In 1983, A Nation at Risk misidentified what is wrong with our public schools and consequently set the nation on a school reform crusade that has done more harm than good.

The diagnosis of the National Commission on Excellence in Education was flawed in three respects: First, it wrongly concluded that student achievement was declining. Second, it placed the blame on schools for national economic problems over which schools have relatively little influence. Third, it ignored the responsibility of the nation’s other social and economic institutions for learning.

As to student achievement: A Nation at Risk based its analysis of declining student achievement entirely on average SAT scores which had dropped by about half a standard deviation from 1963 to 1980. But much of the decline had been due to the changing composition of SAT test takers — in the early 1960s, the preponderance of SAT test takers were high school students planning to apply to the most selective colleges. By 1983, the demographic composition of SAT test takers had mostly stabilized, and average SAT scores were again rising, not declining.[1]

Trend scores from the National Assessment of Educational Progress (NAEP) also show a more complex picture than Risk described. In elementary and middle school math, average scores rose for both black and white students, starting in the late 1970s. This trend might not yet have been fully understood by Risk commission members — they might have concluded that the upturn then barely detectable would be short-lived. But the rise has certainly continued subsequent to 1983. Indeed as the figure below shows, for black students, the improvement has been so dramatic that black fourth grade math scores today are now higher than white fourth grade scores in 1978. In other words, if white math achievement had been stagnant, the black-white achievement gap would have been entirely closed. The continued gap is due to substantial improvement in white scores as well.

Eighth grade math scores have also increased since 1978, although not by as much; for twelfth graders, white scores have been stagnant, while black scores had a big increase from 1982 to 1990, and have been stagnant since.

Reading scores are less positive. For whites, reading performance is not substantially better now than in 1978, at the fourth, eighth, or twelfth grade levels. But it is not worse either. For blacks, reading performance is better, but not nearly as much better as in math.

None of this, however, supports the decline thesis of A Nation at Risk.

Because of the report’s doomsday aura, policymakers have mostly failed since 1983 to investigate the causes of these improvements — the obvious, unasked, question is, what were we doing right from 1978 to 1990 (and since), so we can do more of it?

A belief in decline has led to irresponsibility in school reform. Policymakers who believed they could do no harm because American schools were already in a state of collapse have imposed radical reforms without careful consideration of possible unintended adverse consequences. Not thinking that President Reagan’s rule (“if it ain’t broke, don’t fix it”) applied to what conservatives and liberals alike assumed was an already broken school system, this irresponsibility reached its zenith in the bipartisan No Child Left Behind (NCLB) law of 2002.

I do not suggest that American schools are adequate, that American students’ level of achievement in math and reading is where it should be, that American schools have been improving as rapidly as they should, or that the achievement gap is narrowing to the extent needed to give us any satisfaction. I only suggest that we should approach fixing a system differently if we believe its outcomes are slowly improving than if we believe it is collapsing. And we owe the latter, flawed assumption, to A Nation at Risk.

Perhaps the greatest damage has been done by narrowing the curriculum in an effort to boost math and reading test scores. The trend is most notable since the enactment of NCLB, as schools have diminished attention to history, civics, the sciences, art, music, physical education, character development, and social skills, to make more instructional time available for test preparation in math and reading. This distortion of the historical breadth of American public school goals has been most pronounced for minority and other disadvantaged children. These are the children who most need a broad curriculum, as well as further gains in math and reading.[2]

Risk, to its credit, worried that “schools may emphasize such rudiments as reading and computation at the expense of other essential skills such as comprehension, analysis, solving problems, and drawing conclusions,” and it asserted, “Our concern, however, goes well beyond matters such as industry and commerce. It also includes the intellectual, moral, and spiritual strengths of our people… .”[3] But these caveats were buried beneath the report’s urgent calls to improve the reading and (especially) math skills that purportedly determined the nation’s economic health, and to increase the standardized testing that would spur such improvement.

From an irrational faith in the ability of standardized tests to inspire greater learning, and from an unwillingness to finance more expensive tests that would sample critical thinking as well as basic skills, we’ve again narrowed the curriculum to “minimum competency,” precisely the 1970s standard that A Nation at Risk denounced. From a belief that an alleged decline in student achievement must be attributable to a decline in teacher quality, at best, or to malfeasance (“low expectations”) of teachers, at worst, many districts have attempted to overcome this teacher incompetence by implementing scripted, or nearly so, curricula. We’ve attempted to focus teachers’ attention by a testing regime so rigid that it threatens to destroy teachers’ intrinsic motivation and their ability to address the full range of student difficulties that can only be diagnosed by creative teachers, student-by-student.

Again, this does not suggest that teachers are as well-trained as they should be, as well-motivated as we would like them to be, or as student-oriented as they must be. But it is hard to defend the proposition that teachers, especially those of minority and disadvantaged children, have been sitting around making excuses for poor performance when these children have gained a full standard deviation in test score improvement in a single generation.

As to schools’ responsibility for economic ills:[4] A Nation at Risk claimed that increased market shares for Japanese automobiles, German machine tools, and Korean steel reflected the superior education of those nations’ workers:

Our once unchallenged preeminence in commerce, industry, science, and technological innovation is being overtaken by competitors throughout the world… . [T]he educational foundations of our society are presently being eroded by a rising tide of mediocrity that threatens our very future as a Nation… .[5]

The report claimed that a “long-term decline in educational achievement” was somehow connected to “a steady 15-year decline in industrial productivity, as one great American industry after another falls to world competition.”[6]

Risk then stimulated a spate of similar reports through the late 1980s and early 1990s, all making similar claims that import penetration could be blamed on poor American education.

For example, in 1990, a group of prominent Democrats and Republicans calling themselves the National Center on Education and the Economy followed with another report, America’s Choice: High Skills or Low Wages. It saw skills development as virtually the only policy lever for shaping the economy. It charged that inadequate skills attained at flawed schools had caused industrial productivity to “slow to a crawl” and would, without radical school reform, lead to permanently low wages for the bottom 70 percent of all Americans.

Leading public intellectuals such as Robert Reich focused attention on human capital solutions in a laissez-faire global system. His book The Work of Nations argued that international competition would be won by nations with the most (and best) “symbolic analysts,” not “routine” workers. Lester Thurow’s Head to Head forecast that Western Europe would come to dominate the United States and Japan economically because European schools were superior. Mainstream economists, both liberal and conservative, agreed that rising wage and income inequality were caused by an acceleration of “skill-biased technological change,” meaning that computerization and other advanced technologies were bidding up the relative value of education, leaving the less-skilled worse off.

Yet the response of American manufacturers to this allegedly education-driven import competition was curious. Automakers moved plants to Mexico, where worker education levels are considerably lower than those in the American Midwest. Meanwhile, Japanese manufacturers pressed their advantage by setting up non-union plants in places like Kentucky and Alabama, states not known for having the best-educated workers. High school graduates in those locations apparently had no difficulty working in teams and adapting to Japanese just-in-time manufacturing methods.

The ink was barely dry on the America’s Choice report when Americans’ ability to master technological change generated an extraordinary decade-long acceleration of productivity, beginning in the mid-1990s and exceeding that of other advanced countries. The productivity leap was accomplished by the very same workforce that the experts claimed imperiled our future. No presidential commissions announced that American schools must be superior to those of Western Europe and Japan, as evidenced by our more rapid productivity growth.

Again, the authors of A Nation at Risk cannot entirely be faulted for assuming that poor education had caused a productivity collapse. The big upturn in productivity growth began after Risk was issued. But it did begin, and productivity advances created new wealth with the potential to support a steady increase in the standards of living of all Americans.

And for a brief period, standards of living did indeed increase, because the fruits of productivity growth were broadly shared. As the chart shows, the late 1990s saw increasing wages for both high school and college graduates.[7]

Even wages of high school dropouts climbed. But no presidential commissions praised American schools for producing widely shared prosperity.

The collapse of the stock bubble in 2000, the recession of the early 2000s, and the intensification of policies hostile to labor brought wage growth to a halt. Living standards again began to decline and inequality zoomed — at the same time that workforce productivity continued to climb. White-collar offshoring to India, China, and other low-wage countries signaled that globalization was now taking its toll on computer programmers and other symbolic analysts of the information age.

Today, however, a new cast of doomsayers has resuscitated an old storyline, picking up where A Nation at Risk left off. Forgetting how wrong such analyses were in the 1980s and ‘90s, the contemporary cliché is that however good schools may once have been, the 21st century makes them obsolete. Global competition requires all students to graduate from high school prepared either for academic college or for technical training requiring equivalent cognitive ability. We can only beat the Asians by being smarter and more creative than they are.

The argument got a boost from New York Times columnist Thomas Friedman’s 2005 book, The World is Flat, and has been repeated by the same National Center on Education and the Economy in Tough Choices, a sequel to its 1990 report. The argument has also garnered support from influential foundations such as the Gates Foundation, and its chairman, Bill Gates, and from education advocacy groups such as the American College Testing Program.

The Tough Choices report bemoans the fact that “Indian engineers make $7,500 a year against $45,000 for an American engineer with the same qualifications” and concludes from this that we can compete with the Indian economy only if our engineers are smarter than theirs. This is silly: No matter how good our schools, American engineers won’t be six times as smart as those in the rest of the world. Nonetheless, Marc Tucker, author of Tough Choices (and president of the group that produced the 1990 report as well), asserts, “The fact is that education holds the key to personal and national economic well-being, more now than at any time in our history.”

Administration officials blame workers’ education for middle-class income stagnation. Treasury Secretary Henry Paulson contends that “market forces work to provide the greatest rewards to those with the needed skills in the growth areas. This means that those workers with less education and fewer skills will realize fewer rewards and have fewer opportunities to advance.” Former Federal Reserve Chairman Alan Greenspan frequently blamed schools for inequality: “We have not been able to keep up the average skill level in our workforce to match the required increases of increasing technology… .”

But these 21st-century claims are as misguided as those of the last century. Of course, we should work to improve schools for the middle class. And we have an urgent need to help more students from disadvantaged families graduate from good high schools. If those students do so, our society can become more meritocratic, with children from low-income and minority families better able to compete for good jobs with children from more privileged homes. But the biggest threats to the next generation’s success come from social and economic policy failures, not schools. And enhancing opportunity requires much more than school improvement.

If A Nation at Risk commissioners could not have known that explosive economic growth was just around the corner, today’s education scolds have no such excuse. Workforce skills continue to generate rising productivity. In the last five years, wages of both high school- and college-educated workers have been stagnant, while productivity grew by a quite healthy 10.4 percent.

Rising workforce skills can indeed make American firms more competitive. But better skills, while essential, are not the only source of productivity growth. The honesty of our capital markets, the accountability of our corporations, our fiscal policy and currency management, our national investment in R&D and infrastructure, and the fair-play of the trading system (or its absence), also influence whether the U.S. economy reaps the gains of Americans’ diligence and ingenuity. The singular obsession with schools deflects political attention from policy failures in those other realms.

But while adequate skills are an essential component of productivity growth, workforce skills cannot determine how the wealth created by national productivity is distributed. That decision is made by policies over which schools have no influence — tax, regulatory, trade, monetary, technology, and labor market policies that modify the market forces affecting how much workers will be paid. Continually upgrading skills and education is essential for sustaining growth as well as for closing historic race and ethnic gaps. It does not, however, guarantee economic success without policies that also reconnect pay with productivity growth.

