a wage penalty of about 7% on average across OECD countries, while over-qualification is associated with a larger wage penalty for workers, of about 17% on average across the OECD. The United States is the OECD country where over-qualification carries the largest wage penalty, exceeding 30% (OECD, 2019[67]). Research suggests this penalty has risen over the past few decades (Rose, 2017[69]), and that this penalty is highest for workers who are both over-qualified and work outside of their field of study.

This can be a particular challenge for some higher education graduates. Research suggests that low initial earnings due to under-employment can persist over the long term. Under-employment affects lower-earning fields of study in particular, such as security and law enforcement; parks, recreation, leisure and fitness studies; and consumer and family sciences psychology. By contrast, graduates from Science, Technology, Engineering and Mathematics (STEM) programmes are the least likely to experience under-employment (Burning Glass Technologies and Strada Institute, 2018[70]).

At the same time, American employers report difficulty finding workers with the right skills, whether it be due to lack of experience, lack of “hard skills” or poor “soft skills” (ManpowerGroup, 2018[12]). Skills gaps have been reported in multiple employer surveys nation-wide (IHE, 2019[71]; SHRM, 2019[72]; Adecco USA, 2019[73]). In a 2018 survey conducted on behalf of the Association of American Colleges and Universities (AACU), employers indicated that recent graduates had the skills necessary to succeed in an entry-level position, but few had the skills needed for advancement or promotion within the organisation (Hart Research Associates, 2018[74]). These skills gaps likely result from a combination of factors, reflecting in part the complexity of the relationship between labour supply and demand. Aside from structural changes in the labour market, for example due to the long-term effects of technological change; an overall decline in employer-led, on-the-job training (particularly for entry-level workers) may be a contributing factor in some widely reported skills gaps (Waddoups, 2016[75]; Capelli, 2015[76]).

copy the linklink copied!

2.4. Higher education policy in a federal system

Federal authority in higher education is limited, as states bear the main responsibility and institutions have a high level of autonomy

In common with other federal systems of government, responsibilities for higher education in the United States are shared between the federal government and state governments. State governments have the main responsibility for public higher education in the United States. This derives from Article 10 of the US Constitution (as amended), which declares that powers not explicitly given to the federal government are the purview of state governments (Antonio, Carnoy and Nelson, 2018[77]). State governments are therefore instrumental in establishing public higher education institutions, funding the instructional missions of public higher education institutions, establishing governance structures, setting tuition policies (and often, providing financial aid), and in developing information systems, as well as articulation and credit transfer policies (Eckel and King, 2004[78]).

The federal government plays an important role in collecting and disseminating higher education data, providing oversight of accreditation agencies, and investing in basic research funding as well as student financial assistance under Title IV of the Higher Education Act. For the 2017/18 academic year, it was estimated that all forms of federal financial support for higher education students (e.g. loans, grants, tax credits, federal work-study funding) were worth over USD 240 billion (College Board, 2017[79]). The federal government has leveraged these investments to secure a role in the assurance of quality in higher education by recognising accreditation bodies, and only institutions that are accredited by a recognised accreditor are eligible for federal student financial aid.

Public funding of higher education in the United States draws from federal, state and local sources. Federal funding for instruction is provided primarily through student financial assistance, while most state and local government spending on higher education takes the form of direct transfers to institutions (Scott-Clayton, 2017[40]). As illustrated in Figure 2.10, the distribution of funding between central (federal), regional (state), and local government in the United States is characteristic of federal systems, in which a significant share of funding, sometimes the plurality or majority, is borne by regional and local governments. In 2016, the share of federal funding in tertiary education in the United States was 46%, compared with 41% state spending and 13% local spending.

copy the linklink copied!
Figure 2.10. Distribution across OECD countries of tertiary spending among central, regional and local government sources, 2016
Figure 2.10. Distribution across OECD countries of tertiary spending among central, regional and local government sources, 2016

Note: Data for Canada are not available.

Source: OECD (2019a), Education at a Glance 2019: OECD Indicators, Table C4.2,


As discussed in the next chapters, the post-secondary governance system of each state determines its role in policy processes and the degree of authority it exercises over institutional decision-making. Across the United States, 25 states had at least one co-ordinating board and 25 had at least one governing board, while 14 states had multiple boards for higher education (Fulton, 2019[80]). States with a governing board for higher education generally have extensive authority over system-wide strategic planning, from setting admissions standards and credit transfer rules, to having a substantial degree of influence over academic programming and personnel decisions (Eckel and King, 2004[78]). State-wide co-ordinating boards play a less direct, but still significant, role in the state’s responsibilities for public higher education and, in some cases, oversight responsibilities for independent colleges (Fulton, 2019[80]). Both governing boards and co-ordinating boards typically provide budget recommendations to the state Legislature and articulate a strategic plan for the higher education system.

Higher education institutions must be granted the right to operate in a state, a process that is often legislatively regulated, and states exercise different degrees of influence over institutional operations based on their governance structure and level of autonomy. While the governance structure determines the role of state authorities, higher education institutions across the United States have a well-established legal basis as autonomous organisations. Extensive private financing, through tuition fees, commercial activities and donations, augments this autonomy.

copy the linklink copied!
Box 2.1. Higher education institutions in the United States

There are over 7 000 post-secondary institutions in the United States, according to the National Center for Education Statistics (NCES). Among these, over 4 500 are entitled to award qualifications at the associate’s degree level (ISCED 5) or higher, and these enrol over 98% of all higher education students tracked in national data. Most of these institutions (59%) are relatively small, enrolling less than 2 500 students each, as of fall 2018. Approximately 40% of US institutions are public, and they enrol about 75% of all students. Of the 60% that are private, institutions are either not-for-profit (43%) or for-profit (58%) (NCES, 2019[81]; NCES, 2019[62]).

  • Four-year institutions offer programmes primarily at ISCED Levels 6, 7 and 8. The Carnegie Classification of Institutions of Higher Education identifies five sub-categories of institutions, which can be either public or private. These include doctoral universities, which offer a minimum of 20 research doctoral programmes or 30 professional practice doctoral programmes and enrol about half of all students in the four-year sector. Master’s colleges and universities offer at least 50 master’s programmes, and account for close to 30% of enrolments in the four-year sector. The remaining types include baccalaureate colleges and special focus institutions, which offer a high concentration of degrees in a single field (CCIHE, 2019[82]). Public universities range from selective research-intensive universities, to much less selective institutions focused on serving local labour market demand. Private, not-for-profit universities and colleges are among the oldest institutions in the United States, and are often selective, with some that are highly research-intensive and others focused on a traditional liberal arts education at the baccalaureate level.

  • Two-year institutions offer programmes primarily at ISCED Levels 4 and 5, which include two-year associate’s degrees and workforce-relevant certificates, typically ranging from six weeks to over two years in duration. Public two-year institutions, such as community and junior colleges, generally offer two main types of degree programmes: one is academically oriented and prepares students to transfer to four-year institutions to complete a baccalaureate education, and the other prepares students for direct entry into the labour market. Community colleges also serve as post-secondary providers of career and technical education (CTE), which includes industry-recognised certificates, licenses and certifications.

  • Private for-profit institutions are a newer institutional model, predominantly classified as four-year institutions, although they often offer associate’s degrees and certificates. In recent years, more for-profit college students have pursued associate’s degrees or certificates than bachelor’s degrees (Darolia, 2019[83]). The higher education institutions in the United States with the largest enrolment are for-profit, such as the University of Phoenix Online and Kaplan University, as are many of the country’s smallest institutions (Deming, Goldin and Katz, 2012[84]). The US higher education system also includes post-secondary, non-tertiary institutions (ISCED 4), or non-degree-granting institutions, providing only certificate programmes. This segment of the system is small, accounting for just 1.6% of enrolments in post-secondary education in 2017/18. The majority of these types of institutions are private for-profit institutions.

