Testimony of Melissa S. Kearney

Chairman Brady, Vice Chair Klobuchar, and other distinguished members of the Joint Economic Committee, thank you for inviting me to participate in today’s hearing, “Income Inequality in the United States.” This is an issue on which I have focused my academic research over the past decade and also an area of focus for The Hamilton Project, which I currently direct. Following my oral testimony, I would be happy to take questions.

I. Introduction

As you are aware, as this is the topic of today’s hearing, inequality has increased dramatically in the United States during recent decades. To frame this in a different way, in contrast to earlier periods in our nation’s history, the economic growth experienced since 1975 has not translated into shared prosperity. Consider the period of national economic prosperity between 1947 and 1975. This was a period of strong economic growth in which the growth in family income was roughly the same across the income distribution — families in the bottom 20 percent saw their income grow by 90 percent, as compared to 86 percent for families in the top five percent. In sharp contrast, between 1975 and 2010, income gains were vastly different across families at different points in the income distribution, with each higher income group witnessing a greater increase in family income. Families in the bottom 20 percent of the income distribution saw their income increase by a mere 3.7 percent while those in the top five percent saw an average income gain of 57 percent.

The phenomenon of growing inequality is indeed real, but what does that mean and how should we approach this challenge? There are three main points I want to make about the high level of inequality we are experiencing today:

1. First, the trends in inequality we have witnessed over recent decades are largely the result of structural changes in the labor market that have favored the very highly skilled. There are growing gaps in wages and employment opportunities for those with at least a college education as compared to those without, and there is no reason to think that these labor market forces will be reversed any time soon. The Great Recession has exacerbated these issues, but the dominant forces were present long before the recession and they will remain even when the economy recovers.

2. Second, the growing levels of income inequality have translated into sizable gaps in educational achievement between the children of the rich and the poor. As the rich amass greater shares of the nation’s wealth, the children of these families enjoy even greater advantages relative to their less economically advantaged peers. This has the potential to perpetuate societal divides and erode social mobility. We must meet these structural challenges with a concentrated effort to invest in future generations.

3. Third, income inequality has the potential to interact with poverty in ways that perpetuate disadvantage and exacerbate intergenerational transmission of poverty. If those at the bottom of the income distribution view middle class life as increasingly out of reach, they might forgo the mainstream climb to economic success, perpetuating the cycle of poverty and inequality.

I will now expand briefly on these three main points, all of which lead me to the conclusion that it is of the utmost urgency that we make the necessary investments so that all Americans have the skills and ability to compete in a global labor market. We must ensure that broad swaths of the population are not left behind but rather, can participate productively in the economy and be rewarded with genuine economic security.

II. Inequality Trends and the Returns to Skill

For some decades now, the U.S. labor market has experienced increased demand for skilled workers. Since the late 1970s and early 1980s, the supply of highly-skilled workers has not kept pace with the rising demand for skilled workers, especially among men. The result has been a sharp rise in the inequality of wages.

Since the mid-1970s, those with the highest levels of education – more than 16 years – have seen their wages rise steadily. Those with exactly a college degree or some college have seen some improvement, but the increase in their wages has not kept up with those with more advanced education. High school graduates and those with less than a high school degree saw their real wages fall through the late 1970s and 1980s, but rebound a bit in the early 1990s. Their wages have remained fairly stagnant since then (see Autor, Katz, Kearney, 2008).

Furthermore, as previous Hamilton Project work has emphasized, if we consider that the labor force participation rate among less-educated men has fallen over the same time period, the economic position of the “median male” appears to have eroded by even more than a focus on wages implies.

A second, related trend is what labor economists have referred to as a “polarization” of job opportunities in the United States (Autor, Katz, Kearney, 2008b). The U.S. labor market has witnessed expanding job opportunities in high-skill, high-wage occupations on the one end, and low-skill, low-wage occupations on the other. Employment prospects for middle-skill, white collar workers in white-collar occupations — clerical, administrative, and sales occupations – have weakened, as have those for middle-skill, blue-collar jobs in production, craft, and operative occupations. These trends have led to declining earnings and declining labor force participation rates among individuals with less than a four-year education, in particular, men. Increasingly, individuals who previously worked in these fairly well-paying sectors are now working in low-paying service industries.