American middle-class living standards are threatened, not because workers lack competitive skills but because the richest among us have seized the fruits of productivity growth, denying what were historically considered fair shares to the working- and middle-class Americans, educated in American schools, who have created this new national wealth. Over the last few decades, wages of college graduates overall have increased, but some college graduates — managers, executives, white-collar sales workers — have commandeered disproportionate shares, with little left over for scientists, engineers, teachers, computer programmers, and others with high levels of skill. No amount of school reform can undo policies that redirect wealth generated by skilled workers to profits and executive bonuses.

A Nation at Risk gave renewed currency to the claim, now conventional, that the changing nature of work would require radical changes in education:

Computers and computer-controlled equipment are penetrating every aspect of our lives – homes, factories, and offices… . [B]y the turn of the century, millions of jobs will involve laser technology and robotics. Technology is rapidly transforming a host of other occupations. They include health care, medical science, energy production, food processing, construction, and the building, repair and maintenance of sophisticated scientific, educational, military, and industrial equipment.[8]

This description is literally true; indeed, it explains much of the dramatic rise in productivity we’ve experienced. But the conclusion that these changes would require radical changes in education was flawed. It ignored the obvious reality that technology de-skills many jobs. Retail clerks now routinely use laser technology to scan bar codes; these clerks no longer need basic arithmetic skills.

College graduates are, in fact, not in short supply. Indeed, some college graduates are now forced to take jobs requiring only high-school educations. The Bureau of Labor Statistics projects that, for the next decade, only 22 percent of job vacancies will require a college degree or more. Forty percent will require only one month or less of on-the-job training, and could be filled by high school graduates or, in many cases, by dropouts — retail salespersons and waiters and waitresses, for example.[9]

In many high-school hallways nowadays, you can find a chart displaying the growing “returns to education” — the ratio of college to high-school graduates’ wages.[10]

The idea is to impress on youths the urgency of going to college and the calamity that will befall those who don’t. The data are real — college graduates do earn more than high-school graduates, and the gap is substantially greater than it was a few decades ago.

But it is too facile to conclude that this ratio proves a shortage of college graduates.

The denominator — the falling real wages of high-school graduates — has played a bigger part in boosting the college-to-high-school wage ratio than has the numerator — an unmet demand for college graduates. Important causes of this decline of high school graduates’ wages have been the weakening of labor market institutions, such as the minimum wage and unions, which once boosted the pay of high school-educated workers.

For the first time in a decade, the minimum wage was recently increased. The curious result will be a statistical decline in “returns to education.” But we should not conclude from a minimum-wage increase that we need fewer college graduates, any more than we should have concluded from falling wages for high-school graduates that college graduates are scarce and schools are failing.

Another too glib canard is that our education system used to be acceptable because students could graduate from high school (or even drop out) and still support families with good manufacturing jobs. Today, those jobs are vanishing, and with them the chance of middle-class incomes for those without good educations.

It’s true that many manufacturing jobs have disappeared. Replacements have mostly been equally unskilled or semiskilled jobs in service and retail sectors. There was never anything more inherently valuable about working in a factory assembly line than about changing bed linens in a hotel. What once made semiskilled manufacturing jobs desirable was that many (though not most) were protected by unions, provided pensions and health insurance, and compensated with decent wages. That today’s working class doesn’t get similar protections has nothing to do with the adequacy of its education, but everything to do with policy decisions stemming from the value we place on equality. Hotel jobs that pay $20 an hour, with health and pension benefits (rather than $10 an hour without benefits) typically do so because of union organization, not because maids earned bachelor’s degrees.

It is cynical to tell millions of Americans who work (and who will continue to be needed to work) in low-level administrative jobs and in janitorial, food-service, hospitality, transportation, and retail industries that their wages have stagnated because their educations are inadequate for international competition. The quality of our civic, cultural, community, and family lives demands school improvement, but barriers to unionization are a more important cause of low wages than the quality of workers’ education.

Fortunately, the elite consensus on education as a cure-all seems now to be collapsing. Offshoring of high-tech jobs has deeply undercut the Clinton-era metaphor of an education-fueled transition to the information age, since it is all too apparent that college educations and computer skills do not insulate Americans from globalization’s downsides. Former Clinton economic advisor (and Federal Reserve vice chairman) Alan Blinder has emerged as an establishment voice calling attention to the potentially large-scale impact of continued offshoring. Blinder stresses that the distinction between American jobs likely to be destroyed by international competition and those likely to survive, is not one of workers’ skills or education. “It is unlikely that the services of either taxi drivers or airline pilots will ever be delivered electronically over long distances… . Janitors and crane operators are probably immune to foreign competition; accountants and computer programmers are not.”[11]

These are not problems that can be solved by vouchers, charter schools, teacher accountability, or any other school intervention. A balanced human capital policy would involve schools, but would require tax, regulatory, and labor market reforms as well.

As to the relative responsibility of schools: A Nation at Risk was issued in 1983, a decade after the nation’s postwar narrowing of social and economic inequality had ended. By the time of the report, income was becoming less evenly distributed. The real value of the minimum wage was falling and the share of the workforce with union protection was declining. Progress towards integration had halted and, as William Julius Wilson noted in The Truly Disadvantaged, published only half a dozen years later, the poorest black children were becoming isolated in dysfunctional inner-city communities to an extent not previously seen in American social history.

Social and economic disadvantage contributes in important ways to poor student achievement. Children in poor health attend quality schools less regularly. Those with inadequate housing change schools frequently, disrupting not only their own educations but those of their classmates. Children whose parents are less literate and whose homes have less rich intellectual environments enter school already so far behind that they rarely can catch up. Parents under severe economic stress cannot provide the support children need to excel. And, as Wilson described, children in neighborhoods without academically successful role models are less likely to develop academic ambitions themselves.[12]

These nonschool influences on academic achievement were known to the commissioners who authored A Nation at Risk. The Coleman Report of 1966, still a major document of recent research history, had concluded that family background factors were more important influences on student achievement variation than school quality.[13] In 1972 and 1979, Christopher Jencks and his colleagues had published two widely noticed reassessments of Coleman, Inequality and Who Gets Ahead?, both of which confirmed the Coleman Report’s central finding. Yet the National Commission on Excellence in Education, in preparation for its Nation at Risk report, commissioned 40 research studies from the leading academic researchers in the nation, and not one of these was primarily devoted to the social and economic factors that affect learning.

Most remarkably, A Nation at Risk concluded with a brief “Word to Parents and Students,” acknowledging that schools alone could not reverse the alleged decline in academic performance. It urged parents to be a “living example of what you expect your children to honor and emulate… You should encourage more diligent study and discourage satisfaction with mediocrity… .”[14] This was the report’s only reference to nonschool factors that influence learning.

A Nation at Risk therefore changed the national conversation about education from the Coleman-Jencks focus on social and economic influences to an assumption that schools alone could raise and equalize student achievement. The distorted focus culminated in the No Child Left Behind legislation of 2002, demanding that school accountability alone for raising test scores should raise achievement to never-before-attained levels, and equalize outcomes by race and social class as well.

A Nation at Risk was well-intentioned, but based on flawed analyses, at least some of which should have been known to the commission that authored it. The report burned into Americans’ consciousness a conviction that, evidence notwithstanding, our schools are failures, and warped our view of the relationship between schools and economic well-being. It distracted education policymakers from insisting that our political, economic, and social institutions also have a responsibility to prepare children to be ready to learn when they attend school.

There are many reasons to improve American schools, but declining achievement and international competition are not good arguments for doing so. Asking schools to improve dramatically without support from other social and economic institutions is bound to fail, as a quarter century of experience since A Nation at Risk has demonstrated.

Notes

[1] Willard Wirtz, et. al. 1977. On Further Examination: Report of the Advisory Panel on the Scholastic Aptitude Test Score Decline. Princeton, N.J.: College Board Publications; Albert E. Beaton, Albert E., Thomas L. Hilton, and William B. Shrader, 1977. Changes in the Verbal Abilities of High School Seniors, College Entrants, and SAT Candidates Between 1960 and 1972. See my more extensive discussion in Richard Rothstein. 1998. The Way We Were? The Myths and Realities of America’s Student Achievement. New York: The Century Foundation.

[2] For further discussion of goal distortion in American education, see Richard Rothstein and Rebecca Jacobsen. 2006. “The Goals of Education.” Phi Delta Kappan 88 (4), December.

[3] National Commission on Excellence in Education. 1983. A Nation at Risk. The Imperative for Education Reform. U.S. Government Printing Office (http://www.ed.gov/pubs/NatAtRisk/risk.html). (hereinafter, “Risk”) p. 10, p. 7.

[4] This section was co-authored by Lawrence Mishel, and adapted in part from “Schools as Scapegoats” by Lawrence Mishel and Richard Rothstein in The American Prospect, October 2007.

[5] Risk, p. 5.

[6] Risk, p. 17-18.

[7] “The Productivity-Pay Gap” calculated and illustrated by Lawrence Mishel of the Economic Policy Institute.[8] Risk, p. 10.

[9] Arlene Dohm and Lynn Schniper, 2007. “Occupational Employment Projections to 2016,” Monthly Labor Review, November.

[10] “Returns to Education” from Lawrence Mishel, Jared Bernstein and Sylvia Allegretto, 2007. The State of Working America 2006/2007. Ithaca: Cornell University Press.

[11] Alan S.Blinder. 2006. “Offshoring: The Next Industrial Revolution?” Foreign Affairs 85 (2): March/April

[12] I have discussed these issues in Class and Schools (Teachers College Press, 2004).

[13] Coleman, James S., and Ernest Q. Campbell, Carol J. Hobson, James McPartland, Alexander M. Mood, Frederic D. Weinfeld and Rober L. York, 1966. Equality of Educational Opportunity. Washington,D.C.: U.S. Department of Health, Education, and Welfare, Government Printing Office

[14] Risk, p. 35.

—

Richard Rothstein is a research associate of the Economic Policy Institute.

Richard Rothstein cites evidence that public schools have improved math scores at age 9 and 13, but not age 17. Thus whatever gains are being made in elementary and middle school are being lost in high school. Since 1973, K-12 educational expenditures have more than doubled; on a per-dollar basis, “investing” in public education now shows a thirty-five year trend of steadily decreasing returns. Based on these outcomes, Rothstein asks “what were we doing right … so we can do more of it?” Rothstein’s argument that educational quality should not be a concern due to steadily increasing global economic competition is even less convincing. In order to avoid criticizing public education, Rothstein has been reduced to arguing that education doesn’t matter:

“Workforce skills cannot determine how the wealth created by national productivity is distributed.”

“…the conclusion that [that the changing nature of work] would require radical changes in education was flawed.”

“The quality of our civic, cultural, community, and family lives demands school improvement, but barriers to unionization are a more important cause of low wages than the quality of workers’ education.”

Once upon a time liberals cared passionately about education as foundation for a meritocracy. Rothstein has concluded this Enlightenment ideal is a dead end.

