4
Children’s Education
Education plays a vital role in human development and national prosperity. It is central to citizenship and supports human flourishing in many ways, including the development of the knowledge and skills that will allow children to be successful in the labor market. Earnings from work have been crucial in allowing past generations of families to avoid intergenerational poverty, and that will remain true in the future. Over the last 50 years, earned income consistently lifted the non-work-related incomes of between 70% and 75% of families above the Supplemental Poverty Measure–based poverty line (National Academies of Sciences, Engineering, and Medicine [National Academies], 2019a, Figure 4-1). Building children’s earning capacities is key for enabling them to avoid poverty in adulthood.
Evidence reviewed in this chapter shows that educational outcomes such as years of completed schooling have a strong causal connection with higher earnings and other important measures of life success and well-being, and thus with the ability of children to rise out of poverty when they are adults. There is also a strong association between student achievement—as measured by test scores—and labor market outcomes, although in this case it is more difficult to show a causal connection. Thus, educational institutions that help children from families in poverty to complete more schooling and, perhaps, achieve higher test scores have the potential to dramatically reduce the intergenerational transmission of poverty.
However, the education system is not always as successful as it might be at delivering these educational outcomes. Children who grow up in families in poverty are systematically exposed to lower-quality educational experiences and resources than children from wealthier families. This, in
SOURCE: Data from DeBrey et al. (2021), Tables 501.50, 501.60, and 501.70.
turn, leads to lower levels of achievement, completed schooling, and earnings in adulthood. Resource and achievement gaps are also apparent across children of different races and ethnicities.
Fortunately, recent educational research and evaluation work points to promising ways to increase the chances that children growing up in low-income families will enter the labor market with the skills needed to keep their family incomes well above the poverty threshold.
This chapter begins by reviewing evidence on the importance of education for children’s eventual economic success. It then considers, in turn, four components of the educational process: child care and early education programs for young children; elementary and secondary school; postsecondary education; and, for some, career training. While the chapter focuses on the importance of children’s educational experiences for their development, there is a less direct channel that may be even more important in the long run: Educational successes in one generation may affect children of the next generation by influencing parenting practices, helping parents to provide enriching home environments and in other ways support their children’s educational experiences (e.g., Currie & Moretti, 2003).
HOW EDUCATION AFFECTS THE ECONOMIC MOBILITY OF CHILDREN
The skills that people bring to the labor market are a key determinant of success at finding sufficient employment and earnings to lift themselves and their families above the poverty line. Autor (2019) documents the increasing bifurcation of work in industrialized countries over the past 50 years into occupations that require high levels of education and pay high wages, on the one hand, and low-education occupations that pay relatively little—and often leave workers and their families in poverty—on the other. Traditional midlevel jobs that require only a high school education but pay well are less and less available, and workers who might have secured these jobs in the past are now largely relegated to lower-wage jobs that do not pay enough to lift a family above the poverty line.
Most workers who acquire the skills needed for professional, managerial, and technical jobs do so through formal postsecondary schooling. Workers with college degrees are much more likely to be employed than those with less education, and when they are employed they earn substantially more. Employment rates among 25–34-year-olds in 2019 were 30 percentage points higher (87% vs. 57%) for workers with college degrees than for workers who had not finished high school (Figure 4-1). And while employment rates are generally higher for men than women, both groups show the same pattern of increased employment with higher levels of completed schooling.
Earnings, too, increase steadily with additional years of education: College graduates earn twice as much as high school dropouts and 50% more than workers with high school diplomas but no additional education (Figure 4-2). Autor (2014) shows that even after accounting for tuition costs and the time value of money, college graduates can expect to earn between $500,000 and $800,000 (in 2022 dollars) more than high school graduates over the course of their careers.
Decades of rigorous labor economics research demonstrate that these earnings advantages are not simply reflections of other differences between graduates and nongraduates. Instead, they largely reflect the labor market rewards generated by the knowledge and skills that students gain as they complete more schooling (Card & Giuliano, 1999). On average, each additional year of education causes subsequent earnings to increase by 7% to 12%. Educational credentials can signal persistence, conscientiousness, and other noncognitive traits to employers (as argued, for example, by Caplan, 2018). But the rewards to completed schooling reflect more than credentialism; even a year or two of college, without a degree, raises employment and earnings over what someone would obtain with just a high school diploma. Moreover, the benefits of education are not confined to the labor
SOURCE: Data from U.S. Department of Commerce, Census Bureau, Current Population Survey, Annual Social and Economic Supplement, 1996–2020, Table 502.30.
market—research has also shown that additional years of education reduce criminal behavior (Lochner, 2020) and improve health (Galama et al., 2018; Hamad et al., 2018) and parenting (Currie & Moretti, 2003).
