Limited evidence that a guaranteed basic income improves poverty-related outcomes compared to existing conditional social assistance
Numerous types of guaranteed basic income (GBI) have been tested in high-income countries, yielding limited evidence for each type. Full GBI with gradual withdrawal of benefits (as other income increases) may reduce food insecurity and increase the time youths stay in school after the compulsory age. Supplemental GBI (paid along with existing income assistance programs) may reduce mental distress. GBI with benefit amounts larger than social assistance benefits may improve life satisfaction and subjective financial well-being. The evidence for any type of GBI pertaining to overall health is inconclusive.
What is this review about?
Existing social assistance programs are complex and costly to administer. Recent reforms intended to alleviate poverty by transitioning benefit recipients to paid work through strict requirements and sanctions have not been successful and have sometimes exacerbated poverty and stigma. GBI has been proposed as a simpler solution, but opponents argue that it would be unaffordable (due to benefits being given to more people) and might result in increased poverty for people who are presently eligible for more than one program.
This review examined experiments that intended to predict the impacts of a real world GBI program. We looked at interventions that met our criteria for GBI: unconditional (e.g., not requiring job seeking), paid in cash (e.g., not food vouchers), as well as paid regularly and in fixed or predictable amounts.
Because social assistance programs are intended to alleviate poverty, we looked at any poverty-related outcomes that the GBI experiments examined, including food insecurity, health (mental and physical), and financial, educational, and social impacts.
What is the aim of this review?
This Campbell systematic review synthesizes the findings of twenty-seven studies from high-income countries on the effects of GBI interventions on poverty-related outcomes, compared to “usual care” (including existing social assistance for those who were eligible).
What studies are included?
This review summarizes evidence from 27 studies that analyzed data from 10 experiments carried out between 1968 and 2020 that aimed to explore how GBI would impact people with low incomes. Five experiments took place in the US, two were in Canada, and the rest were in Barcelona (Spain), Finland, and the Netherlands. Nine of the experiments randomly assigned participants to either receive GBI or not (control group) so that observed effects could be attributed to receiving GBI. However, the experiments and the related studies had several other methodological weaknesses, which reduced the quality of the evidence.
What are the main findings of this review?
Analysis of the ten GBI experiments identified four distinct types of GBI that were tested, which formed a framework to categorize and synthesize the findings in the examined studies. A fifth type of GBI, which has been implemented in new experiments that are underway, was also included in the framework to assist in future analyses.
The included studies examine a total of 176 poverty-related outcome variables. Combining quantitative results across studies is only possible if the same variable is being measured. Therefore, most of the findings are synthesized by sorting them according to seven outcome categories and 34 subcategories.
Food insecurity, self-rated mental distress, and post-mandatory school enrollment may be significantly improved when GBI is provided. The evidence on subjective financial well-being and self-rated life satisfaction appears mixed; however, GBI with more generous benefits than existing social assistance appears to yield beneficial results for both of these outcomes.
A supplemental type of GBI, provided in addition to existing social assistance benefits, appears to improve subjective financial well-being, self-rated life satisfaction, and self-rated mental distress.
Receipt of a full GBI (which replaces social assistance benefits) does not appear to improve self-rated overall health.
What are the implications for research and policy?
Although the details of how a full-scale GBI would be implemented have been discussed and debated, the empirical evidence from GBI experiments has been previously assessed as if GBI were a singular type of intervention. The typology developed for this review may assist in the evaluation and synthesis of GBI studies in the future, since numerous new experiments are underway.
The main obstacle to a more complete synthesis in this review was the large number of different outcome measures that studies used. In part, this is because income-support interventions affect many aspects of life (social, economic, health, etc.); however, it may help if researchers use standard validated measures in future studies. At the same time, very broad outcome measures (e.g., general health) may obscure more specific impacts (e.g., on hypertension), so a compromise between general and granular outcomes may provide stronger evidence.
For implementing a full-scale GBI program, a supplemental GBI may be most effective, as well as more prudent, to avoid possible unintended consequences of a broader reform of existing social assistance.
How up-to-date is this review?
Searches for eligible studies were conducted until December 2022. Studies published later would likely include data collected during the COVID-19 pandemic, so they would not be within the scope of this review (on interventions conducted to inform permanent income assistance policies).