Rather than give up our ideals, shouldn’t we reconsider our irrational attachment to government-managed schools? Dan Klein cites economist Robert Solow explaining why he doesn’t like school choice:

It isn’t for any economic reason; all the economic reasons favor school vouchers. It is because what made me an American is the United States Army and the public school system.[1]

There is a fierce loyalty towards the public school system that transcends all reason.

When it comes to market solutions, intellectuals are just plain bigoted. One can still find tens of thousands of volumes in academic libraries citing empirical evidence that communism was as effective at meeting human needs as is capitalism. Paul Samuelson’s thirteenth edition of his Principles of Economics, published in 1989, claimed that “the Soviet economy is proof that, contrary to what many skeptics had earlier believed, a socialist command economy can function and even thrive.”[2] Samuelson was right-of-center within academia. As late as September 1991, the American Political Science Association published claims that “the Soviet Union might be saved by a socialist revolution in the West as our capitalist economy goes into a tailspin.”[3]

After the fall of communism, even the most recalcitrant academics acknowledged that Hayek had been correct that governments cannot manage an economy successfully. And yet nearly twenty years later, the notion that governments can manage an education system remains a respectable notion. Many scholars believe that our experiments in school choice have failed to yield results. Gary Miron, a specialist in the evaluation of the charter school movement, states:

The rhetorical arguments and assumptions about charter schools claim that charter schools will use their autonomy to create focused learning communities and high levels of accountability and that this will result in higher levels of performance. The rhetorical arguments also claimed that charter schools would be innovative, create new professional opportunities for teachers, create new opportunities for community and parental involvement, and have positive impact on other schools by sharing innovative practices and by creating a competitive atmosphere. The rhetorical assumptions about charter schools largely have not been achieved.[4]

To his credit, he notes that the combination of the challenges of starting up a school, the lack of sufficient finances, and constraints on charter school autonomy have prevented charter schools from living up to their original promise. He cites NCLB requirements requiring common outcome measures and traditionally certified teachers, in particular, as unanticipated constraints on charter school autonomy.

After a career in both public and private education, in May of 2002 I moved to Angel Fire, New Mexico, to serve as the founding principal of Moreno Valley High School (MVHS), a charter high school. In a rural area not known for the quality of its education (a UNM-Taos professor told me point-blank that “northern New Mexico students are not capable of passing AP courses), we created an AP program that in the second year of the school ranked us among the top 200 public high schools in the nation and, in the third year, the thirty-sixth best public high school in the nation on the Washington Post Challenge Index. Although the ranking is based on the number of students who took AP tests divided by the number of graduating seniors, our students also achieved a score of 3 or higher at more than twice the rate of the national average. Because of our performance, and our innovative approach to getting there, AP New Mexico co-hosted training by our faculty for AP teachers statewide. Teachers moved from other states in order to teach at MVHS and parents moved from other states so that their children could attend MVHS.[5]

Thus when I read an academic like Miron contrasting the rhetoric of charter schools with the reality, I know that the “rhetoric” of charter schools can be achieved. I know exactly how to do it and could do it again, over and over again, across the country. And I know exactly why it has not been achieved on a broader basis — which has a lot to do with why I am no longer in K-12 education.

Despite a significant Soviet commitment to winning the IT race, by the mid-1980s it was estimated that any decent U.S. university had more computing power than the entire Soviet Union. This was achieved not by creating more rigorous government standards in the IT industry, but rather by allowing millions of college drop-outs to create the companies that changed the world. Indeed, had we restricted entry into the IT field to those with credentials from government-certified licensure programs and only allowed them to create computers according to the 1970s “best practices” as defined by the establishment, the IT industry of today would not be much farther along than it was in the 70s.

When college dropout Steve Jobs saw a new computer interface at Xerox PARC in the 1970s, he didn’t run tests to see how it compared to existing interfaces. He saw it, had a vision for where it could be taken, and set up shop with Wozniak to begin taking it there. Because the barriers to entry and to growth were minimal, he was able to change the world.

Of course, with minimal barriers to entry, there will be many failures. In order to decide whether we want successful educational innovations, we must first decide whether we are prepared to accept many experiments that fail. Had we not allowed the charter schools that we have already allowed, we would not now have the Knowledge Is Power Program (KIPP) and Green Dot charter school chains. If we had allowed more experiments, and given them more freedom, then we would have had more outrageous failures — and perhaps deeper successes. The argument for educational freedom is not that any particular educator will innovate, or even that on average charter schools will be more innovative, but that by allowing freedom some innovations will come into being and, with an environment that allows parental choice, they will win market share. In the extraordinarily harsh environment facing charter school creation, and the limited freedom to innovate that charter schools are allowed (due to NCLB, as Miron notes), it is remarkable that some new and improved charter school chains have come into existence and grow.

The Progressive Policy Institute, hardly a right-wing source, found that in Arizona, the state with the longest and proportionately largest cohort of charter schools, by 2004

… charter school students showed an overall average annual achievement growth roughly three times higher than their traditional public school cohorts. Over four years of elementary school, this difference amounts to about one extra year of growth for charter school students.[6]

They point out that, contrary to expectations, charter schools are not “creaming” the best students; on average they enroll students with lower test scores than public schools.[7] And yet despite their recognition of the achievements of charter schools, they recommended tightening up charter school oversight in Arizona due to several scandals. The tightening up has taken place, with the result that significantly fewer charter schools have opened up in Arizona in the last few years.[8]

The salient metric for evaluating the minor experiments with educational freedom to date is neither whether they have transformed education for all nor whether they have made dazzling innovations. The salient metric is “Have our tiny experiments in freedom allowed a few new and better models to come into being?” They unambiguously have. From this we should rightly infer that we should increase educational freedom to allow more innovations and more growth for the innovations that do take place.

Today governments around the world control K-12 education. At some point, in some nation, that will change. The first nation that creates an educational system that allows educational entrepreneurs significant freedom to innovate will, over time, develop a significant advantage in the global marketplace. I’d prefer that the United States lead this movement rather than follow it.

 

Notes

[1] Cited in Dan Klein, “The People’s Romance: Why People Love Government (As Much as They Do),” The Independent Review, Volume 10, Number 1, September 2005, http://independent.org/publications/tir/article.asp?issueID=42&articleI….

[2] Cited in Mark Skousen, “The Perseverance of Paul Samuelson’s Economics,” http://www.mskousen.com/Books/Articles/perserverance.html.

[3] Cited in Paul Hollander, “Judgments and Misjudgments,” pg. 175 of Lee Edwards The Collapse of Communism, Hoover Institution Press, 1999. The author was Bertell Ollman, writing in PS: Political Science and Politics, September 1991, pg. 40. PS is APSA’s journal of record for the profession. Ollman was given a lifetime achievement award by the APSA in 2001.

[4] Gary Miron, “Strong Charter School Laws are Those That Result in Positive Outcomes,” The Evaluation Center, Western Michigan University, presented at the 2005 Americal Educational Research Association (AERA) meeting, http://www.wmich.edu/evalctr/charter/aera_2005_paper_charter_school_law….

[5] See Michael Strong, “A Tale of Two Charter Schools,” for a more detailed account, http://www.flowidealism.org/Downloads/Two-Charter-Schools.pdf.

[6] Bryan C. Hassel and Michelle Godard Terrell, “The Rugged Frontier: A Decade of Public Charter Schools in Arizona,” Progessive Policy Institute, June 2004, pg. 15, http://www.ppionline.org/documents/AZ_Charters_0604.pdf.

[7] Bryan C. Hassel and Michelle Godard Terrell, op. cit., pgs. 15-16.

[8]See Arwynn Mattix, “Losing Momentum,” Goldwater Institute Today’s News January 15, 2007, http://www.goldwaterinstitute.org/aboutus/ArticleView.aspx?id=1357.

—

Michael Strong is CEO of FLOW, a nonprofit dedicated to “Liberating the Entrepreneurial Spirit for Good.”

After reading his current essay, it occurred to me that there might as well be two Richard Rothsteins writing about the schools. First there’s the Rothstein who trumpets the success of America’s public schools and their teachers. In this essay, and in some of his other publications, Rothstein contends that since World War II the nation’s K-12 education system has continued to perform well above the minimum level needed to sustain a productive and competitive economy. Why then do so many highly knowledgeable people continue to believe that America is saddled with a failing public school system that, in turn, undermines the nation’s economic future? The blame for this confusion, according to Rothstein, rests in the first place with the misdiagnosis of the problem by the authors of A Nation at Risk, the hugely influential 1983 report issued by the National Commission on Excellence in Education.

Rothstein challenges the commission’s assertion, based in part on changes in SAT scores, that student achievement was declining and that American education was sinking under a “rising tide of mediocrity.” He argues that the sharp drop in SAT performance in the 1960s and 1970s cited by the commission should have been attributed to “the changing composition of SAT test takers.” Once the pool of test takers was stabilized, according to Rothstein, average SAT scores for whites and blacks began rising again. Moreover, Rothstein cites a number of other indicators of student academic improvement that the authors of A Nation at Risk either overlooked or failed to anticipate, and that support his contention that the schools were actually accomplishing their academic mission at the time of the report’s publication. He produces trend scores on the National Assessment of Educational Progress (NAEP) showing dramatic improvement in black students’ fourth grade math scores over the past three decades, and adds that blacks made modest gains in reading scores as well. Looking back across this entire period, Rothstein proclaims that our unjustly maligned public schools and their teachers should actually be applauded for the fact that “minority and disadvantaged children … have gained a full standard deviation in test score improvement in a single generation.”

Despite this generally rosy picture, Rothstein is no Dr. Pangloss. He does grant that an unacceptably high rate of academic failure persists among inner city disadvantaged children. In his 2004 book, Class And Schools: Using Social, Economic, and Educational Reform to Close the Black-White Achievement Gap, Rothstein has offered many proposals about how to narrow the still substantial black-white achievement gap. The trouble is that these proposals have little to do with education per se, and are unlikely to ever be implemented. It is here that the other Rothstein emerges. The very public schools that Rothstein lavishes praise on for their crucial role in lifting SAT scores over the past three decades, for raising black test scores, and for somewhat narrowing the achievement gap, suddenly become powerless (and thus blameless) to close the deal.

This other Rothstein concludes that further improvements in academic outcomes for the disadvantaged can not, by and large, be achieved through any current education reform strategy for the schools — not charters and vouchers, not by improving school curriculums or classroom instructional practices, not even by pouring more money into the schools. For Rothstein, school reform essentially has become radical social and political reform. He is convinced that we won’t see further significant gains in academic achievement for the urban poor until we achieve an expansive welfare state. Since this isn’t going to happen, the effect is to let schools and teachers off the hook for failing to raise academic achievement. After all it’s not the schools’ fault that children with poor housing, poor health care, children who come from homes with no intellectual stimulation, “enter school already so far behind they rarely can catch up.” Furthermore, argues Rothstein, “Parents under severe economic stress cannot provide the support children need to excel … .”

This represents, in just one short essay, a remarkable 180 degree turn by Rothstein on the question of how much schools ought to be held responsible for improving outcomes for disadvantaged children. Early on in the essay Rothstein castigates the Nation at Risk commission for ignoring substantial evidence showing how much progress disadvantaged and minority children actually made in the 1960s and 1970s, and he then attributes these gains to the quality of the public schools. At the end of the essay, however, Rothstein criticizes the same commission members for not taking into account the evidence already available in 1983 on “nonschool influences on academic achievement.” The commission is specifically faulted for not considering the widely reported findings of the 1966 Coleman Report that “family background factors were more important influences on student achievement variation than school quality.”

