Two-generation models target low-income children and parents from the same families in hopes of interrupting the cycle of poverty. However, the models themselves vary widely and policymakers and practitioners need guidance on how to best design them to achieve their ambitious goals and capitalize on their multiplier effects. To that end, this brief builds on the insights from the Housing Opportunities and Services Together (HOST) Demonstration to present an updated theoretical framework for two-generation models. The framework emphasizes the importance of using family goals as the lens for targeting individual family members, setting individual goals, and aligning tailored appropriate solutions. This lens also necessitates prioritizing relationship building over programs, designing flexible evaluation approaches, and working for systems change to support families in their efforts.
Why Two-Generation Models?
The concept of intergenerational social mobility lies at the heart of the American dream. National identity is based on a sense that the family into which one is born should not determine one’s prospects for the future, and, by climbing ladders of opportunity, each subsequent generation can do better than the last.
Nevertheless, prospects for intergenerational upward mobility in the United States consistently trail those in other developed countries and have stagnated over the last half century (Chetty et al. 2014; Corak 2013). In parallel, the number of Americans living in poverty has continued to grow, increasing by some 15 million over the last decade (Bishaw 2013). This increase, coupled with the significant number of low-income people who cycle in and out of poverty in any given month, poses tremendous challenges in terms of how to both alleviate suffering and safeguard the resources needed to maintain our social safety net (Cohen 2014).
Over the years, policymakers and practitioners have tried many different approaches to helping families overcome these structural barriers and escape poverty. Some have targeted adults like the welfare-to-work program through Temporary Assistance for Needy Families or efforts to tie services to other benefits like housing vouchers, though many of these approaches were never fully implemented. Other approaches have focused exclusively on the next generation, intervening through early education at a critical life stage with young children in the hopes of altering their life trajectories. However, these kinds of siloed programs have not yielded the magnitude of results needed to make a difference (Chase-Lansdale and Brooks-Gunn 2014).
There is a growing consensus that children cannot thrive in homes where their parents struggle to make ends meet; and low-income parents cannot succeed without meaningful support for their children (Gruendel 2014). In response, many practitioners are exploring two-generation models that explicitly target both low-income children and parents from the same household by combining parent and child interventions to interrupt the cycle of poverty (Chase-Lansdale and Brooks-Gunn 2014; Haskins, Garfinkel, and McLanahan 2014).
The concept of a two-generation approach is not new—the earliest examples date back to the mid-60s with the launch of Head Start (Chase-Lansdale and Brooks-Gunn 2014). But the level of experimentation to improve upon this concept has accelerated rapidly in recent years, spurred by investments from leading foundations including the Bill & Melinda Gates Foundation, the Annie E. Casey Foundation, and the Aspen Institute’s Ascend Network, and even federal agencies, like the US Department of Education, through its Promise Neighborhoods, grant program and the US Department of Health and Human Services’s Administration for Children and Families.1 This rapid roll out across has yielded a tremendous diversity of two-generation programs that vary along innumerable dimensions—their targeting strategies, their settings, their core inventions, their short- and long-term goals (Chase-Lansdale and Brooks-Gunn 2014; Ascend, forthcoming).
However, all two-generation models have two core things in common: (1) they are trying to address extremely complex and intractable social problems; and (2) the complexity of the models makes them extremely difficult to conceptualize and implement effectively. Policymakers and practitioners need a common framework for thinking through these issues and anticipating the challenges that serving adults and children in an integrated way brings to ensure these efforts have the greatest probability of achieving their ambitious goals.
