Economist Art Rolnick's research in which he predicts sizable long-term benefits for individuals and society if early childhood education programs are significantly expanded, reinforced by the Star Tribune's recent full-court press on behalf of his idea, is persuasive. Frankly, his argument is more convincing than I thought it would be. I say this, in part, because Rolnick, who directs research for the Federal Reserve Bank of Minneapolis, has been intellectually honest in acknowledging limitations in early childhood programs and in a key study about one in particular.
For example, in a seminal paper released in March, Rolnick and coauthor Rob Grunewald, also of the Federal Reserve, note that "graduates" of early childhood programs have routinely lost their "advantage in IQ scores over nonparticipants within a few years of completing" their respective programs. The same deteriorating dynamic has frequently held, they report, when it comes to "improvements in cognitive test scores."
Translated, this means that despite the huge amount of money spent since the 1960s in preparing children for kindergarten and further learning, graduates of Head Start and similar programs have tended to lose, by the middle of elementary school, any scholastic progress they may have made because of earlier enrichments.
Likewise, Rolnick and Grunewald, in writing about the famous Perry Preschool program in Ypsilanti, Mich., cite just how few graduates of that enterprise were studied. The Perry findings, it's essential to recognize, have been viewed by many as proof positive that early childhood education works wonderfully. But how many former students, exactly, from the celebrated class of 1962 were followed over the years? The answer is fewer than 60 (in addition to a similar number of now-adults from the original control group). As sample sizes go, especially for a study that has sculpted so many assumptions and convictions, this is an absurdly small number.
This is even more the case given that when the Perry kids were 19 years old (Rolnick and Grunewald don't mention these data, which were reported in Commentary magazine in 1992), 33 percent had dropped out of high school, 41 percent were unemployed, 31 percent had been arrested, and the 25 young women in the group had experienced 17 pregnancies. Yet because the young adults in the control group somehow fared even worse, Perry has been deemed a success ever since. Comparatively speaking, I guess it was (sigh). Rolnick and Grunewald talk of Perry's contribution to "growth in noncognitive areas involving social-emotional functioning."
Still, there is something to be said for comparative, albeit severely dampened, victories if the alternative is legions of low-income children and adults continuing to do as poorly as they currently are. So with caveat piled upon caveat, I'm willing to agree with Rolnick that spending more money on quality early childhood education likely would be a good investment. (I say this, please note, while leaving for another day critical questions about how "quality" might come to be defined, compounded by worries that a too-limited definition would smother flexibility and innovation by further emboldening licensing and other governmental bureaucracies, while at the same time increasing teacher union membership and clout. Not a pretty or helpful combination. Also for another day is a conversation about how the breakdown of marriage is the biggest reason why so many early childhood programs are needed in the first place.)
So willing, in fact, am I to pay attention when a first-rate economist like Rolnick fixes on a problem, I got to thinking about other areas in education in which his powerful methodology might be useful. Staying with the imperative of better serving low-income children, what if he investigated the likely personal and societal benefits of school choice programs?
Keep in mind here that, as opposed to bigger and therefore more expensive early childhood programs, providing more children with a chance to attend private schools could wind up costing taxpayers less money per child rather than more, insofar as private school costs are generally lower than those of public schools. In other words, at issue is not necessarily the return on "investing" more money overall, but rather the return on each dollar spent on vouchers or tax credits compared with the return on dollars spent under current arrangements. This holds whether eventual spending on K-12 education goes up, down or sideways.
To help get started, Rolnick and his colleagues might want to dig deeper into this summing up of school choice research, in 2001, by political scientist Jay Greene:
"In the last few years, there have been seven analyses of random-assignment school choice experiments [for low-income students], from five different programs, conducted by several different researchers. Every one of those analyses finds statistically significant benefits from school choice for those who are provided with opportunities to choose a private school."
In another study, this one of smaller classes in Tennessee, African-American students were described by the RAND Corporation as making "remarkable" progress. But a team of scholars from Harvard, Georgetown and the University of Wisconsin compared four school choice experiments with that Tennessee research: The benefits of vouchers after two years, they found, were over twice as large.
If there is a promised land in the world of school reform it's smaller classes. But here we see how vouchers promise to be more effective -- not to mention loads cheaper.
At the end of the day, of course, the ultimate consideration ought not to be how much money we spend, but how many young lives we save. Yet wouldn't it be nice to learn -- as only an econometrician can teach -- whether all the billions we spend on elementary and secondary education might be better spent if more Minnesota families were afforded educational freedom?
Art, what do you say?