Refund to Savings (R2S)

Financial Behaviors

The Refund to Savings (R2S) randomized experiment, begun in 2012, measures the effect of behavioral economics techniques on savings at tax time by building a saving-promotion scheme into the TurboTax Free File Online product that is available free to low- and moderate-income households. 

The tax refund is the largest check most households receive all year. Because the refund is separate from normal household income, it may present a golden moment to build savings. This research draws on lessons from behavioral economics to design defaults and frameworks that could encourage individuals to save more of their incomes. See R2S_Concept_Paper and R2S_FAQs.

A number of topics have been examined as part of R2S, including:

  • The ways in which households prioritize financial resources in an emergency;
  • The factors that predict income volatility in LMI households, and the ways in which these LMI households try to manage their volatility;
  • The role of health insurance in households’ financial lives;
  • The relationship between student debt ownership and household hardship and the ways that this relationship differs across racial and ethnic groups;
  • The mediating relationship of household assets between financial shocks and material hardships;
  • The reasons for alternative financial service usage.

Household Financial Survey

From 2013 to 2019, Washington University researchers conducted the Household Financial Survey (HFS), and administered to a random sample of low- and moderate-income (LMI) households who use TurboTax Freedom Edition to file their taxes. The HFS was national in scope, conducted twice per year, with each year drawing approximately 15,000-30,000 participants from LMI users of Intuit’s tax filing software immediately after they filed their taxes, and about 5,000-10,000 participants six months after tax filing. The surveys included a comprehensive array of questions about household demographics, financial characteristics, ownership of debt and assets, and experience of financial shocks and hardships. Survey responses were merged with administrative tax records. The resulting dataset has been used to study a broad set of issues related to financial security of LMI households including:

  • The issues surrounding financial inclusion in LMI households with people with disabilities.
  • A comprehensive analysis of the Consumer Financial Protection Bureau’s (CFPB) financial well-being scale in an LMI population;
  • An in-depth analysis of tax behaviors, demographic characteristics, financial behaviors, and employment among LMI participants of gig (or sharing) economy;
  • The impact of expansions of state Earned Income Tax Credit (EITC) benefits on material and medical hardships;
  • An analysis of life satisfaction and optimism among LMI households;
  • The issues surrounding financial inclusion in LMI households with people with disabilities.

Beginning in 2018, researchers expanded the scope of R2S experiments beyond tax time savings to explore behavioral approaches to help households better manage their debt at tax time.

R2S launched as a partnership between Washington University, Duke University, and Intuit Inc., the makers of TurboTax.

Funding Partners: Ford Foundation, Annie E. Casey Foundation, Intuit, Inc., The Intuit Financial Freedom Foundation, The Smith Richardson Foundation, JPMorgan Chase Foundation, U.S. Department of the Treasury


Principal Investigators

Michal Grinstein-Weiss

Michal Grinstein-Weiss


Financial Behaviors: Nudges and Habits

Stephen Roll

Stephen Roll

Co-Director of Research and Policy Innovation

Project Director

Yung Chun

Yung Chun

Director of Data Insight