defining ideas
education next
china leadership monitor
policy review
hoover digest
July 30, 2007

Held Back

No Child Left Behind needs some work

Passed by congress in late 2001and signed by President George W. Bush one year after his inauguration, the No Child Left Behind Act ( nclb) is the most ambitious federal education statute ever. It overhauled and expanded Washington ’s role in education, rewrote the rules, and set out to boost pupil achievement, narrow a host of “learning gaps,” and assure every student a “highly qualified teacher.” The law’s main engine, however, is an historic attempt to impose a results-based accountability regime on schools across the land. After five years of experience with a statute that aims to produce “universal proficiency” (in math and reading, mainly in grades 3–8) by 2014, and with reauthorization looming, it’s time to draw some conclusions about how nclb  has unfolded on the ground — and how it ought to be changed.

Though  nclb is routinely labeled a “Bush” law — in no small part because the White House has proclaimed it a great domestic achievement while prominent Democrats have been far more equivocal — in fact its provisions are a Rube Goldberg-like assemblage of administration proposals, “New Democrat” schemes, liberal nostrums, and proposals and cautions introduced by countless other constituencies, all superimposed upon programmatic habits, architecture, and rules that had accumulated like reefs in federal education policy since lbj worked in the Oval Office. Indeed, from the outset it was clear that implementing this mish-mash would recall the phrase that the late Daniel P. Moynihan used to describe lbj’s multifaceted community action program: “maximum feasible misunderstanding.”


Nclb’s accountability engine  is driven by two pistons: insisting that states adopt systematic standards and testing for schools and districts; then intervening in ineffective schools and districts while also providing immediate relief for their pupils. Congress charged states with defining and adopting the standards and tests while spelling out in considerable detail the remedies that states and districts are responsible for providing. Yes, the statute contains hundreds of other provisions. But if its two main pistons aren ’t firing well, this complex engine won’t budge the massive barge that is American schooling — much less render it a more agile craft.

On the assessment front, nclb requires that all public schools annually test all their students in grades 3–8 in reading and math and that every state measure whether its public schools are making “adequate yearly progress” (ayp) toward universal proficiency in those core subjects by 2014. Each school must meet steadily rising goals in every demographic subgroup: by race, disability, English language status, and so on. Schools are then evaluated on the past year ’s achievement of each category in which they enroll a minimum number of students (that number being determined by the state, subject to federal approval). If a school fails to “make ayp” in any of those categories for two consecutive years, it is judged to be “in need of improvement”; if that school receives federal Title i dollars, it is then subject to a cascade of sanctions and interventions that grow more draconian with each additional year of failure.

The testing issues that arise from nclb have received extensive attention. State and federal officials and platoons of academics have spent five years debating cut scores, proficiency targets, confidence intervals, and so forth. Receiving far less attention has been the remedies side of the equation. Here we address nclb’s own version of the seven-year itch, whereby a Title 1 school that fails to make ayp is subject to a parade of stiffening interventions designed to change it and give new options to its students.

The remedies

What do the remedies actually entail? If a federally aided school (“Title 1 school”) fails to make ayp for two consecutive years, its students are supposed to be offered “public school choice,” enabling them to attend other public schools in their district. Under that provision, the district is to provide each such child with a choice of alternative public schools (including charter schools) that are making suitable progress. Meanwhile, schools identified for improvement must draft or update a multi-year improvement plan and receive technical assistance that addresses the problems that led them to fail to make ayp.

If a school falters for a third straight year, its district is supposed to provide pupils with the opportunity to obtain “supplemental educational services” — essentially free after-school tutoring — from diverse providers, including private firms. This tutoring is to be paid for with a portion of the school ’s federal dollars — a sort of minivoucher. In addition, because the remedies are cumulative, qualifying students continue to be eligible for nclb school choice.

If a school fails to make ayp for a fourth year running, its district is supposed to take “corrective action,” replacing school staff, implementing a new curriculum, reducing the school’s management authority, extending the day or year, appointing an outside expert to advise the school, or reorganizing the school. Pupils in such a school also remain eligible for nclb school choice and supplemental services — and do so as long as their school fails to make ayp.

If it fails to make ayp for a fifth consecutive year, its district must (during year six) prepare a “restructuring” plan for it. This may include reopening it as a charter school, replacing its principal and staff, contracting with a private management company to run it, turning it over to the state, or any other “major restructuring” of school governance. If the school fails to make ayp for a sixth year, that restructuring plan is to be implemented by the beginning of the following (i.e., seventh) school year.

