Response to Diane W. Schanzenbach
本文回应了Schanzenbach关于男孩教育成就差距的观点，讨论了推迟入学等干预措施的效果与争议，强调需基于证据制定政策。
The gender gap in educational outcomes is slowly but surely moving up the academic and policymaking agenda. It is no small thing for a scholar of Diane Schanzenbach's standing to declare so clearly that “we certainly need to pay attention to improving boys' achievement levels.” This is a call that I hope more researchers, policymakers, political figures, and nonprofit organizations will heed. The need for more policies aimed specifically at improving the education outcomes for boys and men is pressing. Progress is being hindered by the false perception that pushing for more pro-male interventions somehow means doing less for girls and women. But as Schanzenbach points out, “improving boys’ school performance does not imply harming girls’ performance. Skill development is not zero-sum, and when students learn more they go on to add more to our shared economy.” The question then is (or should be): How best to improve the outcomes for boys and men? It is important to be careful here. The evidence base here is thin. To confidently propose solutions requires 1) acknowledging the problem, 2) implementing policy or programmatic solutions specifically aimed at solving it, 3) conducting high-quality evaluations of those solutions, and 4) testing whether these replicate at scale. Right now, it can still be a struggle to persuade some stakeholders that the problems faced by boys and men are a big enough problem to move into steps 2 through 4. That said, there are a range of interventions that at the very least offer some real promise, especially an expansion of vocational and technical learning opportunities, including more technical high schools; an increase in the share of male teachers, especially in middle school; and a later school start for boys, the redshirting proposal. To these I would now add the individualized tutoring programs highlighted by Schanzenbach. Where we disagree is over the net value of a later school start for boys. Here I'll address three of the concerns raised by Schanzenbach: the loss of a year in the labor market, the bluntness of such a “one size fits all” policy, and the sharp shift in social norms. One note to keep in mind is that my proposal is to change the default starting age. Parents should have the flexibility to override this default, just as they already do in most jurisdictions. Perhaps they decide that their son would not benefit from a later start, but that their daughter would. This should be up to them. Of the arguments against the redshirting proposal, the loss of time in the labor market is the most serious. On average, a year less of experience in the labor market delays earnings growth and represents an absolute loss of a year's worth of wages, social security contributions, and so on. Even in the UK, where it is more common for students to take a “gap year” before college, the evidence suggests that at the age of 30, students who took a gap year earn less than those who proceeded straight from secondary to higher education, despite performing slightly better in college (Crawford & Cribb, 2012). The question here is whether redshirted boys would see their educational outcomes improved sufficiently to outweigh the potential loss of a year in the labor market. A great deal hinges on what the counterfactual educational trajectory is. The fear of a “lost year” might be less if may boys are already losing time in the labor market or higher education, in part because of their difficulties in secondary education. To be clear, the first step should be some small-scale pilots of various redshirting intervention before considering any larger-scale implementation. In fact, it would be ideal to pilot several interventions alongside each other, including redshirting and the 1:1 in-school tutoring proposed by Schanzenbach—which I strongly endorse—as well as a control group. If redshirting improved the chances of a successful launch from K–12 education into either employment, college or a combination of the two, this would offset the losses associated with a later school start. I agree that my proposed redshirting policy is a “one size fits all” approach. But that criticism applies to any policy that uses the blunt tool of birth date to determine when a child should start school. The overlap in measures of school readiness between one grade and the next is, by definition, going to be very great; likely at least as great as between boys and girls of the same age. Having a different default starting age for boys than girls is a blunt tool. But no blunter than a starting age based on age in the first place. On reflection, I've come to believe that starting boys a whole year later is not necessary and may not deliver the best results. One of the advantages of using a birth date to determine school entry age is that it allows for lots of flexibility. And the evidence for positive impacts from redshirting, including from Cascio and Schanzenbach (2016), does looks to be stronger for the boys who are closest to the age cutoff. It could make most sense to set the default starting age for boys just three or four months later than for girls, rather than a whole year. A more modest approach like this would also reduce some of the risks and costs Schanzenbach outlines, including a potentially higher risk of not completing high school, time lost in the labor market, and additional childcare burdens for parents. The last question I'll address here is the shift in norms. I think that the idea of a later school start for boys could quite quickly become seen as uncontroversial, as it already has in many affluent communities and at more elite secondary schools. As Schanzenbach points out, 1 in 5 children of highly-educated parents have a delayed kindergarten start. In 2022, a prestigious East Coast private K–12 school shared the birth dates of their graduating seniors with me (Reeves, 2022). Almost one in three of the boys (30%) were born before October 2003, the formal cut-off date for school entry for that year, compared to 7% of the girls. This suggests norms on this issue can change quite quickly. I'm encouraged that so many eminent scholars like Schanzenbach are now taking the gender gap in educational outcomes so seriously. She and I agree strongly that policymakers should now be doing the same. The question now is over which policies and programs will help boys and men the most. The mere fact of this debate is a hugely positive step forward. Richard Reeves is President of the American Institute for Boys and Men and Non Resident Senior Fellow at the Brookings Institution (email: [email protected]).