Closing the Achievement Gap

Reihan Salam and Tino Sanandaji
From the Nov. 14, 2011, issue of National Review

During the recent struggle over collective-bargaining rights in Wisconsin, a number of left-of-center observers, including New York Times columnist Paul Krugman, pointed out that students in unionized Wisconsin do better on average than students in non-unionized Texas. The obvious conclusion, or so we were led to believe, is that teachers’ unions lead to better education.

There is, however, a problem with this argument. Drawing on data from the 2009 National Assessment of Educational Progress, the political commentator David Burge pointed out that white students in Texas outperform white students in Wisconsin, black students in Texas outperform black students in Wisconsin, and Hispanic students in Texas outperform Hispanic students in Wisconsin. This may look like a statistical paradox; Wisconsin does better on average, even though all groups do worse in Wisconsin. But there is an explanation: Wisconsin has a considerably larger share of white students than Texas, and white students tend to fare better than black and Hispanic students. This example highlights the increasing importance of demographics to the American education debate.

It is not difficult to understand the sources of the achievement gap. Particularly when confronted with the fact that more K–12 spending hasn’t generally meant better educational outcomes, even defenders of the teachers’ unions often highlight the role of poverty, family disruption, and historical disadvantage in limiting the ability of black and Hispanic students to thrive in school. As a general rule, native-born non-Hispanic whites have reaped the benefits of many generations of relative peace and prosperity. And this long experience of prosperity has contributed to the intergenerational transmission of wealth, tacit knowledge, and social networks that can give one a leg up. Social capital might also contribute to the stability of non-Hispanic white families, in which children are typically raised by both biological parents.

This is in stark contrast to the experience of African Americans, in no small part because of the lingering legacy of enslavement and segregation. The Harvard sociologist Orlando Patterson has written extensively on how the black historical experience has shaped contemporary marriage and family patterns. Today, 72 percent of black children are born to unwed mothers, and a large number of these children are denied the stability and comfort offered by having two parents in the home. African Americans experienced dramatic material progress over the course of the last century, which is not always acknowledged, but it can’t be denied that they still suffer from disadvantages that, say, Mayflower descendants do not.

America’s Hispanic population, much of which has its origins in Mexico, faces its own set of challenges. Family breakdown is not as severe among Hispanics as it is among blacks, but it is still troubling, with 53 percent of Hispanics born out of wedlock. As the economists George Borjas and Lawrence Katz have observed, recent waves of Mexican immigrants have tended to have levels of educational attainment comparable to those of the immigrants from southern Europe who settled in the United States in the 1920s. The problem is that educational attainment for the native-born population has increased considerably since the 1920s, which puts Mexican immigrants at a greater disadvantage in a labor market that increasingly rewards education.

Over time, the gap between the children of Mexican immigrants and the children of native-born parents tends to shrink but not disappear. Perhaps we should not be surprised that those with illiterate grandparents in rural Mexico are on average less likely to succeed than those with grandparents who led a prosperous middle-class life in the suburbs of Boston or Detroit.

All of the above tells us that unionized public-school teachers have the germ of a reasonable point: In many respects, America’s K–12 student population presents greater challenges today than it did in 1970. It is therefore not entirely fair to blame unionized public-school teachers for all of America’s educational woes. But if our K–12 students are having a much harder time because of a complex, interrelated set of social problems, what are the implications for our economic future? What are we to make of the fact that a unionized public-education system in Wisconsin failed to meet these challenges, while the Texas model (when properly measured) has proven more successful?

We take no joy in bringing attention to a neglected problem: the ongoing lowering of the skill level of the U.S. work force. This phenomenon is caused by two factors: the stubborn achievement gap between ethnic minorities and the majority population, and the demographic transformation of the work force, which will soon have a “majority minority” composition. Already, we have seen the average American high-school-graduation rate decline since the mid-1970s. And by some measures, retiring workers have a higher level of educational attainment than those who are just entering the work force.

Ever since the release of A Nation at Risk in 1983, America’s leaders have devoted a great deal of time and attention to public education, spending on which has increased considerably. Yet a large share of the K–12 spending increase has been captured by a dramatic expansion in the number of public-school teachers. As their ranks have increased, and as economic opportunities for women in other sectors have expanded, average teacher quality has deteriorated. This deterioration has happened at the same time that the number of students from difficult home environments and disadvantaged groups has increased, i.e., at the same time that we have needed an increase in teacher quality.