Learning from the Housing Opportunity and Services Together Demonstration
The Housing Opportunity and Services Together (HOST) demonstration was designed to use housing as a platform to deliver two-generation services intended to address the challenges of highly vulnerable families. HOST builds on lessons learned from the Chicago Family Case Management Demonstration, the successful wraparound service model that the Urban Institute and the Chicago Housing Authority (CHA) piloted and tested from 2007 to 2010 to serve 400 families from its Dearborn Homes and Madden/Wells developments. Even during the CHA’s ambitious efforts to transform its public housing, many of the most vulnerable residents were unable to qualify for or maintain a unit in a new mixed-income development or a private unit with a voucher. The three -year Chicago Family Case Management Demonstration sought to help stabilize these families and help them successfully transition to new housing. The model had several different components, including on-site clinical mental health services, a transitional jobs program, financial literacy training, and substance-abuse counseling (Parilla and Theodos 2010). But the centerpiece was the intensive adult case management with a low caseload ratio of 1:25, which allowed case managers to visit each client two to four times a month over 3 years.
This model showed promising gains for a wide range of adults, including “strivers” who tended to be younger, more regularly connected to the labor market, and fairly healthy; slightly older “high-risk” adults with more intermittent work histories and more serious barriers to employment; and “distressed” adults with serious health problems and little or no connection to jobs. Rates of employment increased over time, and adult participants saw significant improvements in their mental and physical health. However, the benefits of the demonstration services did not accrue to the children living in these households (Popkin and Getsinger 2010). At the end of the demonstration, parents and other guardians reported that their teens were struggling in school, engaging in risky behavior, being arrested, and becoming pregnant and parents at rates far above national averages.
To design a next-generation model that would directly integrate supports for youth, the Urban Institute partnered with CHA and Home Forward, the housing authority of Portland, Oregon, to design, implement, and evaluate the HOST demonstration. Using a whole-family wraparound approach combining intensive case management and supplemental supports and services for both adults and children, HOST sites moved from design to implementation over four years to test the viability of a two-generation models in public and subsidized housing.
This brief draws on insights from the demonstration’s four-year formative evaluation—including annual administrative interviews, program observation, and in-depth family interviews—to offer an updated framework for two-generation models aiming to move the most disadvantaged families in the United States down the path to economic security and well-being.
A Theoretical Framework for Two-Generation Models
Unlike traditional social service interventions, two-generation models—by definition—aim to achieve goals well beyond improving outcomes for individuals. The hope is that a two-generation approach will have a multiplier effect that will lead to better outcomes for whole families (see figure 1). The multiplier effect is the result of meeting the goals for both parents and children through a well-aligned set of purposeful interventions. Clearly defined family goals have the potential to help practitioners effectively target individuals within households, set individual goals, align appropriate solutions, and, hopefully, achieve better results for families and make the systems that serve the poor more cost effective over the long term.
In practice, existing two-generation models often have trouble clearly articulating the family outcomes they seek to achieve. Some programs seem to clearly articulate family-level goals but then revert to a set of separate or “simultaneous” parent and child indicators to operationalize this vision, particularly in the short term. Others use broad language about poverty and low income in communities that does not map to specific indicators; and still others essentially skip the step of defining family outcomes altogether (Chase-Lansdale and Brooks-Gunn 2014; Gruendel 2014).
Many two-generation models also try to set the same goal of for all families. In most cases, this decision stems from assumptions behind the careful targeting of families using available administrative data on a variety of factors, from employment, housing status, education, or health. However, both the Chicago Family Case Management Demonstration and the HOST Demonstration—which used administrative data to target hard-to-serve families in public and subsidized housing for intensive case management—found wide variation in families’ strengths and the nature of the challenges they faced, particularly once case managers built trusting relationships with the families (Theodos et al. 2012; Scott, Popkin et al. 2013).
The wide variation in models means that, although ending intergenerational poverty and achieving self-sufficiency are the ultimate goals of most two-generation models, the time frame in which they may reasonably be achieved depends greatly on where families start. Consequently, some families may need to set other goals in the short and medium term to chart their progress.
A typology, adapted from one developed through the Chicago Family Case Management Demonstration research may help practitioners and policymakers think about how to define short to long-term goals for different types of families (Theodos et al. 2012).
- Striver families: Parents in these families have relatively good health, education beyond a high school diploma, work history, and few dependents to support. Children may have a range of strengths and struggles; but, with the right support, these families are poised to escape poverty in the short to medium term, largely through the work of parents.
