President Barack Obama visited Georgia on Thursday to tout his ambitious new proposal for universal preschool. "Let's make sure none of our kids start out the race of life already a step behind," he said. "Let's make it a national priority to give every child access to a high-quality early education." That sounds lovely, but can it work? The Daily Beast's Megan McArdle, the American Enterprise Institute's James Pethokoukis, and Slate's Matthew Yglesias have their doubts. "We have some intriguing examples of amazing preschools," Yglesias writes, voicing the common sentiment, "but little experience with bringing them up to mass scale."
Obama's plan comes from two "amazing preschools"—the Perry Preschool Project, in Michigan, and the Abecedarian Project, in North Carolina. During the 1960s and 1970s, each project created a special, intensive school for disadvantaged children, and then allowed researchers to follow the kids afterward, in order to see how they did relative to a group that didn't get the preschooling. Both groups of children had applied to get into the programs, and were selected by lottery, creating a truly randomized experiment.
The results were clear-cut and, to this day, largely uncontested. The kids who attended preschool achieved more, while developing fewer emotional and intellectual problems. The Perry kids, for example, were less than half as likely to end up in special education and more than one-third as likely to graduate on time from high school. And while gains in IQ dissipated with time, the others persisted through childhood and into adulthood. The Perry kids were half as likely to get in trouble with the law and nearly three times as likely to own a home.
In recent years, neuroscience has helped to provide a theory for why the programs were so successful: The first few years of life turn out to have massive, nearly permanent effects on the actual architecture of the brain. It's not just the acquisition of vocabulary or knowledge of multiplication tables. As Paul Tough documents in his superb new book, How Children Succeed, it's also the ability to block out distraction, to control impulses, and to deal with anxiety—the "soft skills" people need to navigate life.
But results from Head Start, a Great Society–era program that attempts to provide this kind of schooling on a much broader scale, have been less encouraging. The National Head Start Impact Study, for instance, found that the academic benefits of enrolling in the program faded quickly. Within a few years, the kids who had been in Head Start were not doing better academically than those who had not. "There is no measurable advantage to children in elementary school of having participated in Head Start," writes Grover Whitehurst, a former Education Department official in the Bush Administration and now at Brookings. "Head Start does not improve the school readiness of children from low-income families."
The study has its detractors. Some researchers insist that the Head Start kids got more benefits than the study suggests, while others say it wasn't a fair comparison because many of the "control" kids—the ones who didn't go to Head Start—ended up in subsidized private preschool anyway.1 The best answer, probably, is that individual Head Start programs vary enormously and many just aren't that good. They may have weak curricula, call for little involvement from parents, or have less qualified teachers. Just 28 percent of Head Start instructors have bachelor's degrees. By comparison, all the teachers at Perry, which now operates as the Highscope Perry Project, have a college degree. "To get the important outcomes that the Perry Preschool Program got, Head Start needs to do what the Perry Preschool Program did," says Larry Schweinhart, president of Highscope and a longtime researcher. "But it lacks some of the ingredients that made Perry so highly effective."
Skeptics think these disparities show it's tough to replicate the success of Perry or Abecedarian on a large scale: "You should always, always keep scaling problems in mind when you are reading big claims made on the basis of small projects," McArdle writes. She's right. But some not-so-small projects have also succeeded. Scholars who followed children in the Chicago Child-Parent Centers—about two dozen preschools, each serving between 100 and 150 low-income children—into young adulthood found that these students performed modestly better (particularly the girls), were less likely to become criminals, and fared better in the job market.2
In Oklahoma, which already has a universal pre-kindergarten program, the data is not as definitive, and it does not follow the kids as far into adulthood.3 But the findings are encouraging nonetheless. "I suppose someone could argue that these state programs' effects on kindergarten readiness do not prove long-run effects," says Timothy Bartik, a senior economist at the Upjohn Institute for Employment, who has done some of that research. "But these early effects are often of similar size to those achieved in the [Chicago] program, and many of these state programs are similar in design to the [Chicago] program. Therefore, it is a reasonable inference that these large-scale state programs will have long-run effects." One bonus finding from the Oklahoma research: Middle-class kids also seem to benefit from preschool, although not by as much as the low-income kids did.
Reasonable people disagree about these findings. My takeaway is that paying attention to early childhood can help at-risk kids, and in some cases help them a lot. But it takes the right kind of program, which means, among other things, experimenting to see what works best. Fortunately, this seems to be what Obama and his advisers have in mind. Obama's proposal calls for "universal preschool." But the government would accomplish this by providing matching grants to the states, presumably giving them some flexibility to design programs as they see fit.4 The Obama proposal would also reach kids early—earlier, in fact, that many of the early accounts have suggested. As we reported previously, the plan includes proposed investments in programs that also would help infants and toddlers—like putting more money into a widely praised program that sends visiting nurses to new mothers, and strengthening a federal program that helps parents get very young children into quality day care.
Exactly how much money Obama would like to spend on each priority, and how much he would like to spend overall, remains unknown; that information will come when the administration issues its annual budget proposal. But the Center for American Progress Proposal calls for matching state expenditures up to $10,000 per child, which would actually be more than what the Perry project spent in today's dollars. The ten-year cost of the American Progress plan is about $100 billion.
It would probably take that sort of commitment, maybe even more, to produce the best results. And Congress might not be in a mood to find that kind of money right now. But James Heckman, the Nobel economist who has become something of an evangelist for targeted spending on early childhood, has calculated that society reaps a return of about 7 to 10 percent on these "investments"—a term he uses very consciously. "It's costly to build the Hoover Dam, or pave a road, or do research on building a better, more fuel-efficient automobile," Heckman says, but "what do we lose by not doing the program? I think we lose output, we lose value, we lose a lot of dimensions of social function—earnings are only part of it."
Even wary analysts like McArdle agree about the potential for long-term payoffs ("These kids didn't ask to be born to parents who can't give them what it takes to succeed," she writes). The question is whether these programs can realize that potential—a question Obama and his supporters must answer in the coming months.
Update: The Center on American Progress Action Fund came up with a terrific set of charts about early childhood investment:
One other possibility: The effects of Head Start fade because, afterward, kids aren’t getting a good education. That, say scholars like Duke’s Kenneth Dodge, would be an argument for doing more for kids after they get to kindergarten, not doing less for them beforehand.
“It is very rare,” writes Greg Anrig of the Century Foundation, “to find any government program that produces such substantial and carefully documented benefits.”
See the Washington Post’s Suzy Khimm thorough description and research of the Oklahoma program.
Steven Barnett, of the National Institute for Early Education Research, pointed out to me that states like Oklahoma use money to "top off" salaries at existing Head Start programs, so that they can pay enough to attract better trained educators.