Guaranteed income and health in the United States and Canada: a scoping review

Abstract

Although the economic impact of guaranteed income (GI) (recurring, unconditional, and unrestricted cash transfers intended to supplement the income of participants) is well studied, much less is known about how GI may affect health, especially in the context of high-income countries like the United States and Canada. We searched 5 electronic databases for terms related to “guaranteed income” and “cash transfer” through April 23, 2022. Among 5340 records originally identified, 25 met our inclusion criteria and represented 16 unique GI initiatives. Most included studies used a quantitative approach (n = 22; 88%), were published between 2000 and 2022 (n = 21; 84%), and were conducted in the United States (n = 15; 60%). Health outcomes included maternal and child health (eg, preterm births, breastfeeding initiation), healthcare utilization (eg, hospital admissions), mental health (eg, depression), physical health (eg, body mass index), and behavioral health (eg, substance use). Maternal, infant, and child health were the most highly represented health outcomes. Guaranteed-income initiatives generally had significant positive impacts on health outcomes, especially among the most vulnerable recipients. Data were absent on neighborhood-level health outcomes, chronic and infectious diseases, potential unintended consequences, and long-term impacts of GI on health. Studies on the impact of GI on health suggest GI has the potential to positively affect many, but not all, health outcomes. Rigorous assessment of health outcomes is still needed, and additional health outcomes should be considered in the design and evaluation of GI initiatives.

Introduction

Public support for guaranteed income (GI) initiatives in the United States has increased greatly in recent years: as of 2022, there were more than 100 pilot projects in US cities in 30 states, reaching more than 38 000 people.¹ Guaranteed income is defined as recurring, unconditional cash transfers intended to supplement the income of the general population or low-income populations. Unlike universal basic income, a type of cash transfer aiming to provide enough income to meet basic needs of recipients, GI provides only an income floor and supplements social safety net programs. By providing socioeconomic support, GI helps redress systematic exclusion and the disinvestment of minoritized people and/or those experiencing economic insecurity. In addressing economic challenges, GI has the potential to reduce health disparities and improve health equity²; however, the links between GI and health are not well documented, especially in the context of high-income countries.

A brief history of guaranteed income

Despite a renewed focus on GI, direct cash transfer initiatives are not new. In the 16th century, Thomas Paine argued for a tax to establish a fund to be distributed at the age of 21 years to all citizens, regardless of sex or status.² In the 1960s, Milton Friedman argued for a supplemental income to be distributed to citizens below a certain income level, in the form of a negative income tax³; and Dr. Martin Luther King, Jr., as well as several Black female activists, including Johnnie Tillman,⁴ proposed an annual GI to end poverty and racial economic inequality in the United States.⁵ Aspects of these historic influences became reality as part of 1960s and 1970s anti-poverty programs under President Lyndon B. Johnson.⁶ In these programs, older adults, people with disabilities, and widows were issued governmental financial assistance, whereas single mothers, Black Americans, and the unemployed poor were excluded and often blamed for their own poverty.⁷ Advocates continued to call for GI as a means of broadening supports for people not covered under existing anti-poverty programs. In the wake of persistent poverty, growing attention to structural racism, evidence from international settings, and a showing of political feasibility after the introduction of COVID-19 income supports, GI has received renewed interest in the United States.⁸ Similarly, GI proposals began emerging in Canada during 1960s and 1970s. The Old Age Security’s Guaranteed Income Supplement was introduced to reduce poverty among elders and was approved in 1974)⁹ Other proposed initiatives were aimed at supporting 1-parent families with dependent children. Manitoba’s government was successful in launching a basic income known as the Manitoba Basic Annual Income Experiment (MINCOME).⁹ Although MINCOME did not gain political support to transition from pilot to policy, it has shaped policy for GI in present-day Canada, including the Ontario Basic Income Pilot.

As noted, there is now a wealth of global evidence on the impact of cash transfers on health, particularly conditional cash transfers^10,-14; however, specific evaluation of the influence of GI on health has received limited coverage especially in North America. Our understanding of the impact of GI -like initiatives and health largely comes from the original North American experiments implemented in the 1960s and 1970s: MINCOME, Rural Income Maintenance Experiment (RIME), Seattle-Denver Income Maintenance Experiment (SIME-DIME), Gary Income Maintenance Experiment (hereafter, Gary experiment), and the New Jersey Income Maintenance Experiment (hereafter, New Jersey experiment). These experiments reported a combination of null and positive effects on health, leaving room for further investigation. Positive associations with GI -like initiatives included better nutritional intake, improved birth outcomes, and increased health care utilization.

