July 13, 2006

Increasing graduation, part III

I've discussed in-school issues in two earlier entries on increasing graduation (part I, part II). This morning, I'll write a bit on the historical push-pull relationship between schools and labor markets. To focus entirely on what happens inside a school does not fully explain the trends in high school attendance over the past century. Teen attendance and graduation rose over the first two-thirds of the twentieth century with two exceptions (the two world wars), and then graduation with a standard diploma hit a plateau sometime in the 1960s/70s (though we're not exactly sure when that plateau began). While many things changed about schools, the best fit with that data is not what happened inside schools but how the teen labor market changed. The reciprocal relationship between schools and work has real consequences for increasing graduation, though few talk about it.


Here's the gist of the story: as teenagers found full-time work more scarce in the first half of the twentieth century, high school absorbed their time. The only exceptions were wartime labor shortages, when that general trend was temporarily reversed. (High school attendance dropped for girls as well as boys in the early 1940s, so the drop is not explained well by enlistments.) At the same time, the increasing labor-market rewards for high school attendance increased incentives to remain, first for girls seeking clerical jobs and later for all teens. The result was a fairly clear message: stay in school because you can't get a full-time job anyway and you have a better chance at one later.

But in the last third of the century, teens found increasing opportunities in part-time labor. The data on this is not crystal clear, but it suggests that the reorganization of work in the U.S. began to exploit or maybe "work with" the targeting of teens as consumers. Retail outlets and fast-food joints became the management of revolving-door employment. To teens, "dead-end jobs" provided cash for cars, dates, and consumer goods. What's not to like about that, in the short term?

There are two obvious concerns. One is a matter of ideas: what looks like the proliferation of part-time work broke the message that had been created in the first half of the twentieth century. With jobs available in high school, there wasn't the most visible choice anymore of the world of school vs. the world of work. Yes, such a tradeoff was a creation of the early 20th century and not true for everyone. But today, a significant number of high-school students work: 17% of 16-year-old students, 30% of 17-year-old high-school students, and 36% of 18-year-old high school students worked in the spring of 2005, according to the March 2005 Current Population Survey. Technically, that's the proportion who were reported as working the week before the survey, and so the actual percentages are going to be a bit higher if you asked what proportion worked at some point during the academic year. The message my parents told us when growing up—your job is school—isn't the modal view when a good chunk of students are working in a single week in March. (Incidentally, if you want to get quick tabulations from the various Census Bureau surveys, I highly recommend the Data Ferrett program, which is free.)

The second concern we should have is the real tradeoff between substantial work and schoolwork. The small literature on work during school hours suggests that there is a benefit to a limited number of hours: Up to about 15 hours of work per week has tangible benefits for teen students. I suspect this benefit is related to the need to be a little more organized when you juggle school, work, and other commitments. But beyond 15 hours, the studies I've read suggest there are real harms—work hours crowd in on schoolwork, exhaust students, and begin the process of disengagement. All of those complaints about high-school students who sleep through morning classes? Some of that may be teens who party or IM through the wee hours, but a good chunk are students who are trying to work.
Of all 16-year-old students reported by the March 2005 Current Population Survey, 6 percent had worked more than 15 hours the prior week. That proportion jumps to 13 percent of 17-year-old high school students and 24 percent of 18-year-old high school students—who are in the senior year if "on track" for graduation or who may be edging away from school if not.

I don't know of any data source for the hours that teens work, but there are relatively few restrictions on teens' working late at night and trying to go to school early the next day. The variety in state regulations on schoolnight working hours show some of the absurdities of labor regulation for high-school students. Some states have no regulations on when 16- and 17-year-olds can work before a school day. Even those who do have some restrictions only have lax regulations, allowing teens to work until 10 or 11 at night. Pennsylvania allows 16- and 17-year-olds to work until midnight; Louisiana allows 17-year-olds to work until midnight; Massachusetts allows restaurant workers to stay until midnight; and Wyoming allows girls to work until midnight and has no time-of-day restrictions for boys (I guess boys don't need their beauty rest?).

When looking at this data, one is tempted to say one way of increasing graduation would be to restrict the hours that 16- and 17-year-olds can work, both total in a week (no more than 15 hours, if one would follow the existing research) and how late they can work at night (and as a parent, I'd put 9 pm as a reasonable end-point). But there are some problems with such measures, as crudely stated. First, the enforcement of child-labor laws target employers, not the teens—and I could easily see a teen evading a 15-hour restriction by working 14 hours for one employer, 10 hours for another, etc. Second, a 9 pm working limit for 16- and 17-year-olds could harm students who work to get away from gang cultures in their neighborhoods. The call from the National Research Council (1998) for more data collection on teen employment is especially important for disentangling such complicated situations. Unfortunately, I don't know of any actual implementation of those recommendations, so I don't know of better data sources.

I could imagine some tweaking of things like curfews and child-labor laws to minimize overworking teens, but the tweaking needs to be clever. Some not-very-clever ideas with putative but admittedly speculative theories of action:

  • Regulation of child labor should focus more on parents, who should know when and where their children are working. Employers do have to report wages to both federal agencies (Social Security and income-tax withholding) and also to state agencies (unemployment-insurance records), and there easily could be a system to report to parents simultaneously information about work hours along with how the student is doing in school. If you want parents to act based on information, make the feedback as connected as possible: grades along with extracurricular-activity time and work time. (I'm sure some enterprising company is going to sign a few "helicopter parents" up for GPS-and-grade reports.)
  • Graduated driver licenses for minors (such as hours restrictions on when 16-year-olds can drive, restrictions on driving with teens, maybe even raising the license age to 17) might reduce the incentive to work for car expenses and make teens think a little more carefully about working late at night, in addition to saving lives.
  • There could be a more generous minimum-break requirement for employers who hire teens to work after 8 pm. I suppose a reasonable requirement (from a school perspective) would be that employers need to provide an unpaid break of one hour studying time for every two hours of work, and there would also need to be space at the worksite where teen workers could study. The main cost to employers here would be providing space for studying, but I suspect many would be happy to support teens' schooling (and I really don't care about those who only want to exploit teens).

But those are just initial ideas. Unfortunately, the connection between teen work and school performance is largely ignored. That obliviousness has to end if we're to move beyond talking about a graduation-rate crisis, no matter what its magnitude.

Selected readings:

Apel et al., A Job Isn't Just a Job, Crime & Delinquency (2006). This article addresses the possible flow of teens into the informal labor market if there are more restrictions in the formal labor market.

Greenberger & Steinberg, When Teenagers Work: The Psychological and Social Costs of Adolescent Employment (1986). This is the classic work on teen employment.

National Research Council, Protecting Youth at Work (1998) (Executive Summary). Many of the recommendations here parallel the reasoning above—and have been largely ignored, as far as I'm aware.

Listen to this article
Posted in Education policy on July 13, 2006 10:17 AM |