Labor market success appears to depend not only on completed years of schooling, but also on the skills acquired during time spent in school. Many studies have documented associations between test scores, even as early as kindergarten (Chetty et al., 2011), and labor market success. Associations between earnings and high school reading and math test scores appear to be just as strong for men as for women, and they predict earnings as late as age 50 (Watts, 2020).
The labor market rewards for education and skills have grown in magnitude over the past four decades, as documented in Chapter 6. This appears to be partly because technological change and globalization have increased the productivity of highly educated or highly skilled workers, and partly because of the rising monopoly power of employers in the labor market and the weakening of institutions (such as unions and minimum wage statutes) that traditionally protect workers from such power. For an individual child, educational attainment—in whatever form—is more
important than ever for escaping poverty in adulthood, and it is a major driver of intergenerational poverty and mobility.
While a college education is an important path out of poverty, other paths are available to workers without college degrees. These often involve acquiring occupational skills in high-demand industries (such as health care, advanced manufacturing, information technology, construction, and transportation/distribution logistics) that allow workers to earn more than their counterparts without these skills. Such skills can be attained in high school career and technical education (CTE) and in work-based learning programs; in various certificate programs in community (or for-profit) colleges, whether or not they are for-credit programs; and in high-quality sector-based training programs, as noted below.1
ACHIEVEMENT AND ATTAINMENT DIFFERENCES ACROSS RACIAL AND ETHNIC GROUPS
Given the importance of achievement and years of completed schooling for reducing intergenerational poverty, it is alarming to see large gaps on those measures between income, racial, and ethnic groups. Looking at the reading and math scores of children who entered kindergarten in 2010, Reardon and Portilla (2016) found that children from low-income (10th percentile) families were more than a year behind children from high-income (90th percentile) families. Black and Latino students entering kindergarten were about half a year behind their White counterparts in early math achievement (separate data on Native Americans were not available). In the case of kindergarten-entry literacy skills, Latino students again lagged behind their White classmates by about half a year, with a somewhat smaller gap for Black students. Achievement gaps that become evident in kindergarten tend to remain relatively stable or grow slightly after second grade (Kuhfeld et al., 2020; Paschall et al., 2018; Reardon, 2021).
The National Assessment of Educational Progress has tracked reading and math proficiency for decades. Figure 4-3 shows the fractions of 8th graders of different races and ethnicities judged to be proficient in reading; similar patterns appear in 4th grade and in both 8th and 4th grades for math achievement.
Although proficiency rates have increased somewhat for most groups, the rates themselves are very low—generally around 20% for low-income
___________________
1 On average, for-credit community college credentials generate higher labor market rewards than not-for-credit credentials, although both can earn a labor market premium (Baum et al., 2020). Credentials from for-profit colleges also generate lower returns, on average, than those from colleges in the public nonprofit sector (Cellini & Turner, 2019).
SOURCE: Data from National Assessment of Education Progress. https://www.nationsreportcard.gov/ndecore/xplore/NDE
students and for Black, Latino, and Native American/Alaska Native students. Asian students are the only group with proficiency rates above 50%.
These early and persistent achievement differences are no doubt largely due to the differences between the childhood environments experienced by children from more and less affluent families. Higher-income families are able to provide safer and more nurturing home environments and tend to have different parenting styles (Bassok et al., 2016; Phillips et al., 1998). There are racial and socioeconomic differences in exposure to a range of influences that affect cognition and learning—for example, Sampson et al. (2008) and Currie et al. (2014) document racial differences in exposure to violence and pollution (see also Chapters 5 and 9). Higher-income families are also able to invest more in supporting their children’s education, spending an average of $8,000 more annually than lower-income families on educational enrichments such as books, computers, high-quality child care, summer camps, and private schooling (Kaushal et al., 2011). All of these differences contribute to kindergarten-entry achievement gaps across racial/ethnic groups, as well as to differences in learning throughout the educational process. The differences in out-of-school environments make
it very difficult for public schools to equalize educational opportunities across groups.
Importantly, gaps in out-of-school environments tend to accompany gaps in the quality of K-12 schools. While many states have modified their school funding systems in recent years to ensure that schools serving low-income students are as well funded as those serving high-income students, other states have not (Lafortune et al., 2018). Moreover, schools serving low-income students face higher costs (for safety, remedial education, special education, students’ basic needs, and so on), so even equal funding cannot equalize the quality of educational inputs (Duncombe & Yinger, 2005). By nearly any metric, average school quality is lower at schools serving high-poverty populations. As a result, while in principle the formal education system could help to close preexisting gaps, it often seems to magnify them instead (Chetty et al., 2023; Hashim et al., 2023; Reardon, 2011).
