Having marched single-file through years of education, America’s class of 2017 recently climbed onstage to receive their high-school diplomas. Their experiences to that point were remarkably similar for so diverse a group of 3 million minds, the objective of college enrollment almost universal. Yet, after shaking hands with the principal, their paths diverged sharply — as if separate staircases led down the stage’s other side.
Roughly speaking, one fifth of all students were already off track and did not join their classmates crossing the stage. Another fifth will move from their senior year to something besides further schooling. The third fifth will enroll in college but fail to complete it. A fourth fifth will complete some form of college but land in jobs that do not require the degrees they just earned. Only a final, fortunate fifth will successfully navigate the path — high school to college to career — that is our education system’s platonic ideal.
Decades of reform in teacher training, student testing, and standards, as well as school choice and hundreds of billions of dollars in new annual spending, have sought to shrink the ranks of those first four groups and move their members into the last. The goal is admirable, but progress has been slow: Success has reached only a small fraction of the population, and data suggest that a majority of students will never make the move.
Because it refuses to acknowledge its limits, the system gravely underserves the populations most in need of its help. The alternative would be to accept that students’ paths will inevitably diverge and to help each excel on his trajectory. This requires “tracking,” a term with various definitions in education-reform circles but used here to denote the separation of high-school students into different educational programs that aim at different outcomes. One path could lead toward college enrollment, another toward occupational training that leaves a 20-year-old with serious work experience, a marketable skill, and $30,000 in a savings account.
The obstacle to this approach is philosophical, not practical. Compared with the futility of our efforts to boost academic achievement, helping students translate their demonstrated aptitudes into gainful employment is the far more achievable course. It is already the model employed in virtually every other developed country. But it would require Americans to change their conception of what schools can achieve and to relinquish some cherished ideals of egalitarianism that do more harm than good.
Politicians have shown no interest in moving in that direction. Mentions of “apprenticeships” and “vocational training” sprinkle their talking points and budget proposals, but few seem to have contemplated the implications of elevating such programs beyond their second-class status. We need to acknowledge that college is not just “not for everyone” but in fact not for most, that those who are not college-bound deserve to begin their preparation as early as those who are, and that our education policy and resource allocation should prioritize those headed down the rockiest paths instead of those already holding a golden ticket.
Perhaps the best way to understand our misplaced zeal for college and the needed change in mindset is to stare at what blogger Mickey Kaus calls “the most important chart around when it comes to explaining contemporary politics.” The chart shows the average hourly wages of Americans based on their educational attainment, with separate lines tracing the progress since 1970 for each level, from less than high school to advanced degree.
Higher levels of education have always corresponded to higher wages, but whereas the lines moved in parallel through the 1970s, they veer apart after 1980. Wages of college graduates rose over the past generation, those of high-school graduates did not — depending on the inflation adjustment one uses, they have remained flat or have fallen by more than 10 percent. In 1980, median weekly earnings for those with at least a bachelor’s degree were 40 percent higher than for those with only a high-school diploma; by 2016, the premium had risen to 80 percent. Advanced-degree holders did better still, high-school dropouts worst of all.
The diverging lines invite two very different interpretations. In one, the lines depict educational experiences, which cause different outcomes: Attending and graduating from college leads to higher earnings. In the other, the lines reflect preexisting characteristics that correlate with different outcomes: Whatever academic talent and other endowments make one a college graduate also tend to make one a high earner. These two interpretations suggest very different policy responses. If the first is correct, the education system should try to get everyone into college — moving people from lower lines to higher ones. But if the second is closer to the truth, the system should focus on maximizing the fortunes of those headed toward each outcome — and especially on shifting the lower lines upward.
Americans of all political stripes naturally gravitate toward the first, more egalitarian interpretation. We want to believe that anyone given the opportunity to experience a college education can achieve the prosperity that blesses college graduates. And so we expect that an education policy that aims squarely at smoothing every student’s path to college will be to the benefit of all.
Unfortunately, experience has not borne this out. Most high-school seniors are unprepared to succeed in a college environment, and the K–12 system has shown limited capacity to change this fact. Even as per-pupil spending has doubled since the 1970s, average National Assessment of Educational Progress scores for 17-year-olds in both reading and math have remained flat. In states where almost all high-school seniors take the SAT, only about one-third achieve scores that would indicate a likelihood of maintaining a B-minus average at a four-year college.
