7

Moving Forward: The Role of Policy and Key Takeaways

Throughout the workshop, speakers and participants touched on how research can support the development of good policies, and the role that policy plays in either facilitating or inhibiting mobility. To bring these discussions together, Wendy Edelberg and Bob Greenstein (Brookings Institution) held a “fireside chat,” in which they explored existing policies that impact mobility, what types of policies might be most effective moving forward, and how to best translate research findings into policy. To close out the workshop, Courtney Coile (Wellesley College) asked planning committee members to identify key takeaways from the discussions and to share their thoughts on where the field should go from here.

THE ROLE OF POLICY

In this session, Edelberg and Greenstein engaged in a back-and-forth conversation with each other and with workshop participants about the role of policy in reducing income inequality and poverty, and in increasing inter- and intragenerational upward mobility. Edelberg began by emphasizing the importance of upward mobility for policy makers. “History shows us that people with financial means will always find ways to protect themselves and work for the success of their children,” she said, and one of the highest priorities for government is to ensure that those opportunities are available to all people, regardless of financial circumstances.

Most Critical Policies for Children

Edelberg asked Greenstein to identify the policies that can best ensure that a child’s circumstances do not “determine their destiny,” and asked him to comment on how these types of interventions and programs could be made more effective. Greenstein responded by focusing on two areas of policy, the child tax credit and housing policies, which hold potential to improve mobility for children, though he stressed that these were not the only important policies.

Child Tax Credit

There is a strong evidence base to support the notion that providing a fully refundable child tax credit on an ongoing basis could enhance children’s life chances and promote mobility, said Greenstein. A fully refundable credit, he said, ensures that low-income families get the full amount because it is not tied to the level of parental income. Some policy makers have argued that providing cash assistance via the child tax credit could substantially discourage work among the poorest families. Greenstein disagreed with this idea but said that political debate on this point has made it a “front and center” topic in social policy. In 2021, a fully refundable child tax credit was in effect due to the pandemic, but it is unclear what will happen in the future. Further research in this area has real potential to impact policy outcomes in the years ahead, he said.

Housing Policy

Housing has a big impact on children’s well-being, Greenstein continued, including adverse effects from evictions and housing instability. Interventions in this area can make a difference, as was demonstrated by a study which found that children under 13 in families that used a voucher to move to a lower-poverty, higher-opportunity area experienced improved mobility.¹ This evidence led the U.S. Partnership for Mobility from Poverty to propose that vouchers be provided to low-income families with young children, and that the vouchers be accompanied by intensive mobility services to help families navigate issues and succeed in moving to a new lower-poverty neighborhood. This proposal has not become law, but Congress has appropriated money for a demonstration project that will begin soon. This is another area, said Greenstein, where further research could provide

___________________

¹ Chetty, R., Friedman, J.N., Hendren, N., Jones, M.R., and Porter, S.R. 2018. The Opportunity Atlas: Mapping the Childhood Roots of Social Mobility. NBER Working Paper No. 25147. Cambridge, MA: National Bureau of Economic Research.

more data on the impacts of this type of approach, and could improve understanding of which housing mobility strategies are more or less effective.

Human Capital Investments

Greenstein asked Edelberg to talk about human capital investments that can promote mobility—and in particular, the role of fiscal policy in securing or incentivizing those investments. There are a range of programs that support human capital investments, she said, including those that support early childhood education; student loan programs; and efforts to protect workers, such as paid leave and closing the wage gap. Ensuring access to high-quality early childhood education, she said, is a “no-brainer.” Head Start and Early Head Start aim to reduce poverty by providing comprehensive preschool programs to meet the emotional, social, health, nutritional, and psychological needs for children from low-income families. However, only around 36 percent of eligible children ages 3-5 and only 11 percent of eligible children under age three had access to these programs in 2020. These numbers are “far too low,” she said, given the evidence linking the programs to positive short- and long-term impacts, including reducing the intergenerational transmission of poverty. More research, including long-term research, is needed in order to build the evidence base necessary to increase support from policy makers.