Given achievement gaps throughout K-12, it is unsurprising to see parallel gaps in completed schooling. The top panel of Figure 4-4 shows rates of fall college enrollment among previous spring high school graduates, while the bottom panel shows the fractions of all young adults (defined here as people between ages 25 and 29) holding college or advanced degrees as of 2019. Greater than 70% of Asian young adults had completed college, compared with 45% of White, 29% of Black, and 21% of Latino young adults and 14% of young adults in the Native American category. It is encouraging to note that among Black and Latino people as well as White people, these rates were more than 10 percentage points higher than they had been two decades earlier (earlier data are unavailable for Asian people). In the case of Native American people, however, the rates were lower in 2019 than in 2000.
Conclusion 4-1: By imparting skills and other capacities valued by employers, the education system is a key driver of upward intergenerational mobility for low-income children. Large gaps in school achievement and completed schooling persist across socioeconomic, racial, and ethnic subgroups, pose a key challenge for policy makers seeking to reduce intergenerational poverty, and underscore the importance of education-related interventions.
EARLY-LIFE EDUCATION, CARE, AND PARENTING
The early years of life lay the groundwork for a child’s healthy cognitive and behavioral development (Knudsen et al., 2006). From the time of conception, early development is a complex interplay between the child’s genetic blueprint and the early experiences that are essential for subsequent learning and development (Shonkoff, 2010). Children thrive when they are
NOTES: The top panel shows the share of new high school graduates who are enrolled in college the following fall. The bottom panel shows the share of 25–29-year-olds who had BAs.
SOURCE: Graduation rates are for 2019 and are taken from the Digest of Education Statistics, 2019, Table 104.20. Enrollment rates are for 18-24-year-olds in 2015 and are taken from Musu-Gillette et al. (2017), Figure 18.2.
well nourished, protected from disease and violence (see Chapter 5 for more details regarding health), and have caregivers who are responsive to their needs and provide them with learning opportunities from birth onwards (Black et al., 2017). Extensive evidence documents the importance of responsive and stimulating caregiving during a child’s first 5 years, as social and academic skills are developing (Shonkoff, 2010). The importance of these early experiences appears to be universal, but their impacts vary and depend on the family’s culture (Rogoff, 2003).
Early childhood educational programs can enhance children’s early development (Black et al., 2017; Shonkoff, 2010). Early care and education programs in which caregivers form responsive and supportive relationships with the children in their care and provide stimulating and engaging learning opportunities promote healthy development (Hamre & Pianta, 2010).
Children from impoverished families experience lower-quality community-based early childhood education; however, programs like Head Start and public pre-kindergarten have been funded by federal, state, and local governments to ensure access to higher-quality programs for low-income children (Kraus-Friedman et al., 2020; Zigler & Styfco, 2010). In theory, these programs should provide children with the skills they need to succeed in school (Bailey, 2021; Deming, 2009; Garces et al., 2002; García et al., 2021).
Similarly, home-visiting programs promote infant and child health, foster educational development, and help prevent child abuse and neglect by arranging for trained professionals or paraprofessionals to pay regular visits to parents, typically mothers, and to provide coaching on parenting issues (Duncan et al., 2023). The home visitors offer a wide variety of supports, ranging from facilitating access to public services and modeling positive behavior management to addressing discipline issues to promoting stimulating learning activities and interactions (Office of Planning, Research and Evaluation, 2021). Some of these programs have demonstrated long-term impacts (Olds et al., 1997, 1998).
In light of such compelling evidence, early childhood education programs are widely regarded as one of the most effective means to promote success in school for low-income children (Heckman, 2011). Despite the proven potential of these kinds of programs, however, for reasons detailed in Appendix C: Chapter 4, the committee was unable to propose expansions of existing programs or new programs that would meet the evidentiary standards the committee has set for itself. There were two reasons for this: First, many of the programs that showed long-term benefits, such as the Perry Preschool Project, Abecedarian Project, and Nurse Family Partnership programs, were designed and run by researchers. Scaling up demonstration programs to serve hundreds of thousands of children and families, in the face of relentless pressures to cut costs, can significantly dilute program quality. Indeed, the classroom experiences in programs like Head Start, which in the early years showed long-term impacts (e.g., Bailey, 2021), are quite different from classroom experiences in today’s programs (Markowitz & Ansari, 2020).