I can understand Rothstein arguing for one of these two positions — either schools are able to significantly overcome family and neighborhood deficits that children bring to the classroom (and therefore ought to be judged by that standard) or they cannot be expected to overcome the social and economic deficits. But he perversely insists on having it both ways. He wants to credit the public school system and its teachers for any gains made by minorities and the disadvantaged, but also tries to shift blame away from the schools and onto the shortcomings of the larger society, whenever those same disadvantaged children fall behind.

In fact, Rothstein is mostly wrong on both counts. First, there’s a good deal of evidence that actually supports the diagnosis of academic decline in the 1960s and 1970s made by the authors of A Nation at Risk. Diane Ravitch, one of the nation’s leading experts on national testing and standards, assembled some of this evidence in her 1995 book, National Standards in American Education. Ravitch shows that average SAT verbal scores plummeted from a high of 478 in 1963 down to the 420s by the late 1970s. It remained there until 1994, when scores were recentered. (Rothstein doesn’t mention the recentering.) Average math scores were 502 in 1963 and fell to 466 in 1980. Ravitch also debunks the notion that the decline in SAT scores could have been wholly explained by changes in the composition of the test takers. She points out that the ethnic composition of SAT test takers was not fully reported until 1976, so there is insufficient data on the beginning and crucial years of the reported decline. Moreover, the College Board’s own study on composition said that only half of the decline could be attributed to the demographic changes in the pool. And, according to Ravitch, a report by the Congressional Budget Office “concluded ‘that the overall drop in achievement’ entailing ‘sizable declines in higher level skills, such as inference and problem-solving is beyond question.’”

Let’s now turn to the question of how much responsibility schools should bear for such score declines, or for more generally failing to raise the academic performance of disadvantaged children. On this issue, Rothstein is guilty of the same charge of ignoring the findings of sociologist James Coleman that he has leveled against the authors of A Nation at Risk. In Rothstein’s case, however, it’s the second “Coleman Report,” issued in 1982, that is ignored. By that year, Coleman had substantially modified his views about the ability of schools to overcome the social deficits of their students. He now concluded that what schools did matter a great deal and could make a big difference in raising the academic achievement of disadvantaged children. In a study co-authored with Thomas Hoffer and Sally Kilgore, Coleman concluded that Catholic high schools produced greater achievement gains for lower income students than the public schools. These gains were largely due to the higher academic standards, the more orderly classrooms and the greater sense of social solidarity found in the Catholic schools.

Obviously we can’t really expect that the 1982 Coleman report — and the many other scholarly studies that have arrived at similar conclusions about the efficacy of urban Catholic schools — would lead Rothstein to seriously consider vouchers, tuition tax credits or other forms of aid to Catholic schools as a means of offering poor inner city children a better chance of success. Still it’s sad that Rothstein doesn’t even acknowledge the existence of the second Coleman report, which profoundly challenges his own theories about the relationship between specific school effects and the possibility of raising the achievement of disadvantaged students. It’s even sadder that, because of Rothstein’s ideological fixation on the primacy of class, race, and poverty differences as determinants of learning outcomes, he seems blind to specific curricular and instructional reforms within the public schools that have had an impact in raising the achievement of disadvantaged students.

There’s the “Massachusetts Miracle,” for example, which I wrote about in the Winter 2008 issue of City Journal. In the past several years, Massachusetts has improved more than almost every other state on the NAEP tests. In 2007, it scored first in the nation in fourth- and eighth-grade math and reading. The state owes this amazing improvement in student performance to its decision to mandate a rigorous curriculum for all grades, creating demanding tests linked to the curriculum standards, and insisting that all high school graduates pass a comprehensive exit exam.

But I doubt that Rothstein is much interested in these real on-the-ground gains for both white and black students. He certainly hasn’t spoken out to protect the gains against the attempts of Massachusetts’ new Democratic Governor Deval Patrick to turn back the reform agenda. Instead, he’s still waiting for the European-style welfare state that will never come.

As ever, Richard Rothstein is eloquent. He is so compelling that I will not challenge the three premises on which he erects his essay. Rather, I will stipulate the reasonableness of his argument that K-8 performance was not declining in the years preceding 1983 (although I would argue that the level of performance was less satisfactory than his essay seems to imply), agree that A Nation at Risk implicated schools in national problems for which they bear little responsibility (at least in the short- or medium-term), and think it self-evident that the report devoted little attention to the role of institutions other than schools (although I find this point far less troublesome than does he). Our disagreement is where one goes from there. Now, schooling and human capital assuredly have a long-term impact on economic performance, as the Hoover Institution’s Eric Hanushek and several colleagues have recently shown; but such a commonsense observation is a far cry from the fevered language casually tossed about in these debates.

My point of departure is Rothstein’s sensible admission that “I do not suggest that American schools are adequate, that American students’ level of achievement in math and reading is where it should be, [or] that American schools have been improving as rapidly as they should.” He is to be commended for explicitly making this point, as many who favorably cite his work have used it to suggest just those things.

While the bulk of Mr. Rothstein’s essay focuses, not on schooling, but on labor markets, unions, and skewering overheated claims about the restorative powers of education, I am inclined, by experience and expertise, to speak more directly to the question of schooling on this anniversary. To my mind, the key lesson and great shortcoming of A Nation at Risk was the understandable but unfortunate ease with which the commission accepted as a given the familiar institutions and practices of K-12 schooling.

Rather than ask why teacher colleges should hold a monopoly on teacher preparation, why technological advances were not yielding labor-saving practices or new efficiencies, or why schools and classrooms serving very different student populations should be expected to operate in similar ways and in accord with similar rules, the commission focused on recommending more academic courses, more instructional time, and higher standards for teachers.

A consequence was that its calls for improved teacher quality, standards, and assessment were defined in accord with familiar nostrums. While not unreasonable, given the commission’s charge, a document less committed to rallying political support would have been freer to ask why we originally embraced the designs that we regard as familiar, whether those arrangements had grown balky with time, and if they continued to serve the purposes for which they were established. It is those questions that even hard-nosed reformers have too frequently skirted in the past quarter-century.

While calling for a better pool of teachers, A Nation at Risk did not push policymakers to revisit the assumptions of our system for attracting and preparing educators. That was unfortunate because, then as now, the existing teacher pipeline is the result of more than a century of compromises and incremental adjustments responding to the exigencies of another era. During the late nineteenth and early twentieth centuries, the teaching profession relied on a captive pool of female labor, treated workers as largely interchangeable, and counted on male principals and superintendents to micromanage the teaching workforce. Later, collective bargaining agreements, more expansive licensure systems, and local and state statutes and regulations were layered atop these arrangements. Even today, would-be reformers too rarely recognize that mere tweaks are unlikely to deliver satisfying results — and that proposals to promote alternative certification or test-based merit pay are baby steps rather than coherent reform strategies.

Preparation programs continue to presume that most aspiring teachers will decide upon a lifelong teaching career while enrolled in college. This made sense forty years ago, when the typical college graduate would hold five jobs in their career and most teachers were college-educated women with few career choices. Today, however, the average college graduate holds four jobs by the age of thirty — making it hard to be confident that new hires can be retained for an extended period, much less for two decades. Meanwhile, the job of “teacher” has remained remarkably undifferentiated, with the vast majority of teachers within a given subject area or grade level treated as largely interchangeable. All fourth-grade teachers in a district generally cover the same subjects, instruct the same number of students, and take on similar ancillary duties. While this may have made sense when little data was available with which to pinpoint teacher strengths or student needs, today it results in a wasteful utilization of talented educators.

Many of today’s “cutting-edge” efforts to reform teacher recruitment and preparation represent nothing more than repackaging outmoded assumptions. For instance, in perhaps the most widely discussed contemporary critique of teacher preparation, Art Levine’s 2006 Educating School Teachers simply accepted that the default model of teacher recruitment ought to entail driving talented undergraduates into extended preparation as part of a five-year degree. Missing was any recognition that students ready to commit to a career in teaching at age 22 might not be the best suited population for the job, that there is no compelling evidence that modal teacher preparation makes a consistent difference in classrooms, or that this approach may dissuade potentially effective entrants while failing to account for the vast pool of talented working adults that may be interested in pursuing a teaching career.

Similarly, the dangers of trying to paste preferred policies atop existing arrangements has been particularly evident in the realm of choice-based reform. For two decades, choice-based reform has been unwisely and deceptively offered by its proponents as something akin to a miracle cure that will boost student achievement, unleash competition, and advance core democratic values.

Along the way, little attention has been paid to the design of these efforts to deregulate a $500 billion a year industry, fostering a vibrant supply of effective providers, nurturing effective mechanisms for quality control, or understanding the multiplicity of arrangements and practices that stifle even nontraditional schools and service providers. For instance, the choice community has had next to nothing to say about the need for venture capital in education, about the ways in which personnel policies and benefit systems stifle new ventures, or about how consumer choices should impact the compensation and job security of educators and school leaders.

One result is that some who were once enthusiastic proponents of “choice” have reversed course and expressed doubts about the viability of educational markets — without ever having stopped to consider all the ways in which simply promoting one-off choice programs falls desperately short of any serious effort to thoughtfully deregulate schooling or promote a coherent K-12 marketplace. Indeed, some have abandoned the choice bandwagon with the same ill-considered haste that marked their initial enthusiasm.

For decades, we have poured money into schooling while seeing few obvious benefits. Current per-pupil spending in constant dollars more than tripled between 1961-62 and 2003-04, from $2,603 to $8,886. Pupil-to-teacher ratios plunged, from 25.1 students per teacher in 1965 to 15.3 per teacher in 2007. Meanwhile, educational progress has been disappointing, at best, over the past quarter-century. This is the epitome of pushing on a string. In an economy marked by new technologies, labor-saving devices, steady growth in productivity, and an evolving labor pool, we are hiring and deploying educators just the way we did a half-century ago. The result is that new investments have not delivered the hoped-for results.

Ultimately, no one should be surprised that arrangements which have haphazardly taken shape over two centuries are ill-equipped to address the challenges or fully exploit the opportunities of the 21^st century.

Mr. Rothstein is right to point out that there are limits to what any institution can accomplish and that it is unhelpful to make grandiose claims on behalf of schooling. At the same time, it is too easy for such cautions to become excuses for anachronistic and inefficient operations. The lesson I take from A Nation at Risk, then, is that we must reject both excuse-mongering and overwrought hyperbole in favor of a steely willingness to revisit the shopworn assumptions and tired verities that have so long characterized school reform on the left and the right.

—

Frederick M. Hess is director of education policy studies at the American Enterprise Institute and author of Common Sense School Reform.

First, thank you to the editors of Cato Unbound for inviting me to comment on the 25^th anniversary of A Nation at Risk, and then inviting me further to respond to reactions by Sol Stern, Michael Strong, and Frederick Hess.