Institutions adopt diverse approaches to their own governance, subject to state regulation where applicable. Boards of trustees, whose members often represent business and civil society, govern most US colleges and universities, both public and private. For public institutions, these boards are typically appointed by state government, though in some cases (particularly for community colleges), board members may be elected at the state level. The legal bases and revenues of institutions vary across types of higher education institutions. Box 2.1 describes principal institution types in the United States.

The federal role in higher education centres on federal grant and loan programmes aiming to widen access and promote affordability

The US federal government first took on a strong role in the student financial aid system through the 1944 GI Bill, which provided financial assistance for armed services veterans to promote access to higher education, leading to a surge in higher education participation during the post-war period (Antonio, Carnoy and Nelson, 2018[77]). With the 1965 Higher Education Act, the basic formal structure for federal student financial aid was established. Federal financial aid spending for students has grown substantially, at an estimated USD 152 billion in 2018/19 (College Board, 2019[39]). Student, programme and institutional eligibility to participate in federal grant and loan programmes has not been linked to the labour market relevance and outcomes of higher education. Rather, federal student aid programmes provide assistance based upon need (Pell Grant programme) or access to borrowing (Stafford student loan programme), as described in Box 2.2.

copy the linklink copied!
Box 2.2. Overview of three main federal financial aid instruments
  1. 1. Most federal student financial aid takes the form of loans. The primary student loan programme, now called the Stafford Loan, was launched in 1965. Initially, the federal government defined loan eligibility, made interest payments for some loans during students’ studies, and guaranteed lenders against defaults, while private lenders gathered, disbursed, and collected the loan funds. In 2010, the consolidated Federal Direct Loan Programme became the only source of federal loans, spurred by shortfalls in private capital during the 2008-09 recession and concerns regarding private lender practices. The private sector’s role has shifted towards providing services to the U.S. Department of Education in collections, record keeping, and client relations. Students with financial need can access subsidised Stafford Loans, also called direct subsidised loans, which are interest-free during their studies, or unsubsidised loans if they do not have financial need. As of 2008, dependent students can borrow up to USD 31 000 for undergraduate study and independent students up to USD 57 500.

  2. 2. The federal government provides substantial support to students in the form of grants. The Pell Grant is the most important need-based grant aid programme. In 2018/19, 31% of undergraduates received Pell Grants, compared to 38% in 2011/12 The average grant was USD 4 160, though the maximum amount for full-time students with full eligibility was USD 6 095. Eligibility is based on family income, and funds may be used for tuition or other needs. Over half of Pell Grant recipients are above the age of 23 and almost one-quarter are over the age of 30,

  3. 3. There are significant tax benefits in support of students enrolled in undergraduate education. The largest is the American Opportunity Tax Credit (AOTC), which provides up to USD 2 500 in support for education expenses. Part-time students can receive up to USD 2 000 through the Lifelong Learning Tax Credit (LLTC). Additional programmes support savings for education and interest payments on student loans.

Sources: Scott-Clayton (2017[40]), College Board (2019[85]).

There is one exception to the choice not to link student financial aid to labour market outcomes, which relates to vocational programmes, most often delivered by private for-profit institutions. These institutions participate in federal student aid programmes by virtue of leading to what the Higher Education Act describes as “gainful employment in a recognised occupation.” For these programmes, quality assurance and federal regulations have focused on ensuring a link between student aid eligibility and labour market outcomes. In particular, in 2014, the federal government introduced the “gainful employment” (GE) rule, which required institutions delivering programmes with a focus on career preparation, enrolling about 15% of the total student population, to report on their performance using two debt-to-earnings metrics. If programmes failed to meet either one of the metrics, their eligibility for federal financial aid could be suspended (Kelchen and Liu, 2019[86]). The first GE ratings were published in 2017. While two-thirds of programmes with a focus on career preparation were delivered by private for-profit institutions, nearly all (99%) programmes that that failed the standards were delivered by private for-profit institutions (Kelchen and Liu, 2019[86]) (Jump, 2019[87]). The implementation of regulatory programme-level reporting was halted in 2017, and the regulation itself was terminated in 2019, due in part to concerns that it principally targeted institutions based on their for-profit tax status. Moreover, critics of the regulation expressed concern that it would have the effect of reducing access to higher education, most especially for disadvantaged students (Kelchen and Liu, 2019[86]).

Debates continue about whether and how to link student aid eligibility and labour market outcomes. An alternative proposal was to expand GE regulations across more programmes and institutions, removing the link to for-profit status. Though the GE rule was not in place long enough for financial aid eligibility to be withdrawn for any institutions, limited evidence of the effects of introducing the GE rule suggests that it slowed the growth of private for-profit colleges. One quasi-experimental study found that it spurred the closure of for-profit institutions (Kelchen and Liu, 2019[86]); the study also found that institutions may have responded with different strategies, including reducing debt, raising admissions standards, shortening programmes or encouraging students to enroll full-time, and hiring additional staff for student services and support to job placements.

A 2018 Government Accountability Office report found that colleges at risk of losing Title IV eligibility due to high graduate default rates were seeking to influence students’ debt management decisions, often to the students’ detriment (Kreighbaum, 2018[88]). Another similar policy, called “borrower-defense”, currently allows students to have loans forgiven when they have attended an institution that has closed or that misrepresented itself in important ways (U.S. Department of Education - Federal Student Aid, 2020[89]). A recent policy sought to establish a more narrow definition for this misrepresentation, which the U.S. Department of Education (USDOE) estimates will save USD 11 billion over the next ten years (U.S. Department of Education, 2019[90]). One critique of the reform has concerned its short time limit for student to make claims; the USDOE indicates that less than one-third of current claims under the policy would have met the new timeline requirement (Stratford, 2019[91]).

Policy makers at federal and state levels are concerned about the protection of students as consumers of higher education. Thus, political debate continues about how best to support access to student financial aid and track graduate repayment, particularly in the context of rising student debt. Student debt levels are comparatively high by international standards, and a key challenge is the wide variability of returns among graduates. Moreover, there are concerns regarding the timing of loan repayment shortly after graduation, when graduate earnings are relatively low (Dynarski, 2015[45]). To repay federal student loans, there are multiple income-driven repayment plans available to graduates, and about one-quarter of graduates participate in some form of income-driven repayment plan (Britton et al., 2019). Participation in income-driven repayment plans has risen, but remains constrained by complexity, including variation in conditions across multiple plans. Other fee-based OECD jurisdictions such as Canada, the United Kingdom, and Australia, have much more extensive (or universal) participation in income-contingent lending (Barr et al., 2018[92]) (Dynarski, 2015[45]).

Several changes to the federal student aid system, described in Box 2.3, have been introduced in anticipation of the re-authorisation of the Higher Education Act.

copy the linklink copied!
Box 2.3. Higher Education Act re-authorisation proposals relating to federal financial aid

The Higher Education Act is the most important piece of federal legislation relating to higher education, dating originally to 1965. Last re-authorised in 2007, it has been due for an update since 2012 (Harris and Kelderman, 2017[93]). Various proposals have been put forward, advancing different visions for higher education and, notably, for the rules governing federal financial aid.

In 2018, Representative Virginia Foxx (R-NC) introduced the Promoting Real Opportunity, Success and Prosperity Through Education Reform (PROSPER) Act. The bill proposed various measures to alter the federal financial aid system, including:

  • extending financial aid eligibility to programmes with fewer credit hours;

  • shifting from a cohort default rate metric to a repayment rate metric for determining institutions’ continuing Title IV eligibility;

  • eliminating the borrower defense to repayment rules and requiring congressional approval rather than granting discretion to the Secretary of Education for similar rules in the future;

  • simplifying the suite of federal repayment programmes into a choice between the current standard ten-year repayment plan or 15% of income above 150% of the federal poverty line until loans are fully repaid;

  • simplifying the Free Application for Federal Student Aid (FAFSA).