The government also invests in workforce development programs, such as training and reemployment for displaced and disadvantaged workers. Evidence for the effect of specific programs is mixed, she said, but evidence shows that the more successful programs are those that focus on specific populations and sectors. This is a “critical place” to increase data and research, she said. Another area ripe for research is in the area of workers’ bargaining power and how it can support upward mobility. At this moment in time, said Edelberg, the balance of power has shifted, and workers are able to demand more pay and better working conditions. However, whether this continues remains to be seen. The government can do more to support workers’ rights to unionize; it is clear that unions lift wages, reduce inequality, and shape how work is organized. Workers in certain industries—such as agriculture and warehousing—have less bargaining power, and research could help document this phenomenon and support policies to address it.

Lasting Policies

With policy work, said Edelberg, “there is no point at which we declare victory.” A policy that is in place has no guarantee of remaining in place forever, she continued, then asked Greenstein to discuss how to make policies “stick.” He responded that there is a tension between having

policies stick and having the most effective policies. A recent book by Eric Patashnik² examines what types of policy reforms get reversed, and what types persist through changes in government. One insight from the book, said Greenstein, is that a policy enacted on a purely partisan basis is more likely to be reversed than one that had a lot of bipartisan support when first enacted. For example, if a fully refundable child tax credit were to be enacted this year in a Congress controlled by Democrats, he said, it is likely that Republicans would try to repeal full refundability if they regain control. Patashnik also makes the observation that policies can become embedded over time as they develop a constituency. For example, it would be much more difficult for Republicans to repeal the Affordable Care Act now than it was several years ago.

While evidence supporting the effectiveness of a policy matters, it is not definitive; Greenstein said, “simply showing that something is effective doesn’t mean you are going to get the political support to enact it, but it definitely helps.” He gave an example of congressional debate on the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC). WIC is targeted to low-income families, but was only serving a fraction of the eligible population because there was not enough funding. Evidence of its efficacy was accruing, and at a congressional hearing, a panel of Fortune 500 CEOs testified that WIC was the “health care equivalent of a triple-A rated investment.” In the months that followed, there was a bipartisan commitment that Congress would appropriate enough money each year so that all eligible families who applied could receive the benefit. This is an example, he said, of how the evidence of impact can really matter. The chances of enacting a policy such as the refundable child tax credit would be “materially enhanced” if new research could demonstrate positive mid- and long-term effects on children and insignificant effects on parental employment.

Work Requirements

A number of the social policies aimed at improving upward mobility contain work requirements, said Edelberg. One argument often made in favor of work requirements is that people’s participation in the labor force is an essential ingredient to creating upward mobility. However, evidence suggests that work requirements do not have the desired effect of meaningfully increasing people’s participation in the labor force; Edelberg said this is a critical area where researchers can contribute evidence to impact policy making. Greenstein agreed that more research is needed in this area, and

___________________

² Patashnik, E. 2008. Reforms at Risk: What Happens After Major Policy Changes Are Enacted. Princeton, NJ: Princeton University Press.

posited that some work requirements are effective in incentivizing employment while others are not. There is evidence that the Earned Income Tax Credit increases employment, and some research suggests that the employment-increasing effects of the credit are a significant poverty reducer.³ Other types of work requirements—such as those in Medicaid programs—seem to be ineffective. Citing evidence from the Arkansas Medicaid program, Greenstein said these requirements appear to have little noticeable effect on employment but have a major effect on people becoming uninsured or unable to access needed benefits. While the courts have struck down work requirements for Medicaid, there is reason to believe that work requirements will continue to be a major focus of debate in the social program structure. This is another area, he said, in which more research could make a significant impact on policy making.