My initial comment asserted that the analysis of A Nation at Risk was flawed in three ways: it claimed that student performance was declining in 1983, when it was not; it claimed that the nation’s economic problems were the fault of schools, when they were not; and it ignored the contribution that students’ social and economic conditions make to their learning. I concluded that Risk’s irresponsible and out-of-proportion condemnation of public schools led to a quarter-century frenzy of ill-considered reforms, culminating in the No Child Left Behind law which has ushered in a new era of “minimum competency” instruction of precisely the kind that the authors of Risk hoped to condemn.

I have made many of these arguments before, and so am familiar with critics like Sol Stern, and to some extent, Michael Strong, who caricature my argument. In the current issue of Educational Leadership, the magazine of the Association for Supervision and Curriculum Development, my article, “Whose Problem is Poverty?” is a comprehensive response to such caricatures. Briefly, my response is this: there are not “two Rothsteins” but one who has a more complex idea, but still not terribly complex, than Stern and Strong are willing to comprehend. My argument is not that school performance is adequate, or that schools are fully responsible only for student successes and not for student failures. Schools are neither fully responsible, nor fully free from responsibility, for student successes and failures. A strategy to substantially improve student achievement, both for middle and lower-class children, will require improvement in both schools and in out-of-school institutions.

The out-of-school institutions include, but are not restricted to, our health, housing and labor market institutions. They also include cultural factors. For example, both Stern and Strong want to argue with my claim that the SAT score decline in the 1960s and ’70s was primarily a composition effect. What careful analyses actually show is that half, and perhaps up to two-thirds of the decline was attributable to composition.[1] Attributing the other third to half is speculative, but probably due to a combination of school and non-school factors: an increase in less serious elective high school courses, less emphasis on writing, lower standards (grade inflation, automatic promotion, less homework), watered down textbooks (more pictures, less text), social factors (increased divorce rates and mothers working outside the home), too much television watching, and national demoralization, disillusionment, and loss of authority (from the Kennedy and King assassinations and the Vietnam War). All of these were probably responsible, according to the College Board panel to which Sol Stern referred, for some unmeasured portion of the non-compositional part of the decline.

Likewise, the rising test scores of the 1970s and ’80s were also attributable both to school and non-school factors. David Grissmer and his colleagues calculated [pdf] that about half of the narrowing of the black-white test score gap in eighth grade math during this period was attributable primarily to two identifiable demographic changes: higher educational attainment of black mothers, and fewer siblings of black children. Grissmer et al. could not explain the other half, but speculated that it might be attributable to improved schools as well as to social forces like school integration.

Michael Strong is correct to point out that the recent gains posted by black elementary and middle school students do not seem to be replicated in high schools. It seems reasonable to speculate that our high school organization is flawed. But it is also reasonable to speculate that other forces come into play — adolescent music that promotes violence, misogyny, and drugs, not academic success; continued labor market discrimination that makes high school achievement and graduation pay off less for black than for white youth, and so on.

Sol Stern and Michael Strong cannot have the simplistic all-schools or no-schools alternative they seek in analyzing the challenges facing our youth. If we want to substantially raise student achievement, both in-school and out-of-school shortcomings need attention.

Stern, in particular, engages in another caricature. He dismisses any consideration of social and economic reform that would bring students to school more ready to learn, by claiming that such reform is nothing other than “radical social and political reform… , an expansive welfare state [that] isn’t going to happen” and that it amounts to “waiting for the European-style welfare state that will never come.” This is too easy a dodge. There is nothing radical about, for example, ensuring that low-income children have adequate health care, providing an increased number of Section 8 housing vouchers so that more poor families have stable places to live, enforcing our anti-discrimination statutes, providing the parents of minority children with greater economic security (for example, by bringing the minimum wage back up to historic real levels, or giving hotel maids the same union protections that autoworkers once had), completing welfare reform by providing high-quality early-childhood care for working mothers, providing urban youth with high-quality after-school and summer opportunities that might compete with gang recruitment, and so on. These modest proposals, which would have a palpable impact on student achievement, are less radical than Richard Nixon’s domestic program. They do not amount to anything like an expansive welfare state. But Stern charges that the effect of such advocacy “is to let schools and teachers off the hook for failing to raise academic achievement.”

This is absurd. We can “chew gum and cross the street” at the same time. If we want to raise student achievement, we should improve schools and provide children with better health, housing, and economic security. The notion that if we do the latter, we can’t do the former, lets political and corporate officials off the hook. We absolve these leaders from responsibility for narrowing the pervasive inequalities of American society by asserting, contrary to evidence and experience, that good schools alone can overcome them.

Finally, I thank Frederick Hess for his graciousness in acknowledging the accuracy of my analysis of A Nation at Risk. Rather than invent artificial points of disagreement, he appropriately took the space which Cato Unbound offered him to put forward his own views about school deregulation in general and teacher quality in particular. Michael Strong also took the opportunity to use much of his space for advocacy of charter schooling as his preferred reform.

I neither have the space for, nor is a forum about A Nation at Risk the appropriate place for, a response in detail to these suggestions. That would be an entirely separate discussion, for another place and time. I acknowledge that I do not know enough about teacher quality to express an informed opinion about what is necessary to improve it. [2] I have written about choice and charters elsewhere and will make only one brief point here.

Michael Strong makes the appropriate claim that “with minimal barriers to entry, there will be many failures. In order to decide whether we want successful educational innovations, we must first decide whether we are prepared to accept many experiments that fail.” This is the crux of the matter. As I and my colleagues wrote in The Charter School Dust-Up, regulation exists for the purpose of setting a floor on quality. An unintended consequence is necessarily to set a ceiling as well. Michael Strong cites KIPP and Green Dot schools, as well as his own experience, as illustrations of how that ceiling can be shattered. But as we now have accumulating data that, on average, charter and voucher school performance is no better than that of regular public schools, if some students are doing much better in charter schools than they otherwise would have done, others are doing much worse. Our debates about choice have typically ignored the latter group. If we deregulate the teacher market, we will undoubtedly attract some very high quality teachers who might not have made it through the certification hoops. We will also attract some incompetents, who will do serious harm to our children. How many of the latter are we willing to tolerate in order to gain the benefits of the former? Before developing an opinion on Hess’ and Strong’s views, I’d like to hear more about how they propose to address this widely ignored aspect of choice.

Notes

[1] I reviewed these analyses in greater detail in my short book, The Way We Were? (Century Foundation, 1998), p. 51-61.

[2] Actually, my impression is that nobody yet understands teacher quality well enough to make coherent proposals for reform. See my colleague Jennifer Rice’s short book, Teacher Quality; Understanding the Effectiveness of Teacher Attributes (Economic Policy Institute, 2003).

I am delighted that Richard Rothstein acknowledges the ways in which cultural influences undermine teen learning and that “it is reasonable to speculate that our high school organization is flawed.” Analyses of American education that fail to acknowledge that we have an especially severe problem at the secondary level strike me as surreal. For some time TIMSS international comparisons have shown that our performance is mediocre in 4th grade, worse in 8th grade, and worse yet in 12th grade — and it is striking that this pattern of relative decline as one goes up the grade levels holds true both in the NAEP longitudinal trends that Rothstein originally cited as well as in TIMSS international comparisons.

We can’t make any significant progress in secondary education until we acknowledge that teen culture in the United States has changed dramatically in the last fifty years, and that this change has mostly been deleterious both to trends in adolescent learning and adolescent public health. By almost every measure public health in the United States has improved in the past fifty years, with adolescent well-being being the one major exception. Relative to the 1970s there have been some upward trends, but from a parent’s perspective the possibility of a catastrophe is still very real. One out of seven 9th grade girls attempts suicide. More than a third of high school aged girls reported feeling sad or hopeless almost every day for two weeks. Depression is tied to risky teen sex and drug and alcohol use. Accidents, homicide, and suicide are the leading causes of death for 15-19 year olds.

William Damon, one of the leading scholars in the “positive youth development movement,” points out that resiliency in adolescents is associated with “persistence, hardiness, achievement motivation, hopefulness, a sense of purpose, and more.” When I pointed out to him that it is easier to provide young people with these factors of resiliency in charter or private schools than in government-managed schools, he agreed.

In this context, it is not surprising that charter high schools disproportionately attract at-risk students. So while we are all in a hurry to raise test scores, it is worth noting that the first cohorts of young people to leave government schools to attend charter schools might well be those who are profoundly unhappy at their previous school. The research on this aspect of charter schools has only barely begun.[1] Last year Scott Imberman published results from examining individual student records in a large urban district. He discovered that students at non-conversion charter schools (i.e. those that were not merely converted from existing public schools) had better attendance and disciplinary records when they attended charter schools, but that these behavioral improvements vanished when the students went back to public schools. This is helpful empirical evidence to validate the far more common survey evidence that students and parents like charter schools better. Maybe it is important to go to a school where one is somewhat happier and less likely to skip school, misbehave, be depressed, or attempt suicide.

For a parent whose child is miserable, checking out the test scores of prospective educational alternatives is the last thing they are apt to worry about. From time to time I run across condescending comments from educational policy analysts about how parents can’t be trusted to choose their child’s school because criteria such as the child’s happiness often count more highly for parents than do academic criteria. The assumption often seems to be that flighty adolescents are switching schools based on school colors or cafeteria food. In states with caps on the numbers of charter schools, thousands of parents are on waiting lists to get their children into charter schools. While the academic results of charter schools are still coming in, the levels of satisfaction are unambiguously higher. For young people in crisis, and their parents, this is worth something.

I’m skeptical that Rothstein’s proposed social and economic reforms would do much to improve adolescent learning or well-being. In the meantime, there are reasons to believe that allowing parents and students more freedom to choose their educational communities does lead to positive outcomes for those students who escape a system that is damaging to them. We should be united in working to create unlimited charter school expansion in every state. Chains like KIPP and Green Dot will grow, new chains better than the existing generation will come into being, both large charter school chains and more professional boutique charter school chains will serve an ever increasing percentage of the population, and if all students have a choice between these new options and their local government school, then only those government schools that meet the needs of the students will survive.

Once we have a national market in education, with competing chains vying for market share (ideally both for-profit and non-profit national chains would exist), we will gradually see larger scale improvements in adolescent well-being, and as rates of drug and alcohol use, risky sex, depression, suicide, homocide, and accidents decline, we will also see more interest in academics and greater improvements in test scores. If we eliminate NCLB and all requirements for teacher certification, the large chains will develop their own teacher training programs, which will be vastly superior to existing teacher training (key to KIPP’s success is a year-long in-house principal training program, again the tip of an iceberg).

Rothstein is concerned about “tolerating” the less than excellent existing charter schools. But at present fourteen urban districts have graduation rates less than 50%; Detroit’s is less than 25%. Is Rothstein equally concerned about “tolerating” the less than excellent urban districts? Bad charter schools close, whereas bad districts keep going and going and going. Despite smaller per-student budgets and various legislative provisions deliberately intended to hobble charter schools, the charter school movement has spawned a few organizations that have proven their capacity to bring quality education to scale.

Twenty years from now, a higher percentage of the U.S. student population will be attending charter schools, and those charter schools will outperform the present generation of charter schools. Twenty years from now, in those states that allow unlimited charter school growth, a smaller percentage of the U.S. student population will be attending urban districts. In states that do limit charter schools (and which do not pass other larges-cale school-choice options), those districts will, on average, be no better than are the same urban districts today. If Rothstein or anyone else is interesting in wagering against these predictions, I’d be happy to work out formal benchmarks for evaluating these claims.