Source: U.S. House Committee on Education and the Workforce (2017[94]).

In 2019, Senator Lamar Alexander (R-TN) introduced a bipartisan bill, the Student Aid Improvement Act, which proposes:

  • increasing the maximum Pell Grant award and introducing the Short-Term Pell, which allows students to use Pell Grants for short-term skills and job training programs that lead to credentialing and employment in high-demand fields like health care or cybersecurity;

  • allowing incarcerated individuals who are eligible for parole to use a Pell Grant for prison-education programmes;

  • providing permanent mandatory funding, USD 255 million each year, for Historically Black Colleges and Universities and other Minority Serving Institutions;

  • creating a single institutional accountability measure for student loan repayment;

  • simplifying the FAFSA.

Source: U.S. Senate HELP Committee (2019[95]).

The College Affordability Act (CAA), introduced by Representative Robert Scott (D-VA) in 2019, proposes:

  • restoring Gainful Employment and borrower-defense regulations in legislation;

  • streamlining the FAFSA and updating performance goals for the Federal Student Aid office;

  • increasing the size of individual Pell Grants by up to USD 500;

  • supporting a federal-state partnership to eliminate tuition fees at community colleges;

  • introducing a supplemental grant for four-year college students;

  • consolidating repayment assistance programmes and raising the threshold for repaying loans from 150% of the federal poverty line to 250%.

Source: U.S. House Committee on Education and Labor (2019[96]).

Quality assurance in US higher education is based on accreditation, a system of self-regulation operating independently of US states

To ensure that federal investment on higher education is well-spent, federal support to students has been linked to quality assurance processes since the 1950s, when concerns had arisen that low-quality institutions might emerge to take advantage of federal funds under the GI Bill (Kelchen, 2017[97]).

There are two main types of accreditation in the United States: institutional accreditation and programmatic accreditation. Seven regionally organised, membership-based, non-governmental bodies perform institutional accreditation, also called “regional accreditation”. Regional accreditation concerns about 40% of institutions and 85% of US students, making them the most important accrediting bodies in the system (Kelchen, 2017[97]). Regional accreditation focuses on institutions’ governance, financial health, academic resources and facilities, student support, and, to some extent, learning outcomes. Four of the seven US regional accrediting bodies list employment metrics as a possible way for institutions to demonstrate student success, but they do not require information on these metrics (TICAS, 2018[98]).

Seventy-nine national programmatic accrediting agencies provide accreditation for professionally oriented programmes within institutions, especially those that prepare graduates for licensed or regulated professions. These accreditors, which are often professional associations, traditionally use metrics focused on employment, such as the demonstration of a minimum pass rate on entry-to-practice examinations for regulated professions, or job placement rates. In addition, some state agencies also play an accrediting role in public post-secondary vocational education and nursing education.

While the U.S. Department of Education (USDOE) relies on independent organisations to provide accreditation, accreditors must receive recognition in order for their decisions to be considered valid. The USDOE and the Council for Higher Education Accreditation (CHEA) fulfil this role, by recognising accrediting agencies and providing guidelines, resources and relevant data regarding their work.

The recognition process entails reviews by the Department of Education’s Accreditation Group and the National Advisory Committee on Institutional Quality and Integrity (NACIQI), whose members are appointed by the Secretary of Education and Congress, which provide recommendations to the Department of Education (Kelchen, 2017[97]). The NACIQI conducts reviews of accreditation bodies every five years, reviewing accrediting standards and performing site visits. Similarly, the CHEA also recognises regional, national, career and faith-related accrediting agencies, as well as programmatic accrediting agencies. A CHEA-recognised accrediting organisation will undergo a recognition review every seven years (or as approved by the Board).

There have been several debates regarding the impact of accreditation on institutional provision, and its effectiveness in ensuring minimum standards of labour market relevance and satisfactory labour market outcomes for graduates. Accreditation has been described as burdensome, limiting the capacity of higher education institutions to develop innovative programmes responding to student and labour market needs, such as competency-based programmes and micro-credentials. It is also an expensive process for institutions, with four-year institutions estimated to spend as much as USD 3 billion annually for regional institutional accreditation, and another USD 3 billion for programme accreditation (Kelchen, 2017[97]). In addition, as providers that are not institutions of higher education cannot be accredited, the accreditation process has been viewed as potentially limiting the recognition of alternative credentials.

In response, several measures have been proposed to make accreditation more flexible and less burdensome for institutions. The USDOE recently released new regulations governing accreditors and state authorisation of online education providers. These and related reform proposals are described in Box 2.4.

copy the linklink copied!
Box 2.4. Changes to accreditation regulations

In 2019, the federal government introduced new regulations governing accreditors and state authorisation of online education providers, to take effect in July 2020. The intent of the new regulations is to allow greater flexibility to institutions, permit innovations in programme design, and reduce administrative burdens, though there are concerns from some, including the USDOE Inspector General, that oversight will be inadequate. The new rules permit accreditors to provide approvals for colleges and for the federal government to recognise new accreditors more quickly. Accreditors will also have greater flexibility in sanctions, allowing institutions up to four years before imposing sanctions compared to two years previously – there were concerns that two years were insufficient to address problems where they occurred. Institutions will also have more flexibility to introduce new academic programmes or branch campuses without the approval of accreditors. Under earlier policies, states could waive their rules for online providers through reciprocity agreements with other states where the providers are based. The new rules maintained this approach, but without permitting states to enforce their own laws and regulations on top of these agreements.

In February 2020, Representatives Lori Trahan (D-MA), Madeleine Dean (D-PA) and Jahana Hayes (D-CT) introduced the Accreditation Reform Act. The Act would firstly require that NACIQI take part in reviews of the recognition of accrediting agencies to provide greater independence from government. The other elements of the Act focus on transparency surrounding the initial and renewal of accreditor recognition, requiring the publication of information on accreditors and documentation relating to USDOE decision-making.

Sources: U.S. House Committee on Education and Labor (2020[99]); U.S. Department of Education (2019[100]).

On the other hand, there have been critiques regarding the lack of effectiveness of accreditation, particularly in ensuring minimum standards of quality and labour market relevance. One concern is that accreditors do not focus sufficiently on student learning and labour market outcomes. As noted earlier, regional accrediting organisations do not make the reporting of labour market outcomes a compulsory feature of accreditation. National accrediting organisations characteristically require the reporting of job placement rates, but in the absence of a standard methodology for the calculation of job placement rates and infrequent validation, these measures are said to be ineffective in establishing accountability for labour market outcomes (TICAS, 2018[98]). Most failures to secure accreditation result from concerns regarding financial sustainability, rather than poor performance in education. One proposal has been for the federal government to take on the assessment of institutions’ financial sustainability, allowing accrediting bodies to focus on education quality (Kelchen, 2017[97]). The GE regulations, discussed above, aimed at addressing poor graduate outcomes for some of the nation’s higher education programmes

Federal workforce policy creates a workforce system that functions in parallel to higher education, in which higher education institutions play a growing role

Historically, there has been a clear distinction between the higher education and workforce development systems in the United States (Good and Strong, 2015[101]). Although community colleges and vocational schools represent a large share of workforce development training providers, mainly through provision of career and technical education (CTE), workforce development and higher education systems are not always well aligned. At the sub-baccalaureate level, education providers have traditionally been highly responsive to local labour market needs, and as countries mobilise their education and training systems to upskill and re-skill workers, strengthening the alignment between higher education and workforce development systems may become increasingly critical in addressing rapid changes in the labour market.