Because social, economic, and health care landscapes have changed since these initial experiments, the potential impact of current guaranteed-income initiatives on recipients’ health remains unclear and, therefore, warrants more investigation. In this review, we emphasize the United States and Canada to reflect the growing trend of GI pilots across these countries and to inform the movement to establish GI as national policy. Insights from prior studies in these settings can shape future pilots and inform additional health outcomes to be explored for an expanded evidence base on GI and health.

Methods

Study design

We conducted this scoping review in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Figure 1).¹⁵ We used a 5-step approach,^16,17 which included (1) identifying the research question; (2) identifying the relevant studies; (3) study selection; (4) presenting the data; and (5) collating the results. Unlike a systematic review, a scoping review does not assess the quality of the included studies^16,17 but rather is intended to assess the breadth and depth of the spectrum of knowledge in these topical areas.¹⁶ Our overarching research questions were: What health outcomes have (or have not) been studied in relationship to GI initiatives in the United States and Canada? What is the nature of the relationship between health and GI in the North American high-income context?

Definitions

We defined GI as continuous (over time), unconditional (not conditional on specific behaviors), and unrestricted (recipients’ use of income was not prescribed) cash transfers of supplemental amounts to individuals or households.¹⁸ We consider this distinct from universal basic income, in which, although both are types of unconditional cash transfer programs, GI is offered to populations with specific needs (ie, not universal to everyone), and the supplemental amount offered may or may not be enough to cover a recipient’s full amount of need. To account for inconsistent use of terminology related to GI in the literature, we considered other cash transfers that had minimal or no eligibility requirements, including unconditional cash transfers, income maintenance experiments (eg, RIME), and casino disbursements. The RIME, SIME-DIME, and Gary and New Jersey experiments were negative income tax (NIT) programs that provided recurring, unrestricted cash transfers of variable amounts; however, a participant’s cash transfer was at risk if they surpassed income thresholds. Although these programs do not perfectly fit the criteria of present-day GI programs, they were included in our review as sentinel initiatives of their time that also evaluated health outcomes and that inform present-day GI demonstrations. Table 1 provides definitions that discern several cash transfer initiatives discussed here.

We defined health outcomes in terms of physical, mental, and behavioral health; environmental or community health; and health care. Physical health conditions included disability, death, health status (eg, nutrition, self-reported health status), maternal and child health; mental or behavioral health included psychological well-being, mental health, depression, anxiety, resilience and well-being, amd treatment of or support for recovery in mental and substance use disorders; environmental health included community violence, improvement in environmental conditions, occupational health (indicators of health and safety or satisfaction in the workplace); and the health care outcome included health care utilization, access to health care, health insurance, health-seeking behaviors, and quality of care.

Inclusion and exclusion criteria

Studies were eligible for inclusion if they met the following criteria: (1) published in a peer-reviewed journal, gray literature, or scientific conference proceedings and abstracts; (2) published or presented before April 23, 2022; (3) GI was the primary or substantive focus of the article; (4) program or intervention took place in the United States or Canada; (5) health outcomes were measured qualitatively or quantitatively and reported as a primary outcome of the study. All populations were included. We excluded studies that were interventions or programs related to universal basic income (UBI), conditional cash transfers, earned income tax credits, 1-time payments, and studies for which the outcome was unrelated to health or was a social determinant of health (eg, housing, transportation).

Search strategy and screening process

We searched 5 electronic databases—PubMed, the Cumulative Index to Nursing and Allied Health Literature (CINAHL), Web of Science, EconLit, and Scopus—from 1976 (earliest records in databases) through April 23, 2022. Search terms included “guaranteed income”; “cash transfer”, “universal income guaranteed”; “universal allowance”; “supplemental income”; “income maintenance” or “income support.” To capture literature referencing a broad range of health outcomes, we did not specify any health outcomes in the search string; rather, we limited our search to health-focused databases to capture all health outcomes that emerged.

