The Crisis Prompts a Rethink of Education Spending

"If we want to become a strong economy again, the best thing we can do is have an educated workforce." Few would object to U.S. Secretary of Education Arne Duncan's explanation of why Washington is funneling $100 billion to schools and universities as part of February's giant stimulus package. Indeed, other countries are following suit, with Britain, Germany, Canada, China, and others making new education funding part of their anticrisis strategies.

What's far less clear is that this money is going where it's most needed—or likely to have the greatest social and economic payoff. In Germany, the bulk of nearly €10 billion in new school spending is being used to renovate buildings—a bonanza for construction companies and popular with parents and teachers, but unlikely to have much effect on the quality of German graduates. In Britain, Prime Minister Gordon Brown is pushing for more PCs and Web access in schools—another policy that's popular but considered irrelevant by educators. In the United States, a July audit by the Government Accountability Office found that schools were not using the stimulus money to boost student achievement, as promised by Duncan, but to fund their general budgets. And in still other countries, governments are using money to help build new world-class universities—projects that a World Bank study in July warned risk bleeding resources away from more desperately needed areas. "I'm not sure that the people making these decisions even realize the trade-offs involved," says Jamil Salmi, author of the study.

That's particularly unfortunate today, given the economic stakes. According to an April report by McKinsey, the United States' GDP would have been 9 to 16 percent—or $1.3 trillion to $2.3 trillion—higher in 2008 had U.S. high-school graduates attained the average skills of their peers in Canada, Finland, or South Korea. This fall, the Organization for Economic Cooperation and Development (OECD) will unveil a similar study in Paris detailing the losses suffered by other laggards. Andreas Schleicher, author of the OECD study, says that "in a whole row of countries, the economic losses of educational underperformance are significantly higher than the costs of the financial crisis." What's worse, he says, countries pay the price for their mistakes year after year.

The biggest error governments are making is to blindly push for more and better everything at all levels of education: more teachers, flashier facilities, more technology in the classroom, and more elite universities. Saudi Arabia, for example, is currently spending $13 billion on a single graduate school. All such efforts may seem sensible, but studies by academics at Munich's IFO Institute and Stanford, among other places, show that simply spending more on education doesn't yield better results. Kids don't necessarily learn more if they sit in smaller classrooms, in more modern and better-equipped schools, or even if their teachers are better-paid (as opposed to just better). According to Ludger Woessmann of the IFO Institute, merely raising per-student spending has zero effect on achievement. The United States, France, and Germany have upped spending significantly in past decades only to see performance stagnate, while countries like Sweden and Finland have boosted quality through structural reform.

Even building better universities has a lower return than one would think. Peer Ederer, who heads the Lisbon Council's Human Capital Index Project, says the biggest problems university systems face today are high dropout rates and too few skills taught for later careers—problems most current spending proposals do little to address. Too much attention, he says, is given to things like increasing attendance or moving top institutions up the international rankings. Salmi of the World Bank says the huge resources now being spent to create elite schools would be better used for expanding and improving engineering programs at unflashy polytechnics, for example. President Obama's advisers seem to be listening closely; in July, they announced a $12 billion boost for two-year community colleges. But most of the U.S. stimulus money has gone for helping all schools fill in general budget shortfalls.

Studies such as McKinsey's suggest another important way education policy should be refocused. They find that the largest returns on investment come not from funneling more money toward top or even average performers, but toward those who have been left behind. Raising the achievement of the unskilled and excluded would lead not only to individual payoffs, such as higher incomes and more meaningful lives, but also would generate big benefits for economies, such as higher productivity and greater GDP. It would also result in broad social gains—less crime, less welfare spending, and a greater sense of cohesion. "Improving our education to get the economic growth more broadly shared is the one most important thing we can do," says Benjamin Friedman, a Harvard economist and author of The Moral Consequences of Economic Growth. He argues that changing education in this way would be one of the few ways governments could promote both equity and economic growth—not one at the expense of the other.

But it won't be easy. Washington's long history of failed attempts to improve inner-city schools shows that raising the level of the 20 percent or so of U.S. students who leave school functionally illiterate defies straightforward fixes. If there is one thing most experts agree on, however, it's that the earlier the system intervenes, the better. In Germany, where the educational achievements of immigrants and other socially disadvantaged groups are strikingly low, a study by Katharina Spiess of the German Institute for Economic Research in Berlin found out that if you can get immigrant kids to attend kindergarten, they are 25 percent more likely to go on to a pre-university track.