- High-risk families: Parents in these families need to overcome substantial barriers in the short term. Some of these barriers are personal, like low literacy or English language capacity, mental or physical health problems, limited work history, or, simply, too many young dependents in the household to support with wages alone at the beginning of a work trajectory. And some of the barriers these parents face are structural, including policies that exclude those with criminal history and undocumented parents from employment. Because of chronic difficulties making ends meet, children in these families may be under more stress than in striver families. In the short term, family goals may have more to do with stabilization and hardship alleviation while parents work on overcoming their own barriers. However, in the medium to long term, high-risk families may be able to build a path out of poverty either through parents’ efforts, changes in policy, or children’s efforts as they become healthy adults in their own right.
- Severely distressed families: In the remaining families, parents experience the types of chronic physical or mental health problems or disability that make stabilization the primary family goal, not only in the short term, but often into the medium and long term as well. Children in these families may have more problem behaviors, get suspended from school more often, and experience lower-quality health than striver or high-risk children. But they can escape intergenerational poverty should their families receive the kind of support that creates stability within the home, and helps young people stay on track until adulthood.
In reality, there is likely substantial fluidity among these groupings, mostly because of changes to family structure (e.g., marriage, separation, births), health or disability status, or changes in policy (e.g., immigration or federal benefits). However, understanding whole families in this way may help to establish appropriate short-, medium- and long-term family goals that help better measure progress and inform the design of effective whole-family approaches.
Because of their origins in existing programs, most two-generation models decide which family members to target within that framework. For example, in a model like Head Start centered on early childhood education, children under 5 are the primary focus, along with whichever parent is willing and able to engage with the program (Gruendel 2014). In the case of HOST, the default target population for the housing authorities involved were initially the primary leaseholders, with a secondary focus on all of the children and youth living in the household (Scott, Falkenburger et al. 2013).
Similarly, when it comes to setting individual goals, many two-generation models explicitly or implicitly stress one generation, viewing parent outcomes as instrumental to child outcomes, or vice versa. Rarely do programs give both generations equal weight, though most programs set goals for parents and children and aim to serve them both simultaneously (Mosle and Patel 2012).
To take greatest advantage of potential multiplier effects, some researchers theorize that models should focus equally on both parent and child outcomes (Chase-Lansdale and Brooks-Gunn 2014). However, there may be advantages to using family goals as a lens for deciding how to target individuals within families and set individual goals to ensure the greatest benefit not only to the individual but to the family.
For example, a case manager working with a striver family where the parent can find good jobs but can never keep them because of a child’s chronic behavioral issues at school might focus most attention on addressing the child’s issues to empower the parent at work. On the other hand, a coach for a high-risk family might work on normalizing the parent’s immigration status while helping older children find employment to contribute to the household and stay on track to graduate from high school and continue on to college.
This approach fundamentally differs from established two-generation models that uniformly stress either adult or child outcomes (Mosle and Patel 2012) and from recent scholarship suggesting that two-generation models should have adult and child interventions of equal duration and intensity Chase-Lansdale and Brooks-Gunn 2014). It also differs from existing two-generation models that primarily build on existing program frameworks centered on either parents or children at a particular developmental stage (Haskins, Garfinkel, and McLanahan 2014). These approaches may be indeed beneficial in their own right, but they may not align to yield the kind of multiplier effects that move families most effectively toward intergenerational mobility.
In practice, many two-generation models start with a priori assumptions about the programs and services that should be in place based on either the origins of the model or on a one-time logic modeling process using single point in environmental scans, aggregated client assessments, or analyses of administrative data (King, Coffey, and Smith 2013). Then, once decisions on the programs and services are made, providers work to coordinate across the portfolio of partners, often through the a core case manager, to align individual components, minimize burden on families, and maximize the effectiveness of the interventions (Chase-Lansdale and Brooks-Gunn 2014).