Second, the families of children not selected into these programs (including Head Start in its early years) faced much worse conditions than do the families of children today who do not participate in such programs. During those early years, safety-net programs like Food Stamps and Medicaid were not yet universally available, center-based child care choices were more limited, and parental schooling levels and spending on children’s “enrichment goods” were much lower than they are now (Duncan et al.,
2014). It was therefore much easier for early programs to demonstrate long-run effectiveness.
Because of the ambiguity of the evidence, the committee was unable to identify the best ways to expand early childhood education and home visitation programs in order to reduce intergenerational poverty. While it is certainly possible that expanding or reforming our current patchwork of early childhood and home visitation programs would reduce intergenerational poverty, we do not know enough about how to do so in a manner that is very likely to generate long-run benefits.
ELEMENTARY AND SECONDARY EDUCATION
Educational quality can be measured in many ways, but at the most basic level it reflects the extent to which students are receiving instruction that matches their skill level, supports and is responsive to their needs, and helps them advance to the next level of proficiency. This is difficult to do in crumbling buildings (Lafortune & Schonholzer, 2022), with teachers who are poorly qualified or inexperienced (Goe, 2007), or in overcrowded classrooms (Krueger, 1999). Children in households living in poverty are more likely than more affluent children to attend struggling public schools that repeatedly fail to meet accountability standards, are burdened by crumbling physical infrastructure and high levels of violence, and are staffed with less experienced, lower-quality teachers (Kozol, 1991, 2005; Lankford et al., 2002). These schools have fewer enrichment courses, are often unable to offer even the minimum courses needed to prepare their students for four-year colleges, and provide far fewer extracurricular activities.
Moreover, even if instruction is expertly delivered in an up-to-date facility by excellent teachers, students will struggle to learn if they are unstably housed, lack regular and appropriate meals or adequate medical, dental, or vision care, or are threatened by crime and violence on their way to and from school (Rothstein, 2004). Many of these barriers to learning are outside the purview of traditional education, although recent movements to follow a “community schools” model or establish school-based health centers, as well as other expansions of the usual school mission, have attempted to broaden that purview to address students’ multiple needs (Maier et al., 2017). All these challenges are much more difficult to address in schools serving high concentrations of children in poverty. Persistently high levels of neighborhood and school economic and racial segregation therefore represent a real barrier to student progress.
If schools are to provide a high-quality educational environment, they need adequate funding. Traditionally, low-income schools have had fewer resources than schools serving wealthier students, despite their students’ greater needs. Many states have reformed their school finance systems to
ensure that funding in low-income districts equals or even exceeds that in higher-income districts, although others have not. Several recent studies of these reforms show that directing more funding to low-income schools raises students’ test scores in the short run (Jackson & Mackevicius, 2021; Lafortune et al., 2018) as well as boosting their longer-run earnings and health (Jackson et al., 2016; Rothstein & Schanzenbach, 2022). Other studies show shorter-term positive impacts of spending on capital improvements (Lafortune & Schonholzer, 2022); however, longer-term impact estimates for these kinds of investments are not yet available.
Of course, funding is closely related to school segregation. Predominantly non-White schools tend to have less funding, fewer resources, and less skilled teachers (Bischoff & Owens, 2019; Elder et al., 2021). The Supreme Court’s landmark Brown v. Board of Education decision and subsequent court orders resulted in the desegregation of school districts around the country, which led to slow but substantial racial integration throughout the United States (Orfield et al., 2016; Reardon & Owens, 2014; Reardon & Yun, 2003; Reardon et al., 2012). One set of studies using national data compared the educational and occupational attainment of students enrolled before and after courts issued desegregation orders in the 1960s and 1970s, and it found that the resulting desegregation improved educational and occupational attainment among Black adults (Johnson, 2011; Johnson & Nazaryan, 2019). Specifically, each additional year after court-ordered desegregation led to a 1.8-percentage-point increase in the likelihood of high school graduation, and the average effects of 5 years of exposure to court-ordered school desegregation led to about a 15% increase in wages. This study used the same comparisons to examine the ways in which schools may have changed in response to the court orders and found two potential mechanisms: increased per-pupil spending and reduced class sizes. However, Supreme Court decisions since 1991 have made it easier for school districts to be released from prior court orders to desegregate, and they provide limited guidance on maintaining integration after those orders are lifted. This, coupled with rising income inequality and residential segregation, has contributed to school segregation levels that are, by many measures, as high as they were before the school desegregation movement began (Reardon & Owens, 2014; Reardon et al., 2012).