It is thus no surprise that the large gaps in educational attainment and earnings between ethnic groups have proven persistent over time. Broadly speaking, the American labor market can be divided into two pools, one consisting of disadvantaged minorities and one consisting of whites and Asians, whose outcomes tend to resemble each other. While it is normally not a good idea to generalize broadly about race and ethnicity, in some instances it is necessary to examine social problems through the prism of broad group differences. This is especially true for a phenomenon such as poverty, which afflicts ethnic groups in disparate ways.



The sheer size of America’s racial and ethnic disparities may surprise readers. According to the latest statistics from the Census Bureau, the per capita income of whites is 76 percent higher than that of African Americans, and an astonishing 101 percent higher than that of Hispanics. The income gap between whites and Hispanics is larger than the income gap between New York and West Virginia. One recent estimate shows that about one-sixth of whites do not graduate from high school. The corresponding figure is twice as high for African Americans, and nearly three times as high for Hispanics. Even after years of focused attention on the achievement gap, the differences in income and educational attainment between whites and minority groups are roughly the same now as they were a generation ago. African Americans made small gains during this period, but the growing share of Hispanics in the population and their lagging incomes nullified the effect of this advance.



The Left often argues that the root cause of disparities in earnings is structural racism and discrimination. But a new study by economist James Heckman demonstrates that the disparities largely vanish once differences in skills are taken into account. This confirms similar findings in a study by the University of Chicago’s Derek Neal and another by Harvard University’s Roland Fryer. Overall, Hispanics and blacks who attain the same test scores as whites do not earn lower wages and are no less likely to enter college. We believe that disadvantages earlier in life account for the existence of skill disparities. Being assigned to worse schools and having a less stable family environment cause members of minority groups to enter the labor market with fewer skills than whites do, and therefore to earn less. But discrimination in the labor market itself does not appear to be the main explanation for the differences we have discussed.

The achievement gap is not new, but its impact on U.S. economic performance is growing. The reason for this is simply that the number of minority-group members, in particular Hispanics, as a share of the population is rising. To put it another way, the United States, in its demographic composition, is looking less and less like Wisconsin and more and more like Texas.

The gap between whites and Hispanics mattered very little to GDP growth when the Hispanic population represented a small share of the working-age population. In 1940, whites constituted 88 percent of the U.S. population, while Hispanics were only 1 percent. But because of higher birth rates and, to a lesser extent, continued immigration, the Hispanic population is growing rapidly. The Census Bureau estimates that in 2050, whites will be a minority of 46 percent of the population, while Hispanics will have grown to 30 percent.

There is debate among economists regarding whether immigration of less-skilled Hispanics has depressed the wages of less-skilled native-born American workers. But the more important economic impact of this demographic shift is its effect on the overall skill profile of the U.S. work force. Though poorer than native-born Americans, Hispanic immigrants are far richer than they would have been in Latin America. Thus the impact of their immigration on global income has been positive. But if the achievement gap did not exist, this increase in Hispanics’ living standards would be even larger, and American per capita GDP growth would be higher.

It is very difficult to predict the trajectory of future economic growth. We can, however, offer crude estimates by drawing on what we know about the relationship between the skill level of workers and overall economic growth. Per capita GDP growth has already been dampened by the combination of an increasing Hispanic population share and a persistent ethnic gap in average income. In the coming four decades, the effect of demographic change is expected to be even more dramatic.

Let us extrapolate the historical average of 2 percent annual per capita income growth until 2050, and take aging and demographic change into account. Aging will shrink the working-age population, bringing income growth down to about 1.75 percent. Also, whites and Asians will go from two-thirds to a little over half of the working-age population — and the effect of that shift will depend on changes in the achievement gap.

The worst-case scenario is that the gaps between whites and non-whites in education and earnings will not change. In this scenario, the skills and earnings of the American work force decline, and per capita income growth falls to 1.49 percent per year. However, if we assume that policy reform or assimilation will close half of the educational-achievement gap by 2050, and that this in turn will close the earning gap by half, then the average growth rate per capita will be 1.85 percent per year. In this second scenario, the growing Hispanic population not only doesn’t reduce income growth, but actually mitigates some of the effects of population aging.

Comparing the two scenarios vividly illustrates the economic value of closing the achievement gap. The alternative futures available to us are an economy producing $38 trillion per year and an economy producing $44 trillion per year.

Since the American public sector historically has collected around one-third of GDP in revenue, the optimistic scenario could add close to $2 trillion to federal coffers every year, which could make the nation fiscally solvent and go a long way toward lowering public-debt levels. And there would also be gains in terms of lowered crime, healthier lifestyles, and improved family outcomes.