We conducted a hand search of guaranteed-income organization websites and secondary searching of included articles and relevant reviews for additional studies that met the inclusion criteria among the following organizations: Jain Family Institute, Economic Security Project, Mayors for a Guaranteed Income, Stanford Basic Income Lab, and Center for Guaranteed Income Research. Articles were screened based on relevance to the topic. After initial title and abstract screening, full-text articles were obtained of all potential studies. Reviewers independently assessed all full-text articles for study inclusion eligibility and resolved differences through consensus.

Data extraction and analysis

Four reviewers independently used a standardized data abstraction form to capture information on guaranteed-income program characteristics, study design, study population, sample size, and results related to health outcomes for each study.

Results

We retrieved 5340 records via electronic databases (n = 5313) and hand-searched records (n = 27). After de-duplication, we screened 2799 titles and abstracts for inclusion criteria.We excluded a total of 2648 records because they did not meet the study’s inclusion criteria, leaving 151 records for full-text review. Of these, 123 were excluded during the full-text screening phase: 20 were not related to GI as defined in this study; 25 did not have a substantive focus on health outcomes, 79 did not qualitatively or qualitatively measure health outcomes or report health outcomes as a primary outcome of the study, and 2 were excluded because a full-text record was unavailable. Thus, 25 records met our inclusion criteria and were included in the analysis (Table S1).

Characteristics of included studies

Table 2 presents summary characteristics of the 25 included records. The majority (84.0%) were published between 2000 and 2022. Of the 25 records included, 22 used a quantitative approach, 1 used qualitative approach, and 2 used a mixed-methods approach. Of the studies that applied quantitative approaches, 5 were randomized control trials (RCTs), 9 were quasi-experimental, and 6 were cohort studies. Most studies (60.0%) involved guaranteed-income initiatives in the United States. A total of 16 unique GI programs were represented. Health outcomes assessed included maternal and child health (eg, low-birth-weight births, preterm births, breastfeeding initiation), health care utilization (eg, hospital admissions), physical health (eg, body mass index [BMI]), mental health (eg, depression), and behavioral health (eg, substance use).

Guaranteed-income program characteristics

Table 4 shows the characteristics of guaranteed income pilots and experiments included in this review. The earliest guaranteed-income study with published health results was the Mothers Pension’s Program (1911-1931). Coverage of guaranteed-income programs ranged from state-wide (eg, Alaska Permanent Fund, AK, United States)¹⁹ to city-wide (eg, MINCOME in Dauphin, Manitoba, Canada)²⁰ to a narrow selection of individuals or households within a specific area of a city (eg, Stockton Economic Empowerment Demonstration [SEED] in Stockton, CA, United States).²¹ Most programs disbursed monthly income supplements ranging from US$300 to $1000 in US-based studies and from CAN$81.41 to CAN$764.40 in Canadian studies (Tables 3–5). For example, mothers enrolled in the Baby’s First Years program receive $333/month for the 52-month duration of the program. Distribution amounts ranged from CAN$81/month received by Healthy Baby Prenatal Benefit (HBPB) participants²² to US$9000 in 1 year from the Alaska Permanent Fund.¹⁹ Four programs used an NIT approach: SIME/DIME, RIME, the Gary experiment, and the New Jersey experiment. Two programs distributed a percentage of revenue: 1 distributed casino dividends to Native Americans residing on the reservation where the casino was built, and the other distributed a percentage of revenue from an oil pipeline to Alaska residents. A large proportion of programs directed guaranteed-income benefits to pregnant women (eg, Magnolia Mother’s Trust), mothers with young children (eg, Mother’s Pension Program), and low-income families (eg, income maintenance experiments in the United States and Manitoba in Canada).

