The Social Safety Net as an Investment in Children

London School of Economics and Political Science (LSE): Public Lectures and Events
Speaker: Professor Hilary Hoynes (UC Berkeley)
Host: Professor Coretta Phillips (LSE)
Date: October 23, 2025

Episode Overview

This episode features Professor Hilary Hoynes, Chancellor’s Professor of Economics and Public Policy at UC Berkeley, delivering the annual LSE Social Policy Lecture. The central theme: reimagining the social safety net not simply as charitable relief, but as a long-term investment in children's wellbeing and societal prosperity. Professor Hoynes surveys recent research—much of it her own—on the longitudinal benefits of programs like cash assistance, tax credits, food assistance, and public health insurance. She argues that focusing on children's outcomes reveals robust returns to social spending that have often been obscured by fixation on short-term costs and parental labor market responses.

Key Discussion Points and Insights

1. Reframing the Social Safety Net (04:40–14:00)

• Traditional economic evaluations of social programs have prioritized measuring costs—especially effects on parental labor supply—while undervaluing long-term benefits, particularly to children.
• The field has seen a recent shift, with a surge in research quantifying benefits such as improved education, earnings, and health for children receiving assistance.

“If the program generates a lot of benefits, maybe we’re willing to tolerate some costs along the way... We really need to bring them both together in order to think about optimal policy.”

—Hilary Hoynes [12:40]

2. Historical Context and Comparative Evidence (14:00–25:00)

• Analysis of OECD data: higher spending on family benefits is strongly correlated with lower child poverty rates across developed countries.
• The US spends relatively less than peers, with persistently high child poverty.
• Literature review: Most published papers historically have studied program incentives and costs, much less the potential long-term returns.

3. Theoretical Foundations (25:00–32:00)

• Fetal origins/Barker hypothesis (UK-born): Environmental shocks during pregnancy or early childhood can have profound, lasting effects.
• Chronic stress theory: Resource instability and stress undermine child development and long-term potential.
• Positive versus negative shocks: Much prior work focused on harmful shocks (famines, pollution), but increasing positive resources (via social policy) can have equally lasting positive effects.

“If bad things are bad for the long run, could positive things like increasing social supports be positive in the long term?”

—Hilary Hoynes [31:10]

4. Empirical Evidence by Program Type

A. Cash Assistance – Mothers’ Pension Program (32:00–38:00)

• Study by Anna Aizer et al: Compared children of mothers accepted vs. rejected from early 20th-century US cash welfare.
• Findings: Those accepted had greater longevity, lower rates of underweight (malnourishment), higher educational attainment, and increased income in later life.
• Long-term effects are significant though not transformative—small, persistent gains aggregate over lifespans.

B. Cash Infusions from Native American Casinos (38:00–40:00)

• Randy Akee et al: Casino revenues distributed as basic income in Native American communities led to higher education and reduced criminal activity among exposed children.

C. Tax Credits – Earned Income Tax Credit (EITC) (40:00–45:00)

• EITC is a large, work-conditioned US tax benefit, with effects channeled via both higher household income and increased maternal labor supply.

• Andrew Barr et al: A simple yet powerful “natural experiment” found that children born just before year-end (and thus eligible for a child tax credit sooner) had higher adult earnings, especially boys (+8–9% effect).

  “$1,300 in the first year of life, a modest treatment, leads to a quantifiable, not life-changing, but a quantifiable effect on earnings 25 years later.”
  —Hilary Hoynes [44:30]

• Larger, aggregated evidence demonstrates EITC boosts later educational attainment, test scores, and adult income.

D. Medicaid – Public Health Insurance for Children (45:00–54:00)

• Janet Currie & John Gruber; Wary et al: Medicaid expansions substantially reduced uninsured rates for low-income US children.
• Effects observed: Improved birth outcomes, lower teenage and adult mortality (esp. for internal health causes), fewer ER visits for chronic conditions (as late as age 25), higher education and income, reduced criminal justice involvement; even improved birth outcomes for the next generation.

“The rate of return on Medicaid providing health insurance for pregnant women and small children is so high that the program fully pays for itself in the long run through subsequently higher taxes and lower social spending, which is kind of amazing.”

—Hilary Hoynes [22:50]

E. Food Assistance – SNAP/Food Stamps (54:00–58:00)

• Due to staggered county-level rollouts, researchers can estimate effects of early-life access.

• Findings: Access before school age is linked with higher human capital, better adult health (lower rates of metabolic syndrome), more upward mobility, and lower premature mortality.

  "The way to read this graph is: as I move to the left...as they get more access to food stamps... my human capital improves. It’s almost kind of linear—a dose response."

  —Hilary Hoynes [55:10]

5. Key Puzzles and Research Gaps (58:00–60:00)

• Not all programs generate long-term gains. Example: Disability-targeted child benefits (SSI) show no discernible effect on adult earnings/education, possibly due to the severity of disadvantage.

• Unknown mechanisms: Surprisingly, recent RCTs of basic income in the US (e.g., "Babies’ First Years") show little immediate impact on child development or parental behavior, raising questions about how cash translates (or doesn't) to long-term success.

  "...a lot of our hypotheses about where the mechanisms are in terms of how these additional resources translate to better education and earnings in the long, long run just might not be right..."

  —Hilary Hoynes [60:20]

6. Calculating Returns and Policy Implications (60:00–63:00)

• Marginal Value of Public Funds: New frameworks attempt to quantify both "leaky bucket" (costs/lost tax revenue due to reduced labor supply) and the downstream benefits (higher earnings, education, longevity).