Nevertheless, given the importance of gaining the kind of nuanced understanding of individual families required to set reasonable and appropriate family goals, there may be distinct advantages to delaying the selection of programs and services. First, taking a “families first” approach would help ensure families have access to the high-quality interventions that closely align with their needs and have the appropriate timing, duration, and intensity (Chase-Lansdale and Brooks-Gunn 2014; Haskins, Garfinkel, and McLanahan 2014). For example, all families might objectively need job training, but it might be a second or third order priority for a majority of families. Basic literacy, English language proficiency, or child care services may need to come first. If decisions about programs are made up front using only employment data, there may be no resources left to fund the kinds of interventions that would make the biggest difference. And coordinating unaligned programs and services may have only a limited impact on families.
Second, delaying programmatic decisions could potentially free up vital discretionary resources for individual families. Not all solutions are programs or services. Sometimes families simply need short-term discretionary money, such as to pay the utility bill that will get them evicted if unpaid, to fix their car so that they can get to work, or to pay the fees for an immigration lawyer to apply for an adjustment of status. Very little cash assistance is available through government programs in the wake of welfare reform (Congressional Budget Office 2015; Edin and Shaefer 2015), thus having these discretionary resources in place so a wraparound program could help to stabilize a situation before the family goes into crisis as well as help them make meaningful progress toward their goals.
It may also be important to think about policy changes as part of the solution set for families. At the program level, there are many rules and regulations within the control of service providers that they can adjust to help families succeed. In the context of HOST, for example, providers can allow families more time to adjust after their wages go up to learn how budget and more successfully cope with the rising cost of rent as they move toward self-sufficiency. But for other programs, it might mean everything from changing office hours to eligibility criteria to formulas for allocating resources.
One of the most compelling rationales for an integrated two-generation approach to service delivery is the multiplier effects for parents and children. In particular, having both parent and child participate in coordinated services will hopefully lead to multiplier effects that could not be achieved when parents and children are enrolled in separate and uncoordinated programs. The hope is that children and parents who participate in coordinated programs may participate at higher levels and have greater success than children or parents participating on their own. Spurred on by their children’s success, parents may pursue more credentials education and obtain a better job (Sommer et al. 2012). Further improvement in children’s development might follow, for example, in the domains of school success and social competence (Duncan, Ziol Guest, and Kalil 2010). Ultimately, the benefits of these multiplier effects would accrue to not only the parent or child participating in an intervention, but to the family as a whole.
For policymakers and practitioners, the potential advantage of using a two-generation service model may stem from its potential to transform the systems that serve the poor. Short-term gains may surface through the closer alignment and coordination of a myriad of programs and services that independently serve parents and children and through the calibration of the use of programs and services versus more flexible cash resources and policy solutions (Chase-Lansdale and Brooks-Gunn 2014; Pierre, Layzer, and Barnes1996; Scott, Falkenburger et al. 2013). Over the long term, more successfully interrupting the multi-generational cycle of poverty with solutions well aligned to family and individual goals would mean that up-front investments in two-generation models could significantly reduce or even eliminate the investments required to support the third and subsequent generations.
Implications for the Field
At the beginning of the four-year demonstration, the HOST sites started out with a conceptual framework different than the one presented here—one more focused on standardized, aggregate and parallel outcomes for adults and children. However, the focus on intensive, whole-family case management provided new insights into how two-generation models may be conceptualized to better harness their power and potential. Case managers organically began to think in terms of family goals and strategic targeting and attention to individuals within families. This brief’s accompanying publication on the details HOST implementation provides additional detail on what this work looked like in practice (Scott, Popkin, and Saxena, forthcoming).
On the policy level, this updated concept has several concrete implications for the field:
- Put people before programs: The ultimate success of two-generation models may hinge on being able to set appropriate family and individual goals. The only way to set these goals is for staff to take the time to first build relationships and establish trust with families. This process may take significant time and resources up front, but it will hopefully yield a much stronger set of solutions for individual families and more efficient use of resources for service providers.