Guaranteed income and health outcomes

Adult mental health

Three studies assessed the association between receipt of GI and adult mental health.^21,23,24 For SIME/DIME, the association with health outcomes including mental health were inconclusive because of poor measurement or study design.²⁴ In contrast, the SEED pilot reported improvements in measures of recipients’ anxiety, depression, and emotional well-being that were partially attributed to a decrease in financial strain.²¹

Adult physical health

Six studies assessed outcomes related to adult physical health.^21,24,-28 Findings were consistently null among the 3 NIT experiments that reported on adult health outcomes, including workdays lost to illness,^24,25 nutritional intake,²⁶ and prevalence of chronic conditions that limited daily living activities.²⁴

The SEED pilot used the Short Form Health Survey 36 to evaluate overall health and well-being. After 1 year, the treatment group outpaced the control group in the domains of energy/fatigue, emotional well-being, and pain.²¹ In a study of elderly Supplemental Security Income (SSI) recipients, a decrease in mobility limitations was observed after a $100 increase in the maximum monthly SSI benefit.²⁸ Among parents receiving the Universal Child Care Benefit, decreased prevalence of overweight and obesity was observed among mothers, but not fathers, of young children.²⁷

Maternal health and birth outcomes

Twelve studies assessed outcomes related to birth outcomes and/or maternal health^{19,22,24,29,-37} and these were mainly from studies providing cash transfers to pregnant women, such as HBPB, Magnolia Mothers Trust, or Baby’s First Years. Several studies reported significant decreases in low-birth-weight births^{22,31,32,34,35} and preterm births^22,34 among babies born to mothers receiving GI. The Gary experiment attributed the better birth outcomes to greater quality of maternal nutritional intake.³¹

Recent evidence suggests that GI may be associated with stronger cognitive development in infants. Babies born to pregnant individuals who received guaranteed-income payments from Baby’s First Years had greater brain activity in mid- and high-frequency brain waves compared with babies born to nonrecipient pregnant individuals.³⁰

Two studies reported increases in short-term fertility among women in beneficiary households receiving Alaska Permanent Fund payments, especially for women with more socioeconomic challenges.^19,29 There was no association observed with fertility among recipients of MINCOME payments or SSI.^20,36 Of 3 other studies assessing maternal health and birth outcomes, 2 found no association with maternal health^36,37 and 1 found no association with birth outcomes.²⁰

Child health

Eleven studies assessed outcomes related to child health,^{20,23,-25,35,-41} with most reporting positive child health outcomes and several offering data on the long-term impacts of GI. In the historic Mother’s Pension Program, male children of women with low income were followed from birth to death. Children of recipient mothers were less likely to be underweight and lived approximately 1 year longer than children of nonrecipient mothers with similar characteristics.³⁸ Among children with disabilities receiving SSI, researchers observed delayed incidence of new chronic conditions and fewer new chronic conditions through age 3 years.³⁶ Two studies assessed the use of preventive care in evaluating the impact of HBPB among indigenous populations and found increased likelihood of recipient children receiving early childhood vaccinations.^35,41

Three studies assessed health outcomes among Native American children in households receiving casino disbursements.^23,40,42 Native American children in recipient households reported lower prevalence of mental health and substance use disorders compared with non–Native American children in nonrecipient households, and this pattern continued into adulthood. Among children in households receiving casino disbursements, 1 study reported a decrease in behavioral problems (eg, conduct and oppositional defiant disorders) but no change in anxiety and depression.⁴⁰ Another study reported higher BMIs among children from the poorest households compared with children from wealthier households.⁴²

Similar to findings from casino disbursement studies, better behavioral outcomes were observed among UCCB recipients; however, this association was only observed among girls.³⁷ There was no association of the UCCB benefit with children’s general health, hyperactivity and emotional disorder scores, or physical aggression.³⁷ The SIME/DIME and the New Jersey experiment reported no association between cash transfer receipt and child health outcomes.^24,25

Health care access and utilization

Outcomes varied among the 5 studies that assessed outcomes related to health care access and utilization.^{20,25,36,43,44} The New Jersey experiment reported no associations with the number of days in a hospital or the number of doctor’s visits among heads of household participating in the program.²⁶ Among mothers participating in Magnolia’s Mothers Trust, a 25% increase in health insurance coverage was observed.⁴⁴ Recipients of Alberta Income Support, a province-specific cash transfer similar to unemployment, reported more hospital admissions for diabetes-related care, primary care physician visits, internal medicine, or endocrinology visits but lower likelihood of diabetes or retinal screenings.⁴³ Among families with disabled children receiving SSI, each US$1 increase in supplemental income was associated with US$3 less Medicaid spending.³⁶

Other health-related outcomes

Four studies measured other health-related outcomes, including better self-rated health⁴⁵ and self-care. Among mothers receiving UCCB, researchers observed more consistent parenting, but this association was restricted to mothers classified as having low education (high school diploma or less).³⁷ Pregnant women who received HBPB reported being better able to prepare for their babies. Both HBPB and Magnolia’s Mothers Trust recipients reported being able to engage in self-care.^33,46  Table 6 synthesizes all health findings from the guaranteed-income studies in the United States and Canada from 1976 to 2022.