If my predictions are correct, then I can see no morally justifiable excuse for continuing to provide political cover for these failed dinosaurs. Children’s lives are being ruined, at great cost to the public, through institutions that, despite the occasional heroic leader, cannot be fixed. We already know how to allow for the creation of better alternatives. I believe that increasing educational freedom will provide profound benefits to our economy and, more importantly, to the happiness and well-being of those at every level of our society. But because most people are not ready to believe in the most positive outcomes, I’m content to argue to allow for consensus on the elimination of charter school caps. As more and more parents and students escape, more and more will want to escape. And those who have supported their tormenters will increasingly experience remorse not having supported more educational freedom sooner.

Notes

[1] Though it is not focused on charter schools, Les Gallay and Suet-Ling Pong, “School Climate and Students’ Invervention Strategies,” Paper presented at the Society for Prevention Research Annual Meeting, Quebec City, May 2004, presents evidence that when “school climate” is good, adolescents tend to intervene with their peers to prevent risk behaviors. They acknowledge that very little research had been done on school climate’s impact on behavioral issues.

Michael Strong and Richard Rothstein have more in common than they are ever likely to acknowledge. Each of them seems to have largely given up on the prospect of significant achievement gains for disadvantaged children within the nation’s existing education system. Instead of offering a practical answer to the forum’s question — “Can the Schools Be Fixed?” — Strong and Rothstein propose education utopias that just happen to coincide with their broader political beliefs.

Michael Strong assures us that “Once we have a national market in education … we will gradually see larger scale improvements in adolescent well-being, and as rates of drug and alcohol use, risky sex, depression, suicide, homicide, and accidents decline, we will also see more interest in academics and greater improvements in test scores.” (Will teenagers still have to worry about acne?) Moreover, Strong assures us that charter school enrollment is going to take off in the next twenty years and those school districts that don’t jump on the bandwagon are going to see no improvement in student outcomes.

I am impressed with Strong’s confidence in his predictive powers, but until he shows us some evidence on the ground I think I will stick with our imperfect system and place my bets on the kinds of changes called for in A Nation at Risk — for example, restoring a content-based curriculum and maintaining high academic standards and expectations. For all of its faults, “A Nation at Risk” was mostly right about the collapse of expectations in the schools and in spotting the decline in achievement levels of American students. It was prescient in calling for restoring an academic curriculum. Such mundane reforms helped bring about the “Massachusetts Miracle.” With no vouchers and very few charters in the state, Massachusetts students nevertheless surged to first in the nation in fourth and eighth reading and math on the NAEP tests. (See my City Journal article, “School Choice Isn’t Enough.”)

In his current response, Richard Rothstein says that I failed to “comprehend” his argument about what we can expect from the existing public schools. He also claims that I “caricatured” his position by suggesting there were “two Rothsteins.” I will admit to indulging in a bit of hyperbole for emphasis. My concern was that Rothstein was being slippery in simultaneously giving the schools credit for any achievement gains, but then absolving those same schools whenever it became clear that student achievement, particularly in the inner cities, was still appallingly low. Now, given a chance to explain this conveniently shifting perspective, he resorts to banality: “Schools are neither fully responsible, nor fully free from responsibility, for student successes and failures. A strategy to substantially improve student achievement, both for middle and lower-class children, will require improvement in both schools and in out-of-school institutions.” But Rothstein doesn’t say when he would apply the “not fully responsible” standard to schools and when he would grant that schools are responsible for student failure. His standard is a constantly moving target.

Moreover Rothstein hasn’t even told us which in-school improvements he thinks should be part of the strategy for raising student achievement. For example, is he in favor of the federal Reading First reading program which encourages scientifically based reading instruction in the early grades? How does he feel about the fact that the Democrats in Congress (his party, I assume) irresponsibly cut this program that was working to raise reading achievement for disadvantaged inner city students?

By contrast, Rothstein gets very specific, but also very utopian, when it comes to telling us which “out-of-school” social and economic changes will bring about higher academic improvement. He wants universal health care for all children, Section 8 housing vouchers for their families, raising the minimum wage, unionizing maids, and many other social and economic interventions. But even if America experiences the political earthquake that might make such changes possible, how does Rothstein know that his very expensive interventions will even work to bring about significant achievement gains? Certainly our past experience with War on Poverty programs is not very reassuring in this regard. In any event, and while he’s waiting for lefty in the political arena, shouldn’t Rothstein be putting some effort into rallying support for instructional strategies that have been proven through scientific research to bring about significant improvement for poor children?

Dr. Reid Lyon, who headed two decades of NIH-sponsored research in reading science, has said that if urban school districts were guided by the NIH findings in selecting programs for the classroom, the black-white achievement gap in reading could be substantially narrowed. I would put my bet on that strategy, rather than either Michael Strong’s free market utopia or Richard Rothstein’s social welfare utopia.

I found Mr. Rothstein’s discussion of the various critiques somewhat less convincing than his initial essay primarily on two counts. Allowances should be made in light of the fact that he was asked to address three discordant responses and, given that Mr. Strong and Mr. Stern have already weighed in, I will try to fashion my reply accordingly and then extend the discussion a bit.

In putting meat on his proposal for the kinds of social supports he deems necessary to support schools, Mr. Rothstein wound up importing the better part of a full-scale domestic policy agenda under the justification “it’s for the kids.” Mr. Rothstein’s policy recommendations regarding health care, expanded Section 8 housing, increased early childhood care, providing parents of low-income children with greater “economic security,” and so forth may or may not be worthwhile policy goals (to my mind, some are and some are not), but the sticking points are more often program design and ensuring efficacy. Moreover, for those of us skeptical of public bureaucracies, there are deeper concerns here about permanently expanding the reach of government and entrusting bureaucrats with fashioning and providing an array of new or expanded services. Simply asserting that these efforts are less radical than Nixon’s proposed domestic program and that they are essential if children are to succeed in school is an unconvincing response to one who regards Nixon’s domestic efforts as highly problematic and is dubious that the proposed programs would yield the intended results.

While Mr. Rothstein takes some pains in his response to acknowledge the necessity that schools must do better and that this requires qualitative improvements in teaching and learning, it is unfortunately true that many who latch onto his arguments or who make parallel arguments do not add that proviso. The concern is that framing proposals for social reforms as educationally necessary can amount to little more than a rhetorical device that seeks to sidestep thorny debates about fiscal constraints, tax burdens, unintended consequences, and competing priorities.

Indeed, arguing “it’s for the schools” or “it’s for the children” has too often become a convenient rallying cry for broader agendas that may or may not be primarily about schools and schooling. To take but one example, education icon Jean Anyon, a professor in the Department of Education at Rutgers University, argues in her widely cited Ghetto Schooling, “Educational change in the inner city, to be successful, has to be part and parcel of more fundamental social change.” Anyon calls for “an all-out attack on poverty and racial isolation” that requires new regulation of teachers, expanded social services, new federal and state spending on cities, a higher minimum wage, and “improved” teaching and learning. To fund all of this “school reform,” she advocates cutting defense spending and agricultural subsidies, and imposing new taxes on corporations, Social Security, capital gains, and executive pay.[1] Now, thoughtful readers may accept or reject these proposals, but it strikes me as ludicrous to suggest that they ought to be primarily weighed on the basis of their implications for school improvement.

If the Rothstein agenda reads something like a domestic policy agenda rather than a school improvement agenda, it may be because educational advocacy can readily become a politically useful way to package contested social agendas rather than tackle mundane questions like how to provide schools that promote quality teaching and learning. Either way, to my mind, we have now drifted far from any discussion of A Nation at Risk or the forum’s question of “Can America’s Schools Be Fixed?”

With that, let me turn to the second issue — Mr. Rothstein’s relatively relatively sweeping dismissal of “deregulation” as a handmaiden of improvement. He briefly tries his hand at consequentalist response — mentioning Jennifer Rice’s 2003 booklet on teacher quality while omitting mention of more current and compelling work on teacher licensure by scholars like University of Washington’s Dan Goldhaber, Stanford University’s Susanna Loeb, or Harvard University’s Tom Kane — before pointing out that sensible deregulation must be coupled with attention to quality control. This is a useful (if obvious) reminder, and reminds us that rethinking anachronistic models is not just a question of whether but of how. Indeed, the need to push more aggressively into these questions remains a critical one for education thinkers; a splendid recent example of this kind of thinking is provided by Checker Finn’s fall 2007 paper “Quality Control in a Dynamic Sector.”

Interestingly, Mr. Rothstein does not suggest the same degree of caution when it comes to gauging the merits of his various social welfare proposals; there, he relies upon assertion rather than evidence in charging that they are critical to educational improvement. When it comes to revisiting routines regarding teacher recruitment, school governance, or other educational questions, however, Mr. Rothstein seems to demand a different standard of proof to justify alternative arrangements. This twinned desire for grand remedies and for sure things, to my mind, reflects the challenge posed by the forum question and the difficulty that we have in grappling with it, and helps explain the desire to turn to other proposals (relating to unionization, health care, or the minimum wage) as promising solutions.

I’ll try to make my position painfully clear. I am highly skeptical that we can “fix” American schooling in any straightforward sense — especially through public policy or government activity. Indeed, adopting an approach that envisions “fixing” 90,000+ schools or 15,000 school districts lends itself to mechanistic solutions that routinely build upon existing institutions and arrangements. When we address educational shortcomings, we are not fixing any one thing — but are seeking to ensure that thousands of terrible schools, and tens of thousands of mediocre schools, are doing a far better job of educating students of variable skills, attitudes, and circumstances. These circumstances help explain why we have so many times been disappointed by shiny new solutions (included many of those proffered by A Nation at Risk) that have promised dramatic improvements if only we would implement them properly. The need is not for one answer fastidiously repeated, but for a variety of answers that can be adopted where and when appropriate and in accord with local needs and the tools at hand.

This is where deregulation becomes essential — not in the sense of promising a glittering new future if only the old rules no longer apply — but by allowing us to revisit arrangements governing who can teach, how we pay teachers, or who operates schools. At work is not a conviction that there is anything magical about deregulation or alternative arrangments — but the banal observation that it is hard to change public institutions and that our existing arrangements, whatever their merits given the labor markets, demands, and available tools of the mid-20th century, are not equal to the challenges of today. Making the necessary improvements, and doing so in a time frame that can be measured in years rather than generations, requires new practices in terms of instruction, management, and governance. Such an approach necessarily forswears the grandiose ambition of “fixing” America’s schools in any particular window and instead asks two questions: how we can promote improved teaching and learning, and what policymakers and reformers can do to help provide the teachers, arrangements, systems of schools, and culture that can encourage learning.

I see little evidence that conventional state or district school systems, even those with the most energetic and savvy leadership, are capable of pursuing the requisite changes through established arrangements, personnel, and routines. As I noted in my initial essay, this is where I think A Nation at Risk’s narrow focus on tweaking existing practices (such as teacher testing and high school student course-taking) constituted an unfortunate missed opportunity. I do agree with Mr. Stern that there is a clear role for research and research-based practice when it comes to teaching and learning, and that the pursuit of disciplined research and the application of that research (especially in the case of reading) can make a substantial difference.