This employment polarization is widespread across industrialized economies. It is not a uniquely American phenomenon and has occurred in Europe to at least as large a degree (Goos, Manning, and Salomons, 2009; Autor, 2010.) This implies that there are global economic forces that have led to a restructuring of the labor market. It is important to acknowledge that while the large share of income accruing to the top one percent is striking, trends in inequality are not simply reflecting something peculiar at the very top of the distribution. The more relevant issue for public policy is the growing gap between those with high levels of skills and education and the rest of the population.

We need policies that focus on the disparities across education and skill groups and in particular, on ways to upgrade the earnings capacity of the majority of Americans, in particular those at risk of falling into an intergenerational trap of poverty. Ultimately, we need to equip our workforce with the skills that are demanded and rewarded in today’s global economy.

III. Income Gap Between Families at the Top and the Bottom of the Income Distribution

This experience of rising income inequality has led to sizable income gaps in experience and achievement between children at the top and bottom of the income distribution. This class of highly-educated, high-income individuals are marrying one another, and showering advantages on their children. There is nothing alarming about that per se. The problem is that children born into the bottom half of the distribution are falling behind. As those near the bottom of the socio-economic distribution fall farther behind their more advantaged peers in terms of resources and opportunities, their chances of moving up on the socio-economic ladder are threatened. A recent Hamilton Project policy memo, “Thirteen Economic Facts about Social Mobility and Education,” clearly highlights these trends in a very accessible way.

In terms of the specific issue of education gaps, we see that it appears early in life and persists through to differences in college completion rates. By age four, children in the highest income quintile score, on average, in the 69th percentile on tests of literacy and mathematics, compared to children in the lowest income quintile who score in the 34th and 32nd percentile, respectively (Waldfogel and Washbrook 2011).

Scholars and policy makers have increasingly come to appreciate the importance of both cognitive and non-cognitive skills (cf. Heckman, Stixrud, and Urzua, 2006) and that many of these skills are developed early in life. This has led to an emphasis on early childhood interventions. When we talk about early childhood environments, we need to think about both preschools and home environments. Numerous studies have shown that higher-educated, higher-income parents spend more time with their kids, and more time in educational activities in particular (e.g., Guryan, Hurst and Kearney, 2008; Kalil, Ryan and Corey, 2012).

It is also important to note that these income gaps in educational performance have grown over time. Trends in student achievement – as measured by math and reading test scores – reveal that while the racial gap has declined significantly over time, gaps between low and high-income students have increased (Reardon, 2011).

Importantly, this growing gap does not appear to be attributable to widening inequality itself. Rather, the data indicate that the increase in the income gap is driven by the increased attainment of families above the median income level (Reardon, 2011). And by the time these students reach college, many from low-income and disadvantaged backgrounds just don’t have what it takes to persist and graduate from college. Work by Bailey and Dynarski (2011) documents a growing income-based gap in college completion.

To be clear, this growing gap in educational performance is not a result of eroding support for public education. In fact, recent empirical evidence finds that the opposite occurs. Various researchers have shown that in areas with greater levels of local income inequality, public school spending increases (Boustan et al., 2012; Corcoran and Evans, 2012; Gordon 2013).

Rather, this is about diverging attainment largely driven by the kids at the top of the income distribution benefiting from numerous advantages that are simply increasingly out of reach for those from the middle and the bottom. Princeton Sociologist Sara McLanahan has called attention to the “diverging destinies” of children from low-income, often single-parent homes, and their more economically advantaged peers. Research has shown that the divide in marriage rates between those of lower and higher economic classes has amplified the disparities for children. Poor children are much less likely to have access to the incomes, family and friendship networks, and time investments from two stably married parents than their rich parents, and this perpetuates their heightened likelihood of ending up poor (McLanahan, 2004).