In Figure 2, we summarize the various relationships examined and offer an overview of the expected direction of relationships between GI and health outcomes based on the included records in this review. This visualization and forthcoming framework analysis provides associations of GI on proximal and distal determinants of health.

Discussion

This scoping review examining the association between GI and health outcomes in the United States and Canada presents a wide range of health outcomes and describes substantial heterogeneity in implementation and evaluation of GI programs. Maternal, newborn, and child health were the most highly represented health outcomes, and pregnant women, mothers, newborns, and young children were also the most common target populations for guaranteed-income initiatives, including Baby’s First Years, HBPB, Magnolia Mothers Trust, Mother’s Pension Program, and the Universal Child Care Benefit. Guaranteed-income initiatives generally had a significant positive impact on health outcomes in these populations, with the largest impact seen among the most vulnerable recipients.

The impact of GI on other populations and health outcomes is less clear. Because our current understanding of the relationship between GI and health is largely informed by studies among pregnant women, we need to carefully consider the generalizability of findings to other populations. Furthermore, most guaranteed-income experiments have not been designed to focus on health outcomes and, therefore, health has not been rigorously studied. For example, the most influential NIT experiments reported either largely null health outcome results attributed to poor measurement and study design (SIME/DIME)²⁴ or did not report any health outcomes apart from nutritional intake (RIME).²⁶

This scoping review provides insight into potential mechanisms through which GI might operate on health and social determinants of health. One of the most cited mechanisms was that GI alleviated adversity and stress associated with financial strain, which led to better mental health outcomes.²¹ Forget²⁰ suggested that the improvement in accident and injury rates reported by MINCOME was related to the alleviation of stress induced by income insecurity, illustrating how GI could affect mental and physical health. Several studies proposed that better birth outcomes could be attributed to the ability to consume healthier foods during pregnancy because of the cash transfers,^22,27,33 another pathway through which GI could influence physical health. For infants and children in households receiving cash transfers, avoidance of maternal stress during sensitive development periods for newborns and early-life shocks in childhood could influence cognitive, behavioral, and physiological, and development. Findings reported by Aizer et al.,³⁸ who studied the Mother’s Pension Program, and Ko et al.,³⁶ who studied SSI, are supported by life course theory, which posits that there may be critical or sensitive periods during an individual’s development that can shape health outcomes later in life.^47,48 Cowan and Douds^19,30 proposed that cash transfers provided by the Alaska Permanent Fund Dividend alleviated economic barriers to reproductive autonomy, which could explain fertility increases. Similarly, autonomy and freedom from “forced vulnerability” and “time scarcity” were cited as mechanisms for improved well-being in the SEED.²¹

This scoping review suggests areas for further examination of the relationship between GI and health. Recent guaranteed-income pilot programs have demonstrated more robust investigation of health outcomes. For example, Baby’s First Years measured infant brain wave activity to directly assess cognitive function in treatment and control groups.³⁰ The SEED used qualitative methods to explore how and why health outcomes were affected and validated measures to examine anxiety and depression.²¹ Health outcomes that have not been rigorously studied but are plausibly linked to GI include neighborhood-level health outcomes; long-term outcomes such as development of chronic diseases, incidence of infectious disease such as COVID-19, and sexually transmitted infections; and other potential unintended consequences on health. Notably, the differential impacts of GI on health outcomes of specific sub-populations warrant further examination, particularly by recipient age, education attainment, position in the household, housing status, social isolation, immigration, carceral, and employment status. Research on GI should prioritize the study of health outcomes, using rigorous quantitative and qualitative methods that examine both the extent to which health is affected and potential mechanisms of action. More advanced statistical methods, such as mediation analysis, can be used to examine mechanisms of action and the extent to which social determinants of health mediate relationships between GI and health.

