In Ohio and across the nation, policymakers are contemplating sizeable increases to public outlays for early childhood programs, including expanded preschool, childcare, and other support services. Polls indicate that early childhood programs enjoy broad support, and proponents of early childhood programs often cite as evidence for expansion the positive, long-run effects of the boutique Perry Preschool program (it served just fifty-eight low-income children during the 1960s). But will greater expenditures in early childhood programs generate big returns? Or could they backfire?
A new study by university researchers Michael Baker, Jonathan Gruber, and Kevin Milligan offers a cautionary tale. They examine the short- and longer-run outcomes of children participating in North America’s largest universal childcare program. Starting in fall 1997, Quebec began offering large public subsidies, open to all parents, for center- or home-based childcare programs serving youngsters up to four years old. As of 2011–12, the program cost $2 billion per year and subsidized roughly 80 percent of a family’s child care costs. Quebec has been the only province to adopt such an expansive childcare policy. For example, from the mid-1990s to 2008, Quebec children in center-based childcare jumped from 10 to 60 percent; during that same period, the rise was just 10 to 20 percent in the rest of Canada.
The study examines both the immediate effects of the program on two- and three-year olds, as well as longer-run impacts for those exposed to the Quebec childcare program. Baker and colleagues examine a range of outcomes, including non-cognitive indicators based on surveys gauging parent perceptions about their child’s behavior; standardized exam results; health outcomes (self-reported in surveys by twelve- to twenty-year-olds); and criminal activity. To conduct the evaluation, they compare the outcomes of children exposed to the Quebec program to Canadian children who were not, while using statistical methods to control for various demographic differences.
First, the short-run findings on children in Quebec’s childcare program. Consistent with a prior analysis in 2005 from this research team, they find significant negative effects of universal childcare on young children’s non-cognitive outcomes in areas such as anxiety and aggression; they also uncover negative impacts on the only cognitive measure available, the Peabody Picture Vocabulary Test. Though not analyzed in this paper, Baker and colleagues cite other work showing that the negative results are driven by children from two-parent families who, on average, fare worse in the childcare program (disadvantaged children from single-parent families see improved outcomes).
Second, the study examines the longer-run outcomes. On non-cognitive measures, the analysts find persistent, negative effects. At ages five through nine, children exposed to the childcare program exhibited more aggression, anxiety, and hyperactivity and they were less likely to get along with their teacher. Meanwhile, on cognitive measures—national math, reading, and science exams given to thirteen- and sixteen-year-olds—the analysts find no clear impacts on test scores (the estimates are negative in all subjects but not statistically significant). However, the analysts do report a positive effect on the international PISA math exam, but no impacts in reading and science. The researchers conclude, “Overall the evidence on the long-run impact of the Quebec Family Plan on test scores is mixed.” As for health and criminal activity outcomes at ages twelve to twenty, the study uncovers negative results. In surveys, those exposed to the Quebec program reported worse physical health and life satisfaction, and an analysis of national crime data concludes that the program led to higher rates of criminal activity (both accusations and convictions).
All told, this research paints a sobering picture about how Quebec’s universal early childhood program has affected children, save perhaps for its least advantaged children who seem to have benefitted. The authors don’t speculate in this report on mechanisms driving the results, though in their previous evaluation they write, “We also find suggestive evidence that families we study became more strained with the introduction of the program. This is manifested in increased aggressiveness and anxiety for the children.”
Early childhood programs that target the neediest children likely make sense—such as Head Start or Ohio’s income-based Early Childhood Education Grant. Yet as this study suggests, policymakers and advocates of universal early childhood programs should also beware of the potential impacts to children when broad-based public subsidies shift the care of our youngest kids away from parents and more to professionals.
Source: Michael Baker, Jonathan Gruber, and Kevin Milligan, “The Long-Run Impacts of a Universal Child Care Program,” American Economic Journal: Economic Policy (forthcoming). An open-access version is available here.