• For SNAP, benefits to the family can reach $56 per $1 of government cost—not that the program "pays for itself" fiscally, but these private and public returns have long been underappreciated.

• Hendren and Sprung-Kaiser: Comprehensive studies consistently find larger rates of return for programs that reach children earlier.

  "If we hadn’t done all of this work that shows kids who have access to food stamps have higher education, are less likely to die young... we'd have a marginal value of public funds that's less than one."

  —Hilary Hoynes [62:00]

Audience Q&A Highlights

Conditional vs. Unconditional Assistance (58:21)

• Hoynes: Most studied US programs are "conditional" (income and/or work requirements), not universal basic income. Evidence for unconditional long-term effects is scarce but growing, especially via universal child benefits seen in some cross-country studies.

Long-term Recovery for Malnourished Children in Conflict Zones (61:23)

• These children face high long-term risk for metabolic syndrome and chronic stress, even if conditions suddenly improve; research shows neither risks nor recoveries are ever "all or nothing."

SNAP Work Requirements and Administrative Burdens (63:37)

• Tighter work requirements often kick recipients off more than they increase employment. More flexible administration can mitigate this, but policy changes tend to add burdensome hoops.

The Role of Research in Polarized Political Environments (69:45)

• The impact of academic research on policy is limited when leaders operate in an "informationless environment" and public trust in experts erodes.

Methods: Dealing with Confounders and Generalizability (72:57)

• Economists exploit natural experiments and discontinuities (birthdates, weights, etc.) to rule out many confounders, but these designs sometimes limit generalizability and fail to reveal nuanced mechanism or context effects.

Intrinsic vs. Instrumental Value of Investment in Children (76:33)

• There are both intrinsic (dignity, reduced stress, reduced stigma) and instrumental (higher earnings, less crime, better health) benefits; economics tends to quantify the latter but both matter in policy debates.

Social Impact Bonds and Measuring Success (79:29)

• Social impact bonds are powerful for piloting innovations and pre-specifying metrics, but short evaluation windows present challenges for measuring long-term gains that accrue over decades.

Extending Evidence to Global South (82:34)

• Many safety net programs in low and middle income countries are new, so evidence thus far is limited to near-term outcomes; as programs mature, opportunities for long-term studies will grow.

Family History and Health Outcomes (83:22)

• Some US administrative data does not allow tracking family health history across generations, but future longitudinal data linkages may make this possible.

Why Policymakers Resist “No-Brainer” Investments in Children (83:55)

• Barriers: Political time horizons (electoral cycles), accounting conventions (infrastructure = investment; child benefits = spending), and bureaucratic structures obscure the logic of long-term returns. Some suggest legislative mandates to extend cost windows for program evaluation.

  "All tax bills need to be costed out by the Congressional Budget Office... There's a fixed window—either five or ten years. So... they're trying to think about how they would use the information they have from existing studies in order to do costs, like, generational costs."

  —Hilary Hoynes [84:49]

Notable Quotes

• "We really need to bring both costs and benefits together in order to think about optimal policy."
  —Hilary Hoynes [12:40]

• “If bad things are bad for the long run, could positive things like increasing social supports be positive in the long term?”
  —Hilary Hoynes [31:10]

• “$1,300 in the first year of life... leads to a quantifiable effect on earnings 25 years later.”
  —Hilary Hoynes [44:30]

• “The rate of return on Medicaid... is so high that the program fully pays for itself in the long run through subsequently higher taxes and lower social spending, which is kind of amazing.”
  —Hilary Hoynes [22:50]

• “If we hadn’t done all of this work... we'd have a marginal value of public funds that's less than one.”
  —Hilary Hoynes [62:00]

Summary Table of Key Evidence (Approximate Timestamps)

| Segment | Key Program | Main Long-Term Outcomes | Notable Studies | |---------------------------------------|----------------------------------|------------------------------------------------|----------------| | [32:00–38:00] Cash assistance | Mothers’ Pension Program | Longevity, education, income | Anna Aizer et al. | | [38:00–40:00] Basic income (casinos) | Native American casino revenues | Education, reduced crime | Randy Akee et al. | | [40:00–45:00] Tax credits (EITC) | Earned Income Tax Credit | Test scores, education, income | Barr et al., McInnes et al. | | [45:00–54:00] Health insurance | Medicaid | Health, earnings, lower mortality, less crime | Currie & Gruber, Wary et al. | | [54:00–58:00] Food stamps | SNAP/Food Stamps | Human capital, metabolic health, longer life | Hoynes et al. | | [58:00–60:00] Disability cash (SSI) | Supplemental Security Income | No consistent long-term gains | Miller et al. |

Concluding Thoughts

• Investments in children through safety net programs demonstrably yield widespread, lasting social and economic returns, even if effects are modest and slow to mature.
• There is a pressing need to integrate both cost and benefit analyses in policy evaluation, look beyond parental work responses, and push for political/administrative reforms to recognize social spending on children as true “investment.”
• Enduring challenges remain: understanding mechanisms, improving data for better causal inference, extending findings to new contexts, and, crucially, persuading policymakers to act on long-term evidence in an era of short-termism and skepticism.

“I just want to end here by saying I hope this has given you a glimpse into this emerging literature... spending more of a society in basic safety net programs can both generate improvements and also sort of pay back that tax investment in the long term by these improvements in outcomes.”

—Hilary Hoynes [63:00]