Abstract
To fill gaps in the safety net, municipalities have experimented with giving low-income residents a guaranteed income: regular cash transfers that can be spent without restriction. Combining survey evidence from a randomized experiment (n = 1,074) with longitudinal in-depth qualitative interviews (n = 56), we evaluate a two-year guaranteed income program in Compton, California. Recipients indicated that smaller, steadier transfers helped them keep up with bills, while less frequent, larger transfers enabled financial planning. Most households took actions requiring restraint, such as catching up on bills or paying off debt, but rewarded these efforts with spending to meet family wants and create exceptions in routines to spend time on memorable activities. Impacts of the transfers depend on residents’ household situations. For single-parent households (mostly single mothers), receiving unconditional money led to more work and higher earnings but also higher debt. In contrast, dual-parent households earned less, constrained their spending, and reduced their debt.
In the years since Making Ends Meet (Edin and Lein 1997), it has become more widely accepted that an essential problem for many American families with children is simply that they do not have enough money to get by. Too often, budgets do not add up even for those with full-time employment and access to public benefits such as government health insurance and the Supplemental Nutrition Assistance Program (SNAP). Contributions from relatives and friends can help fill gaps, but the kin of struggling families are often struggling themselves. Side jobs can help, but they take time away from parenting and are not always feasible due to unpredictable work schedules (Kwon et al. 2026, this issue). As Kathryn Edin and H. Luke Shaefer (2015) and H. Luke Shaefer and colleagues (2020) document in their description of the historical transition of social support programs, from Aid to Families with Dependent Children (AFDC) to Temporary Assistance for Needy Families (TANF) to the Earned Income Tax Credit (EITC), the cash safety net for the families most in need of support has weakened considerably over the past three decades as the funds once designated for assisting them have been diverted to the working poor and other uses.1
A recent response to the erosion of the cash safety net has been a patchwork of guaranteed income pilot programs. The programs are often funded by local governments or philanthropy and provide steady cash supplements to low-income households at regular intervals, usually monthly, over a limited period. Some explicitly target mothers. For example, the Magnolia Mother’s Trust, which was launched by Springboard to Opportunities in 2018, provided $1,000 a month to one hundred women in Jackson, Mississippi. The Bridge Project in New York provided $1,000 a month to five hundred mothers, beginning in 2021. Rx Kids in Flint, Michigan provided $500 a month for a baby’s first year and effectively ended infant poverty in the city. Baby’s First Years, which was structured as a randomized controlled trial (RCT), provided 1,000 mothers of newborns with $333 per month for the first fifty-two months (Noble et al. 2025). The study found increased spending on child-specific goods out of the cash transfers relative to other income, despite no requirements on how the money was spent, though the magnitude was small (Gennetian et al. 2022). Others, like the recent programs in Stockton, Compton, Los Angeles, and Chicago, target a broader range of low-income households (see Balakrishnan et al. 2025; Vivalt et al. 2024).2
Here, we examine the effects of a large RCT of a guaranteed income program in Compton, California (Balakrishnan et al. 2025), which randomized the frequency of the cash transfers. Compton residents face relatively high rents, low wages, and disproportionately high rates of unemployment and poverty.3 By combining survey data with longitudinal in-depth interviews of randomly chosen program participants, we analyze the impacts of the transfers and the meanings associated with this money for recipients living in different types of households. Our focus is on the response of single-parent or single-mother households as compared to that of two-parent households, other households with minors, and households without minors.
The Compton study offers a chance to ask questions that complement Making Ends Meet, which used semistructured, in-depth interviews to collect detailed information on expenditures and the income needed to meet them (Edin and Lein, 1997). The interviews revealed that poor single mothers who relied on job earnings or welfare, and most often both, did not have much room to breathe, as evidenced by high rates of material deprivation and the extensive juggling needed to get by. In our study, we ask: what are the spending priorities when an extra sum of money materializes in the checking accounts of financially stretched households? How much freedom or flexibility is actually gained, and for whom? Which trade-offs are made? What is the social meaning ascribed to this money, and how does that influence how it is allocated? Here, we ask these questions of unconditional cash transfers, which, unlike the EITC, one of the focal programs in It’s Not Like I’m Poor (Halpern-Meekin et al. 2015), are not attached to labor or earnings and were initially unexpected, and so unencumbered by existing obligations and claims.
The Compton program is unique in its randomization of the frequency of the cash payments. Half of the recipients were randomly chosen to receive smaller transfers every two weeks, while the other half received larger sums once per quarter, keeping the same yearly total. The high-frequency treatment arm is closer to regular installments of income payments or food assistance programs like SNAP, while the low-frequency arm is closer in timing to programs like the EITC. The phrase “making ends meet” is often assumed to refer to the problems of paying rent, keeping food on the table, and taking care of other recurring expenses. In other words, how can money be stretched to cover basic needs? But often families’ needs exceed this narrow definition: making ends meet can require large chunks of money for big expenses, sometimes with urgency and all at once (Morduch and Schneider 2017; Kansikas et al. 2023). The car may need fixing. The water heater might need replacing. Families might need to cope with the death of a loved one. The EITC—paid out in a lump sum at tax time—is powerful partly because it allows families to absorb these expenses (Romich and Weisner 2000; Goodman-Bacon and McGranahan 2008; Halpern-Meekin et al. 2015; Abbott and Tach 2026, this issue), and it also allows families to invest in a “durable” commodity of another kind—special and memorable experiences for parents and children. Like the EITC, which recipients refer to with names such as “get ahead money” and “the kids’ money” (Halpern-Meekin et al. 2015), guaranteed income can facilitate different kinds of expenditures and may be ascribed with specific meaning (Zelizer 1995) that may vary depending on how it is designed, including how frequently it is disbursed.
Making Ends Meet put a focus on the distinct circumstances of single mothers. Here, we also find that the impacts of cash transfers depend on household structures—and that single parent households, largely single-mother households in our sample, show a distinct response. For example, while recipients in general do not significantly adjust their labor supply, single-parent households significantly increase their weekly labor supply in hours—perhaps due to the fact that many of the safety net programs they are eligible for, such as the EITC and (for a lucky few) childcare subsidies, are conditioned on work (and in the case of the EITC, the more you work and earn the more you receive, up until the phaseout rate).
Single-parent households appear to be more constrained: they cannot easily use the money to invest more in parenting time and less in wage-based labor, even if they would prefer it. The results we see among single parents stand in contrast to a recent RCT of cash transfers in Texas and Illinois, which finds negative labor supply responses to cash transfers (Vivalt el al. 2024). Consistent with the increased labor supply result, we find a significant increase in income among single-parent households relative to the control group and all other household types, which largely show a significant decrease in income and expenditures relative to the control group. Additionally, while recipients in dual-parent households pay down a substantial amount of debt, single-parent households accumulate significantly more debt compared to dual-parent households. Scott Fulford (2023) describes how relief provided during the COVID-19 pandemic created a chance for households to reset (for example, by reducing debt or increasing savings). Our results suggest the chance to reset may not have been available to single-parent households in Compton.
We did not find pronounced differences in average impacts between the low- and high-frequency treatment groups in a subsample of households with minors. However, the qualitative interviews reveal that participants experience cash differently—and form different aspirations and spending goals—depending on how often it is received. We find that recipients receiving quarterly transfers paid down more debt and reduced expenditures relative to controls, while those receiving twice-monthly transfers accumulated more assets. As a result, recipients of the twice-monthly transfers reported substantially higher net assets (assets – debt) than those receiving quarterly transfers.4
From the qualitative interviews, we find that many recipient households blended the transfers with regular income and allocated the transfers to a combination of overhanging needs—past debt and unpaid bills—and immediate needs. However, some households treated the cash separately from income—using it as a chance to do something different. The qualitative interviews reveal that households regularly made hard choices to spend limited money on paying debt and catching up on bills, but coupled these payments with spending on enjoyable activities, especially for their children. Spending on these items and experiences was seen by recipients as being critical for strengthening their relationships with their children. The ability to have these positive experiences also provided a helpful balance to the difficult choices needed to allocate limited resources to bills and debt repayment—suggesting that they may be complementary activities rather than substitutes. For parents, this spending created exceptions or pauses in otherwise stressful routines, allowing them to spend meaningful time with their children and create durable memories that can be continuously revisited, even in times of scarcity (Halpern-Meekin et al. 2015; Ananat and Gassman-Pines, personal communication). Receiving the unconditional transfers afforded recipients the dignity and flexibility to meet basic needs and enjoy new experiences and time with their families.
