Wealth Thresholds in College Saving: Evidence from the CollegeBound Boost Guaranteed Income–CDA Trial

• In Event: Designing Early Wealth-Building Policy for Equity, Scale, and Impact

Abstract

This paper uses 24‑month results from the CollegeBound Boost randomized controlled trial in Saint Paul to test whether integrating guaranteed income and baby-bond–style deposits with a universal child development account (CDA) platform increases college saving across the lower and lower-middle wealth distribution. CollegeBound Boost is embedded in CollegeBound Saint Paul, which automatically opens a seeded CDA for every child born in the city; the trial enrolls 1,065 families with incomes at or below 300 percent of the federal poverty guideline and randomly assigns them to CDA-only control, Deposit-only, or GI+Deposit for 24 months. To the authors’ knowledge, nearly all other guaranteed income pilots in the United States limit eligibility to 200 percent of poverty or less and primarily serve families with very low or negative wealth. In contrast, the 300 percent threshold in Saint Paul yields a somewhat higher-wealth sample, with many families holding at least some positive net worth, making it possible to examine how GI+asset policy operates once families are above bare survival but still far from middle-class wealth.The 24‑month CollegeBound Boost results provide a clearer and structurally informative signal for college saving. In the higher-wealth group (net worth ≥ 35,000), 80.8 percent of GI+Deposit households report saving for college compared with 60.2 percent of control households in weighted analyses (p = .0093). Among lower-wealth households (< 35,000), 61.4 percent of treatment households save versus 58.2 percent of controls (p = .4493), a substantively small and statistically null difference. Heterogeneous treatment-effect models show that the GI+Deposit × wealth35k interaction is statistically significant (p = .022), indicating that the same GI+asset dose is functionally more powerful once households cross a structural wealth breakpoint.Within this RCT, guaranteed income and baby-bond–style deposits meaningfully increase college saving only among families with sufficient baseline wealth to respond, suggesting that future guaranteed income policy should move beyond income-only eligibility and explicitly account for wealth when targeting and calibrating benefits.

Author

• William Elliott, University of Michigan
  Presenting Author