On the other hand, I am skeptical of what I take to be Mr. Stern’s implication that most existing schools and school systems are capable of effectively implementing that research in the manner he suggests. Here I find compelling Mr. Strong’s dictum that we be open to new arrangements and new ways of operating schools. Ultimately, however, I find Mr. Strong’s focus on schools and school management somewhat constraining, and believe it is essential that we devote as much energy to thinking about how the status quo shapes the educational labor market, the delivery of professional development, the pursuit of research, and the employment of technology as to the issue of “deregulating” school management and governance through choice-based reforms. I think that examining teacher and administrative licensure, teacher preparation, charter school compensation, or the use of research and technology by private schools or charter schools schools, for instance, suggests choice-based reform is not a one-size-fits-all tool for allowing educators and reformers to explore better approaches to familiar frustrations.
Notes

[1] Jean Anyon, Ghetto Schooling: A Political Economy of Urban Educational Reform (New York: Teachers College Press, 1997), p. 13.

Frederick Hess’ latest contribution misunderstands my position in one respect: I would never suggest, and have never suggested, that the social and economic reforms I mentioned should be supported primarily because they would raise student achievement. There are many reasons, for example, to want to ensure that children have adequate health care besides the fact that it would enable them to miss less school because of illness. My only point in this regard is that it is fanciful to think that if only they had better teachers, children will have high rates of success despite poor health including more frequent illness, more frequent iron deficiency anemia, lead poisoning, and asthma; greater family economic stress, including inadequate housing resulting in high rates of mobility, and living in unsafe neighborhoods with high rates of crime and drug abuse. Academic progress is not the only, or even the best reason to fix these things, but if we don’t fix them, expectations that better teachers alone will overcome them make no sense.

Mr. Hess is skeptical that public policy can fix these problems. That is his right, notwithstanding the ability of other industrial democracies to do a somewhat better job in narrowing inequality. But if “the poor shall always be with us,” so too will low test scores and other aspects of disadvantaged students’ inadequate achievement. It is unreasonable and irresponsible to expect schools and teachers to overcome social class differences while exempting every other institution and public official from taking action to do so.

With regard to my claim (and that of Jennifer Rice in her book) that we have made little progress in defining teacher quality, Mr. Hess cites work of Dan Goldhaber, Susanna Loeb, and Tom Kane. What their research has in common is a willingness to use standardized basic skills test scores in reading and (more often, or) math as their sole dependent variable. That we are so willing to entertain the legitimacy of such definitions is one of the tragic legacies of the frenzy stimulated by A Nation at Risk — a frenzy, as I pointed out in my opening essay, that would have horrified Risk’s authors who were terribly concerned about critical thinking skills and other outcomes of education — the “fostering of a common culture,” for example.

Even in their own terms, definitions of teacher quality based on students’ standardized test scores are flawed. Teachers whose students get high test scores (or value added) in math are not necessarily the same as those whose students get high scores in reading. And it is purely speculative to assume that teachers who get high scores in one of these are also teachers who inspire high achievement in history, or science, or citizenship, or inquiry, or democratic habits. I’d say that a teacher is of high quality only if she achieves balanced success in all of these. The research of Goldhaber, Loeb, or Kane cannot help me identify such teachers.

I do note that none of the Cato Unbound respondents, Frederick Hess included, really addressed the question I posed previously. I pointed out that there is now overwhelming evidence that, although there are some wonderful examples of superior charter and private schools flourishing in a deregulated environment, average achievement of charter and private schools does not seem to be substantially better. So I asked:

If we deregulate the teacher market, we will undoubtedly attract some very high quality teachers who might not have made it through the certification hoops. We will also attract some incompetents, who will do serious harm to our children. How many of the latter are we willing to tolerate in order to gain the benefits of the former? Before developing an opinion on Hess’ and Strong’s views, I’d like to hear more about how they propose to address this widely-ignored aspect of choice.

Mr. Hess responds that “sensible deregulation must be coupled with attention to quality control.” To me, ‘quality control’ sounds awfully much like ‘re-regulation.’ This response begs my question. Does Mr. Hess claim that deregulation is not really the answer, but rather that we simply need to substitute a new set of regulations (euphemistically, ‘quality control’) for the old ones?

I find myself in accord with Mr. Rothstein on several counts. We agree that it is naïve to imagine that underlying social conditions will not influence student performance, that it would be a mistake to debate the merits of any proposed domestic policy agenda primarily in terms of the educational consequences, and that it is a mistake to judge teacher efficacy solely in terms of reading and math scores. I know of no responsible thinker who would deny that poverty, crime, cultural depredation, or parenting influence academic outcomes. Indeed, the challenge for schools is to help counter the inevitable inequalities in a democratic nation so that handicaps due to education and wealth do not become a baton casually passed from one generation to the next. In this sense, Mr. Rothstein is obviously correct that broader changes in social well-being will influence educational outcomes (although it appears that we disagree about the kinds of policies likely to yield desirable outcomes). Gratifying throughout this exchange has been Mr. Rothstein’s explicit acknowledgment that — whatever the status of our broader social indicators — we can and should expect our schools to perform better than they currently are. At the same time, Mr. Rothstein is right to caution that such recognition should not extend to casual, utopian expectations that even high-performing schools can overcome any hurdles that social conditions and culture throw in their way.

In the meantime, Mr. Rothstein has suggested a variety of domestic policy initiatives which he believes will improve national well-being and deliver ancillary educational benefits. As I noted previously, I find some of his proposed measures reasonable and others less so, but we agree that these proposals can and should be debated broadly and not primarily as educational reforms. In short, while economic and social developments can make the work of teachers and schools easier or harder, they are not likely to be debated or adopted based on their educational impact. Consequently, just as A Nation at Risk focused on educational responses to educational concerns, I find it advisable to focus on how we may improve schools and schooling — whatever the societal context — rather than on how fiscal, monetary, trade, or social policy might yield social changes that could facilitate school improvement.

With that, I shall turn to Mr. Rothstein’s request that I clarify what I mean when I say that “sensible deregulation must be coupled with attention to quality control.” To answer Mr. Rothstein as directly as possible, yes, I am suggesting that new policies governing choice-based arrangement or the teaching profession will require new kinds of regulation. Market advocates in nearly every sector — from trucking to airlines to telecommunications — have recognized that “deregulation” is a term of art.

In just about every realm except education, even the most far-reaching deregulatory proposals have featured extensive attention to specifying just how regulation would unfold, what it would entail, and what new provisions it would require. Unfortunately, the education debates have tended to proceed without this commonsense caution — yielding polarized debates between those who cling tightly to today’s familiar arrangements and those who find themselves proclaiming the primacy of “choice.” This, I have long contended, has been a problematic consequence of our tendency to focus on the more popular notion of “educational choice” rather than the more meaningful, but less palatable, notion of “educational deregulation.”

The implicit assumption by choice proponents has been that creating charter school laws or voucher programs will be enough to spur the creation of new schools and programs and that competition will ensure quality and responsible use of public funds. In truth, of course, relatively “free” markets are plagued by such concerns. School choice is no elixir. The expectation that mere adoption of charter schooling or voucher plans will prompt the emergence of a raft of promising competitors is not borne out by theory or experience. Proposals to promote school choice have done little to eliminate the hindrances posed by licensure requirements or state reporting systems or to replace them with more sensible counterparts. Similarly, proponents of choice-based arrangements have not done enough to refine authorization processes or protocols or to develop more nimble, market-friendly mechanisms for policing the expenditure of public dollars.

Now, I presume Mr. Rothstein’s subsequent question to be: “So what kinds of new regulation do you propose to replace the old?” Here is where I think we have fallen short, focusing for too long on tweaking old arrangements rather than working to construct new ones (and the primary concern I have about the legacy of A Nation at Risk). Such an effort is not the work of one scholar, but should be a central preoccupation of those in the education community seeking to retool schooling, teaching, and learning for the challenges of a new century. Unfortunately, as I conceded in a previous essay, that effort has not been pursued with sufficient seriousness to date.

While there has been enormous effort among advocates to make the moral case for parental choice, to design saleable programs, and to demonstrate that it can serve both students and the larger democratic community, these efforts have been marked by lax attention to market design. After all, in fields like foreign policy and economics, it is taken for granted that vacuums will not naturally or automatically be filled by effective or virtuous actors. Whether dealing with nascent democracy in Iraq in the 2000s or nascent markets in Eastern Europe in the 1990s, reformers have struggled to establish the institutions, norms, and practices that foster the emergence of healthy markets. Proponents of choice programs have too rarely paid attention to such considerations.

As for my own views, what might constitute desirable new arrangements is something I have explored to some extent in volumes including Common Sense School Reform (Palgrave Macmillan 2004) and Educational Entrepreneurship (Harvard Education Press 2006), and which I not dwell upon here, except to say that there will be a place for appropriate oversight, transparency, and accountability when it comes to the public dollars and public purpose at stake in K-12 schooling. For me, the purpose of deregulation is not to break that link between public investment and requisite oversight but to enable us to reimagine how schooling should be delivered and how (and by whom) that oversight ought to be provided.

Since Richard Rothstein has nothing to say about the criticisms I made in my last post, I see nothing useful in raising those points again. I would only note that he now takes his determinist view about the possibility of in-school reforms raising the achievement of disadvantaged children to new heights (or should I say to new lows). Consider this remarkable statement by Rothstein: “It is unreasonable and irresponsible to expect schools and teachers to overcome social class differences while exempting every other institution and public official from taking action to do so.”

No one is suggesting that schools “can overcome class differences.” However, what some of us in the school reform community do believe is that schools have not even come close to fulfilling their potential for improving outcomes for the disadvantaged. As I pointed out in my last post, there is a strong scientific consensus that the black-white gap in reading achievement in the early grades can be substantially narrowed with the use of programs that follow the protocols of scientifically based reading research. In the current issue of Education Week, E.D. Hirsch reiterates his long-held claim that, “there is still no definite, coherent academic curriculum in the early grades” and that this lack of content knowledge is the source of a significant part of the low achievement of American students. Until the schools start paying attention to what cognitive scientists have discovered about the best instructional methods for teaching reading and the need for a knowledge-based curriculum in the early grades, it is useless to speculate about how much of the academic achievement gap is produced by “social class differences.”

Richard Rothstein is concerned that

If we deregulate the teacher market, we will undoubtedly attract some very high quality teachers who might not have made it through the certification hoops. We will also attract some incompetents, who will do serious harm to our children. How many of the latter are we willing to tolerate in order to gain the benefits of the former?

My response to this is that we first have to acknowledge that we have already attracted numerous incompetents, who are already doing serious harm to our children. Indeed, based on thirty-five years of stagnant NAEP 12^th grade scores, the devastating data on adolescent well-being, and graduation rates below 50% in many urban districts, I should think that Rothstein should be horrified at the serious harm we are already doing to our children. For decades we’ve known that education majors are among the lowest scoring academic majors at universities. We’ve created a system in which the intellectual development of our young people is supervised by one of the least intellectually capable category of college graduates in our society, and Rothstein is concerned that without the education degree “we will attract some incompetents”?