These findings provide a window into the possibilities offered by guaranteed income, while also highlighting the contexts—especially of overhanging debt (Dwyer 2018; Tach and Sternberg Greene 2014) and parental responsibilities (Chaudry 2004)—that lead different types of households to use the resources in ways that extend beyond basic material concerns.
THE CONTEXT FOR GUARANTEED INCOME PROGRAMS
Today’s guaranteed income pilots build on a history of cash-assistance programs in the United States and Canada. The best-known precursors are the negative income tax experiments of the 1970s, also known as income maintenance experiments, which took place in Seattle (SIME); Denver (DIME); rural North Carolina and Iowa (RIME); Gary, Indiana; New Jersey and Pennsylvania; and Manitoba, Canada (Mincome; Forget 2011). While the programs differed in their target populations, each provided guaranteed minimum incomes to households below the poverty line and were evaluated via randomized experiments where the control group received the existing mix of public benefits.
The programs, which initially received bipartisan support (most famously from economists James Tobin and Milton Friedman), lost backing when evidence suggested that they created large work disincentives (Robins 1985). The early evidence on labor supply responses was consistent with the fact that cash benefits were large relative to earnings and that benefits fell sharply when recipients earned more. Gary Burtless (1986) estimated that men in two-parent families reduced their work effort by about 7 percent, while women in two-parent families and single mothers reduced their hours by 17 percent. These labor supply reductions were accompanied by reduced household incomes. However, David Ellwood (1988) countered that part of the large labor supply response was likely due to under-reporting of work, which the program also incentivized. Among other concerns, economists have argued for caution in generalizing from the studies since the samples tilted toward poorer populations—which could have accentuated the negative labor supply responses—and were not broadly representative (see Moffitt 2003; Widerquist 2005; and the studies reviewed in Munnell 1987).5
The recent wave of guaranteed income programs signal a rebirth of interest in cash-based programs. However, this time the interest has been driven by the global rise of conditional cash-transfer programs (CCTs) in the 1990s, interest in universal basic income (UBI), touted by some as a solution for a world where workers lose out to automation (Lowrey 2018), and greater attention to the gaps in existing benefits programs that were made evident by the economic shocks associated with the COVID pandemic.
The randomized evaluation of Mexico’s Progresa, a conditional cash-transfer program launched in 1997, showed broad improvements in household well-being (Levy 2006). Subsequent studies in global settings, often focused on unconditional cash transfers (UCTs) that have conditions only on eligibility requirements and not how the cash is spent, find positive impacts across many outcomes (Crosta et al. 2024; Daidone et al. 2019) and reasonable cost-effectiveness. These findings refute the concern that recipients would quit working (Banerjee et al. 2017) and spend more on alcohol and tobacco (Evans and Popova 2017)—prominent concerns among policymakers and economists.
Unlike UBI, which is envisioned as blanketing an entire population with regular cash transfers (Hoynes and Rothstein 2019), today’s guaranteed income programs (including global CCTs) target only the households that are determined to be most in need, usually through a means test. In a typical US pilot program, recipients receive monthly cash transfers of $500 to $1,000 with no strings attached over one to five years (Vivalt et al. 2024). The transfers are meant to supplement earnings and other benefits but—unlike the early vision for the income maintenance experiments of the 1970s—are not in themselves sufficient to fully replace other income.
THE COMPTON PLEDGE
In December 2020, the City of Compton, California introduced the Compton Pledge—at the time, the largest city-led guaranteed income pilot in the US. At the time, Compton was a city of almost 96,000 residents, 26 percent of whom identified as Black or African American and 71 percent of whom identified as Hispanic or Latino. Located in southern Los Angeles County, Compton has long faced economic struggles. As middle-class Black families moved into Compton in the 1950s and 1960s, it became a center of Black culture in Los Angeles; followed by White flight in the subsequent decades. Since the 1990s, it has seen an influx of Hispanic residents. Today, almost one in five Compton residents live in poverty—almost double the national poverty rate—and many more struggle to make ends meet in a city with high costs of living and, in particular, high rents.6 The unemployment rate at the time of the start of transfers, in part due to the impacts of the pandemic, was just below 16 percent (while the national rate was 6.4 percent).
The two-year unconditional cash transfer pilot program was launched by Mayor Aja Brown under the Compton Community Development Corporation (CCDC) between January and March, 2021 and lasted until April 2023. The political energy that drove the Compton Pledge was created by the COVID pandemic and the urgent need to provide more for low-income residents. Households were eligible to participate in the program if they lived in Compton, had at least one household member aged twenty-three to fifty-seven, and had an income below 220 percent of the federal poverty threshold—the same cutoff used for the EITC. Households receiving Supplemental Security Income (SSI) or Social Security Disability Insurance (SSDI) benefits were excluded from the program to avoid the potential loss of benefits due to the receipt of the cash transfers.7 Transfers lasted two years and were disbursed on either a twice-monthly or quarterly basis. The magnitude of the transfers varied with the number of children: households with no children received $3,600 per year, households with one child received $5,400 per year, and households with two or more children received $7,200 per year.
The quantitative evaluation of the program was structured as a randomized controlled trial. To recruit individuals into the study, contact information was obtained from the Compton 2018 and 2020 voter lists, the Compton Public Housing list, and community organizations.8 These lists were supplemented with random-digit dialing. Using email, SMS messaging, voice calls, and mailers, individuals were invited to participate in a well-being study for the City of Compton by completing a short survey between January and March 2021, which was used to assess their eligibility for the program.
From the list of eligible individuals, 2,100 households were randomly selected to participate in The Compton Well-Being Study. These households were randomly assigned to a treatment group (N = 698) or a control group (N = 1,402), stratified by gender of the participant, and this information was provided to the CCDC to use for their cash disbursements. Half of the treatment group was randomly assigned to receive twice-monthly transfers (N = 347) and the other half to receive transfers once per quarter (N = 351), with each group receiving the same total sum of transfers.9 Participants assigned to the treatment group were informed by the city that they were selected to receive cash with no conditions attached. They were also informed about the magnitude and timing of the transfers, including their end date. Participants could choose to receive the transfers by check in the mail, debit card, direct deposit, PayPal, or Venmo. Out of the 698 households assigned to treatment, 625 enrolled in the program and received payments totaling $6,204 per year, on average, with a 99.9 percent payment success rate.10
METHODS
Our quantitative and qualitative analyses draw on distinct and overlapping sets of participants from the Compton Well-Being Study—the sampling procedure, analytic methods, and sample characteristics are described in this section and in the online supplement, which also reports the sample characteristics for the full study sample, including those who did not participate in the survey or qualitative interviews.11
Quantitative Sample and Procedures
Sample Selection Procedure and Attrition
Eighteen months after the start of transfers, all 2,100 study participants were invited to complete a follow-up survey. The survey had a 51 percent response rate, resulting in a sample of N = 1,074 households (the quantitative sample). While the overall response rate was low, there was no differential attrition by treatment condition, the treatment arms remain balanced on baseline demographics and income, and attrition was not predicted by these characteristics (Balakrishnan et al. 2025). We extend the analysis reported in Balakrishnan and colleagues (2025) by studying the impact of transfers as a function of household composition and by drawing on extensive longitudinal qualitative interviews with recipients.
We identify households with minors using the baseline survey, which was conducted before the start of transfers (from February to March 2021), and household composition using the household roster included in the midline survey, supplemented by information from the qualitative interviews. We construct indicators for four household composition groups: single-parent households in which one parent and their child(ren) are present, where child includes son or daughter, stepson or stepdaughter, or foster child (single-parent households; n = 262); two-parent households (n = 242); other households with minors (for example, grandparent with grandchild; n = 309); and households without minors (n = 261).12
Primary Outcome Variables
The main outcome variables in the quantitative analyses are income, labor supply, expenditures, assets and debt, as well as time use. The focus on these outcome variables reflects the priorities of economists, who have largely led the study of unconditional cash transfer programs. We measure labor supply as hours worked in formal and informal jobs, or self-employment. We also have binary indicators of labor force participation and participation in unpaid work. Using an abridged time use question, we ask participants to report the average time spent per day on unpaid childcare or eldercare in the prior week. We calculate household income as the sum of the participant’s own labor market earnings; earnings of other household members; income from rent payments, dividends or interest; benefits including unemployment insurance, Social Security, SSI, CalWORKS, and CalFresh, SNAP, or WIC; and income from other sources.
