Reflections coming out of the recent AEI forum debating pre-K

On Wednesday, February 17, I participated in a forum at the American Enterprise Institute. The forum, organized by AEI Research Fellow Katharine Stevens, was entitled “Does pre-K work? A look at the research.”

Forum participants, in addition to me, were Bill Gormley of Georgetown, Russ Whitehurst of the Brookings Institution, and Dale Farran of Vanderbilt University. Professor Gormley is well-known for his years of research on Tulsa’s pre-K program, which finds that this program has high benefits relative to costs. Dr. Whitehurst is a prominent critic of pre-K research, arguing that the evidence is fragile. (For a response to Whitehurst’s previous comments on pre-K research, see my previous post.)  Professor Farran is a lead researcher on the Tennessee pre-K experiment, which has attracted attention because the results suggest fade-out of test score effects.

Readers can view the video here.  Bill Gormley argued that there is much good evidence that pre-K can work. Russ Whitehurst argued that pre-K has been oversold as a magic bullet. Dale Farran argued that many of the state pre-K programs are not as high quality as model programs, and that we need to know more about what determine pre-K quality.

I wonder if all viewers of the video will understand that the general research consensus is that pre-K can work. For example, see this recent research consensus letter, signed by over 500 researchers.

My main argument at the forum was that we know enough to move forward with a significant expansion of pre-K, but we should provide the funding and services needed to make that pre-K high-quality. We should err on the side of overinvesting in quality, while answering the many remaining research questions by evaluating alternative program designs.

Policymakers’ decisions about pre-K expansion are subject to two types of errors.  The first error is to expand a low-quality pre-K program, whose few benefits are outweighed by costs. The second error is to NOT expand a high-quality pre-K program, which could have helped many children attain greater adult success.

Pre-K is not a magic bullet that solves all problems of income inequality, but there is good evidence that high-quality pre-K can increase adult earnings by 10% or more. This does not solve all social inequities, but it makes progress at a high benefit-cost ratio.  This argument is documented in my recent book, From Preschool to Prosperity, available for free download at the Upjohn Institute website, or for the big spenders, for $0.99 as a Kindle book.

The real research issue for pre-K is not whether pre-K can work – we know it can – but how to design a pre-K system so that we can learn more about what best makes for high-quality pre-K.

We need to expand pre-K towards universal access for all children at age 4. That system should over-invest in quality features such as paying enough to hire and retain quality teachers, and providing teacher coaching services.  The benefit-cost ratio for even small improvements in quality is huge. If we get a better teacher, this benefits many kids in each class affected by that teacher, and many classes over time.

Because the benefits of quality improvements are large, we need to evaluate experiments with many different designs of pre-K, which explore different ways of hiring and training teachers, and different approaches to delivering pre-K.

State and local governments have considerable incentives to expand pre-K on their own, because it will boost growth and help many families.  But because quality is hard to measure, state and local governments lack sufficient incentive to invest in high-quality. In addition, state and local governments lack sufficient incentive to invest in evaluation, as evaluation’s benefits are national. Therefore, we need local initiatives, but combined with quality and evaluation being supported at the national level, by some combination of the federal government and national foundations.

About timbartik

Tim Bartik is a senior economist at the Upjohn Institute for Employment Research, a non-profit and non-partisan research organization in Kalamazoo, Michigan. His research specializes in state and local economic development policies and local labor markets.
This entry was posted in Early childhood program design issues, Early childhood programs, National vs. state vs. local, Uncategorized. Bookmark the permalink.