The public workforce development system in the United States consists of a large and highly decentralised network of local and regional workforce development agencies in each state. The federal government plays a key role in the public workforce development system through funding transfers to states, who are the primary actors in delivering programmes. The 2014 Workforce Innovation and Opportunity Act (WIOA) is the principal federal workforce development legislation. Funding under the WIOA (USD 4.8 billion in 2018) equates to approximately half of all federal Department of Labor spending on mandatory workforce development funding to states. With the Workforce Investment Act of 1998, the predecessor for WIOA, states were required to create local workforce investment boards in order to adapt policies to the needs of local communities. As a result, there is considerable variation in workforce development policies and programmes across the country, including how they interact with higher education institutions and agencies (Good and Strong, 2015[101]).

copy the linklink copied!
Box 2.5. The Workforce Innovation and Opportunity Act: The role of state and local workforce investment boards

Under the 2014 Workforce Innovation and Opportunity Act (WIOA) and its predecessor, the Workforce Investment Act of 1998 (WIA), each state must have a State Workforce Investment Board, which should include the Governor, members of the state Legislature and representatives of business, labour and educational organisations, economic development agencies and community-based organisations. Below this level, local jurisdictions can form Workforce Investment Areas, directed and supervised by Local Workforce Investment Boards (LWIBs), with representatives from business, labour and community organisations, and local government.

LWIBs oversee the local implementation of the public workforce development system in the United States. Their main responsibilities are setting strategic workforce development priorities that comply with federal and state regulations and are responsive to local and regional labour market needs. There are more than 560 LWIBs across the United States. LWIBs oversee federal grants through WIOA, particularly by directing the almost 3 000 American Job Centers (AJCs). AJCs are responsible for not only delivering WIOA programming, but also co-ordinating access to a much wider array of programmes. Workforce development efforts focus, in large part, on so-called core and intensive services of career planning and job search assistance, particularly for disadvantaged populations. Training is the third component of workforce development efforts.

Source: Wolff (2015[102]).

One of the key linkages between the higher education and workforce systems is information about workforce needs. Labour market information (LMI) is critical not only for the effectiveness of the workforce development system, but also as important input for strategic planning and forecasting in higher education. The U.S. Department of Labor and the Bureau of Labour Statistics are key federal actors in this area, providing current and projected occupational employment statistics. They collaborate extensively with states and local agencies, including local workforce investment boards, which assist in the development of state-wide employment statistics systems. In addition, one of the most important sources of data on the labour market outcomes of higher education graduates comes from Unemployment Insurance (UI) wage records, which are an important element of the LMI system, operated largely at the state level (Workforce Information Advisory Council, 2018[103]).

Under the Workforce Innovation and Opportunity Act, the US Secretary of Labor must “seek, review, and evaluate” recommendations from a 14-member Workforce Information Advisory Council (WIAC) on the evaluation and improvement of the workforce and labour market information system (Workforce Information Advisory Council, 2018[103]). Current reforms would seek to help state and local workforce boards better understand labour market demands, to better inform students in their career and education decisions, to improve the labour market responsiveness of education and training providers, and to facilitate evidence-based policy development. The WIAC released recommendations in 2019, which included enhancing UI wage records; investing to expand information on occupations, skills and credentials; increasing support to states for collaborative work on LMI; tackling barriers to data sharing; and strengthening the workforce LMI system through improved use of technology.

An important federal tool to support worker training is the Carl D. Perkins Career and Technical Education Act (The Perkins Act). In 2018, total federal appropriations under the Perkins Act were equal to USD 1.209 billion, almost entirely through transfers to states (Congressional Research Service, 2018[104]). Lower-income states receive relatively more funding. States can determine how to allocate the funds between secondary and post-secondary programmes, and develop their own formulae to distribute the funds between providers. The Perkins Act, originally named the Vocational Education Act, was first established in 1963. It was most recently re-authorised in 2018. Now referred to as Perkins V, the re-authorisation gave more power to states and local agencies while strengthening system alignment, seeking in particular to align performance metrics across other federal programmes such as the WIOA and the Every Student Succeeds Act (JFF, 2018[105]). Some key elements of Perkins V are described in Box 2.6.

The alignment between post-secondary education and workforce needs is well articulated through career and technical education (CTE) at the sub-baccalaureate level. The emphasis on work-based learning is an integral component of CTE and, in response to industry needs, some community and technical colleges are also beginning to offer applied bachelor’s degrees, demonstrating growing demand for specific, higher-level technical skills with a direct link to the labour market. Additionally, a taskforce on workforce policy, initiated by the current administration and led by the CEOs of Apple and IBM, is launching a campaign to promote multiple post-secondary pathways to develop in-demand, career-relevant skills, including “industry-recognised stackable credentials and certifications” (U.S. Department of Commerce, n.d.[106]). Importantly, this articulation between higher education and the labour market is bolstered by alignment between education and workforce development policy. Often referred to as the “education-to-workforce pipeline”, the alignment between education and workforce needs requires co-ordination between multiple actors, including the primary and secondary school (K-12) system, CTE programmes, and higher education institutions (Cushing et al., 2019[107]). Some states do better than others in connecting these parts of the education and workforce development systems, particularly in terms of strategic co-ordination. For example, in Ohio, the work of state agencies in charge of higher education and workforce development policy is co-ordinated by the Governor’s Office for Workforce Transformation, which reports directly to the Lieutenant Governor. These and other comparative aspects across states are discussed in Chapter 3.

copy the linklink copied!
Box 2.6. Strengthening Career and Technical Education for the 21st Century Act

In 2018, the Carl D. Perkins Career and Technical Education Act was amended by the Strengthening Career and Technical Education for the 21st Century Act. Referred to as Perkins V, the legislation has increased funding for career and technical education (CTE) for the first time in 30 years. Notably, Perkins V provides states with greater flexibility to allocate funds to rural areas and for innovative programmes, as well as the discretion to set their own performance targets beyond a set of minimum thresholds, aligned with accountability metrics used under other federal laws.

Compared to its predecessor, Perkins V places greater emphasis on work-based learning, including pre-apprenticeships and apprenticeships. Work-based learning is defined as “sustained interactions with industry or community professionals in real workplace settings, to the extent practicable, or simulated environments at an educational institution that foster in depth, first-hand engagement with the tasks required in a given career field, that are aligned to curriculum and instruction” (von Zastrow, 2018[108]).

In addition, states must develop plans for co-ordinating CTE with activities under the Workforce Innovation and Opportunity Act (WIOA) and the Every Student Succeeds Act (ESSA); for example, in supporting the professional development of CTE teachers and post-secondary faculty. Local CTE providers must undertake needs assessments to ensure their study programmes are aligned to local labour market needs. Perkins V also introduces a national competitive grant programme aimed to spur evidence-based innovations in CTE.

Under Title I of Perkins V, states must:

  • “provide and support equal access to at least one sequenced program of study integrating core academic and technical training, including employability skills, across secondary and post-secondary education that leads to an industry-recognized credential and meets local or state industry needs;

  • contribute to attainment of higher-order reasoning and problem-solving skills, work attitudes, and employability skills;

  • provide activities to prepare CTE participants (including special populations) for high-skill, high-wage, or in-demand sectors;

  • provide career exploration and development activities, in collaboration with local workforce development boards, agencies, or one-stop delivery systems, including in middle grades (Cushing et al., 2019, pp. 5,15[107]).”