Whereas children who were born to the most-educated women are gaining resources, in terms of parents’ time and money, those who were born to the least-educated women are losing resources. This is largely driven by the divergence in the rates at which these women are having marital versus non-marital births. The rise in non-marital birth rates cannot be ignored. Study after study has shown that children born into two-parent families have tremendous benefits. One especially stark difference is in rates of poverty. Census data show that the rates of poverty are five times as high among children headed by a single female head of household (31 percent), as compared to children living in married-couple families (6.3 percent) (DeNavas-Walt, Proctor, and Smith 2013).

Furthermore, differences in family structure exacerbate income inequality. More educated, higher-income couples have much higher rates of marriage and lower rates of divorce. The Pew Research Center finds that in 2010, nearly two-thirds of Americans with college degrees were married, and only 47 percent of those with a high school degree or less were married. In the 1960s, the odds of being married were just about equally likely across the education distribution at around 70 percent (Cohn et al., 2011).

On this issue of family structure, we see the feedback effect from declining economic opportunities, in particular for men. A decrease in the economic prosperity of men lead to lower rates of marriage among disadvantaged populations (Wilson, 1987). This further hinders the well-being of children in low-income families. The dissolution of the family is appropriately considered a marker of social and economic problems, as opposed to the cause.

IV. Interaction of Income Inequality and Poverty

The above section focused on the fact that families are diverging in ways that leave kids in the bottom half of the income distribution at a clear and distinct disadvantage.

A separate question is whether income inequality itself affects the economic decisions and outcomes for those near the bottom of the income distribution. To be clear, this is not about income gaps, but about how being poor could actually be worse if one lives in a more unequal place, say, by increasing rates of social and economic isolation. The question is whether the behaviors and outcomes of economically disadvantaged individuals are different based on the level of inequality.

There is a recent body of evidence about the interaction effects between being poor and living in a more unequal place, where inequality appears to negatively affect those who are near the bottom of the distribution. It means as bad as it is to be poor, it is even worse to be poor in a more unequal place. My coauthor Phil Levine and I have been investigating this question for the past few years and have come to some striking conclusions (Kearney and Levine, 2013a and 2013b). Plenty of previous work has documented that correlations showing that places that are more unequal tend to have worse social outcomes, such as higher rates of teen childbearing and lower rates of high school completion. But it is crucial to figure out whether income inequality is actually causing these worse social outcomes. Our research suggests that when it comes to two important issues – educational attainment and young, non-marital childbearing — it does.

We find that economically disadvantaged youth – both males and females – are more likely to drop out of high school if they live in a place where lower-tail income inequality is greater. This means that when the gap between the bottom of the income distribution (as captured by the 10th percentile) and the median is wider, economically disadvantaged youth are more likely to drop out of high school. In a related study, we find strikingly similar results showing that economically disadvantaged girls and young women are substantially more likely to become unmarried mothers if they live in a place where lower-tail income inequality is greater.

These findings are consistent with the view that economic despair or economic marginalization can lead individuals to simply drop out of the mainstream climb to economic success. Simply put, these individuals do not see promising opportunities, and so they essentially give up. We cannot let this happen. This will lead us to an extremely unproductive cycle of the intergenerational transmission of poverty.

Rates of social mobility have, to date, remained fairly constant in the United States. Given labor market and demographic trends, this is a relief. But, as usual, the devil is in the details and rates of social mobility vary tremendously across the United States. A fascinating new study (Chetty, Hendren, Kline, and Saez, 2013.) shows that in that some cities – such as Salt Lake City and San Jose –rates of mobility are comparable to countries with the highest rates of relative mobility, such as Denmark. Other cities – such as Atlanta and Milwaukee – have lower rates of mobility than any developed country for which data are currently available.

These researchers document that differences across places in the level of upward mobility are not explained by differences in average income levels, economic growth rates, or the concentration of income in the top 1 percent. They also demonstrate that the differences across places is not directly due to race at the individual level: geography matters regardless of race.

