Does Savings Affect Participation in the Gig Economy? Evidence from a Tax Refund Field Experiment

Abstract

This paper investigates how saving the federal tax refund affects gig economy participation for
low-income online tax filers in the six months following tax filing. Using longitudinal survey
and administrative data, we leverage random assignment in a unique refund savings experiment
as an instrument for refund savings. We find significant heterogeneity in estimated effects that
are consistent with life cycle models on consumption and savings. Specifically, refund savings
reduced the likelihood of low-income students working in the gig economy, but increased the
likelihood of more economically vulnerable households working in the gig economy.