Context, school composition, and funding are obviously important. In addition, students will learn more when instruction is more effectively delivered. Efforts to identify the active ingredients of school quality have had mixed success, however. Some “structural” factors that are easy to measure and screen for, such as teacher experience and educational credentials, fail to show consistent links with student achievement (e.g., Papay & Kraft, 2015). However, smaller class sizes in the early grades—which are closely related to school resources—have been linked to better outcomes (Chetty
et al., 2011; Krueger, 1999), and recruiting science, technology, engineering, and mathematics (STEM)-trained college graduates to teach math and science classes has been shown to correlate with better student grades and test scores in these subjects (Backes et al., 2018).
The No Child Left Behind Act of 2001 refocused federal education policy to hold schools accountable for student performance, and some evidence indicates that this accountability had modest but positive impacts on student outcomes. A related movement has been the rapid growth of charter schools, which provide alternatives to public schools that can test different instructional or organizational strategies. Evidence indicates that on average, charter schools are neither better nor worse at promoting student achievement (CREDO, 2013) than conventional public schools. However, a subset of charter schools known as “no excuses” schools have been found to have substantial positive effects on students’ test scores and 4-year college enrollment, relative to traditional public-school alternatives (Angrist et al., 2016). This is particularly important because these types of charter schools disproportionately serve low-income urban students, use a very different educational model (more school time, drilling, testing, and emphasis on behavior) than is typically found in schools serving middle- and upper-class students.
Both charter and traditional public schools have tested several new strategies for increasing student learning, many of which show promise. For example, rigorous evaluations have been conducted of tutoring programs that provide for frequent one-on-one or small-group interactions with struggling students (e.g., Fryer et al., 2016). Although these programs are generally too new to have demonstrated long-term impacts, early evidence of short-run impacts is promising, suggesting that personalizing instruction to students’ specific skills and needs may be beneficial.
The nature of the match between students and their teachers may also matter for student learning. Causal evidence shows that having a Black elementary school teacher has positive effects on high school graduation and 2-year college enrollment for low-income Black boys—a population of particular interest in this report (Gershenson et al., 2022). These effects appear to be generated by some combination of role modeling, fewer disciplinary actions, and higher teacher expectations (Gershenson et al., 2016; Lindsay & Hart, 2017). Moreover, a recent study found that enrollment in Ethnic Studies courses improved rates of high school graduation and engagement, and may also increase college enrollment (Bonilla et al., 2021). These courses focus on social justice, anti-racism, stereotypes, and social movements in U.S. history led by people from various racial and ethnic minority groups, spanning the period from the late eighteenth century to the 1970s.
The content of course offerings and the labor market skills they can impart to students also appear to be significant. For students who might not go directly from high school to college, CTE in high schools offers an alternative path to better jobs over time, although it can also divert students—particularly low-income students and students of color—from a path toward traditional college enrollment. Many young adults who begin full-time work after completing their schooling return to take advantage of vocational training and adult programs, often through community colleges, that provide job skills outside of a traditional academic setting. Evidence indicates that these programs, when implemented well, can also increase earnings (Brunner et al., 2021; Hemelt et al., 2021).
As seen in the case of charter schools, school policies beyond direct instruction can have large impacts on student outcomes. Research shows that harsh school discipline, such as suspensions and expulsions, leads to long-term negative outcomes for children. One study, for example, found that students who were randomly assigned to schools with higher suspension rates were more likely to be arrested or incarcerated as adults, more likely to drop out of high school, and less likely to attend a 4-year college. The effects on arrest and incarceration were substantially larger for Black, Latino, and male students, and especially for Black and Latino males (Bacher-Hicks et al., 2019; also see Chu and Ready, 2018). No studies have examined the long-term effects of interventions to reduce harsh school discipline; however, Appendix C: Chapter 4 discusses some promising strategies with documented shorter-run beneficial impacts.
POSTSECONDARY EDUCATION
Although the bottom panel of Figure 4-4 (above) shows widely disparate rates of college completion across racial/ethnic groups, the top panel shows that the rates of fall college enrollment across different groups of recent high school graduates are more similar. Thus, Figure 4-4 points to a key challenge in seeking to reduce intergenerational poverty through postsecondary education: Many of those who enroll fail to graduate with 4-year degrees.
An important factor underlying differences in enrollment rates is success in secondary school. One recent study found, for example, that Black students are more likely to enroll in college than are White students with similar family backgrounds and prior academic performance (Ciocca & DiPrete, 2018).