We measure household expenditures in the thirty days prior to the survey for the following goods: food and drinks at home and outside the home; alcohol, cigarettes, and tobacco; clothing; housing (rent, mortgage payments, utilities, internet, and phone bills); health care; childcare and eldercare; vehicles; and transportation. The value of household assets includes cash, retirement account balances, the value of businesses, gifts from relatives, and loans to relatives, as well as the total value of durable goods across a variety of items commonly owned by households.13 The value of household debt includes amounts owed on student loans, credit cards, medical debt, gifts to relatives, loans from relatives, and any other debt (for example, vehicle loans, legal bills, and so on).14
Analysis
To examine the heterogeneous treatment effects by household type on our outcomes, we regress our outcome variables on an indicator for treatment condition, collapsing across the two treatment arms, and interact it with an indicator for household type. To examine the effect of transfer frequency on our outcome variables, we include separate indicators for each treatment arm. Because of sample size limitations, for this analysis we subset the sample to households with minors rather than estimating heterogeneous treatment effects by household type. We thus examine differences in our outcomes by transfer frequency among households with children. All regressions include controls for baseline household income; labor supply; CTC amount; number of people and children in the household; and respondent race, age, and sex (see the online supplement, Quantitative Methods, for additional details).
These results provide only a snapshot of the impacts of cash transfers since they were evaluated at one point in time while living arrangements and needs are dynamic. We complement this analysis with the qualitative interviews, which are longitudinal, capturing changes over time as well as aspects of the experience and interpretation of the transfers that are not included in the survey.
Qualitative Sample and Procedures
We recruited fifty-six participants from the RCT to take part in a longitudinal qualitative study that included three rounds of in-depth, semistructured interviews with participants from the two treatment arms and the control group.
Sample Selection Procedure and Attrition
We recruited forty-two participants from the experimental groups (twenty-one from the high-frequency treatment arm, twenty-one from the low-transfer frequency arm), and fourteen from the control group. We randomly selected participants with the following constraints: equal representation of Black/African American and Hispanic/Latino respondents and equal representation of households above (up to 220 percent) and below the federal poverty line in each cell (high frequency, low frequency, and control). Additionally, we selected participants to ensure diversity in gender, age, and household composition. To meet these criteria, we attempted to contact ninety-one participants before reaching the target of fifty-six consented participants, resulting in a take-up rate of 62 percent.
The first round of interviews included fifty-six households, with forty-three interviews conducted on Zoom due to the COVID pandemic. The second round included fifty households (an 11 percent attrition rate), and the third round included forty-six households (an 8 percent attrition rate from round two). Overall, we conducted 152 interviews with fifty-six unique participants.
Interview and Analysis Procedure
Interviews were semistructured, lasted approximately ninety minutes, and were conducted in in English and in Spanish in participants’ homes, public places in Compton, or on Zoom. The first round of interviews was conducted approximately twelve months after the start of transfers (February–June 2022); the second round was conducted, on average, six months after the first interview (August–December 2022); and the final round was conducted after the end of transfers, on average, twelve months after the second interview (April–August 2023).
The first interview covered a range of topics, including family background and household composition; social relations; current and recent employment; sources of income and financial support; and use of government programs, benefits, and services. Subsequent interviews covered changes from the prior interview and included new focal modules (for example, debt and savings; material hardship; division of household finances; housing and household composition; health, stress, and well-being; guaranteed income uses and perceptions; social supports; and perceptions of employment opportunities). Because of our interest in understanding the experience of unconditional cash transfers, including how they were treated vis-à-vis other income sources, many of our questions were focused on household finances, uses of the transfers and their impacts, and the meaning ascribed to the transfers. See the online supplement, Qualitative Methods, for additional details.
RESULTS
In the following sections we discuss the results of the randomized-controlled trial, followed by our analysis of the longitudinal qualitative interviews.
Sample Characteristics
We start by describing the overlapping yet distinct quantitative and qualitative samples.
Quantitative Sample
The quantitative sample relies on data from 1,047 respondents who completed both the baseline and follow-up surveys. Respondents in this sample were, on average, thirty-five years old and living in a household with around five other people. The majority of respondents in the survey sample identified as Hispanic or Latino (67 percent, with 29 percent identifying as African American or Black; no households identified as both), reflecting the demographics of Compton, and 78 percent identified as female. Female recipients were much more likely to have young children living at home (80 percent) than male respondents (61 percent). Households in the quantitative sample had an average of 1.8 minors per household at baseline and an annual average income of $27,555. Control and treatment arms were balanced on age, ethnicity, race, gender, income, household size, and number of minors in the household. Descriptive statistics are reported in table S.1 of the Supplementary Materials.
Splitting the sample into our four household composition types—single-parent households; two-parent households; other households with minors; and households without minors—we see significant differences in demographic and socioeconomic variables across the four groups (shown in table S.1). Recipients in single-parent households are significantly more likely to identify as female (91 percent) than those in two-parent households (78 percent), other households with minors (76 percent), or households without minors (66 percent). They are also significantly less likely to identify as Hispanic (57 percent) and more likely to identify as Black or African American (39 percent) compared to two-parent households (84 percent Hispanic and 12 percent Black or African American) and other households with minors (73 percent Hispanic and 23 percent Black or African American). On average, single-parent and two-parent households have similar numbers of minors at home (~2.4).
We see significantly lower baseline employment rates in single-parent households (52 percent) relative to other households with minors (66 percent) and without minors (62 percent). Respondents worked on average twenty hours per week. Conditional on being employed, they worked on average thirty-three hours per week. Single-parent households also reported significantly lower monthly income than two-parent households—due to a combination of lower recipient earnings and contributions from others in the household—and received significantly more benefits transfers (for example, from SNAP) relative to the three other household types.
Qualitative Sample
Of the fifty-six respondents who participated in the qualitative portion of the study, twenty-seven identified as African American or Black and twenty-nine as Hispanic or Latino; twenty-seven (48 percent) reported household income below the federal poverty line at baseline; and forty-two (75 percent) identified as female and had a mean age of thirty-five, matching the characteristics of the full Compton Well-Being Study.15 Of the thirty-four households with minors, twelve were single-parent households (all were single mothers), and seven respondents were living with minors who were not their children. Almost half of households with minors had more than one financially independent economic unit in the household, which we refer to as households within households.
Our qualitative interviews, and a follow-up survey focused on household structures, reveal varied and fluid arrangements in terms of the relationships present in a home, how spaces are occupied, and how finances are shared. Many households in our sample are multigenerational, often including a father or mother, an adult son or daughter, children, and siblings. Hope Harvey (2026) refers to shared households as doubled-up, while we use households within households to describe discrete financial units within a home (see also Harvey et al. 2021). In reality, the boundaries of financial dependence or independence are porous and change over time—contributions to household finances vary from week to week or month to month as needs or means arise, and in-kind contributions from others (for example, borrowing a car, childcare) have material impacts. These arrangements are much more common among single-parent households (54 percent) and other households with minors (80 percent) than they are among two-parent households (33 percent) or households without minors (37 percent), likely due to differences in financial or other needs.
In the next sections, we examine how differences in household composition—which come with different aspirations, priorities, needs, and constraints—shape the effects and the experience of unconditional, temporary, and unexpected and therefore unencumbered cash transfers on various aspects of recipients’ material and social realities. In the next section, we describe the results of the quantitative analyses, which reveal systematic differences in the impact of the transfers by household composition type. We then turn to the qualitative analyses, which provide insight into how unconditional cash transfers are experienced, prioritized, and ascribed social meaning. While our quantitative results focus on discretely defined household types, we provide examples of more fluid arrangements throughout the qualitative analyses.
Overall Impact of Cash Transfers by Household Composition: Quantitative Sample
Using the survey results, we assess the causal impacts of the transfers on various outcomes by household composition. Table 1 shows the effect of treatment, relative to the control group, for single-parent households (column 2), two-parent households (column 3), other households with minors (column 4), and households without minors (column 5). Columns 6–8 show the p-values for the pairwise comparisons between treatment effects on single-parent households and the other three types.16
Overall Treatment Effects by Household Composition
Work and Earnings
Among economists and policymakers, a central concern with unconditional cash transfers is that they may crowd out incentives for wage labor without increasing time spent on other productive activities, such as education. This crowding out may be more prevalent in two-parent households, where there is greater flexibility for one of the parents to reduce their labor supply while the other works full time. Single parents usually lack that flexibility and have strong incentives to maintain their labor supply, in part due to incentives created by the EITC program and because they may have lower household incomes to begin with.
