Spending, Time, and Social Network Patterns from the Stockton Experiment

Background and Purpose: Persistent inequality and rising levels of income volatility have prompted renewed interest in guaranteed income while challenging entrenched constructions around deservedness and assumptions about low-income households. In response, Stockton, CA was the first city since the 1980s to pilot-test guaranteed income by providing $500 per month for 2 years to 125 randomly selected individuals from census tracts at or below AMI. This paper presents findings from one strand of the larger RCT to answer: How did people use the money? What drives their decision-making? What were the barriers to uptake?

Methods: We utilized a sequential mixed-methods design to understand patterns aggregate spending data masked. The $500 was disbursed on a pre-paid debit card and each month aggregate spending data were categorized into merchant category codes corresponding to the type of transaction. These trend signals informed analysis of ethnographic field notes and narrative data from 50 in-depth interviews with participants in the treatment group. Interviews focused on understanding strategies, processes, and pooling behaviors surrounding how participants made sense of and engaged with the $500. Interviews occurred in participant’s homes or a community-based agency, lasted 1.5-3 hours, were digitally recorded and professionally transcribed. In keeping with iterative MM design, we combined thematic analysis at the semantic level, with grounded theory for latent analysis. All coding occurred in Dedoose. Final codebook included process, values, focus, and theoretical coding.

Findings: The largest spending category was food, followed by Big Box stores, transportation and utilities. 40% of the money was transferred into cash or pre-existing accounts. Narrative data demonstrates that transferring money was driven by preference for existing financial relationships, the need to spread the money around fragile networks, a lack of trust following decades of predatory lending and watching Ontario’s basic income trial end prematurely. Material and immaterial needs within pooling networks, and the ways that financial scarcity generated time scarcity drove how people interacted with the $500. In reciprocal pooling networks, the money spilled over to stabilize food insecurity across households, patch holes in the safety net, functioned as a form of paid care work, and generated more time for relationships and meaningful activities that generated dignity and increased quality of life. Women who ordinarily spent much of their life and time on unpaid care work noted that alleviating financial stress created an infusion of time that allowed them to prioritize themselves and their desires in ways previously ignored.

Conclusion and Implications:

First, these findings indicate that attempts at scaling up guaranteed income requires attention to the lack of trust in communities targeted for risky products, and an understanding of the unpaid care work stressors strain if policymakers wish to avoid gaps in take-up and the possibility of new forms of inequality from those unwilling to engage with systems they do not trust. Second, the data points to a need for understanding how guaranteed income can function alongside other programs, policy, and safety net initiatives that support unpaid care work and alleviate stress within financially strained networks.