Two-year colleges offer a wide range of associate degrees and certificates (along with noncredit programs), many of which have significant labor market value (Backes et al., 2015; Holzer & Baum, 2017). But they can also be complicated institutions for students to navigate, and they
often provide too little structure or guidance (Bailey et al., 2015; Holzer & Xu, 2021; Scott-Clayton, 2011). Students need to not only make many high-stakes academic decisions (major or program, classes to take each term, whether to transfer, and if so, where), but also navigate a wide range of needlessly complex processes (transferring credits if they take classes at more than one institution, navigating financial aid, arranging for parking, obtaining support services). These challenges are not intellectually or academically important, but failing to complete them can mean not being able to continue in a program.
Among those who aspire to earn a bachelor of arts degree, students starting at 2-year rather than 4-year institutions are less likely to complete their studies. More general differences in postsecondary success are driven by a range of factors, including weaker academic preparation, as a result of having attended lower-quality K-12 schools; a lack of family financial resources and the necessary information for choosing the institution that best suits their needs; the need to work full-time while attending college, making it very difficult to be a full-time student; a lack of social capital and guidance on accessing available resources or studying effectively; and a lack of support services at their current institutions (Baum & McPherson, 2022; Holzer & Baum, 2017). Low-income students of color may face cultural and social barriers at predominately White institutions. All these factors lead to higher dropout rates, lower grades, and a lower probability of success in majors leading to highly compensated employment (Bleemer & Mehta, 2021). It should be noted, however, that minority-serving institutions (MSIs) offer culturally relevant support and encouragement, and there is evidence that they are more successful than non-MSIs at facilitating upward mobility for low-income students of color (Espinosa et al., 2018; National Academies, 2019b).
Cost is another important barrier for low-income students as they seek access to colleges in general and to high-quality colleges in particular. The National Center for Education Statistics (2019) finds that in 2015–2016, the average out-of-pocket net price for a full-time, low-income student was $7,100 per year, which constitutes 70% of the total income of a very low-income family of three. The primary federal program designed to make college more affordable is the Pell Grant, which offers funding of up to $6,495 (in 2022 dollars) to students from lower- and moderate-income families. Although the research evidence on Pell grants is mixed, substantial increases in the value of these grants have generated notable increases in degree attainment (Denning et al., 2019; Dynarski et al., 2022b). Other recent evidence (Angrist et al., 2022) shows that generous and well-targeted financial aid, especially when it allows higher-achieving low-income students to enter 4-year programs to which they would otherwise lack access, can substantially increase the rates of bachelor of arts attainment.
CAREER TRAINING
Much of our K-12 schooling system is oriented toward providing students with skills and other capacities that will enable them to attend and graduate from 4-year postsecondary universities. At the same time, there are many well-paying industry-specific occupations that do not require a bachelor’s degree, are in strong local demand, and offer opportunities for advancement. Placing students from disadvantaged backgrounds into these jobs can help to reduce the likelihood that they will remain mired in persistent intergenerational poverty. So too can educational opportunities aimed at enabling adults to participate in retraining for these jobs in mid-career.
Community colleges can provide some of the needed training through traditional degree programs. But so too can other kinds of training-related interventions, and strong evaluation evidence points to a number of promising approaches. The first consists of CTE pathways in high schools. The second involves sectoral programs that provide occupational skills training resulting in credentials that are valued by prospective employers in local labor markets.
While evidence on training programs in general is mixed, there is clear evidence that high-quality training programs that target certain high-demand sectors of the economy that need particular occupational skills can generate strong labor market returns (Katz et al., 2022). Sometimes these programs (such as Project Quest, which provides training in health care occupations) are offered at community colleges, while others (such as Per Scholas for Information Technology training) use other providers. Completing these programs can take anywhere from 6 months to 2 years, and they can cost between $5,000 and $12,000. But the best of these programs are clearly cost-effective.
Conclusion 4-2: The vast U.S. education system is a potentially important factor in enabling individuals to escape from poverty. However, it fails to equalize educational opportunities for students across socioeconomic and racial/ethnic groups. Research points to many possible ways to improve the quality of educational experiences offered to students in K-12 and postsecondary school settings, to create high-quality job training programs, and to prepare young people for the labor market.
EDUCATION INTERVENTIONS
There are many possible interventions for promoting child and youth learning in educational settings as well as encouraging young people to complete more years of education. Our discussion of such interventions is presented in the same order as above: K-12 schooling; postsecondary
schooling; and career training, with many of the details relegated to Appendix C: Chapter 4. For reasons discussed above, the committee does not offer policy and program ideas for the early childhood period.