University of Chicago economist James Heckman estimates that preschool programs for disadvantaged kids cost $10,000 per student a year—not cheap by any means—but generate a 16 percent annual rate of return. Heckman figures that a quarter of this gain goes to the kids, who he says are more likely to get better grades, stay in school, get a diploma, go on to college, and enjoy higher lifetime earnings. But 75 percent of the "profit" accrues to society, in the form of lower crime rates, fewer welfare payments, and greater economic productivity, as well as better integration of minorities and immigrants. "Trying to remediate later is much more costly and often ineffective," says Spiess. Yet many of the world's school systems try to "fix" the problem of the unskilled and uneducated when it's too late—with welfare payments, high-school equivalency courses, or programs like Germany's ghettoized Hauptschulen, low-level schools that train for menial jobs.

Raising the achievement of previously excluded groups is thus no softhearted social policy, but an economic imperative. One country that has systematically pushed this maxim is Canada, which saw a big surge in immigration in the 1990s. In Toronto, where more than 40 percent of schoolchildren speak a mother tongue other than English and more than a third come from socially disadvantaged families, the entire school system is concentrated on raising these kids to the level of their peers. Though it still struggles with high dropout rates among some groups, Toronto has been able to erase the achievement difference between migrants and natives—in marked contrast to cities in Germany and France, where the gap has been widening. It's one reason why Canadian students get some of the highest scores on international achievement tests.

Given how obvious all this sounds, why haven't more countries already figured it out? One answer is that determining what kind of education spending is most economically and socially effective—as opposed to merely popular—requires the kind of close self-scrutiny that is foreign to most school systems. That said, Britain, Norway, and several American states have finally begun to conduct "value added" testing, which measures students' progress from the time they enter a school until they leave. Such testing, which measures advances instead of achievement, is controversial. For one thing, it highlights the fact that top institutions often rely on "creaming"—admitting only the best of the best, who then, unsurprisingly, do well.

When the emphasis of testing shifts from outright achievement to relative gains, the results can be stunning. In Florida, for example, school districts that have introduced such testing have found that at some of their "best" schools (by conventional rankings), students' skills actually dropped between grades, while lower-ranked schools were much better at raising students' abilities from their (admittedly lower) starting points. Such analysis may be uncomfortable for some, but it is critical, for once the schools or programs that provide the biggest boost are identified, they can be targeted for spending and their methods replicated.

The logical next step would be to identify which individual teachers give their students the greatest boost. "Forget class size, curriculums, budgets—the most effective policy is good teachers," says Eric Hanushek, an education specialist at Stanford University's Hoover Institution. The OECD study finds per-student spending has little relation to student performance. It's teacher quality—which involves much more than money—that makes the difference. According to Hanushek, students taught by the worst 10 percent of teachers achieve only 50 percent of average gains, while those in the best teachers' classes improve by 150 percent—even after adjusting for the differences between good and bad schools or students. Finland and South Korea have the world's highest-achieving high-school students—thanks in large part to a focus on teachers: improving their selection, upgrading their training, and concentrating on how they can best help individual students keep up.

So why haven't more countries followed their lead? The answer is politics. Spending on elite or middle-class schools is much more popular than redirecting scarce funds to migrants and the underclass. Parents and teachers' unions also love mostly irrelevant policies like lowering class size and hiring more teachers. One reason, beyond their intuitive appeal, is that the benefits of such measures, such as nicer or less crowded classrooms, are often immediately apparent. Focusing on early intervention and preschool, by contrast, may not deliver dividends for 20 years.

Teachers' unions also often object to any disruptions in the current system. When the British Education Ministry first tried to introduce value-added testing, it couldn't overcome stiff resistance from teachers and schools until Parliament made it illegal to use the results to determine salaries or budgets. Something similar happened in Belgium, and in Germany it is taboo to compare individual schools' or teachers' performance. Yet without such testing and evaluation, education policy will remain vague and approximate, based on folk wisdom and political expediency rather than evidence and facts. Some of the new policies, such as Obama's community-college plan, are big steps in the right direction. Yet most of what governments around the world are calling "education investments" still fails to meet the effectiveness test. That's despite the fact that redirecting spending toward society's underachievers is one reform that would pay for itself many times over. Indeed, to put it another way, the truth about such measures is stark: they represent one change today's crisis-stricken governments simply can't afford not to make.