Balakrishnan and colleagues (2025) find no significant effects of the transfers on overall labor supply. However, we find a more nuanced pattern of results when disaggregating by household type (table 1). The transfers lead to a significant increase in the weekly labor supply in hours of single-parent households (unconditional on working) relative to controls (6.94 hours per week, p < .01) and relative to other households with minors (p < .01) and without minors (p < .001). This latter group significantly decreases their hours relative to the control group (−4.94 hours per week, p < .01). The results suggest that receiving the transfers may make it easier for single-parent families to participate in full-time employment. In keeping with this possibility, single-parents report substantively higher car assets relative to the control group and two-parent households, though this difference is not statistically significant (table S.4).
All else equal, total income should rise by approximately $600 for all households with minors and $300 for households without minors—the average transfer amounts per month for the different household types. Perhaps unsurprisingly then, total income in the past thirty days, including the transfers, is significantly higher among single-parent homes ($747) after the transfers. However, this is not the case for two-parent households (−$3), other households with minors (−$162), or households without minors (−$196), where income effects, even including the transfers, are negative and significantly lower than for single-parent households. This finding suggests that the transfers displace other income above and beyond the amount of the transfers, and may be consistent with directional, albeit nonsignificant, differences in labor market effects. This pattern extends to income without transfers, where we see a small, nonsignificant increase for single-parent households ($234) relative to controls and strong and significant decreases for all other household types (−$450 to −$584) relative to both control and single-parent households.17 Lower rates of employment, income, and spending relative to the control group should be interpreted against a context where income and spending were generally rising (Fulford 2023). A decrease in labor supply, income or expenditures may mean a slower increase in the treatment group relative to the control group. Indeed, as shown in table S.6, there is a substantial increase in control group income between the baseline ($2,565) and midline survey ($3,341), which gets compared to a relatively smaller increase in income among cash recipients ($2,390 at baseline, $2,848 at survey).
Spending and Debt
Consistent with the negative impact on income, recipients in two-parent households and other households with minors spent significantly less on a broad range of spending categories (excluding major durables) in the thirty days prior to the survey relative to the control group (−$439 and −$407, respectively). These negative effects were not concentrated in specific spending categories but rather distributed across diverse needs such as food, transportation, healthcare, and clothing (table S.3). Consistent with the possibility that two-parent families receiving transfers are able to reallocate some of their time to childcare or eldercare, we see a small but significant decrease in monthly spending on these categories relative to the control group (−$44, p <.001).
At the same time, we see large yet nonsignificant positive effects on net assets among two-parent households and other households with minors that are driven by large and statistically significant decreases in debt (−$6,423 and −$6620, respectively) relative to both the control group and single-parent households. The decrease in debt is primarily driven by reductions in student loans and credit-card debt (see table S.4 for the Assets and Debt breakdown). In contrast, single-parent households appear to accrue debt ($3,783), though this estimate is not statistically significant and lacks precision, and to increase their assets relative to controls and two-parent households through the accumulation of durable goods such as refrigerators and other appliances, though these differences are not statistically significant (table S.4).
Many recipients struggle with overhanging debt, a condition broadly felt in low-income communities in the US (Dwyer 2018; Tach and Sternberg Greene 2014). Lower expenditures among some recipient households relative to the control group could actually indicate a slower increase in expenditures relative to the control group, and this could be partly driven by the desire to pay off debt.18 Despite their reduced income, two-parent households may have been able to pay down debt through a combination of the transfers, lower monthly expenditures, and by drawing down savings and checking accounts or selling durable goods, which could in turn ease their earnings pressure.
Differences in the Impact of Cash Transfers by Frequency of Payments
We also examine whether the impacts of cash transfers on households with minors vary with the frequency of the transfers. Prior research finds that bonuses, which tend to be large and given just once a year, are spent differently than raises, due to what behavioral economists call mental accounting—in other words, different sums of money are placed in different mental buckets and are earmarked for different kinds of expenses (Thaler and Benartzi, 2004), an idea also explored by Viviana Zelizer (1995) through a social lens.19 Similarly, prior research on the EITC suggests that its lump-sum feature encourages recipients to focus on long-term economic goals rather than short-term needs, facilitated through the alleviation of what psychologists call the bandwidth tax, imposed when families are forced to put all their energy into scrambling and strategizing to meet urgent short-term financial needs (Mani et al. 2013; see also Sykes et al. 2015 and Halpern-Meekin et al. 2015). Thus, even when the same total amount of money is transferred, differences in how the sums are aggregated and disbursed may affect their impacts.
In table 2, we subset our data to households with minors and show effects for bimonthly (that is, twice monthly) transfers in column 2, quarterly transfers in column 3, and the contrast between the two in column 4. We present results of the interaction with single-parent households in table S.5.20
Treatment Effects by Frequency Arm (Households with Minors)
The fourth column of table 2 shows limited statistically significant differences between the treatment arms among households with minors. We find that recipients who receive transfers twice-monthly accumulate significantly more assets (and have higher net assets) than those in the quarterly treatment arm (p < 0.02). We also find notable and distinct impacts of transfers on different outcomes relative to control for each transfer frequency condition. The low frequency (quarterly) transfers are associated with a significant positive effect on weekly hours worked conditional on labor market participation (6.3 hours per week) relative to the control group, which could be due to more part-time workers taking time before returning to the labor force. This latter interpretation is supported by the absence of a positive effect when we do not condition on work. Echoing the overall results, there is a negative impact on monthly income in both treatment arms, which is statistically significant only for the twice-monthly (high frequency) treatment arm—perhaps because twice-monthly transfers can more readily substitute for income. Recipients in the quarterly treatment arm spend significantly less in the thirty days before the survey (−$462, p < 0.001) than controls and pay down significantly more debt (−$3,793, p < 0.05).
Summary of Quantitative Results
Overall, these findings indicate that different types of households have heterogeneous financial needs, and experience cash transfers differently. Single-parent households may be more financially constrained and may thus use the cash to increase the hours they work each week and secure or increase their incomes. At baseline, 52 percent of single-parent households were employed, suggesting that they may face substantial barriers to work, in part because of childcare needs. It is possible that the positive effect of transfers on some expenditure categories—for example, larger spending on durable goods, which includes computers, phones, or even cars—facilitates additional work and earnings and increases long-term savings and may be worthwhile even if they require taking on debt in the short-term (table S.3).
Our results also suggest that single-parent households may have less flexibility when it comes to adjusting their monthly expenditures. They may be under more pressure to increase or maintain spending compared to other households with minors or households without minors. They may also take the transfers as an opportunity to purchase needed appliances and meet other unmet or overhanging needs. Two-parent households may have more flexibility to reduce work during this period, potentially spending more time with children, which could explain lower spending on childcare and eldercare (table S.3). Two-parent households and other households with minors also spend their transfers to pay down debt, an option that may not be as readily available to, or desirable for, recipients in single-parent households (or households without minors) who actually accrue debt relative to the control group during this period.
Experience and Social Meaning of Unconditional Cash Transfers: Qualitative Sample
While the quantitative results show important systematic differences among household types that suggest different strategies and constraints shaping the use of unconditional cash transfers, they say very little about the actual experience of the transfers and how they relate to household arrangements. We turn to the qualitative results to understand how transfers were perceived, experienced, and treated relative to other income sources, focusing primarily on single-mother households but drawing on other household types where relevant.
Regular, Unencumbered, and Temporary Cash Transfers
The cash transfers from the guaranteed income programs being tested in the US are distinct from many other forms of public benefits in that they are regularly occurring payments—over a certain number of years—that are, unlike the EITC, not attached to employment or earnings, at least after initial selection into the program. The random assignment of many of these programs also means that the funds are unexpected and have not yet been encumbered by existing financial obligations when recipients find out about the transfers. Dionne, an African American single mother of three who had just turned forty-three, described her initial surprise at learning about the Compton Pledge, “I didn’t believe it. And so when I got it, I’m like, this is so crazy because I needed it so bad. With this money I try to do more of the things that I can’t do with my paycheck, cause my paycheck is already taken on, you know, for the rent, the house bills and stuff like that.” The possibilities afforded by the unencumbered quality of the transfers were echoed by Renata, a forty-year-old Hispanic woman living with her husband and three children, who said, “I felt a relief, cause I was able to use it like I’m telling you because it wasn’t on my budget, so I was able to afford a few more things, you know?” Other recipients, like Jamila, a single mother of a six-year-old boy, emphasized that the transfers came just in time: “Yeah, I feel like it came through on a clutch. Like it didn’t come the same day my paycheck came, so it came like when I really needed it.” The timing of the transfers vis-à-vis other sources of income payments came up repeatedly as a factor influencing how the cash was experienced and allocated.