- Tailor evaluation approach to fit diverse families: Traditionally, the evaluation framework for two-generation models reflects the program-level logic modeling that is done before the launch of services. However, using a family- rather than program-centric lens necessitates a different approach to evaluation. One-size-fits-all outcome measures that do not align with actual family or individual goals almost ensure finding little or no effect of the integrated work with families. Formative evaluation using the “plan, do , study, act” approach can be particularly helpful for two-generation programs.2
- For example, case managers might work with families to define and operationalize personalized family and individual goals as their relationship develops. Under ideal circumstances, researchers would use these self-defined outcomes to group families together, and individual parents and youth by common outcomes to analyze changes pre- and post- intervention. This approach allows measurement of relative, rather than absolute, progress toward goals. At a minimum, researchers need to consider using whatever data they have available on families’ strengths and barriers to approximate these groups and facilitate a more nuanced analysis of changes over time, much like the typologies used for families in this brief or for the Chicago Family Case Management Demonstration (Theodos et al. 2012).
- Advocate for systems change: Some policies are outside of the control of local organizations and may require advocacy on the behalf of and in partnership with low-income families. Providers may want to consider advocating for changes in federal safety net programs, from formulas that provide lower per capita subsidies for larger families to the benefits cliff that families transitioning to self-sufficiency often face. In addition, it may be important for two-generation practitioners to work in their local communities to address the structural barriers that make it difficult for families to succeed, including a lack of good jobs and inadequate medical and mental health services, transportation systems, child care, education, and training.
Lessons learned from the next set of deliberately designed and implemented two-generation models may inform both the practice of local practitioners and the work of policymakers, thus allowing for the construction of sustainable, integrated systems that yield better outcomes for all low-income families and more efficient use of public resources in the fight against intergenerational poverty.
- For more information on rural two-generation initiative, see Doug O’Brien, “HHS Federal Partners Announce Rural IMPACT Demonstration Designees,” Family Room (blog), US Department of Health and Human Services, September 28, 2015.
- This cycle is a systematic series of steps developed by Walter A. Shewhart of Bell Laboratories in New York and developed further by Dr. W. Edwards Deming. The cycle is widely used from the health sector to education to social services to support the continuous improvement of products or process.
Ascend. Forthcoming. “Making Tomorrow Better Together, Report of the Two-Generation Outcomes Working Group,” Washington, DC: Ascend, Aspen Institute.
Bishaw, Alemayehu. 2013. “Poverty: 2000 to 2012.” American Community Survey Briefs. Washington, DC: U.S. Census Bureau, 2013.
Chase-Lansdale, P. Lindsay, and Jeanne Brooks-Gunn. 2014. “Two-Generation Programs in the Twenty-First Century.” Future of Children 24 (1): 13–40.
Chetty, Raj, Nathaniel Hendren, Patrick Kline, Emmanuel Saez, and Nicholas Turner. 2014. “Is the United States Still a Land of Opportunity? Recent Trends in Intergenerational Mobility.” In The Quality of Opportunity Project, 1–36. Cambridge, MA: National Bureau of Economic Research.
Cohen, Miriam. 2014. “Women and Welfare in the United States.” Journal of Women's History 26 (2): 180–90.
Corak, Miles. 2013. “Income Inequality, Equality of Opportunity, and Intergenerational Mobility.” Journal of Economic Perspectives 27 (3): 79–102.
Duncan, Greg J., Kathleen M. Ziol Guest, and Ariel Kalil. 2010. “Early Childhood Poverty and Adult Attainment, Behavior, and Health.” Child Development 81 (1): 306–25.
Edin, Kathryn J., and H. Luke Shaefer. 2015. $2.00 a Day: Living on Almost Nothing in America. New York: Houghton Mifflin Harcourt Publishing Company.
Gruendel, Janice M. 2014. “Two (or More) Generation Frameworks: A Look across and Within.” New York: Center for the Study of Social Policy.
Haskins, Ron, Irwin Garfinkel, and Sara McLanahan. 2014. “Introduction: Two-Generation Mechanisms of Child Development.” Future of Children 24 (1): 3–12.