Consideration of study designs is also critical in accumulating unbiased evidence for the impact of GI on health outcomes. Randomized controlled trials (RCTs) are the gold standard study design in evidence-based research because of their ability to achieve exchangeability from balancing measured and unmeasured confounders.⁴⁹ Similarly, quasi-experimental designs are emerging as an alternative to RCTs. Both RCTs and quasi-experimental designs reflect a substantial proportion of guaranteed-income demonstrations in our review and current directions of most programs. However, RCTs and quasi-experimental studies are resource intensive and often require a treatment and control group.⁵⁰ Given the promising results of guaranteed-income programs, having a no treatment control group is approaching unethical by principles of equipoise—the genuine uncertainty regarding a comparative treatments superiority—and should be reexamined to provide equitable opportunity in future experiments.⁵¹ Future guaranteed-income demonstrations could apply innovative study designs, such as a stepped-wedge design with sequential crossover of control participants to receive the intervention, to assess unbiased and precise causal effects while minimizing resources and ethical concerns^50,52,53.

Strengths and limitations

This scoping review, to our knowledgem is the first to synthesize the literature on the association between GI and health, as distinct from basic income, and with a focus on high-income contexts of the United States and Canada. Although previous reviews looked at basic income,^10,13 our focus on GI may better inform the landscape of cash transfers as they have been implemented in the United States and Canada. Given the increasing popularity of GI in North America, this study is timely and can inform future implementation and evaluation of guaranteed-income initiatives to maximize positive effects on health outcomes.

This study has limitations. Some of the cash transfers examined in this analysis did not meet all the required components of GI. For clarity, we describe each cash transfer program in detail in Tables 3–5. Inconsistent usage of the term “guaranteed income” may have led to the exclusion of relevant cash transfer programs; however, the authors consulted experts in the field to develop a comprehensive search strategy designed to capture cash transfer programs with GI components. Furthermore, as with other reviews, our study may be affected by selective publication and publication bias. To address this risk, we included unpublished references, including white papers, in our search, which reported negative and null findings (eg, MINCOME results) and may be less subject to publication bias. Publications of even null findings could prevent bias and increase precision about what health outcomes GI can and cannot affect.

Because of the heterogeneity in program structure, study design, and outcomes evaluated in included studies, a meta-analysis to quantify the effect of GI was not possible. Heterogeneity in GI initiatives will be a challenge and present opportunities for future analyses. As more GI programs come on board, leaders would be advised to replicate the amount, frequency, andduration of income distributions with similar study populations (eg, programs focusing on maternal health) to deepen evidence of effect and mechanisms. Studies evaluating health and other outcomes should also consider specific characteristics of study participants and their community context, including social (eg, social isolation), mental or physical health (eg, aging), and environments (eg, structural vulnerabilities), to understand if an intervention has differential impact on subpopulations. Additionally, a meta-analysis can be possible in a future study with more replication of program structures, designs, outcome definitions, and measurement in GI initiatives.

Despite an assessment of the quality of studies or selection bias that a systemic review would provide, this scoping review helps us understand the breadth of health outcomes that have been studied and where the gaps are, and can set a foundation for developing a consistent set of health outcomes that might be assessed across the ongoing and upcoming GI initiatives.

Conclusion

This scoping review synthesized the extant literature on the associations and mechanisms through which GI may be associated with health outcomes. These initiatives can affect health directly and through important social determinants of health. Rigorous assessment of health outcomes associated with GI are still needed. Potential additional areas of study should explore neighborhood-level health outcomes, chronic and infectious diseases, the potential unintended consequences on health, as well as long-term impacts of GI.

Acknowledgments

We thank the Johns Hopkins Urban Health Institute and Robert Wood Johnson Foundation for funding this project. We also thank the Johns Hopkins SOURCE for providing research support for our team. In addition, we would like to extend gratitude to the Baltimore City Guaranteed Income Steering Committee, whose interest in advancing research on health and guaranteed income inspired our manuscript.

Funding

This project was supported through a grant from the Johns Hopkins Urban Health Institute, and the Robert Wood Johnson Foundation Interdisciplinary Research Leaders grant 79125. Johns Hopkins SOURCE supported for Dr. Nishimura's work.

Conflict of interest

None declared.

Supplementary material

Supplementary material is available at the American Journal of Epidemiologic Reviews online.

References

Author notes