    In addition, states may use Title I funding to:

  • “support local relationships among education, business, and one-stop centers, including sector partnerships, to align programs of study with industry demand;

  • support work-based learning experiences for CTE students and non-academic support services for disadvantaged and/or special populations (such as child care or transportation);

  • support integration of employability skills into CTE programs and programs of study;

  • expand opportunities to participate in dual/concurrent coursework, early college, Advanced Placement/International Baccalaureate coursework, and CTE pathways and certification exams established through articulation agreements with post-secondary institutions” (Cushing et al., 2019, p. 15[107]).

Source: U.S. Department of Education (2020[109]).

Student and labour market information, an area of state initiative, has become an increasing focus of federal policy

Across OECD countries, governments have invested in improving the quantity and quality of information available on the labour market outcomes of higher education graduates. In the United States, the federal government plays an important role in providing information about the student population, institutional performance and graduate outcomes. The National Center for Education Statistics (NCES) collects and publishes a wealth of information about higher education, primarily through the Integrated Post-secondary Education Data System (IPEDS). All Title IV eligible institutions must complete surveys each year addressing topics such as institutional characteristics, student enrolment, and number of degrees and certificates conferred. The NCES also develops and implements several longitudinal surveys, including the Beginning Post-secondary Students Longitudinal Study (BPS) and the Baccalaureate and Beyond Longitudinal Study (B&B).

The United States has a demand-driven system of higher education, shaped by institutional autonomy and student choice. In this context, aligning the provision of higher education to labour market needs depends heavily upon students making informed choices, and institutions responding to student choices and labour market information in the design and revision of higher education programmes. In addition, as the cost of higher education continues to rise, there has been increasing public debate in the United States about the value of a college education (Federal Reserve, 2019[110]; Strada Education; Gallup, 2019[111]). This has led to renewed efforts by states and the federal government to improve information about the costs and returns to higher education and the accessibility of this information.

Information about students in public institutions, as well as earnings and employment data, have been available to states for many decades, and have been used to generate information about the labour market outcomes of graduates. By contrast, the federal role in connecting education to labour market outcomes has largely been prohibitive, as the Family Educational Rights and Privacy Act (FERPA) of 1974 barred the use of personal information that did not lead to the improvement of educational programmes. This delayed the linking of higher education and labour market information both at the state and national level. In recent decades, the federal role has expanded through a series of initiatives. These include the Statewide Longitudinal Data Systems (SLDS) programme, where the federal government has subsidised the development of state data systems, as well as the College Scorecard.

Under the SLDS and Workforce Data Quality Initiative, many states have created post-secondary longitudinal data systems and public-facing websites where students, policy makers and others can explore the earnings outcomes of higher education graduates in their state by institution, field of study and other graduate characteristics. Some states, such as Virginia, Texas and Tennessee, also collect graduate earnings data by programme level, including certificate programmes at the sub-baccalaureate level (Dorrer, 2016[112]). Post-secondary longitudinal data systems link administrative data on earnings from Unemployment Insurance (UI) wage records with data from higher education institutions on post-secondary graduates. It is estimated that administrative earnings data generally capture about 80% of the workforce (Pena, 2018[113]). Wage records do not provide information on all graduates who move out of the state, those who are self-employed, or federal civilian and uniformed military service members. These remains important gaps in state post-secondary data systems. States participating in a federal data sharing agreement are able to access wage data for graduates who live in other states to a limited extent. This system, formerly known as the Wage Record Interchange System (WRIS), has facilitated the exchange of wage data among participating states for the purpose of tracking individuals who have participated in workforce investment programmes in one state and then subsequently secured employment in another state, including employment and training programmes delivered by post-secondary education institutions. The purpose of the agreement was to assess and report on state and local employment and training program performance, and evaluate training provider performance. In 2020, the State Wage Interchange System (SWIS) Data Sharing Agreement replaced the WRIS Agreement (U.S. Department of Labor, 2019[114]). Importantly, the new data sharing agreement contains the US Department of Education, it meets the confidentiality requirements of the Family Educational Rights and Privacy Act (FERPA) for education records, and authorises record sharing for purposes other than training programme evaluation. Launched in 2013, the College Scorecard is the most prominent federal intervention specifically relating to student-oriented information on higher education and, particularly, labour market outcomes (Box 2.7).

copy the linklink copied!
Box 2.7. The College Scorecard

The College Scorecard is a website that publishes data on the costs of education and labour market outcomes of recipients of federal student aid at the level of individual post-secondary institutions, based on data from the National Student Loan Data System (NSLDS). The aim of the Scorecard is to improve the transparency of costs and likely outcomes of different college options to students and their families. A subset of data focusing on undergraduate study is provided through a consumer portal, while more comprehensive data are made available as file downloads or through API access. Key data available include:

  • completion metrics;

  • median earnings of graduates one year after graduation;

  • debt and loan repayment metrics;

  • average annual cost of attendance (net of grants and scholarships);

  • students’ average family incomes;

  • characteristics of students’ communities in aggregate based on ZIP codes;

  • transfer student metrics.

The coverage of Scorecard data has improved over time. In 2019, new data tracked first-year earnings and debt information college graduates at the four-digit level of the Classification of Instructional Programmes. This means, for example that the Scorecard can provide data on graduates from a history programme, but not on sub-fields such as American history. Over time, programme-level income data will extend to ten years after graduation, like the institution-level Scorecard income data.

Improving how the data are communicated to students has been a priority for the U.S. Department of Education, which has used the College Scorecard as a platform for private sector firms to develop innovative tools for. For example, the USDOE has been working with Google to help direct prospective students to the data when they are searching for information on potential colleges.

Sources: Rothwell (2015[115]); Schneider (2017[116]).

Some gaps remain in the College Scorecard. For instance, it cannot distinguish earnings between campuses of multi-campus institutions (Rothwell, 2015[115]). It only tracks programme graduates, which contributes to an important gap in data regarding those who do not complete their programmes. Most importantly, the Scorecard can only track students who receive federal aid, and these students are more disadvantaged than the average college student. Since 2008, under the Higher Education Act, the federal government is prohibited from developing, implementing or maintaining a database of personally identifiable information on individuals receiving federal financial assistance, “or any other system that tracks individual students over time”. The College Transparency Act proposes to revoke this restriction, as outlined in Box 2.8, as a means of improving the College Scorecard and other data.

New efforts are underway to address the limitations of state systems and the College Scorecard. For example, the U.S. Census Bureau’s Post-Secondary Employment Outcomes (PSEO) pilot project matches graduate transcripts from partner institutions with the national Longitudinal Employer Household Dynamics (LEHD) database, allowing for the outcomes of graduates to be analysed several years after graduation. Aggregate data on employment and earnings are published annually according to institution, degree level and degree major. The PSEO initiative differs from the College Scorecard programme, as the outcomes of all students are tracked, rather than federal aid recipients only, and outcomes data are tracked only for those who graduated from an in-scope institution with a certificate or degree. The PSEO project also has the important benefit of allowing graduate outcomes to be tracked out of their state of graduation, therefore overcoming a key limitation of existing state post-secondary data systems. To date, a few institutions in four states (Texas, Colorado, Michigan and Wisconsin) have participated in the initiative.

Another important challenge lies in the variety of information sources about higher education options and outcomes, which can be difficult to navigate for students. Beyond the College Scorecard, various federal platforms provide public access to institutional data from IPEDS, such as the College Navigator, the Net Price Calculator Center, and the College Affordability and Transparency List. To address this problem, there have been proposals to better co-ordinate the provision of information, for instance through the 2016 Strengthening Transparency in Higher Education Act, and to strengthen outcomes data through the 2019 College Transparency Act (see Box 2.8).

copy the linklink copied!
Box 2.8. Proposals to improve the quality and presentation of higher education data

The College Transparency Act (CTA) is a bipartisan, bicameral bill introduced in 2019 by Senator William Cassidy (R-LA). The CTA proposes eliminating the federal ban on student-level data and creating a student unit record system with more comprehensive outcomes data. Advocates for the legislation include public colleges and universities, who want to improve the data used to inform their programming and planning processes as well as the information provided to students to guide them in their educational choices. Community colleges have a particular interest in this legislation, primarily because the proposed changes would capture transfers on a more comprehensive basis and improve the data on graduation rates at community colleges (AACC, 2019[117]).