King, Christopher T., Rheagan Coffey, and Tara C. Smith. 2013. “Promoting Two-Generation Strategies: A Getting-Started Guide for State and Local Policy Makers.” Austin, TX: Ray Marshall Center for the Study of Human Resources and the Lyndon B. Johnson School of Public Affairs at the University of Texas at Austin.
Mosle, Anne, and Nisha Patel. 2012. “Two Generations, One Future: Moving Parents and Children Beyond Poverty Together.” Aspen, CO: Aspen Institute.
Parilla, Joe, and Brett Theodos. 2010. “Moving "Hard-to-House" Resident to Work: The Role of Intensive Case Management.” Washington, DC: Urban Institute.
Pierre, Robert St, Jean I. Layzer, and Helen V. Barnes. 1996. “Regenerating Two-Generation Programs.” Cambridge: Abt Associates.
Popkin, Susan J., and Liza Getsinger. 2010. “Reaching the Next Generation: The Crisis for Cha's Youth.” Supporting Vulnerable Public Housing Families. Washington, DC: Urban Institute.
Scott, Molly M., Elsa Falkenburger, Marla McDaniel, Amy Khare, and Susan J. Popkin. 2013. "Host Year 2: Implementation and Expansion." Washington, DC: Urban Institute.
Scott, Molly M., Susan J. Popkin, Marla McDaniel, Priya Saxena, and Reed Jordan. 2013. “Serving Host Families: The Challenges to Overcome.” Washington, DC: Urban Institute.
Scott, Molly, Susan J. Popkin, and Priya Saxena. Forthcoming. “Making a Two-Generation Model Work in the Real World.” Washington, DC: Urban Institute.
Sommer, Teresa Eckrich, P. Lindsay Chase-Lansdale, Jeanne Brooks-Gunn, Margo Gardner, Diana M. Rauner, and Karen Freel. 2012. “Early Childhood Education Centers and Mothers Postsecondary Attainment: A New Conceptual Framework for a Dual-Generation Education Intervention.” Teachers College Record 114 (10): 1–40.
Congressional Budget Office. 2015. “Temporary Assistance for Needy Families: Spending and Policy Options.” Washington, DC: Congressional Budget Office.
Theodos, Brett, Susan J. Popkin, Joe Parilla, and Liza Getsinger. 2012. “The Challenge of Targeting Services: A Typology of Public-Housing Residents .” Social Service Review 86 (3): 517–44.
This brief was funded The Urban Institute thanks the Open Society Foundations, Paul G. Allen Family Foundation, Kresge Foundation, W. K. Kellogg Foundation, US Department of Housing and Urban Development, Annie E. Casey Foundation, and National Institutes of Health. We are grateful to them and to all our funders, who make it possible for Urban to advance its mission. We also thank the Chicago and Portland HOST communities that opened their doors to us and shared valuable information and life experiences.
The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Funders do not determine our research findings or the insights and recommendations of our experts. Further information on the Urban Institute’s funding principles is available at www.urban.org/support.
The authors also thank our colleagues, Abigail Baum, Mary Bogle, Elsa Falkenburger, Sarah Gillespie, Chantal Hailey, Chris Hayes, Reed Jordan, Marla McDaniel, Amanda Mireles, and Priya Saxena.
About the Authors
Molly M. Scott is a senior research associate in the Metropolitan Housing and Communities Policy Center. In recent years, much of her work has focused on providing practical research support for nonprofits and local government agencies in planning, implementing, and testing innovative, place-based strategies to improve the lives of the ever-more-diverse low-income population.
Susan J. Popkin is a senior fellow and director of the Neighborhoods and Youth Development initiative in the Metropolitan Housing and Communities Policy Center at the Urban Institute, a research program that focuses on the ways neighborhood environments affect outcomes for youth and on assessing comprehensive community-based interventions. A particular focus is gender differences in neighborhood effects and improving outcomes for marginalized girls.
Jasmine K. Simington is a research assistant in the Metropolitan Housing and Communities Policy Center, where her work focuses on assessing housing needs and conditions, school-based wraparound services, and community-based participatory research.