Translating Evidence into Policy

Kathleen Mullan Harris (University of North Carolina at Chapel Hill) recalled that earlier in the workshop, Greg Duncan (University of California, Irvine) noted that an understanding of the processes of mobility does not necessarily equate to an understanding of what types of policies should be enacted to impact mobility. Harris asked Edelberg and Greenstein to comment on how research findings can be better translated for policy makers. Edelberg referred to the Hamilton Project⁴ as an example of a place that does such work. Its core mission, she said, is to reach out to academics and experts and help them translate their findings for a policy audience; the Hamilton Project helps academics go beyond simply including a section on “policy implications,” and make the policy implications and their policy ideas more concrete and actionable. Greenstein offered another perspective, saying that academic researchers who are not familiar with the political process should not strive to “go beyond what they know.” When researchers do have a significant sense of policy implications, he said, they can and should spell them out for the reader. He encouraged researchers to acknowledge any uncertainty on the details of the policy implications, but to “make what contribution” they can.

Greenstein noted that an academic paper may not be the ideal place for researchers to reach policy makers, because of the structure, style, and audience for these papers. A better approach may be to issue a policy brief that accompanies the academic paper; he noted that many research centers engage in this practice. Partnerships with organizations such as the Hamilton Project

___________________

³ Hoynes, H. 2019. The earned income tax credit. The ANNALS of the American Academy of Political and Social Science, 686(1), 180-203.

⁴https://www.hamiltonproject.org/

or the Center on Budget and Policy Priorities can also be useful for researchers hoping to make an impact on policy discussions. These organizations can work to connect researchers with policy makers, and can facilitate fruitful conversations between the two. He said that there are people in Washington who are “happy to help” make the connection between the people doing the research and the people making the policy. Finally, Greenstein said that engaging university communications staff can help draw attention to research findings; “often it takes getting one journalist” to take an interest and “all of a sudden you’re in the middle of the policy debate.”

Differing Goals of Policies

Policies in this area, said Coile, can look similar but be directed at different goals, such as reducing poverty, reducing inequality, or increasing social mobility. Coile asked Edelberg and Greenstein to give their perspectives on these different goals and how they think about the tradeoffs between types of policies. Edelberg responded that these goals are generally overlapping, but some policies, such as those directed at helping people suffering from a temporary setback, are aimed at reducing shorter-term inequality. For example, if a certain segment of the population experiences a dramatic decrease in income during a recession, the government may provide temporary assistance to ensure that they can weather the storm. These policies are directed at inequality rather than long-term mobility, but “of course they are totally related.” Greenstein said that discussions of mobility have often treated the “ladder” and the “safety net” as if they are entirely separate things. One of the key developments in recent years, he said, has been evidence that significant poverty reduction, especially in early childhood, affects long-term mobility. However, there is still not widespread understanding among policy makers about this relationship between the safety net and the ladder, and public education is this area is an important need.

KEY TAKEAWAYS

At the end of the workshop, Coile asked the planning committee members to share their thoughts on the key points made and insights provided, needs that were identified, issues that may have been missed, and implications for policy.

General Observations on Social Processes and Structures

Planning committee members shared some of the key observations they made over the course of the workshop. Several members spoke about the centrality of the family unit in mobility; for example, Snipp said that Greig’s

data about family members paying back student loans underscored the importance of family. Logan said researchers need to look at how families have changed over time and how these changes impact mobility. A great deal of research has focused on family form and how families function as an economic unit; this research should be incorporated into measures of mobility. Mobility is “essentially a family process,” said Harris, and data and theories should reflect this relationship. Harris also emphasized the importance of considering mobility as a life course process. The life course looks different for different cohorts, different populations, and at different times, she said, and people acquire human, financial, and social capital at different times across their life course. Methods for measuring and understanding mobility should consider and capture these differences.

Small noted the importance of structures and institutions. He highlighted Sharkey’s discussion of how policies and institutions create boundaries and space that separate neighborhoods in ways that impact mobility, and he recalled Lee’s discussion of trust and underreporting and how understanding of the Asian American experience is enhanced as a result of people’s understanding of governing institutions. Small also said that integrating research on structures, including structural racism, is critical for understanding mobility. Coile agreed, recalling Brown’s observation that the ability to understand the effects of structural racism has been hampered by the current inability to measure it. Coile said that she also agreed with Brown’s call for creating a public data resource on measures of structural racism, which could be used by other researchers to help explain outcome measures.