How could so many reforms and so much money have accomplished so little? Because, as far as we can tell, and as counterintuitive as it might seem, school quality appears not to contribute that much to academic achievement. In 1966, the “Coleman Report,” which was called for by the Civil Rights Act, concluded that, when socioeconomic factors are controlled for, “differences between schools account for only a small fraction of differences in pupil achievement.” Revisiting the question 50 years later, Johns Hopkins researchers found that “expenditures and related school inputs have very weak associations not only with test scores in the sophomore and senior years of high school but also with high-school graduation and subsequent college entry.”
“The results,” they concluded, “are mostly in line with the whispered result that has become the apocryphal characterization of [the Coleman Report]: ‘It’s all family.’”
This might overstate the case — certainly, there is evidence that good schools and good teachers can lift achievement. But such results often come with the caveat that we do not know how to replicate them. For instance, while some charter schools serving disadvantaged students have achieved notable success, Bruno V. Manno and Chester E. Finn Jr. highlighted in National Affairs the “painful tradeoff states face between charter-school quality and quantity.” Likewise, at EducationNext, Jay Greene described a “general disconnect between test scores and later life outcomes.” Better test scores do not necessarily extend to higher graduation rates or college enrollment, let alone college completion. We should continue striving to improve our K–12 schools, but we should not be under the illusion that higher spending will yield higher achievement, or expect nationwide college preparedness to jump anytime soon.
The experience of college enrollees further confirms what the high-school data predict. Fewer than 40 percent of enrollees at four-year colleges graduate from their first school on schedule; fewer than 30 percent of enrollees at two-year schools graduate in even three years. Those figures are closer to one-half if transfers and delayed graduations are included, but reliance on such circuitous paths does not indicate a system functioning well. A recent Brookings Institution report put it bluntly: “Despite continual increases in college attendance, college degree attainment has stagnated.” The percentage of 25-year-old Americans with at least a bachelor’s degree was lower in 2015 than in 1995. A U.S. Census Bureau report, looking at people between the ages of 25 and 29 over a longer period of time, also highlighted a divergence between the sexes. Women have made substantial progress, but this is owing at least in part to changing social norms. For men, “college completion reached 27 percent in 1976 and 1977, but subsequently fell, and did not rise above 27 percent again until 2011.” As of 2015, it remained below one-third.
If it seems as if America’s colleges and universities are poorly suited to the average American 18-year-old, perhaps that’s because they were never designed to serve him. We think of the post–World War II GI Bill boom as having opened college to the masses, but even at its 1949 peak, only 20 percent of men between the ages of 18 and 24 were enrolled in higher education. It would be an extraordinary coincidence if a program designed decades if not centuries ago for the edification of the wealthy and the training of a narrow segment of the population also happened to meet the needs of a typical 2010s work-force entrant.
The disconnect is on stark display in the massive gap between these institutions’ self-perception and the demands of the public. In Gallup’s 2017 survey of more than 700 college and university presidents, only 1 percent strongly agreed with the statement that “most Americans have an accurate view of the purpose of higher education”; overall, they were four times more likely to disagree than agree. College presidents believe their goal is to help students “become educated citizens, be exposed to new points of view, to become lifelong learners,” said Terry Hartle, senior vice president at the American Council on Education, to Inside Higher Ed. As for the general public, “often all they care about is jobs.”
Indeed, a 2015 survey by the think tank New America found that the top three “reasons to go to college” for prospective students were “to improve my employment opportunities” (91 percent), “to make more money” (90 percent), and “to get a good job” (89 percent). Unfortunately, college presidents appear successful so far in resisting such bourgeois priorities: A separate Gallup survey found that only one-third of Americans believe that college graduates are well prepared for success in the work force — and that’s if you are fortunate enough to graduate.
If our education system has not shown the ability to move a significant share of the population into the ranks of college graduates, we need to give more credence to the second interpretation of the income-by-education-level chart. Most Americans’ educational trajectories appear to develop independent of their school experiences, which places the onus on our schools to meet students where they are and help them prepare for success with the academic outcomes toward which they are headed.