Key initiatives of the CTA include:

  • creating a secure student-level data network within the NCES using strong security standards and data governance protocols;

  • accurately reporting on student outcomes including enrolment, completion and post-college success across colleges and programmes, and providing information disaggregated by race, ethnicity and gender to identify inequities in students’ success;

  • feeding aggregate information back to states and institutions so they can develop and implement targeted, data-informed strategies aimed at supporting student success;

  • requiring a user-friendly website to ensure the data are transparent, informative, and accessible to students, parents, policymakers and employers.

Sources: AACC (2019[117]); U.S. Senator Cassidy (n.d.[118]).

The Strengthening Transparency in Higher Education Act was introduced in 2016 by Representative Virginia Foxx (R-NC). The bill focused on improving the presentation of data to students and their families when making educational choices. It would consolidate user-oriented information into a single College Dashboard that would be user-tested to improve its design. Specifically, the College Dashboard would include completion rates of students receiving Pell Grants, students classified as having a disability, and students receiving assistance under Department of Defense tuition assistance programmes. Importantly, it would also show information for all full-time students, not just first-time students who are starting post-secondary education for the first time. The bill also required that all institutional financial aid web pages link to the results of the Net Price Calculator, providing students with information on costs of attendance, available financial aid, and the share of students at institutions receiving financial aid.

Source: U.S. House Committee on Education and Labor (2015[119]).

Despite its growing role in the provision of information on graduate outcomes, the federal government has not taken a direct role in the provision of assistance to states in how to generate accessible, usable, trusted and relevant information. This is contrast to some other federal systems, like Canada, where a new body was created, the Labour Market Information Council, with representation from the federal government and each province and territory, to assess labour market information needs and support the development of effective tools of use across the country (see Box 3.12 in Chapter 3).

States have several opportunities to strengthen the labour market relevance and outcomes of higher education

Section 2.4 has provided an overview of the role of the federal government in US higher education, which is concentrated in specific areas of policy. This suggests that the federal system of higher education in the United States empowers states to take a leading role in funding higher education institutions, and to choose whether and how to place a focus on the labour market relevance of programmes and the outcomes of graduates. States are also permitted to create student aid programmes that complement federal programmes, and may choose to take account of the labour market and outcomes of higher education in designing their programmes in a way that federal programmes are not designed to do. It also shows that states are permitted by federal law, and more recently supported by federal programmes, in playing a very substantial role in creating and disseminating labour market information for learners. However, it also shows that the nation’s system of accreditation through self-regulation limits states’ ability to systematically use quality assurance as a means to orient state higher education systems towards labour market outcomes.


[117] AACC (2019), College Transparency Act of 2019, American Association of Community Colleges, (accessed on 2 March 2020).

[23] Abraham, K. and M. Kearney (2019), “Explaining the Decline in the U.S. Employment-to-Population Ratio: a Review of the Evidence”, NBER Working Paper No. 24333, The National Bureau of Economic Research,

[15] Acemoglu, D. and P. Restrepo (2017), “Robots and jobs: Evidence from US labor markets.”, NBER Working Paper No. 23285,

[73] Adecco USA (2019), The American Skills Gap is Real, (accessed on 3 February 2020).

[49] Andrews, R., J. Li and F. Lovenheim, Michael (2016), “Quantile Treatment Effects of College Quality on Earnings”, Journal of Human Resources, Vol. 51/1, pp. 201-238, (accessed on 17 May 2019).

[77] Antonio, A., M. Carnoy and C. Nelson (2018), “The United States pf America: Changes and Challenges in a Highly Decentralized System”, in Carnoy, M. et al. (eds.), Higher education in federal countries : a comparative study, SAGE Publication, Thousand Oaks, California, (accessed on 3 June 2019).

[18] Atalay, E. et al. (2018), “The Evolving U.S. Occupational Structure”, University of Michigan, (accessed on 21 May 2019).

[13] Autor, D., D. Dorn and G. Hanson (2013), “The China syndrome: Local labor market effects of import competition in the United States”, American Economic Review, Vol. 103/6, pp. 2121-2168,

[52] Autor, D. et al. (2017), “The Fall of the Labor Share and the Rise of Superstar Firms”, No. 10756, IZA Institute of Labour Economics, Bonn, (accessed on 12 March 2020).

[64] Backes, B., H. Holzer and E. Velez (2015), “Is it worth it? Postsecondary education and labor market outcomes for the disadvantaged”, IZA Journal of Labor Policy, Vol. 4/1, pp. 1-30,

[92] Barr, N. et al. (2018), “Reflections on the US College Loans System: Lessons from Australia and England”, IZA Discussion Papers, No. 11422, IZA - Institute of Labour Economics, (accessed on 16 March 2020).

[63] Blagg, K. et al. (2017), Rethinking Consumer Information in Higher Education, Urban Institute,

[70] Burning Glass Technologies and Strada Institute (2018), The Permanent Detour: Underemployment’s Long-Term Effects on the Careers of College Grads, (accessed on 4 February 2019).

[76] Capelli, P. (2015), “Skill Gaps, Skill Shortages, and Skill Mismatches: Evidence and Arguments for the United States”, Industrial & Labor Relations Review, Vol. 68/2, pp. 251-290,

[21] Card, D. (2009), “How immigration affects U.S. cities”, in Making Cities Work: Prospects and Policies for Urban America, Princeton University Press,

[14] Carnevale, A. (2016), The New ’Good Jobs’ - CityLab, (accessed on 12 March 2020).

[65] Carnevale, A. et al. (2017), Major Matters Most. The Economic Value of Bachelor’s Degrees from The University of Texas System, Georgetown University, (accessed on 20 May 2019).

[56] Carnevale, A., S. Rose and B. Cheah (2013), The College Payoff: Education, Occupation, Lifetime Earnings, Georgetown University, (accessed on 20 May 2019).

[54] Carnevale, A., S. Rose and A. Hanson (2012), “Certificates: Gateway to Gainful Employment and College Degrees”, Georgetown University Center for Education and the Workforce,

[61] Carnevale, A. et al. (2018), Our Separate & Unequal Public Colleges: How Public Colleges Reinforce White Racial Privilege and Marginalize Black and Latino Students, Georgetown University Center on Education and the Workforce., Washington DC, (accessed on 12 March 2020).

[57] Carruthers, C. and T. Sanford (2018), “Way station or launching pad? Unpacking the returns to adult technical education”, Journal of Public Economics, Vol. 165, pp. 146-159,

[82] CCIHE (2019), 2018 Update Facts & Figures - Descriptive Highlights, Center for Postsecondary Research, Indiana University School of Education, (accessed on 13 January 2020).

[44] Cellini, S. and N. Turner (2018), “Gainfully Employed? Assessing the Employment and Earnings of For-Profit College Students Using Administrative Data”, NBER Working Paper No. 22287, National Bureau of Economic Research, Massachusetts, (accessed on 30 August 2018).

[47] Chakrabarti, R. and M. Jiang (2018), Education’s Role in Earnings, Employment, and Economic Mobility, Federal Reserve Bank of New York, New York, (accessed on 25 July 2019).