Need for New Data, New Models, and New Theories

Many planning committee members identified the need for new data, new models, and new theories in order to drive the field of social mobility forward. Snipp noted that there has not been a major sociological study of social mobility since the 1970s. This may be one reason for the dearth of major methodological innovations in quantitative sociology, he said; although there have been many tweaks to existing methodologies, there have not been revolutionary developments along the lines of the introduction of path analysis of the log linear models in the 1970s. However, the lack of a comprehensive study has also led to increased interest in the area and new efforts to create data infrastructure and data linkages—the importance of which, Snipp said, was “impossible to overstate.” Recalling the discussion from the data infrastructure session, Snipp said that some parts of the federal government wanted to make data more available while others wanted to restrict access; who will prevail in this disagreement will not be up to the social scientists or even to the agencies, as this is ultimately an issue

that will have to be settled either in the courts or in Congress. Coile agreed with Snipp that while there is substantial promise on the data front, there are also challenges in terms of access and privacy, as well as tensions inherent in the fact that private companies collect data for purposes other than research. Snipp reminded the workshop participants about the National Academies consensus study underway on “The Scope, Components, and Key Characteristics of a 21st Century Data Infrastructure”⁵; the study will explore the possibility of developing a consistent set of standards across all federal statistical sources, think about ways of incorporating data from the private sector and state and local governments, and so on.

Harris identified several future needs for mobility research. She said there is a need for more research on topics including subjective social status (in addition to components and forms of mobility such as education, income, occupation, and wealth), the timing of investments, and the interaction between individuals and structural systems (e.g., educational, legal, criminal justice, and health care). Harris emphasized the importance of data on wealth, which she said can be difficult to get but for which there are some good extant sources in terms of home ownership, residence, and credit information; she recalled Darity’s comment on the importance of getting information on the wealth of parents and grandparents, which would be more difficult to get. Intergenerational data are also a particularly major need, according to Harris, and this can be addressed by adding simple questions to surveys (e.g., place of birth for the individual, the parent, and the grandparent; citizenship; and nativity) or through data linkages projects. Recalling the presentations by Brown and Sharkey, Harris said that researchers are making good progress on structural measures by being “incredibly entrepreneurial and resourceful,” and there is a need to build infrastructure to make these data available to all. Finally, said Harris, many speakers emphasized the importance of qualitative data and the importance of integrating qualitative data within quantitative studies.

Snipp and Coile concurred with the need for more qualitative data. Coile added that she observed a need for data that allows researchers to see variation in smaller areas and groups. Recalling Small’s presentation, she pointed out the need for hyperlocal data on places (e.g., at the block level), as well as the need to be able to adopt a broader comparative approach and compare across areas. Coile said that it is also important to have the data necessary to understand the experiences of subgroups (e.g., different ethnicities within the larger group of Asian Americans).

While acknowledging the data needs, Small said that “data alone can’t produce science.” Given the vast amount of data that are and will become

___________________

⁵https://www.nationalacademies.org/our-work/the-scope-components-and-key-characteristics-of-a-21st-century-data-infrastructure-a-workshop

available, he said, new theoretical ideas and models can be developed to increase understanding of mobility. Logan agreed with Small, saying that “some of our mobility research has been too guided by the data” and noting the importance of thinking theoretically about the real meaning of mobility and how it can be measured accurately. Logan added that there are currently many ways to define and think about mobility—for example, occupational, wealth, and income mobility—and a theory that can integrate these different types and examine the relationship between them is needed.

Harris recalled the integrative model that Grusky presented and said that this is “a goal that mobility research should aspire to.” At the same time, there is a need to identify and explore the complexities of a model of mobility; as with any type of integrative model, she said, “you just sort of chip away at it.” Coile agreed with Harris and indicated that the conversation about the integrative model was “an exciting development.” Coile also said that the “convergence” across the fields of sociology and economics was a welcome development, both with regard to causality and the greater attention being paid to models, theory, institutions, and contextual factors.