Where possible, we structure the policy and program ideas in a way that enables us to provide a rough estimate of their costs. Smaller or larger scale versions of the policies or programs would reduce or increase these cost estimates accordingly. As discussed in Chapter 1, we characterize the evidence on some of the programs or policies as “strong” and denote them with an “*.” This indicates that the program’s or policy’s impact on intergenerational poverty is supported by random-assignment evaluation evidence that has been replicated across several sites or by compelling quasi-experimental evidence based on national or multi-state data or a scaled-up program.
K-12 Policy and Program Ideas Based on Direct Evidence
K-12 Spending in Low-Income School Districts
Recent impact studies have found that directing increased school funding at under-resourced districts improves both student achievement and rates of completed schooling, both of which have been linked to reductions in intergenerational poverty (Jackson & Mackevicius, 2021). This argues in favor of increasing federal funding for school districts with the highest concentrations of low-income students (details in Appendix C: Chapter 4):
- Increase K-12 school spending in the lowest-resourced districts. Increase annual spending by $1,000 per pupil in the 20% of districts with the lowest average family incomes. These districts serve one-third of free and reduced-price lunch (a proxy for poverty) students in the country. The committee estimated that this would cost $15 billion, with the assumption that states would use some of this money (or their own money that this would supplement) for other purposes.
Racial disparities are relevant for virtually any intervention aimed at reducing intergenerational poverty, so the committee looked for evidence about types of programs specifically designed to reduce them. It found three related to K-12 schooling that passed the committee’s direct evidence test:
- Increase teacher workforce diversity, based on strong evidence of the positive effects of Black teachers on the high school graduation and college enrollment of Black students.
- Reduce exclusionary school discipline practices, based on strong evidence that such exclusionary school discipline increases students’ chances of dropping out of high school and their contact with the criminal justice system in young adulthood and reduces their college enrollment.
- Increase access to Ethnic Studies courses, based on strong evidence of the positive effect of Ethnic Studies course-taking for high school graduation.
The evidence supporting these three policy areas is detailed in Appendix C: Chapter 4. The committee was unable to identify specific evidence-based ways of implementing them, nor could it determine how responsibilities for funding and implementing these policies should be allocated across federal, state, and school district entities responsible for public education. That said, the committee felt that the strength of the direct evidence supporting these policies warranted bringing them to the attention of policymakers.
K-12 Policy and Program Ideas Based on Indirect Evidence
A complementary approach to improving K-12 education outcomes is to focus on specific educational practices and policies that school districts could adopt, given additional resources, to achieve their educational goals. The federal government has little control over the spending decisions of states and districts. This can be a virtue, as local policy makers are often in the best position to judge local needs. In any event, because the education system is constantly evolving, evidence of long-term effects of specific practices or policies is scarce. That said, a number of promising educational strategies have proved effective in promoting short- to medium-run gains in student achievement (evidence is detailed in Appendix C: Chapter 4):
- Introduce or expand high-dosage tutoring for struggling students, with educated young adults serving as tutors and following carefully crafted instructional plans;
- Improve teacher quality, focusing on teachers of Black, Latino, and Native American students. This might be done, for example, through university programs that encourage and facilitate the certification of STEM undergraduate majors and Black, Latino, and Native American undergraduates as public-school teachers;
- Reduce class sizes, particularly in the early grades; and
- Expand high-quality (“no excuses”) charter schools.
Other curricular programs are still in the earlier stages of the evidence life cycle; there is only short-run evidence of the effectiveness of pilot
programs that have not yet been implemented or studied at scale. This category includes small high schools, double-dose algebra courses, and reading curricula that emphasize phonics for early readers. Again, the committee reviews several of these programs in Appendix C: Chapter 4.
Last, there is strong evidence that integrating schools by race/ethnicity, and probably by socioeconomic status (SES) as well, brings both short-run benefits in achievement and attainment and long-run improvements in adult life outcomes for Black students and low-SES students. As pointed out above, with the decline in court-ordered desegregation plans, schools are as segregated now as they were before the desegregation movement. Reversing this trend is a complex challenge, affected by both logistical hurdles and a complex legal environment. The committee is unaware of specific interventions that have been shown to be effective at increasing integration and that could be implemented at the federal level. Nevertheless, this is an important issue, and the committee believes that continued experimentation in this area (e.g., via changes in local school assignment processes) might yield evidence that could lead to meaningful reductions in intergenerational poverty.