The human resources director of a large urban district, who had previously been a great principal, once explained to me that his job largely consisted of dealing with teachers who came to work drunk. In a district with 30,000 teachers, about one came to work drunk every day. And then there are the infamous “Rubber Rooms” where NYC places its hundreds of incompetent teachers whom it can’t fire, costing the district tens of millions per year. And we can all look forward to the results of the “Ten Worst Union-Protected Teachers” contest. I know it is unseemly to point out that there are now numerous incompetent teachers, but I have known hundreds of students whose lives were “seriously harmed” in the existing system, including attempted suicides. Many young people experience school as a degrading prison, and occasionally the teachers are part of the problem.

The salient question is will we do more harm, on average, allowing parents and students greater decision-making powers in education than we are already doing. It seems unlikely that parents and students would do much worse and highly likely that they will do far better over time once we have created a competitive education market.

There are several features of a prospective competitive market in education that seem to confuse some observers. First of all, education is often somewhat of a natural monopoly in many places; insofar as families have to bear the costs (in time and money) of transporting their children to non-neighborhood schools, there is often a significant implicit tax associated with all but the geographically closest schools. As a consequence, most local education markets are oligopolies rather than competitive markets.

In order to overcome the quasi-natural monopoly of local schooling, we need national educational chains that compete for the opportunity to create new schools, with national brand-name appeal and associated capitalization and focused R&D to support their particular brand advantage. KIPP, Green Dot, and others are the beginning of this trend, but they are all still developing their ability to bring quality to scale. John Merrifield explains why large scale school choice is a crucial prerequisite to reaping the benefits of innovation and how the few existing choices are largely irrelevant.

Second, existing public and private schools with campuses have an enormous competitive advantage over new charter and private schools that have to pay rent or mortgage costs, often allocating 10-20% (or more) towards their facility. Thus except for those few states that allocate capital budgets for charter schools, new charter schools must pay for their teaching staff with a state per-pupil allocation lower than the public schools and then subtract from that facility expenses that are typically paid for through bond measures. Established private schools may have campuses that were donated many decades ago or that were paid for through capital campaigns which are only possible due to an extended alumni base. In effect we have a local oligopolies with existing “firms” highly subsidized relative to new market entrants.

Thus in evaluating the performance of charter schools on average, one should note that not only are they educating a more at-risk population, they are also doing so with less funding. Bringing charter schools to full funding parity with government-managed schools, including capital costs, will help to advance the rate at which charter schools succeed and create a competitive environment.

The third feature of the existing education system that prevents an innovative market from developing is the fact that the government system acts as a dominant operating system, similar to that of Microsoft, but with a higher market share that is legislatively financed and enforced. Because of requirements such as teacher certification, statewide textbook adoptions, and state standards in curriculum and testing, it is difficult and costly for any single school to create and finance an alternative. We will need large chains with their own R&D budgets with freedom to create their own educational system in order to receive the ultimate benefits of choice.

It is worth examining the discrepancy between the research community’s understanding of the dynamics of school innovation by means of markets vs. the entrepreneurs’ approach to innovation. Let’s examine teacher training as a case in point.

When I read Rothstein’s note that “nobody yet understands teacher quality well enough to make coherent proposals for reform,” I am reminded of the extraordinary statist biases of the research establishment. What he means of course, is that none of those academic researchers sifting through data “understands teacher quality well enough.” It is simply not true that “nobody” understands teacher quality. Those entrepreneurs who discover “teacher quality” appropriate to their academic programs will find the “teacher qualities” they need to support their programs. (For instance, I almost always avoided education majors and preferred to hire liberal arts graduates, lawyers, business professionals, etc.)

In order to understand how absurd Rothstein’s notion is (though he is in the good company of the entire educational research establishment), note how bizarre it would have been for academic researchers in computer science or business management to attempt to have established “coherent proposals for reform” that would have identified the entrepreneurs who created the IT revolution. Picture, if you will, thousands of earnest academics holding conferences and publishing papers on the credentials needed to identify “qualified” IT professionals in 1970. One simply doesn’t discover people like Steve Jobs and Steve Wozniak or Linus Torvalds by means of top-down academic research.

To take a different example, most European nations have state-supported churches and very little church attendance. The U.S. has instead a vibrantly entrepreneurial religious sector and is, not coincidentally, the only developed nation that has remained intensely religious. Imagine, in an alternative universe, thousands of earnest European research scholars sifting through the data on existing church attendance in Europe in an attempt to discover “coherent proposals for reform” that would increase religiosity. They would never find the Joseph Smiths, William Seymours, and Billy Grahams that have created and sustained much of our religiosity. Seymour, a former slave, created Pentacostalism in 1906; today there are now estimated to be more than half a billion adherents globally, and nearly 20 million more every year. One doesn’t analyze the data to find the qualifications needed to create a successful church. Why should we expect the data to reveal who is qualified to create a successful school?

As Hayek explained long ago, competition is a discovery process. In order to allow parents the opportunity to discover the right teacher quality standards, we need to create a level playing field through which various educational entrepreneurs will select various kinds of teachers. If their organizations are large enough, and if we can break the existing government school certification standard, these entrepreneurs will be able to provide customized training programs that are increasingly suited to the needs of the parents and students they serve. Existing academic teacher training programs will largely become obsolete.

The notion that researchers are going to discover “teacher quality” is itself part of the problem. The fatal conceit that professors analyzing data can discover “teacher quality” is exactly analogous to the fatal conceit that the government analysts can create a five-year plan that produces enough shoes or tractor parts. In a dynamic entrepreneurial market, the relevant information is always intimately entwined with the particular vision of a particular entrepreneur in a particular place appealing to a particular demographic. In order for learning to be as dynamic and alive in the U.S. as is religion, we have to let educational entrepreneurs define their own quality standards, and let parents and students decide what “quality” is.

Stern regards the possibility of significantly better education as utopian, and has decided that government imposed standards are a more pragmatic path. It may be that, occasionally, some improvements result from the approach that he advocates; even the Soviet Union had a few successes here and there. But as Lisa Snell at Reason points out, the Massachusetts Miracle lauded by Stern has resulted in a large and increasing gap between low-income and high-income students. This is no more inspiring than Rothstein’s cheerleading for NAEP longitudinal score growth in which 4^th and 8^th graders improve but 12^th graders get worse.

It seems especially odd that Stern should consider unlimited charter school growth to be “utopian.” A survey done and marketed by PDK, staunch public school advocates, shows that from 2000 to 2007 support for charter schools increased from 42% to 60%. More significantly, the rate at which public-school parents are coming to favor charter schools (up 23 points) is growing even more rapidly [pdf] than that of those who do not have children in school (up 16 points). When given an opportunity, even union teachers are desperate to escape to charter schools: When the United Federation of Teachers opened up their first charter schools in NYC, they received 1561 applications for 27 positions. In L.A., it was reported that nearly 700 union teachers had applied for leaves of absence to teach in mostly non-union charter schools, and only 176 had returned. As one such teacher explained, “I felt like a spark trying to ignite that kept sputtering out.” It is thus only mildly surprising that Barack Obama has openly endorsed charter schools. When the leading candidate for the teachers’ union party openly supports charter schools during a hotly contested primary race, unlimited charter school growth can hardly be considered “utopian.”

Max Planck is credited with saying “Science progresses one funeral at a time.” As bit-by-bit we continue to get more educational freedom through diverse small victories, the old romance with government-managed schools will continue to decrease. As bit-by-bit these small freedoms result in a few more improvements, the trends will accelerate. At some point we will have our Berlin wall moment, and the long reign of government-managed schooling will collapse. And finally we will experience a “Silicon Valley of Education,” in which many thousands of educational entrepreneurs compete to provide constantly improving forms of education to diverse niches.

But change will not come from those who research education, because they are always taking a frozen snapshot from two years ago, and arguing against greater freedom because they have not yet identified how to do it right going forward. But the vast majority of researchers have never been the change agents, nor are they likely to discover the change agents, nor are they even very capable of understanding the full set of conditions required for change agents to flourish.

It is endlessly odd to me that advocacy for innovation, or the conditions required for innovation, should be considered ideological. It is endlessly sad to me that education, where innovation is most needed for improving human well-being, is the realm in which there is the most dogged opposition the conditions needed to innovate. In every realm of human endeavor, allowing millions of people the opportunity to discover their own solutions, and create institutions devoted to the ongoing development and dissemination of those solutions, has out-performed management by politicized government entities. Schooling is the one stronghold in which, oddly, even many conservatives continue to believe that government can do it better.

Steve Wozniak, who spent years teaching in public schools after founding Apple and becoming wealthy, concluded

… schools close us off from creative development. They do it because education has to be provided to everyone, and that means that government has to provide it, and that’s the problem.

Wozniak is not an ideologue; he is a brilliant, generous innovator with great integrity. If government provides schooling, it can’t be innovative as an institution nor can it encourage our children to be creative. The sooner the next generation realizes this, the sooner we will no longer be a nation at risk.

It’s sad that Michael Strong has to end what has been a useful debate by egregiously distorting my position on the best strategies for school reform. He starts his critique by claiming that I believe that “the possibility of significantly better education [is] utopian.” In fact, throughout this exchange, and in all of my writing on education, I have made it clear that we can, indeed, “significantly” improve education outcomes by adopting certain instructional and curriculum reforms. I have also argued that school choice programs have improved education for the disadvantaged. The only time I used the word “utopian” in my previous posts was in characterizing Strong’s statement that free markets in education will lead to “lowering rates of drug use, risky sex, depression, suicide, homicide, and accidents” for the nation’s teenagers. Indeed, I don’t know of a better operational definition of “utopian.”

Strong strays even further from the facts and reality when he resorts to citing Lisa Snell’s article in Reason to show that I misinterpreted the “Massachusetts Miracle.” Snell’s article makes the claim that the extraordinary test score gains in Massachusetts are somehow less of a triumph for the state’s curricular reforms because white students made greater gains than blacks, and therefore the gap between low-income and high-income students grew. This is supremely silly and raises the question of whether free-market education experts are now joining the ranks of the politically correct on questions of race and schooling.

Actually, if Strong is really interested in a useful example of a state with a huge academic achievement gap between white and black students he should look to Wisconsin. On the 2007 NAEP tests in 4^th and 8^th grade reading and 8^th grade writing Wisconsin had the highest gap between black and white students of any state in the country. Wisconsin’s black students also had the lowest absolute scores in the nation on those three tests. Need I remind Strong and other supporters of free markets that Wisconsin also has the biggest voucher program in the country?

Strong also distorts my position on charter schools. I didn’t say that it’s utopian to expect “unlimited charter school growth.” What is utopian is Strong’s belief that urban districts in states that don’t have charter schools will not see improvements in education outcomes. To me this is wildly speculative and utopian, because not a shred of evidence is offered to support this prediction. I also think it’s mildly utopian for Strong to put his hopes for expansion of charter schools on the election of Barack Obama. All I can say is, lots of luck.

Back on earth, there are possibilities for significantly improving education outcomes for minority students if we do the right thing in the classroom. The right thing involves, at the least, using methods for teaching kids how to read that have been validated by scientific research. This should be as controversial as using the findings of science for treating diabetes. Yet I haven’t heard either Michael Strong or Richard Rothstein endorse this obvious and easily implemented school reform.