Such a conclusion can become emotionally charged if read as a negative judgment of those on the lower lines. But it needn’t be read that way. The factors that influence an individual’s trajectory might be any combination of innate and environmental. Blame society if that helps. For education policy, what matters is that we have not found the school-based tools that will reshape trajectories. You can argue for early-childhood interventions or for anti-poverty programs — but while our schools await success on those fronts, we need a different plan of attack.
None of this should surprise conservatives, and it would not if it were stated in the abstract. America’s K–12 public-school system, after all, is among the world’s largest government programs. Its budget rivals that of the Department of Defense. It operates through a unionized work force of 3 million teachers, to whom parents are required by law to send their children every day. In no other context would conservatives expect such a program of mass social engineering to correct for the differences among individuals, their families, and their communities. If anything, it should be a tremendous relief that the system does not inadvertently make matters worse.
But many conservatives, facing liberals who insisted that society must achieve an equality of opportunity that comes close to equality of “life chances” (and, by implication, equal outcomes), became enamored of the idea that they could outflank their opponents by emphasizing education reform. Both accepting the definition and expecting public schools to meet it were mistakes.
The original, achievable meaning of “equality of opportunity” requires that “desirable positions in the public and private sectors [be] open to all and . . . awarded on the basis of merit,” wrote David Azerrad in National Affairs last year. “The liberal understanding of equality of opportunity as equality of life chances is neither possible nor desirable. . . . The first step is to overcome the flawed race-of-life formulation, since it inexorably points to the expansive liberal definition of equality of opportunity. Life is not a race. It is a journey whose goal is happiness. . . . We are not all racing toward the same finish line, but all pursuing happiness in our own way.”
If our commitment to equality of opportunity returned to its emphasis on the elimination of public impediments rather than on attempting to correct for differences among individuals, we could give our schools a more achievable task. “What we need,” Charles Murray has written in the Wall Street Journal, “is an educational system that brings children with all combinations of assets and deficits to adulthood having identified things they enjoy doing and having learned how to do them well.” This calls for “completely different agendas than the ones that dominate today’s rhetoric.”
The critical component in this shift should be tracking: offering dramatically different programs of secondary education depending on whether the student will proceed next to college or directly to a career. This model is standard elsewhere in the developed world but roundly rejected in the United States. In Learning for Jobs, the Organization for Economic Cooperation and Development (OECD) features a chart showing that most developed countries have between 40 and 70 percent of their secondary-school students in vocational and technical programs. A footnote explains that the United States is omitted from the chart entirely because of “the rather different approach to vocational education and training in U.S. high schools.”
Americans have long resisted tracking. In 1892, Harvard University president Charles W. Eliot exclaimed, “I refuse to believe that the American public intends to have its children sorted before their teens into clerks, watchmakers, lithographers, . . . and so forth, and treated differently in their schools according to their prophecies of their appropriate life careers. Who are we to make these prophecies?” When the nation pioneered universal secondary education in the early 20th century, its “comprehensive high schools” emphasized a broad education in the liberal arts. During the 1960s and ’70s, in keeping with so many other areas in which idealism trumped practicality at the expense of the purported beneficiaries, the notion of tracking was squelched entirely.
This particular bout of American exceptionalism is a mistake. What sense does it make to treat the vast majority of high-schoolers as if they were prospective college graduates when they are not; to pretend that the sudden divergence of outcomes after high-school graduation did not in fact begin long before? Indeed, the best way to understand the U.S. system is not as trackless, but rather as committed to a single track tailored toward those most likely to succeed anyway.
One common objection to tracking is that students from disadvantaged backgrounds will be more likely to land on (or be steered toward) a career track. But this is a description of society and an implicit condemnation of the current system, not a plausible criticism of tracking. After all, students best suited to a career track are precisely those least well served by its absence and experiencing the worst outcomes today. A tracked system could offer them a better chance at economic success, increasing in turn the odds that their own children land on the college track a generation later. It will speed social progress and improve countless lives along the way.
More generally, it is true that a tracked system will unavoidably place some students on a career track who might have succeeded in college. But the career track is not a death sentence. It can lead in many cases to a more fulfilling (and even more remunerative) career than might the college track, especially for its most talented students. And individuals would have opportunities to shift tracks both during their education and much later; as the New York Times notes, “it is not uncommon to find executives in Europe who got their start in apprenticeships.” No public-education system will serve every student well, but the share finding themselves mismatched will be far lower if programs at least attempt to meet the needs of the majority.