Postsecondary Education Policy and Program Ideas Based on Direct Evidence
Interventions to improve postsecondary attendance and completion for low-income students can focus on the demand side, or institutions of higher education; as well as the supply side, or the students. Interventions focused on higher education might pursue three different goals:
- Increasing the fraction of students who attend college, through both demand-side changes that make it less expensive for colleges to enroll low-income students and supply-side interventions that provide incentives for students or further financial support to attend college;
- Improving instruction and student support services at institutions with large low-SES enrollments by providing additional resources to raise completion rates; and
- Incentivizing students to enroll at higher-value institutions and in higher-value programs of study within institutions.
They include these interventions:
- Increase federal funding for higher education by $10 billion annually for supply-side programs like financial aid that targets low-SES students (while limiting crowding-out of state and local funding). For instance, spending $10 billion on college scholarships (with
- Increase federal funding by $8 billion to $10 billion a year for institutional supports to improve completion rates among low-income students. Spending an additional $8 billion on proven student support programs (with $4 billion for community colleges and $4 billion for 4-year programs) would cover a half million community college students.
$5 billion each spent on community and 4-year programs) could support an extra one million full-time community college students per year (spread evenly across two or three yearly cohorts at an annual cost of $5,000 per year) and a half million for 4-year programs (also spread evenly across yearly cohorts at a cost of about $10,000 per year).
Postsecondary Education Policy and Program Ideas Based on Indirect Evidence
Other approaches to help achieve the three broad goals defined above—but with less rigorous research support to date—could include these:
- Increase maximum Pell awards, with limits imposed on states or institutions regarding offsetting these increases with other cuts in aid;
- Provide matching federal funds for state higher education allocations, conditional on a maximum tuition threshold and a minimum level of low-SES enrollment;
- Expand support for MSIs, which currently raise attainment of college degrees but to a lesser degree earnings;
- Simplify financial aid applications, which are intrusive and difficult for students and their parents to complete, by limiting the information required to that already collected by the IRS;
- Adjust federal aid formulas and the Integrated Postsecondary Education Data System data on specific colleges and programs to provide more information to students applying for admission (for instance, on their “expected family contributions” before they apply to college and on required grade point averages in specific institutions if they wish to major in certain fields);
- Target aid to programs with high labor market value—through grants to institutions that provide such programs; and
- Expand “Gainful Employment” regulations to limit the eligibility for receiving federal student financial aid to attend institutions or programs that show poor outcomes in their graduates’ post-program earnings and debt-to-income ratios.
Career Training Policy and Program Ideas Based on Direct Evidence
Much of the evaluation research concerning career training programs is based on random assignment, which provides strong, direct evidence on a number of promising approaches to career education and training. The first area of research involves CTE pathways in high schools. The second includes sectoral programs that provide occupational skills training leading to credentials that are valued by prospective employers in local labor markets.
High School Career and Technical Education
- Provide (through reforms of the federal Perkins Act and by allocating additional funds) both formula and competitive funding for states and localities to expand high-quality career and technical education. Three models of CTE can be prioritized: (1) Career Academies, which provide education and training focused on a specific high-demand economic sector (such as health care, IT, or finance) within comprehensive high schools; (2) technical high schools, or newer approaches (like Innovation Pathways in Massachusetts) to create programs similar to technical high schools within comprehensive schools; and (3) Grade 9–14 pathways, modeled after P-Tech, which would combine work-based learning and work experience with rigorous academics.
Postsecondary Sectoral Training Programs
“Sectoral programs” train people for well-paying jobs in specific industries and occupations that do not require a bachelor’s or associate’s degree for which there is strong local demand and that offer opportunities for advancement. The programs that the committee discussed scaling up combine intensive screening, career readiness services, specific occupational skills training that yields credentials valued by prospective employers, and post-training counseling in the interest of maximizing job retention and advancement.
Several such programs—including Year Up (for youth), Per Scholas (for youth or adults), and Project Quest (all for community college students)—have been evaluated using rigorous methods and found to be effective. The estimated impacts of these programs on earnings are generally large and persist over time. In the interest of reducing intergenerational poverty, the committee discussed scaling up sectoral programs that have proven impacts for youth:
- Offer sectoral training for youth. Offer scaled-up versions of Year Up, Per Scholas, and other proven sectoral training programs to 250,000 youth each year who come from low-income families and appear unlikely to earn postsecondary credentials. This would cost roughly $7.5 billion annually.
In addition, the committee suggests considering the following, which would help low-income youth indirectly by raising the incomes in their households (while providing models of labor market success and information on how to achieve success):
- Offer sectoral training for low-income parents. Each year, offer scaled-up versions of Project Quest, Per Scholas, and other proven sectoral training programs to one million low-income adults with children, which would indirectly reduce intergenerational poverty by raising the incomes of parents and households. This would cost roughly $10 billion annually.