[60] Chetty, R. et al. (2017), “Mobility Report Cards: The Role of Colleges in Intergenerational Mobility”, NBER Working Paper No. 23618, National Bureau of Economic Research, (accessed on 26 July 2019).

[85] College Board (2019), Pell grants: recipients, maximum Pell and average Pell, (accessed on 31 December 2019).

[39] College Board (2019), Trends in Student Aid 2019, Trends in higher Education Series, (accessed on 12 March 2020).

[79] College Board (2017), Trends in College Pricing 2017, College Board, New York,

[104] Congressional Research Service (2018), Reauthorization of the Perkins Act in the 115 th Congress: The Strengthening Career and Technical Education for the 21 st Century Act, Congressional Research Service, Washington DC, (accessed on 16 March 2020).

[53] Crouzet, N. and J. Eberly (2019), “Understanding Weak Capital Investment: the Role of Market Concentration and Intangibles”, NBER Working Paper No. 25869, National Bureau of Economic Research, (accessed on 12 March 2020).

[107] Cushing, E. et al. (2019), Developing a College- and Career-Ready Workforce: An Analysis of ESSA, Perkins V, IDEA, and WIOA, College & Career Readiness & Success Center, American Institutes for Research,

[83] Darolia, R. (2019), What Happens to Students When the Federal Government Sanctions Colleges? – Third Way, (accessed on 18 January 2020).

[16] Deming, D. (2017), “The Growing Importance of Social Skills in the Labor Market”, The Quarterly Journal of Economics, Vol. 132/4, pp. 1593-1640,

[84] Deming, D., C. Goldin and L. Katz (2012), “The For-Profit Postsecondary School Sector: Nimble Critters or Agile Predators?”, Journal of Economic Perspectives, Vol. 26/1, pp. 139-164,

[43] Deming, D. et al. (2016), “The Value of Postsecondary Credentials in the Labor Market: An Experimental Study”, American Economic Review, Vol. 106/3, pp. 778-806,

[112] Dorrer, J. (2016), “Using real-time labor market information to achieve better labor market outcomes”, Lumina Issue Papers, Lumina Foundation,

[45] Dynarski, S. (2015), “An economist’s perspective on student loans in the United States.”, No. 5579, Center for Economic Studies and Ifo Institute (CESifo), Munich, (accessed on 12 March 2020).

[78] Eckel, P. and J. King (2004), “An Overview of Higher Education in the United States: Diversity, Access, and the Role of the Marketplace”, in Forest, J. and P. Altbach (eds.), The International Handbook of Higher Education, (accessed on 3 June 2019).

[51] Engbom, N. and C. Moser (2017), “Returns to Education through Access to Higher-Paying Firms: Evidence from US Matched Employer-Employee Data”, American Economic Review: Papers & Proceedings, Vol. 107/5, pp. 374-378,

[110] Federal Reserve (2019), Report on the Economic Well-Being of U.S. Households in 2018 - Higher Education, (accessed on 11 March 2020).

[41] FRBNY (2019), Quarterly Report on Household Debt and Credit - 2019:Q3. November., Federal Reserve Bank of New York, (accessed on 12 March 2020).

[80] Fulton, M. (2019), An Analysis of State Postsecondary Governance Structures, Education Commission on the States,

[19] Giannone, E. et al. (2017), “Skilled-Biased Technical Change and Regional Convergence”, University of Chicago, (accessed on 12 March 2020).

[101] Good, L. and E. Strong (2015), “Reimagining Workforce Policy in the United States”, in Van Horn, C., T. Edwards and T. Greene (eds.), Transforming U.S. Workforce Development Policies for the 21st Century, W.E. Upjohn Institute for Employment Research, Kalamazoo, MI, (accessed on 16 March 2020).

[93] Harris, A. and E. Kelderman (2017), House Republicans Press for Higher-Ed Overhaul in 2018, The Chronicle of Higher Education, (accessed on 16 January 2020).

[74] Hart Research Associates (2018), Fulfilling the American Dream: Liberal Education and the Future of Work. Selected Findings from Online Surveys of Business Executives and Hiring Managers.,

[6] Hermansen, M. (2019), “Occupational licensing and job mobility in the United States”, OECD Economics Department Working Papers, No. 1585, OECD Publishing, Paris,

[17] Hershbein, B. and L. Kahn (2018), “Do recessions accelerate routine-biased technological change? evidence from vacancy postings”, American Economic Review, Vol. 108/7, pp. 1737-1772,

[59] Hoxby, C. (2018), “The Productivity of U.S. Postsecondary Institutions”, in Hoxby, C. and K. Stange (eds.), Productivity in Higher Education, National Bureau of Economic Research/University of Chicago Press, (accessed on 7 August 2019).

[71] IHE (2019), Survey: Employers Want ’Soft Skills’ From Graduates, Inside Higher Ed, (accessed on 3 February 2020).

[46] Itzkowitz, M. (2018), Want More Students to Pay Down their Loans: Help Them Graduate, Third Way,

[105] JFF (2018), Leveraging Perkins V to support college and career pathways, Jobs for the Future,

[87] Jump, J. (2019), Ethical College Admissions: Beyond Gainful Employment, Inside Higher Ed., (accessed on 3 February 2020).

[97] Kelchen, R. (2017), Higher Education Accreditation and the Federal Government, Urban Institute , Washington, DC, (accessed on 10 May 2019).

[86] Kelchen, R. and Z. Liu (2019), “Did Gainful Employment Regulations Result in College and Program Closures? An Empirical Analysis”.

[88] Kreighbaum, A. (2018), GAO: Colleges, Consultants Game Rules to Lower Default Rates, Inside Higher Ed., (accessed on 19 January 2020).

[8] Krueger, A. and O. Ashenfelter (2018), Theory and Evidence on Employer Collusion in the Franchise Sector, IZA/ National Bureau of Economic Research, (accessed on 12 March 2020).

[24] Lawrence, R. (2018), “Recent US Manufacturing Employment: The Exception that Proves the Rule”, 18-002, Harvard Kennedy School of Government - Faculty Research Working Paper Series, (accessed on 12 March 2020).

[12] ManpowerGroup (2018), 2018 Talent Shortage Survey: United States., WI: ManpowerGroup, Milwaukee, (accessed on 12 March 2020).

[38] McFarland, J. et al. (2018), The Condition of Education 2018, US Department of Education - National Center for Education Statistics., Washington DC, (accessed on 12 March 2020).

[35] Mitchell, M. et al. (2018), Unkept Promises: State Cuts to Higher Education Threaten Access and Equity, (accessed on 12 June 2019).

[22] Molloy, R., C. Smith and A. Wozniak (2017), “Job Changing and the Decline in Long-Distance Migration in the United States”, Demography, Vol. 54/2, pp. 631-653,

[68] Montt, G. (2015), “The causes and consequences of field-of-study mismatch: An analysis using PIAAC”, OECD Social, Employment and Migration Working Papers, No. 167, OECD Publishing, Paris,

[20] Moretti, E. (2012), The new geography of jobs, Houghton Mifflin Harcourt.

[62] NCES (2019), Digest of Education Statistics, National Center for Education Statistics, (accessed on 17 January 2019).

[81] NCES (2019), Integrated Postsecondary Education Data System (IPEDS), 12-month Enrollment component 2018-19 provisional data, National Center for Education Statistics - US Department of Education, Washington, D.C., (accessed on 5 October 2020).

[33] NCES (2018), Digest of Education Statistics, 2018, National Center for Education Statistics - US Department of Education, Washington, D.C., (accessed on 11 March 2020).

[37] NCES (2018), Tuition costs of colleges and universities, National Center for Education Statistics, (accessed on 13 June 2019).