The qualitative results reveal uses of the transfers that go beyond the narrow concerns of typical economic analyses, which tend to focus on labor supply, income, and broad spending categories. The unexpected nature of the transfers created, for some recipients, the desire to do something different with the money. Dionne expressed how receiving the money expanded her spending aspirations: “When I get this money, I try to think outside the box. Do I need a deep freezer? You know, do I need tires for the car? Does my son need a new wardrobe because he grew. It makes you think about the other things you need that you don’t normally think you need, that you can change with a big lump sum, you know?” Ximena, a twenty-seven-year-old Hispanic single mother living with her nine-year-old-son described the distinctiveness of the funds: “I treat it differently. I see it as money that I wouldn’t have. So I try to take care of it better or use it for things that are gonna boost me up a little bit from my position that I’m in.” She used the transfers to purchase appliances, pay for school, and settle debts with her sister. The transfers were treated as “special monies” in the sense of Viviana Zelizer (1995) and Jennifer Sykes and colleagues (2015). At the same time, given the weight of recurring expenses, many recipients did not earmark the money or wall it off into mental accounts (Thaler 1985). Almost half of our sample blended the transfers with their regular income and used it to keep up with basic needs and regular expenses.
The vast majority of recipients found this windfall to be beneficial; however, some saw it as insufficient relative to their basic unmet needs and the cost of living. Chantel, who was living at home with her teenage son at the time, explained that “The first six months I was saving it. We started having to dig into it to pay bills and stuff. Now it’s like every time I get it, before it even hit my card, I already know what it’s about to be used for.” A sentiment echoed by Derrick, a twenty-five-year-old African American man who was living at home with his parents: “$300 is essentially nothing. It doesn’t last long here. I blinked and everything’s so expensive now.”
The temporary nature of the program meant that many of the recipients were reluctant to become too reliant on the transfers. When asked how the transfers had changed her spending, Carmen, who was raising her daughter with her grandmother, said “I don’t like to make commitments that I know I won’t be able to make, where they tell me that tomorrow they won’t give me anything.” Ericka, a thirty-eight-year-old Hispanic mother of five who lives with her husband in his parents’ back house, expressed a similar reluctance to become reliant on this money: “My mentality was, I can’t, the money’s not guaranteed. You don’t know when it’s going to stop coming. It’s been a great help, it’s just you don’t count with it like, ‘I’m going to receive it and so I’m going to use it for this’. No, you don’t count for it.”
The recipient experiences of the Compton Pledge program provide a window into how low-income households situated in a setting with high costs of living and facing significant debt burdens and unmet or overhanging needs understand and spend extra or “special” money. Even when the money is put toward paying regular bills rather than being earmarked for special purposes, the money itself is still seen as an independent cash flow, with a particular social meaning and context, as discussed by Viviana Zelizer (1995).
Spending Priorities for Cash Transfers
The qualitative interviews offer insights into how single mothers and other types of households receiving the transfers considered this money in relation to their other finances and prioritized different categories of spending. In particular, the interviews revealed four primary approaches used by Compton residents to organize or allocate the transfers: eighteen recipients blended the money with other income sources and used it for regular purchases, over half of recipients earmarked at least a portion of the transfers for specific expenses or spent the transfers on bigger purchases that they could not otherwise afford, and almost half of the forty-two recipients put a portion of the transfers into savings. While some recipients were able to use the cash for future-oriented purchases and investments, such as education, most entered the program with a combination of overhanging needs, and used the cash to cover past expenses carried forward in the form of current debt owed on credit cards, or to utilities, landlords, and family members, and to meet basic needs (see Dwyer 2018; Tach and Sternberg Greene 2014).
Addressing Overhanging Needs or “Helping us play catch up”
Most recipients entered the program with multiple forms of debt. In the survey, the control group reported owing, on average, $19,142 for student loans, medical debt, and credit-card debt, among other sources of debt.21 In the interviews, many recipients explained that they used the money to catch up on overdue bills and to pay off debt owed on credit cards and to family members. Recipients often attributed their accumulated debt to challenges brought on by the COVID pandemic.
Bernice, a thirty-seven-year-old African American mother living with her husband and son, told us that she used the funds to start “catching up on some bills, because I’m the extension queen, so [I pay bills in] stages and stages. …” The use of the transfers to pay past-due bills was echoed by many single-mother households. Cherice, a thirty-three-year-old single-mother of three who worked as a cashier, explained that her “main thing is bills.” Similarly, Jamila echoed, “Bills! When it comes, it is time to pay bills.” Ximena said that she used the money to pay “outstanding credit-card bills. … I’ve been able to pay that off and it’s helped my credit score so I’m very happy about that.” LaTasha, a thirty-eight-year-old African American single mother of three, focused first on “the big bills. Then I got car insurance. Then the bills I was short on. I was short on the credit card. Trying to keep afloat. That’s how it goes.”
Recipients also noted that the transfers allowed them to avoid new debt, or break the cycle of old debt. Candice, a thirty-nine-year-old African American single mother of four, said, “I’m grateful I don’t have to do that, or haven’t had to do that in at least a year. That’s stressful when you get into a loan and then it starts just piling up and then the interest rate and it’s like, this loan is not going down. …” Keisha, a financially independent woman living with her mother, two sisters, two brothers, and a niece, described her experience with debt prior to the transfers as “living in a circle”—using her income to pay loans that she took out to pay bills. In addition to paying outstanding bills and credit-card debts, recipients used the transfers to pay off student loans, car-related debt, and loans from friends and family.
Many recipients had savings goals that they were unable to meet due to this debt cycle. The transfers allowed them to establish emergency funds for the first time. However, contributions to these savings accounts were highly contingent, depending on income earned and bills accrued in the period prior to the transfers, and less common among single mothers.
Basics First or “At the end of the month … we don’t have anything … now I can buy groceries.”
Many households with children used the funds primarily to meet their immediate basic needs, such as paying bills and purchasing food, but also to purchase extra or better groceries for themselves and their children. Ericka described using the transfers to buy a bit more at the grocery store: “It’s been a little bit better, because we felt a little bit more relieved. At the grocery, you could buy a little extra because my kids, they eat a lot, you know. … I was able to stock a little bit more on extra snacks and stuff for them.” Renata explained that “it’s been different, you know, because I’m able to afford a little bit more. … I used to buy meat at El Super. It was cheaper and low-quality meat. Now I am able to spend a little bit more to buy meat with more quality, less fat.” The use of the transfers for groceries first was a common theme among the single mothers, as was using the transfers to meet basic needs before considering other expenses. Chantel explained that she uses the transfers for groceries most of the time, “As soon as it hits the card, I go on Walmart to order groceries. If it’s not groceries, it’s like a detergent for me to wash my kids’ clothes and fabric softener.” Cherice emphasized her priorities as follows, “Everything has to go where it has to go. I handle my business first and whatever else I have, me and my kids can do whatever we want to do, but my pledge money is getting spent just as hard as my hardworking money.” While the money might in the end go to meet similar needs, it remains psychologically distinct from “hardworking money.”
Although spending on groceries and basic needs was a prominent theme of the qualitative interviews, when we compare households receiving the transfers to the control group in the quantitative analyses, we see what appears to be a slower increase in spending relative to low baseline levels at the start of the COVID pandemic. This is especially the case among two-parent households and other households with minors, which is consistent with their use of transfers and consumption spending to pay down debt.
Big, Irregular Purchases or “Her going to school—the initial moving in—that’s a big chunk of money.”
A smaller number of recipients used the cash to start emergency funds or to make large purchases that they otherwise would not have been able to afford. For example, several recipients, including Ericka, were able to purchase or pay off debt from the purchase of refrigerators and other appliances, while others used the cash to purchase used cars, make necessary, expensive car or truck repairs, or to renovate their bathrooms or other parts of their homes. For example, Angel, a thirty-two-year-old Hispanic mother of two, was able to buy a Honda Accord with the help of the transfers. “Someone I knew was selling it and he was like ‘you can give me half now and half later’ So I said, I can’t pass on that. The pledge money helped me give him that first half.”
Other recipients put the transfers toward moving. Ximena also used the transfers to purchase a car, “I was able to get a car because mine broke down,” and later was able to use the transfers to pay for her last month’s rent deposit, and also her sister’s, “That was the goal, you know, to put people in a better position, And here we are, two years later, with my own place.” Angel used it to build her credit and take out a mortgage: “Without the credit I had been building, I wouldn’t have been in consideration for getting a loan.” Bernice put it toward savings earmarked for buying a house.