What May Have Been Missed

Members identified a number of issues that either were not discussed at all or received only a brief mention. Pfeffer noted that some issues were deliberately excluded from the workshop in the interest of time, but that these issues would be important to explore in future conversations. International perspectives were not well represented during the workshop, said Pfeffer, and there is a lot to learn from other countries. Another area for future exploration is the intersection between gender inequalities and social mobility; although Lee reported interesting findings on gender at the intersection of race and ethnicity, this is a topic that has lagged behind and deserves more empirical research. Finally, said Pfeffer, the distinction between social mobility and social inequality was touched on a few times, but further conversation would be useful to explore the relationship between mobility and inequality, and how “in a society where the rungs of the ladder are further apart, it is harder to climb the rungs.”

Logan pointed out that there were many discussions of policy at the workshop, but few mentions of politics. Politics is inextricably linked to mobility, he said; the distribution of resources, the places people live, and where investments are made are all political issues. For example, cities that were very segregated in the 19th century made the first early investments into water systems; choices about how to exclude some people from clean water resulted in persistent racial differences in waterborne diseases. On the other side, said Logan, research shows that families who were enslavers at the end of the Civil War maintained their economic position after the end

of slavery; this phenomenon is a function of power and the political process. There is a political and historical process that “makes place good and makes place bad,” said Logan, and one cannot talk about the importance of place without talking about the impact of politics on place. In discussions of mobility, Logan said it is important to bring politics more fully into the conversation, and to clearly articulate what it means for policy and programs.

Snipp said that not much was heard during the workshop about the importance of social media and what can be learned from those resources. He indicated that, to his knowledge, not much thought has been given to how those data might be used to learn about social mobility; however, network models and the kinds of things on which those data are being brought to bear might have some promise and may be worth thinking about in the future.

Informing Policy

Small said he was struck by the distinction in Duncan’s presentation between problem research and program research. There has been great progress on problem research—that is, understanding the phenomenon of social mobility—and the future is “even more promising than I thought,” he said. However, program research—knowing what to do—has been more limited and there is space for much more work in the future. Researchers can design or evaluate ideas that are part of the policy conversation, he said, but more importantly, they can contribute to the discourse and identify the questions that should be asked.

Edelberg made three points about how research can inform policy. First, she said, there is a need for clarity about what is meant by “mobility.” The word mobility is often used to refer only to upward mobility; there are a number of policies designed to give people the opportunities and the resources necessary to move up the mobility ladder. However, policies designed to improve mobility can also be directed at making the rungs on the ladder closer together and potentially moving people at the top of the distribution down. According to Edelberg, “the rules have been rigged” in such a way that the children of wealthy parents stay wealthy. She said that policies need to remove unfair advantages (e.g., legacy preferences in college admissions), and the conception of mobility needs to include not only pulling people up from the bottom of the distribution but also seeing people at the top of the distribution move down. Edelberg explained that this distinction is important for two reasons. First, there are different policies that can be more or less successful along each of these lines. Second, this distinction is the reason why people sometimes talk at cross purposes regarding whether or not the United States has more or less mobility than other countries, or more or less mobility now than it used to.

The second point about policy that Edelberg made was the importance of considering and researching the most basic questions. She noted that policy debates often center around questions that researchers may consider ill-informed or simplistic (e.g., do policies without work requirements encourage people to stay poor?). By taking these questions seriously and providing straightforward answers, Edelberg said that researchers can shift the policy discourse, provide answers to those who want them, and provide “cover” to policy makers who are being confronted with ill-informed arguments. Finally, said Edelberg, the battle for good policy is never won. Because policies can be repealed, it is important to remain vigilant: “we will never just unfurl the banner saying: ‘Mission Accomplished.’”

CLOSING OF THE WORKSHOP

Upon the conclusion of the discussion, Coile and Majmundar thanked the planning committee members, the speakers, and the workshop participants for their work and input. In addition, Coile thanked the Gates Foundation for their support of the workshop, and the staff at the National Academies for their work in putting the workshop together. Coile adjourned the workshop.

This page intentionally left blank.