Education savings plans,matching contributions,and household financial allocations: Evidence from a Canadian reform

Despite a large literature that investigates the effects of education savings plans on schooling outcomes, much less is known about how parents finance this saving and how it affects other household spending, including spending broadly related to child-rearing. To provide new insights on this issue, this paper estimates the spillover effects of contributions to an education savings plan on household spending and retirement saving behavior in Canada. The analysis uses a cross-sectional, nationally-representative household expenditures survey (N = 28,978) and, in an instrumental variables design, exploits policy-induced variation in education saving resulting from changes in federal matching contributions for low-income and middle-income households. The analysis finds that education saving has little effect on total household expenditures, donations, and home investments, and crowds in retirement saving among households with strictly positive net contributions in both types of saving vehicles. Only out-of-house childcare expenditures decrease as a result of education saving, suggesting that households substitute between educational investments and current spending on other child-rearing commodities. However, the dollar value of this substitution is small. These findings offer new insights into how households finance education saving for their children and suggest that different types of tax-preferred saving vehicles may be complementary rather than substitutes.