In practice, tracking would begin in ninth or tenth grade, at which point each student could elect a college or career track. Critically, while the school would provide a recommendation, the choice would be the family’s. No high-pressure test dictates the future; no institutional discrimination tramples on the judgment of those who know the student best.
While the college track would look much like today’s high school, the career track would be very different. Further academics would be compressed and then placed in a curriculum that emphasized practical skills and offered opportunities to explore career opportunities. By eleventh grade, time might be evenly divided between classroom learning and vocational training; twelfth grade would generally be replaced by the first year of an apprenticeship.
Such things are easier to describe than to implement. Schools would require new curricula, new teachers, and new facilities. But models in other countries suggest that this can be done, and they suggest a place to start. The New York Times reports that, in Switzerland, “nearly 70 percent choose the vocational track, with programs for about 230 occupations. Beginning in tenth grade, students rotate among employers, industry organizations, and school for three to four years of training and mentoring.” Germany’s apprenticeship system offers recognized qualifications in 350 occupations, and 80 percent of its young adults are employed within six months of completing their education, compared with 48 percent in the United States.
Presumably, the U.S. cannot and would not want merely to replicate some other country’s system; the point is that reform in this direction is neither untested nor implausible. Already, isolated programs in the U.S. are validating the concept. In City Journal, Steven Malanga has described a variety of experiments under way, including efforts in New York City that now reach 26,000 students. Siemens has created a four-year program for high-school seniors in North Carolina, in partnership with a local community college, that leaves each student with an associate’s degree in mechatronics, no student debt, and a job paying more than $50,000 per year. But it is no coincidence that Siemens is headquartered in Munich; “compared with Europe, America has few firms that participate in apprenticeship and other job-training programs,” reports Malanga. Nor should policymakers be surprised that American firms have had limited enthusiasm for such programs when efforts promoting them are not accompanied by the reforms that would elevate their status and position them to succeed.
A major benefit of a tracked vocational system is the additional resources it makes available because students are simultaneously learning and working. In the U.S. system, a student might move through eleventh and twelfth grades at a cost of $15,000 per year, then spend two years in community college at an additional cost of $10,000 per year, and then at the age of 20 begin looking for a full-time job. He will then have consumed $50,000 in public resources and forgone substantial earnings of his own. By contrast, a German on a career track has much of his training cost funded by the company at which he works, on top of which he might begin to earn wages.
With no net increase in education spending, the U.S. could offer every tenth-grader additional classroom learning, a subsidized three-year apprenticeship for which he might also be paid, and a savings account with $20,000 to $40,000 awaiting him upon completion. That offer might not appeal to the fifth of students headed down the idealized path today. But the other 80 percent would be wise to consider it, and we owe them that option.
Reforms in this direction would begin with the elimination of federal standards, mandates, and funding streams that treat academic achievement and college enrollment as primary goals. Career tracks should be presented as equally worthy, evaluated on their own terms, and eligible for at least comparable per-student funding. Partnerships with private employers should be not only permitted but encouraged and, where appropriate, subsidized with public-education dollars. Ultimately, the education system should commit to spending at least as much on a 15-year-old whose next seven years will be spent in a combination of school, apprenticeship, and employment as it spends on one headed to a four-year public university. If anything, given their respective earnings horizons upon completion, it is the career-tracked youngster who deserves the more generous support.
Such reforms are not incompatible with other education priorities. Testing and accountability remain good ideas, so long as we are testing and holding schools accountable for the right things. School choice could become even more important; in the short term, innovation at the high-school level will be more needed than ever. In elementary and middle schools, meanwhile, charters should continue striving to overcome the challenges facing disadvantaged students in particular, so that the college track is a viable option for as many as possible when the time comes.
For some reason, Americans have decided that “unskilled” is less insulting an adjective than “uneducated.” Tell someone he can’t do anything, sure, just don’t suggest he did not sit in school long enough. A society in which everyone can become skilled, regardless of how well educated, would come much closer to the ideal of offering genuine opportunity to all.
– Mr. Cass is a senior fellow at the Manhattan Institute.