Some recipients invested the transfers in their own or their children’s education. Rosa used it to pay the portion of her tuition that was not covered by financial aid. Candice, a single mother, used it to pay for her daughter’s gymnastics class, “She had to take some time off because it was becoming too much for me to have to pay. She’s been back for probably a year now. [The pledge] gave me a little bit of wiggle room.” Dionne used the pledge to buy her son’s school uniform and to pay for his summer school, as well as a car repair, “I actually got my car fixed to the point where the service engine light is not on.”
In the qualitative interviews, spending on mobility, whether cars or trucks, was particularly salient. Many of our participants have long commutes to work, and Compton has limited access to convenient grocery stores or health facilities—it is a “multiply-deserted” area (Satcher 2022)—and so residents require cars to maintain their jobs and buy affordable groceries. Some respondents also described needing the car to drive their children to school due to concerns about gang-related violence.
Differences in Cash Transfer Spending by Household Composition
In the qualitative interviews, single-parent households were more likely to report spending on groceries and bills, while two-parent homes were more likely to allocate the transfers to larger items, such as car maintenance and furniture, family outings, or savings. More broadly, the evidence from both the qualitative and quantitative data suggests that household needs are multiple and diverse, and one of the important attributes of cash is that it is flexible and can be used for many kinds of expenses and to meet many different needs. Low-income households maintain complex financial portfolios and arrangements to make ends meet (Collins et al. 2009; Shaefer et al. 2015). The characteristics of temporary, unconditional cash transfers may enter into these portfolios in varied ways, and may carry distinct meaning from income or public benefits. While this multiplicity of strategies and allocations is a positive feature of money, it also makes it harder to detect spending on specific uses in quantitative analyses (Duncan et al. 2008).
It’s Not Bills or Treats, but Bills and Treats
One theme that emerged from the interviews that was not evident in the quantitative expenditure analysis was the coupling of bill payments—financial choices that are often seen to be more worthy or necessary—with spending on enjoyment, especially among households with children. As discussed in Kathryn Edin and Laura Lein’s book, Making Ends Meet (1997), providing treats to children may be seen as part of being a good parent (see also Sykes et al. 2015 and Halpern-Meekin et al. 2015). Indeed, we find evidence that it strengthened relationships between parents and children: going out to eat was also a way to spend time together as a family. It may also help recipients make hard choices—for example, reinforcing the allocation of limited resources to pay back large debts—by acting as a reward for doing so. Spending on enjoyment may also confer a sense of dignity and financial freedom in a context that is otherwise full of financial constraints.
Reflecting the complementarity between spending on bills or debts and spending on treats, recipients often referred to the two expenses in a single sentence. Ana, a single mother of three children, recollected that the cash “was a big help because at that time I didn’t have no money. And those $600, when they hit the account, I was able to take my kids out to eat, I was able to catch up on some of my bills.” Juan, a forty-eight-year-old Hispanic father of two who was recently separated and living alone at the time of the interview, used it to “pay rent, and to take my daughters out.” Bernice, who earlier described herself as the “extension queen” in reference to her late bill payments, described using the transfers for “catching up on some bills and, like, going out to eat.” Angel was stressed at the time about American Express debt that she could not get out of, so she decided to use the transfers to pay off that debt. She said, “So I told myself the first thing you’re gonna do, you’re getting rid of that stress. So I paid that off. … So I said, okay, you guys, we’re gonna treat ourselves. So I paid one card and we did Vegas.”
Dionne, who has three children, reflected on the complementarity between bills on the one hand and more fun things on the other. “At first it was more for bills and stuff like that. But then, as it started relieving some of that stress, I could use it for more fun things. I think people sometimes forget that you can pay bills and stuff, but you can also have fun. Not so much fun, that I take my son out instead of paying the bills. I’m not trying to ball out with it, I’m not buying Jordans or anything like that. But to do educational stuff, take him to the museum, take him to have dinner as well. Every time people think about giving you money, it’s always for a bill. …” Recipients without children also showed this pattern, often treating themselves to self-care or clothing as a reward for paying down debt or catching up on bills.
Countering the conservative fear that money from guaranteed income and other unconditional cash transfer programs will be wasted on inessential goods, participants reveal that spending on meals out or other treats is fundamental to the health of their familial relationships and, in a behavioral sense, may function as a reward for making other difficult choices (Thaler and Sunstein 2008), and thus reinforce spending on bills and paying down debt.
Creating Experiences to Build Relationships
For households with children, spending on memorable experiences created exceptions in weekly routines that strengthened the relationships between parents, often mothers, and their children. Recipients reported that their children became more open with them and that they were able to spend quality time with their children during critical ages. For example, Cherice used the money to take her kids out to eat. “We were able to do small, necessary things that make the kids happy, such as stopping at McDonald’s after school, or the donut shop. As a parent, you got kids so you know, the small things make your kids happy … a Happy Meal, different little things, a sweater for school. It makes a difference, it helps my whole family.” Ana also recalled being able to take her children out to eat, “My kids wanted something simple, like McDonald’s, but at that time I didn’t have any money or anything, and my unemployment had gotten denied a couple of times, so I had maxed out. … And so when the money came in, I was just really excited about it. The kids could have some pizza or McDonald’s or something.”
LaTasha described her experiences with her three children, as well as her memory of these experiences as a child: “I try to, because it’s not every month I can take them out on a date. We’ll go to the movies and we’ll go out to the beach and walk along the Long Beach Pier. Give them some time. … I love it. It’s simple stuff. And I was like, they gonna remember that. Just like I remember simple stuff with mine, you know, even with my grandfather, if it was pouring manure out on the grass. … It’s like about that quality time we spend with each other.”
Ximena emphasized the time that the transfers created. “It has given me the ability to have more time on my hands and not have to worry about where I’m gonna get this extra money? Because I’m a mother, you know, so my child requires a lot of time. So with that money, I’m like, alright, I don’t have to work the hours after school when he needs me the most to pick him up, make his lunch, whatever. So I’m actually there for him and it feels a lot better.”
The transfers were used to create new experiences, reduce stress, and facilitate parent-child time, all of which work to support children’s development.
Scarcity and Stability
Recipients attributed decreases in anxiety, stress, and worry to the transfers, which they described as: “something consistent I can depend on,” “a comforting feeling to know it’s gonna come,” and a source of hope. Lucia, a twenty-eight-year-old Hispanic mother of two who lives with her husband in his parents’ home, said, “Now we don’t have to worry as much month to month. We’re more stable in that sense. Before I would constantly have to look at my account and be like, okay, where are we at now? And my husband would always reassure me we’re gonna be fine. And since we’ve gotten this, I honestly feel like now we don’t really talk or argue about things like that, financial stuff.” Chantel, who was struggling to make ends meet with low-paying gig work for herself and her son, said, “If I wasn’t getting it, I just would have gone deeper into a hole. It kept me afloat instead of letting me sink, you know?”
Some recipients also described a sense of freedom and clarity conferred by the stability of the transfers, as well as an aspiration to change their situation—an experience described by Anandi Mani and colleagues (2013) as escaping a “scarcity mindset.” Destiny, a forty-one-year-old African American woman who had been living with her parents for years, described the transfers as creating a “freedom for flow.” “It’s alleviated [stress], also awakened, in a way, some things, freedom for flow. Some, you know, things that have been stagnated. … So for me, in a way it’s kind of like … how do you keep this going for yourself in a way? Or where would you be when this is done? How do I use this money to show me how to make money?” Diego, a twenty-five-year-old Hispanic man living with his parents and siblings, described the impact of the transfers: “When you have money, or have enough money to where you’re able not to worry about it, you’re kinda, living stress-free. But you know, people say that money doesn’t buy you happiness. It’s true. But money does buy you clarity, stability, you know, less stress.” Jamila described how the transfers changed her perspective on financial planning, “It opened my eyes to budgeting and saving my money more, because I had this extra money coming in.” However, several recipients also raised concerns about the end of transfers and needing to “get unused to that dependency because it’s gonna go away.”
Is Cash Experienced Differently as a Flow or a Lump Sum?