There is no question that liberals care as much about the plight of minorities as conservatives do. Their fault is not a lack of good intentions. Rather, it is coalition politics.

In Special Interest, Stanford political scientist Terry Moe offers an exhaustive account of the political influence of America’s 4 million unionized public-school teachers, and how it has been deployed to block education-reform efforts for the past few decades. Union leaders have thwarted attempts to deploy staff in a more efficient manner and to offer incentive-based compensation. Their solution to every problem in education is more money. And of course, any increase in resources is channeled toward either hiring more teachers, thus creating more loyal union members, or increasing compensation for teachers, ideally in a way tied to length of service and not quality of performance.

Though unionized public-school teachers are a force to be reckoned with in both political parties, they’ve grown particularly powerful with the Democrats. At recent Democratic national conventions, delegates from teachers’ unions have outnumbered delegates from California. Unionized public-school teachers are massively overrepresented in the Democratic grassroots, which has undoubtedly contributed to what we might call the neutering of the Obama administration’s education-reform efforts, and helps explain the president’s counterproductive insistence that state-level reform efforts have “buy-in” from teachers if they’re to have any hope of securing federal Race to the Top funds.

According to the OECD, the United States currently spends over $1 trillion per year on education, more than 8 percent of its national income. This makes the U.S. the second-highest spender on education among industrialized nations, whether we measure by share of national income or absolute dollars per pupil. Expenditure per pupil in elementary and secondary school is now in excess of $10,000 per year. Adjusted for inflation, this is two and a half times the sum that was spent per pupil in 1970, according to the Digest of Education Statistics. Despite this spending increase, reading and math test scores were virtually flat over the same time period, while, as mentioned above, high-school-graduation rates actually declined.



We know by now that increased funding will not miraculously close the achievement gap. We also know that there is no single recipe for improving educational outcomes for minority students. Reducing the influence of the teachers’ unions seems to be an important first step, as Texas’s experience suggests. Doing so gives schools the political breathing room they need to deploy their resources with the interests of students, as opposed to union leaders, foremost in mind. Allowing schools through trial and error to discover effective new instructional methods also holds great promise. But there remains the question of what exactly is to be done.

Closing the achievement gap through new methods and new policy is not impossible, as gains made in recent years by black students demonstrate. In a recent study, Harvard economist Roland Fryer found that African-American students who entered high-quality charter schools in the Harlem Children’s Zone scored as well in mathematics as white children nationally.

The work of James Heckman, widely considered America’s foremost education researcher, offers important insight on how to spread these gains. He finds that education reforms can be cost-effective, but only if they meet certain criteria. First, interventions should address the needs of children very early in life. As we age, our ability to absorb knowledge decreases. A school system that targets 16-year-olds after a decade of deficiency is not likely to succeed.

Second, interventions should mainly inculcate behavioral norms, not cognitive skills — for example, social skills, an ability to control aggression, and a propensity to follow rules. Such norms are particularly lacking among disadvantaged children from broken homes and neighborhoods with weak social bonds. One reason to focus on these norms and social abilities is that they are far more malleable than cognitive skills, and thus more susceptible to the influence of well-designed educational programs. Moreover, social skills and norms of conduct are as important for success in school and in the labor market as cognitive skills.

Finally, Heckman contends that programs must be targeted toward the truly disadvantaged in order to be cost-effective. Universal pre-school programs geared toward the middle class, such as those championed by President Obama, do not appear to work as well. In the case of middle-class children, families and communities already do an excellent job of norm- and character-formation, so universal pre-school education is largely a waste of scarce resources. It is for poor students that public schools must pick up the slack, and for whom the programs are effective. Such a strategy has been aggressively pursued by the highly successful KIPP charter-school network.

All of this is more easily said than done. Shifting resources from failed approaches to more successful ones will start political fights that will make Wisconsin’s recent battles look easy. One factor encouraging change, however, has been the rise of innovative educational technologies, which are quickly rendering traditional classrooms obsolete and undermining the power of the teachers’ unions. Another is the growing frustration among African Americans and Hispanics, who are keenly aware that traditional public schools are failing their children. This, in turn, has forced a growing number of Democrats to reconsider their party’s alliance with the teachers’ unions, and to consider more innovative approaches to improving educational performance.

What hasn’t happened, and what needs to happen, is for middle-class voters to recognize that the achievement gap isn’t some sentimental side issue that shouldn’t concern serious people. Rather, it is absolutely central to America’s economic future.