Factors that do appear to be strong correlates of local area social mobility rates include levels of income inequality, measures of economic and racial residential segregation, K-12 student test scores and high school dropout rates, social capital indices, and rates of single parenthood. Areas with a smaller middle class and areas with sizable residential segregation by income class have lower rates of upward mobility. Areas with higher rates of social mobility tend to be characterized by higher student test scores (controlling for income levels), lower dropout rates, and higher spending per student in schools. The authors highlight that two of the strongest correlates of mobility rates are the fraction of religious individuals (a positive relationship) and the share of children raised by single parents (a negative relationship). These correlations are fascinating, but the research is not yet at the point of being able to identify the actual causes of social mobility. This is a crucial area for research going forward.

The bottom line is that inequality exacerbates the economic marginalization that comes with being poor. This has important implications for social mobility and potentially for understanding variation in social mobility across places in the United States. We see that places with high levels of income inequality and income segregation have lower rates of social mobility. These areas of the United States should caution us about what will happen if we allow income inequality to continue on its current path, with all the social, economic, and cultural polarization that comes along with it. There is the real threat that low rates of social mobility will become the norm in the United States. We will become a nation where those who are born poor will find it increasingly difficult to engage productively in mainstream economic life, and so they will remain poor, and their children will remain poor. This flies into the face of the American Dream and a nation built on the promise that you can move up the income distribution through hard work and ingenuity.

V. Conclusion

In summary, trends in income inequality have been largely driven by structural labor market trends that have favored the highly skilled and highly educated in terms of both higher wages and increased job opportunities. Those with less than a college degree have really seen their job opportunities erode, and their wages remain stagnant. Marriage patterns have amplified these differences for family income, with the highly educated marrying each other, amassing income and wealth, and investing a great deal of financial and other resources into their children.

The worry is that without effective and targeted policies, the less-advantaged will fall behind and we will see a damaging feedback effect – with those born in the bottom of the income distribution increasingly stuck there, dropping out of the mainstream climb to educational and economic success.

There is no silver bullet for conquering the challenge of inequality in the United States, but we have many policy levers available to us and many possibilities for investing in our future generations. The fact that rates of social mobility have remained roughly constant in the face of these global labor market forces that have so disproportionately rewarded the highly educated is a testament to the effectiveness of policies that work against them.

• The Earned Income Tax Credit has been shown to be a very important program that has encouraged work among single mothers and led to long-term improvements in the well-being of families and children.

• SNAP (the food stamp program) is the quintessential safety net program and has proven to be responsive to weak economic conditions in exactly the way a true safety net program should be. Researchers have documented the long-term health and economic benefits of this program to low-income children and individuals (Almond, Hoynes, Schanzenbach, 2011; Hoynes, Schanzenbach, Almond, 2012). A recent Hamilton Project paper by Diane Whitmore Schanzenbach highlights opportunities to strengthen SNAP.

• Expanded access to higher education has been important to keep the supply of college educated workers from falling even further behind the demand, which has yielded large benefits to individuals and also kept the earnings inequality from rising by even more than it has. Again, The Hamilton Project has released a series of papers in the last six months offering policies for expanding college access and affordability that could be considered.

And we need to do more.

• We need effective intervention in early childhood, perhaps most importantly expanded access to high quality child care and preschool.

• We also need programs aimed at decreasing the prevalence of non-marital births and improving the home environment and parenting practices of poor families.

• We need to keep adolescents and older teens engaged in school. For the one-quarter of American youth inclined to drop out of high-school, we need to find ways to equip them with the skills necessary to become productive workers and citizens. We need to think seriously about a role for vocational training or employment apprenticeships.

In general, we must maintain and expand programs and policies that constitute investments in the health, education, and skills of individuals from broad swaths of the population. These types of investments are crucial if we are to remain a nation of shared prosperity where economic security is the reward for anyone who is willing to work hard and provided for those who are truly unable.

Thank you for the opportunity to testify. I would be happy to take questions at this time.

To watch the full webcast, click here>>