The specific impact of the transfers in alleviating stress likely depends on how well they fit a household’s specific needs (Collins et al. 2009; Kansikas et al. 2023). While some households face challenges aggregating sufficient funds for unexpected or large purchases, other households are instead unable to smooth their earnings between payment periods to meet their basic needs. While we did not see strong differences between the groups receiving high- or low-frequency transfers in the quantitative results, our qualitative interviews show that recipients identified savings and financial planning as a benefit of lump-sum transfers, while they pointed to the flexibility and timing with respect to bills, rent, and other basic needs as benefits of the twice-monthly transfers. Receiving the same sum of money in aggregated chunks or as a steady flow altered the aspirations and goals recipients made.
Alejandro, a thirty-four-year-old married father of four who lives in an Accessory Dwelling Unit (ADU) in the back of the house where his parents and siblings live, received the transfers on a quarterly basis. He said, “I knew if I would have got it every month, I wouldn’t be able to save cuz I would think, heck, I’ll get it again next month.” Similarly, Jamila, who received her transfers twice monthly, thought it would be easier to manage quarterly payments, “I think if I were paid every three months, I probably would’ve balanced my money better. Because it would just come in every two weeks, I always had something to pay for.” And LaTasha found the quarterly payments helpful because she didn’t get too used to the payments, “If it was consecutive, like if I had got it every month, I probably would have got a little comfortable.” This sentiment was also echoed by Dionne, “I think getting it every three months, you won’t become dependent on it. Getting it every two weeks, you depend on it. But it depends on who the person is and their goal.”
In contrast, Candice found the quarterly wait too long, “Every three months, that’s a long time to wait. If you don’t work, then every two weeks is better.” Similarly, Ana, who received quarterly payments, thought that “monthly would be easier. I get paid monthly so I already kind of know how to manage the money monthly. It was kind of hard for me to hold on until the next three months.” And Ximena, who also received the transfer quarterly, thought that receiving it at the start of each month would have been most useful because it would have focused the spending on rent and come “when things are the lowest.” Keisha, who received her transfers every two weeks, described the frequency as “Perfect. Because you get two payments. It’s enough to cover a bill or something, but it’s not enough to splurge.”
Despite the varied needs and preferences of different types of households, there has been limited opportunity for households to choose how to structure the benefits they receive since the elimination of the Advance EITC in 2010, which allowed households to receive a portion of the payments throughout the year. Control over when and with what frequency to receive benefits allows households to select the payment frequencies that best fill existing gaps in their economic circumstances or allow them to achieve specific goals (Duncan et al. 2008).
DISCUSSION
Outside of the United States, conditional and unconditional cash transfer programs have been implemented at national scale by federal institutions over long durations (Richterman et al. 2023). CCTs require that recipients fulfill certain conditions, such as enrolling children in school and taking them to health clinics (Attanasio et al. 2015; Baird et al. 2011; Riccio and Miller 2016), but they place no restrictions on how the money is spent.
In contrast, US guaranteed income programs, like the Compton Pledge, are mostly pilot programs or small-scale demonstrations run by municipalities, counties, or local nonprofits. The programs are designed partly to gauge the impacts and feasibility of unconditional cash transfers and to show the potential impacts of larger programs. Unconditional programs, by not overseeing and managing the spending of the poor, have the potential to be both efficient and transformative by allowing families facing different constraints and needs to choose strategies that meet their particular circumstances (van der Naald et al. 2026, this issue). To date, however, these programs are too small and too temporary to make a dent in the failings of federal or state-level policy for the poorest Americans (see Bruch et al. 2026, this issue). Still, they open a window into the possibilities and limits of unconditional cash-based approaches to social provision—ideas that may eventually translate to larger public programs, as they have in other countries, often through the vehicle of a universal child allowance. Unlike the income maintenance programs of the 1970s—which were inspired by the idea of a negative income tax, a system designed to replace most other benefits with a single, simplified cash assistance program—today’s guaranteed income programs are mainly seen as supplements to existing programs rather than substitutes. If the idea of guaranteed income is to be taken further, its relationship to existing programs (such as SSI and SSDI) will have to be worked out.
Entry into the Compton program was decided via randomized assignment among eligible populations, allowing evaluation using an RCT. Stacia West and Amy Castro (2023) find reductions in income volatility, greater financial resilience, and positive impacts on mental well-being in an RCT of the Stockton cash transfer program (n = 331), at least before the COVID pandemic. The Compton Pledge program was introduced during the pandemic and had mixed results. While it led recipients to pay back debt and purchase large durables, there were also important negative treatment effects (on earned income and spending) and no effects on psychological well-being (n = 1074) (Balakrishnan et al. 2025), echoing the findings of other recent pilots (Vivalt et al. 2024). Relative to the often positive and strong results from studies in low-income and middle-income countries (Reynolds et al. 2017), evidence of the impact of cash transfer programs in the US is generally weaker and mixed. This may be, in part, because the cash transfers are smaller as a fraction of income and expenditures and because American households may face different constraints, including high debt burdens and different family structures. Additionally, unconditional cash transfers are fungible and can meet many different needs and wants, diminishing the statistical power to detect impacts on narrowly defined metrics. The qualitative data highlight the importance of certain outcomes seen also in the RCT (such as purchases of durable goods and debt payoff), while also revealing others that are harder to quantify—especially improvements in family and parenting relationships that may have longer-term consequences but which are often secondary in RCT studies (see Krause et al. 2025). These future-focused priorities are consistent with parents’ focus on investments in mobility documented in studies of the EITC and working-family finances (Chaudry 2004; Sykes et al. 2015; and Halpern-Meekin et al. 2015).
A major contribution of Making Ends Meet was to focus on the specific challenges and conditions of single mothers. Our analysis shows that single-parent households (who, in the Compton sample, are nearly all single mothers) face distinct conditions. They tend to have less room to adjust work and childcare arrangements (partly because of work incentives attached to the EITC), and they face tighter budget constraints and have more unmet needs. Consistent with this, we find that the cash transfers led single parents to significantly increase the number of hours worked each week relative to the control group and other households with and without minors. The transfers also led to an increase in income for single-parent households relative to all other household types, who experience instead a decrease in income and expenditures relative to the control group. We also find that different households have different spending needs and priorities. While recipients in dual-parent households pay down a substantial amount of debt and show a significant decrease in their monthly expenditures relative to the control group, single-parent households have less ability to make the necessary spending cuts and end up with more debt. Single-parent households instead take advantage of the cash transfers to accumulate durable goods, purchases that are otherwise difficult given the calls of daily life. These purchases may be investments that lead to longer-term gains, such as securing a vehicle to support work-related activities.
The qualitative interviews reveal an important complementarity between spending on overhanging and immediate needs, and on enjoyment, with the two forms of spending often mentioned back-to-back. Households that allocate limited resources to pay debt, get up to date with bills, or cover immediate needs often pair these expenses with spending on treats and activities, especially for their children. Being able to build positive experiences for children can make it easier to simultaneously undertake the belt-tightening needed to pay overdue bills and repay debt—one type of spending may thus reinforce the other. Additionally, this spending creates durable memories with children that can be revisited in times of relative scarcity.
Overall, the results from Compton and other US programs are mixed. One explanation is that the programs so far have been relatively constrained in terms of the number of recipients and the duration and magnitude of the transfers. As pilots, the US programs described here have features that can add to and subtract from impacts. First, the funds are provided to only a subset of potentially eligible households. This means that general equilibrium effects (for example, price rises when increasing demand for local goods exceeds supply) may not be present in the pilots, but they could be if the programs were implemented at scale (as Egger et al. 2022 show in Kenya; see Jones and Marinescu 2022 for an estimate of the labor market effects of the Alaska Permanent Fund). It also means that recipients may feel stigma or pressure to share the transfers, since the funds only go to selected households (as shown by Clara A. Yoshino and colleagues [2023] in a broad overview); this might be mitigated at scale when many more people receive funds.
Second, the limited duration of the programs (one to three years) also means that money may be perceived and spent differently than if it was guaranteed for a longer duration or institutionalized as an ongoing policy where it would come to be anticipated (as in the discussion of the EITC in Romich and Weisner 2000). A longer duration would also make it easier for recipients to invest over time and take bigger risks.
Third, many of the programs were initiated in response to the COVID pandemic of 2020–2022—a context of sharp economic downturn coupled with newly generous public programs like federal stimulus payments and child tax credits (Parolin et al. 2023; Fulford 2023). As a result, studies that show no impacts of large one-time cash transfers on key economic and psychological outcomes might have produced different outcomes in a different context (Jaroszewicz et al. 2023; Pilkauskas et al. 2022; Jacob et al. 2022). Indeed, the stimulus payments may have obscured differences between the lump-sum and twice-monthly transfer conditions, since households received generous lump-sum transfers from the government during this period. Relatedly, several negative treatment effects (that is, on labor, income, and expenditures) should be understood as a slower increase among treated households relative to control households as they recovered from the most severe impacts of the COVID-19 pandemic, rather than an absolute decline. In particular, the transfers may have allowed less constrained households to take more time to reenter the workforce while continuing to spend time with children, search for better jobs, and address overhanging financial obligations.
