In a new Brookings Institution paper, Grover J. “Russ” Whitehurst and Ellie Klein claim that the Obama administration’s proposal for a new federal universal preschool program significantly overstates how much it would cost to enable all four-year-olds from low- and moderate-income families to attend free pre-K.

Titled “Do we already have universal pre-school?” the report argues that more young children already attend early childhood programs than the administration claims and that a more reasonable estimate of the cost per child for that smaller pool of beneficiaries would yield a substantially lower price tag than the administration is proposing. Instead of the $66 billion in additional federal spending over ten years in the Obama plan, Whitehurst and Klein conclude that the same goal could be achieved with no more than $40 billion in federal outlays. Setting aside whether a $2.6 billion annual gap in federal spending between the administration and Brookings estimates is sizeable enough to warrant the degree of consternation expressed by Whitehurst and Klein, the paper omits a crucial consideration and makes several claims that require a response.

The word “quality” appears just once in the Brookings report with respect to preschool, and only when it quotes the first sentence of the White House webpage describing Obama’s policies on early childhood: “Expanding access to high quality early childhood education is among the smartest investments that we can make.” The Brookings authors highlighted the words “access” and “investments,” which they take to be the two primary considerations in pre-K policy. But almost all scholars and advocates steeped in the topic consider the fourth and fifth words of that sentence—“high quality”—to be equally important. Research has demonstrated that the quality of early education influences the academic and social development of children. Furthermore, high-quality programs have been demonstrated to save public education dollars over the long run. But high-quality preschool is more expensive—and much less readily available—than low-quality preschool. That’s a basic, well-documented reality that Whitehurst and Klein never acknowledge.

Omitting the central topic of quality skews any analysis of government costs toward low-ball figures and poor policy outcomes. Whitehurst himself has long emphasized the central importance of focusing on the effectiveness of public programs in assessing whether taxpayers are receiving their money’s worth. But the implicit assumption in this paper is that any pre-K program that exists is good enough. The Brookings authors note that Florida’s Voluntary Pre-K program, for example, is at the low end of the cost spectrum nationwide at just over $2,000 per child, without mentioning research highlighting the poor quality of the services provided through the program.

The figure Whitehurst and Klein chose as the average cost per child of universal half-day pre-K is $5,500, which they based on half the current average expenditure nationally for K–12 students in full-day schools. But why should the pre-K program be limited to half days? The vast majority of low-income and working parents are employed in full-time jobs that don’t conclude at lunchtime. Indeed, single parents receiving public assistance are required to hold full-time jobs that leave them unable to retrieve children from half-day programs. The Brookings authors say that it would be “a politically improbable leapfrog from a situation in which a significant proportion of families have no access to a situation in which all those families have access to high cost full-day programs.” A paper that is focused on purportedly overstated costs in Obama’s proposal cannot simply wave away some of the price because the authors think it unlikely. Furthermore, if they are to make a case for fewer hours, they should wrestle more directly with the greatly diminished benefits of half-day rather than full-day services.

The authors also take the Obama administration to task for highlighting pre-K enrollment statistics covering both 3- and 4-year-olds rather than just the latter age group. Because Preschool for All would only extend a year before kindergarten rather than two, Whitehurst and Klein argue that the administration is providing a worse portrait of the status quo in order to boost political support for its plan. But 3-year-olds, especially from low-income families, would also benefit in a variety of ways from high-quality preschool. Their relatively low levels of enrollment are a genuine problem that provides important context for extending pre-K services as broadly as possible as early as possible.

Whitehurst and Klein write: “Combining three- and four-year-olds to estimate pre-school uptake is akin to combining 17- and 18-year-olds to estimate college enrollment rates.” Given that virtually no one believes it would be desirable for most 17-year-olds to attend college, while research clearly supports the benefits of quality preschool to 3-year-olds, that analogy is simply silly.

Finally, the Brookings authors raise a number of concerns about the data the Obama administration and others have used to estimate pre-K enrollment levels. But even according to their own calculations, they find that only 50 percent of 4-year-olds from the lowest socioeconomic quintile participated in pre-school. That conclusion seems dissonant with a paper titled “Do we already have universal preschool?” No, we do not.

Ten-year cost estimates of any major policy proposal inevitably entail imprecise assumptions and a multitude of uncertainties, so they are fair game for debate. But it is also essential to carefully weigh the potential benefits of proposals, based on the best research available examining comparable smaller-scale programs. The impact on the quality of services children receive from curtailing spending below Obama’s proposed level deserves at least as much scrutiny as the dollar amounts themselves.