[27] NCES (2017), Table 303.20. Total fall enrollment in all postsecondary institutions participating in Title IV programs and annual percentage change in enrollment, by degree-granting status and control of institution: 1995 through 2016, National Center for Education Statistics, (accessed on 5 June 2019).

[3] OECD (2020), Average wages (indicator), (accessed on 12 March 2020).

[26] OECD (2020), Population with tertiary education (indicator), (accessed on 12 March 2020).

[2] OECD (2020), Trade Union, (accessed on 12 March 2020).

[4] OECD (2020), Wage levels (indicator), (accessed on 12 March 2020).

[34] OECD (2019), Benchmarking Higher Education System Performance, Higher Education, OECD Publishing, Paris,

[50] OECD (2019), “Employment, unemployment and inactivity rates of 25-34 year-olds, by educational attainment (2018)”, in Education at a Glance 2019: OECD Indicators, OECD Publishing, Paris,

[25] OECD (2019), Education at a Glance 2019: OECD Indicators, OECD Publishing, Paris, (accessed on 4 15 2020).

[32] OECD (2019), “Relative share of public, private and international expenditure on educational institutions, final source of funds (2016)”, in Education at a Glance 2019: OECD Indicators, OECD Publishing, Paris, (accessed on 10 May 2020).

[1] OECD (2019), OECD Employment Outlook 2019: The Future of Work, OECD Publishing, Paris,

[67] OECD (2019), Skills Matter: Additional Results from the Survey of Adult Skills, OECD Skills Studies, OECD Publishing, Paris, (accessed on 10 April 2020).

[10] OECD (2018), Labour Force Survey by sex and age - indicators, (accessed on 12 March 2020).

[29] Page, L. and J. Scott-Clayton (2016), “Improving college access in the United States: Barriers and policy responses”, Economics of Education Review, Vol. 51, pp. 4-22,

[113] Pena, C. (2018), Workforce success relies on transparent postsecondary data, National Skills Coalition,

[36] Ripley, A. (2018), Why Is College in America So Expensive?, Emerson Collective,

[69] Rose, S. (2017), Mismatch: How Many Workers with a Bachelor’s Degree Are Overqualified for Their Jobs?, Urban Institute , Washington, (accessed on 21 May 2019).

[9] Rossi-Hansberg, E., P. Sarte and N. Trachter (2018), “Diverging Trends in National and Local Concentration”, Federal Reserve Bank of Richmond Working Papers, Vol. 18/15, pp. 1-39,

[115] Rothwell, J. (2015), Understanding the College Scorecard, (accessed on 16 March 2020).

[116] Schneider, M. (2017), Reforms to Increase Transparency in Higher Education - What to do: Policy Recommendations for 2017, American Enterprise Institute, (accessed on 16 March 2020).

[58] Schneider, M. (2015), Higher Education Pays: But a Lot More for Some Graduates Than for Others, College Measures, (accessed on 20 May 2019).

[40] Scott-Clayton, J. (2017), Undergraduate Financial Aid in the United States, American Academy of Arts & Sciences., Washington DC, (accessed on 12 March 2020).

[72] SHRM (2019), Employers Say Students Aren’t Learning Soft Skills in College, Society for Human Resource Management, (accessed on 3 February 2020).

[7] Starr, E., J. Prescott and N. Bishara (2019), “Noncompetes in the U.S. Labor Force”, University of Michigan Law & Economics Research Paper, Vol. 18/013,

[111] Strada Education; Gallup (2019), From College to Life: Relevance and the Value of Higher Education, (accessed on 11 March 2020).

[55] Strada; Gallup; Lumina Foundation (2019), Certified Value: When do Adults without Degrees Benefit from Earning Certificates and Certifications?,

[91] Stratford, M. (2019), DeVos finalizes ’borrower defense’ rules as critics plan lawsuits, Politico, (accessed on 31 December 2019).

[98] TICAS (2018), Of metrics and markets: Measuring post-college employment success, The Institute for College Access and Success,

[11] U.S. Bureau of Labor Statistics (2020), Labor Force Statistics from the Current Population Survey., (accessed on 12 March 2020).

[5] U.S. Bureau of Labor Statistics (2020), Table 3. Median usual weekly earnings by age, race, Hispanic or Latino ethnicity, and sex, (accessed on 12 March 2020).

[28] U.S. Bureau of Labour Statistics (2019), College Enrollment and Work Activity of Recent High School and College Graduates Summary, (accessed on 15 April 2020).

[31] U.S. Census Bureau (2019), Educational Attainment in the United States: 2010 - Current Population Survey, 2010 Annual Social and Economic Supplement, (accessed on 12 March 2020).

[30] U.S. Census Bureau (2019), Educational Attainment in the United States: 2018 - Current Population Survey, 2018 Annual Social and Economic Supplement., (accessed on 12 March 2020).

[106] U.S. Department of Commerce (n.d.), American Workforce Policy Advisory Board, (accessed on 5 May 2020).

[109] U.S. Department of Education (2020), Perkins V, Perkins Collaborative Resource Network, (accessed on 2 March 2020).

[100] U.S. Department of Education (2019), Secretary DeVos Finalizes Higher Education Regulations that Promote Innovation, Protect Students, and Reduce Regulatory Burden, (accessed on 5 May 2020).

[90] U.S. Department of Education (2019), U.S. Department of Education Finalizes Regulations to Protect Student Borrowers, Hold Higher Education Institutions Accountable and Save Taxpayers $11.1 Billion Over 10 Years, (accessed on 5 May 2020).

[89] U.S. Department of Education - Federal Student Aid (2020), Borrower Defense, (accessed on 5 May 2020).

[114] U.S. Department of Labor (2019), State Wage Interchange System (SWIS) Data Sharing Agreement, (accessed on 6 May 2020).

[99] U.S. House Committee on Education and Labor (2020), H.R.5768 Accreditation Reform Act of 2020, (accessed on 5 May 2020).

[96] U.S. House Committee on Education and Labor (2019), H.R.4674 College Affordability Act, (accessed on 5 May 2020).

[119] U.S. House Committee on Education and Labor (2015), H.R. 3178 Transparency Fact Sheet, (accessed on 18 March 2020).

[94] U.S. House Committee on Education and the Workforce (2017), H.R.4508 PROSPER Act, (accessed on 5 May 2020).

[95] U.S. Senate HELP Committee (2019), The Student Aid Improvement Act of 2019, U.S. Senate Committee on Health, Education, Labor and Pensions, (accessed on 2 March 2020).

[118] U.S. Senator Cassidy (n.d.), College Transparency Act: One Pager, (accessed on 18 March 2020).

[42] Urban Institute (2017), Understanding College Affordability, (accessed on 12 March 2020).

[108] von Zastrow, C. (2018), Perkins V: Expanding Opportunities for Work-Based Learning, Education Commission on the States: EdNote, (accessed on 2 March 2020).

[75] Waddoups, C. (2016), “Did Employers in the United States Back Away from Skills Training during the Early 2000s?”, Industrial & Labor Relations Review, Vol. 69/2, pp. 405-434,

[48] Webber, D. (2014), “The lifetime earnings premia of different majors: Correcting for selection based on cognitive, noncognitive, and unobserved factors”, Labour Economics, Vol. 28, pp. 14-23,

[102] Wolff, K. (2015), Governing the Public Workforce System: The Structure and Priorities of Local Workforce Investment Boards, Social Policy Research Associates, (accessed on 2 March 2020).

[103] Workforce Information Advisory Council (2018), Recommendations to Improve the Nation’s Workforce and Labor Market Information System., US Department of Labor, Washington DC, (accessed on 16 March 2020).

[66] Zimmerman, S. (2014), The Returns to College Admission for Academically Marginal Students,

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2020

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at