Finally, the magnitudes of the transfers in the US pilots vary but tend to be smaller relative to income or expenditure than transfers in low- and middle-income countries. For example, the average transfer in a guaranteed income study in Kenya was equivalent to two years of per capita expenditure (Haushofer and Shapiro 2016), while the Compton transfers corresponded to 21 percent of mean household income.
Despite the mixed and heterogeneous quantitative impacts of cash transfers on different types of households, recipients’ experiences of the cash benefits were largely positive. While economists have focused on labor market participation, income, and consumption effects (for example, Vivalt et al. 2024; Balakrishnan et al. 2025; Bartik et al. 2024), recipients saw a broader range of impacts, including opportunities to strengthen family relationships. The transfers were particularly valuable as extra money that arrived free of existing financial obligations or work requirements. For some recipients, the transfers brought an aspiration to do something different. The most ambitious aspirations were not realized in the course of the two-year program and in a context with overhanging needs (including significant debt and late bills) that crowded out the ability of recipients to save or invest in education, a new apartment, a business, or other goals. But the transfers did confer a sense of freedom, stability, and relief—and gave recipients the opportunity to preserve some of these feelings in the shared experiences and memories created with their children.
FOOTNOTES
↵1. For a discussion of how the safety net has become increasingly focused on working parents with children, see Joseph van der Naald and colleagues (2026, this issue).
↵2. Other programs include the Compton Pledge in Compton, CA (over $500 a month to 700 residents for two years); Stockton Economic Empowerment Demonstration ($500 a month to 125 residents for two years); BREATHE in Los Angeles ($1000 a month to 1000 residents for three years); the Cook County Promise Guaranteed Income Pilot in Chicago ($500 a month to 3,250 families in Cook County for two years); and RISE in Cambridge, MA ($500 a month to 120 single caretakers).
↵3. The poverty rate in Compton was 18 percent in 2022, measured by the federal poverty line, a measure that does not adjust for the relative high cost of living in the city (US Census Bureau, n.d.).
↵4. The absence of pronounced differences between treatment arms in the survey data may be, in part, due to the lump-sum COVID pandemic relief payments, including the expanded Child Tax Credit (CTC) and the stimulus checks, that many households received around the same time (Parolin et al. 2023).
↵5. Apart from the results on labor supply, the early findings indicated an uptick in divorce for recipients, reinforcing Senator Daniel Patrick Moynihan’s disenchantment with the programs (he had helped design them while working for President Richard Nixon), and contributing to the withering of political support (Matthews 2014).
↵6. “Official Poverty Measure,” 2022 US Census Bureau (Shrider and Creamer 2023).
↵7. Transfers did not affect recipients’ eligibility for the EITC, CTC, subsidized health care, the Women, Infants and Children program (WIC), the Low Income Home Energy Assistance Program, and Housing Choice Vouchers, CalWORKS, and CalFresh since they were kept below the federal government’s gift maximum of $15,000 a year and because the city obtained waivers for state-specific benefits.
Beyond these eligibility rules, it is plausible that we might see an impact on benefits receipt if the control group puts relatively more effort into obtaining or maintaining benefits, or the treatment group puts in relatively less effort. We find small and nonsignificant negative effects of treatment on benefits, suggesting that this dynamic, if it did take place, was not common or large (Balakrishnan et al. 2025).
↵8. Specifically, the Coalition for Humane Immigrant Rights of Los Angeles and One Fair Wage.
↵9. To hold the net present value of the transfers constant across the treatment arms, quarterly transfers were disbursed so that the average amount disbursed to each quarterly household at each timepoint equaled the average amount disbursed to each twice-monthly household at that same time point.
↵10. The implementation report of the Compton Pledge was published in May 2023 (Fund for Guaranteed Income, 2023).
↵11. The online supplement, which includes tables S.1–S.6, can be found at https://www.rsfjournal.org/content/12/1/220/tab-supplemental.
↵12. There were 179 households who indicated that minors were present at baseline but did not specify the relationship at midline. These households are categorized as other households with minors. Additionally, there were 21 households that indicated a child was present at the midline but who indicated no minors at baseline. These households are assigned to households without minors.
↵13. We used the US Bureau of Labor Statistics Consumer Expenditure Survey to assign values to the following categories: washing machine, clothes dryer, dishwasher, microwave oven, vacuum cleaner, home entertainment system with television and audio, gaming console, gym equipment, air conditioner, valuable jewelry or watches, musical instruments, power tools, computer or tablet, mobile phone, car or truck, motorcycle, bicycles.
↵14. Note that we consider home equity (value of home minus mortgages) separately.
↵15. In the qualitative sample, 36 percent reported working a full-time job, 36 percent reported working multiple jobs, and 13 percent were unemployed. Thirty-four (60 percent) respondents indicated that they were living in households with minors at baseline, which again matched the overall distribution in the quantitative sample (64 percent with minors), though this number dropped to 32 by the first interview.
↵16. Prior cash transfer studies have shown differential impacts by recipient gender (DeMel et al. 2012; Crosta et al. 2024). In the Compton sample, female recipients are more likely to have children, and are overrepresented in single-parent households.
↵17. It could be that the control group makes additional efforts to obtain other benefits and that the treatment group, while receiving the transfers, makes a relatively smaller effort to obtain benefits. However, we do not find a significant effect of treatment on benefits, including CalWORKS, CalFresh, SNAP, WIC, SSI or Old-Age, Survivors, and Disability Insurance. Instead we find only marginal and nonsignificant negative effects.
↵18. While we do not have estimates of baseline expenditures, and so cannot know whether negative impacts represent absolute declines or slower increases in expenditures, those impacts plausibly follow the pattern observed for income. An overall increase in spending (for both the treatment and control groups) is consistent with depressed spending during the baseline, which coincided with the hardest economic moments of the pandemic.
↵19. For further discussion of saving from anticipated money versus existing money, see Richard Thaler and Shlomo Benartzi (2004).
↵20. We are not sufficiently powered to look at the interaction of the frequency arms with all four household types.
↵21. Note that we report debt in the control group, since we do not have a baseline measure of debt and the control group is likely a better reflection of debt in this population in the absence of the cash transfer program.
- © 2026 Russell Sage Foundation. Constantino, Sara M., Ajay Chaudry, and Jonathan Morduch. 2026. “Guaranteed Income Programs: Single Parents, Spending, and Debt.” RSF: The Russell Sage Foundation Journal of the Social Sciences 12(1): 220–46. https://doi.org/10.7758/RSF.2026.12.1.09. We are grateful to Sidhya Balakrishnan, Sewin Chan, and Johannes Haushofer for their collaboration on the quantitative survey and analyses. In Compton, we are thankful for Mia Hernandez-Perez, Courtnee King, and Kaytlin Ernske’s support in conducting qualitative interviews, both in person and remotely. At NYU, we would like to acknowledge Georgia Assy Hillel, Junyi Ouyang, Julia Raab, Rachael Eplee, Yeji Lee, and Kaytlin Ernske, who provided project management and assistance with transcription and coding. This work benefited from collaboration with the Jain Family Institute, and we are grateful for research support from Ege Aksu, Marcella Cartledge, Roberta Costa, Sara Restrepo, and Yunjie Xie. The qualitative analysis was supported by funding from the Russell Sage Foundation (grant number 2010-28280; ROR: https://ror.org/02yh9se80) and the Mastercard Center for Inclusive Growth. We are also grateful for support from the NYU Financial Access Initiative. The quantitative surveys and analysis received generous support from the Robert Wood Johnson Foundation, J-PAL, and private donors. The RCT is registered in the AEA RCT Registry as AEARCTR-0007621. We thank our implementation partner, the USC Center for Economic and Social Research (CESR), for their work on survey data collection. We are grateful for insightful feedback from Kathy Edin, Elizabeth Ananat, Carolyn Barnes, Sandra Danziger and the other contributors to this issue. Direct correspondence to: Sara Constantino, at sara.constantino{at}gmail.com, 473 Via Ortega, Stanford, CA 94305, United States.
Open Access Policy: RSF: The Russell Sage Foundation Journal of the Social Sciences is an open access journal. This article is published under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
REFERENCES
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.