On paper, this all proceeds in an orderly and familiar top-down sequence, with federal rules spelling out what states are to do, states telling districts what to do, and districts working with their individual schools. State education departments are charged with setting standards, creating tests, intervening in faltering districts, and generally watching over matters. That hierarchy of responsibility — from Washington to state capital to local school system to schools — has been the basic architecture of federal education policy for decades. But it was never designed to support a results-based accountability system, to make effective repairs to faltering schools, or to function in an environment peppered with education novelties like charter schools, home schooling, and distance learning. Yet nclb’s architects never paused to ask whether a hierarchy decently suited to distribute money according to certain formulae and rules could also handle the challenges of a very different and much more aggressive federal role.

The nclb remedies show the influence of at least four different theories of education change. One is to boost achievement by helping children to exit weak schools for stronger ones — and to obtain remedial tutoring from education providers other than the (ineffectual) school itself. A second is creating market-style competitive pressures (via the exodus of students and dollars) to motivate weak schools or districts to improve. A third is to effect desired improvements by furnishing low-performing schools with technical assistance, guidance, and support in developing and implementing their own reform plans. And the fourth, in the event that the preceding steps do not yield the desired gains, is that states or districts are forcibly to “restructure” faltering schools. Alternatives. Competition. Help. Intervention. Over the seven-year time-frame elaborately constructed by nclb, and with various options at each step, all four types of education reform — and the quartet of change theories that undergird them — were to be imposed on schools “in need of improvement.”

Thus, nclb’s remedies are more an assemblage of reform ideas than a coherent scheme. They were adopted with scant attention to how they fit together, the resources or authority they would require, or whether they could be sensibly deployed through the available machinery. The result is messy and confusing at best, a patchwork produced by much compromising on Capitol Hill, all part of an 1,100-page cornucopia of programs, incentives, interventions, and sops to ideologies, advocates, and interest groups.


How well are these complex remedies working? Based on the research conducted for a forthcoming book, here ’s how we size things up. Eight points are key.

First, with the gift of hindsight, educational accountability under nclb turns out to be less about any conventional notion of school improvement or “reinventing government” and more about fealty to a noble, even millennial aspiration. Rather than simply seeking to ensure that schools and school districts serve their students effectively, nclb’s authors set the extraordinarily ambitious goal that every single American child will be proficient in reading and math by 2014. In so doing, they took the language and mechanisms of standards-based education reform and married them to a policy agenda that owes more to Great Society dreams and the civil rights initiatives of the     ’60s than to any contemporary vision of disciplined education governance. In short, educational accountability à la nclb is more a form of moral advocacy than a sensibly designed set of institutional improvement mechanisms and incentives. Indeed, nclb’s aspirations call to mind the grandly ambitious “National Education Goals” that the nation’s governors and the first Bush administration crafted in 1990. Those millennial targets, including the declaration that the U.S. would lead the world in math and science by the year 2000, quietly sank from view and are now mostly remembered for their naïveté and earnest unseriousness.

Yet the political calculus and compromises that produced nclb meant that its soaring aspirations were coupled with outdated machinery, weak sanctions, and uncertain interventions. Rather than a sharp slap to the face of districts and schools that lose students, for example, nclb choice results in no consequences for any educator at the school or district level, and the only pressure it brings on districts is the mandate that they may have to spend some of their Title 1 dollars for transportation. The tutoring minivouchers known as “ses” may mean that a small portion of Title 1 money flows to private providers, but there’s no evidence that this spending has cost a single educator a job or pay raise — and most of the ses outlays, even by for-profit providers, flow back to local teachers who teach after hours for these firms. In short, as designed, the rhetorical nods that nclb’s proponents pay to competition are empty flourishes.

Unrealistic goals focus employees on compliance and on keeping out of trouble. Failure is inevitable.

Effective behavior-changing regimes are rooted in realistic expectations. Such pragmatism leads self-interested workers to take goals seriously, to focus relentlessly on outcomes, and to employ the levers at their disposal to produce those outcomes. When goals are patently unreachable, the logic of accountability changes in important ways. If managers and workers know they are unlikely to reach the target, their primary motivation becomes avoiding trouble when they fail. Since unrealistic goals make failure inevitable, they have the perverse effect of focusing employees on compliance and on keeping out of trouble — rather than on the ostensible goals. We sense — and fear — that nclb’s aspirational framework has created a system in which the prospect of likely failure by many schools gives educators more reason to focus on obeying rules and following procedures than on delivering results.

Second, almost everywhere, compliance-style activity is underway as state and local officials attempt, sometimes cynically and sometimes in good faith, to fulfill nclb’s formal requirements, fill out the forms, and keep the money flowing. nclb’s remedies don’t actually require states, districts or schools to do better academically; they require only that states and districts comply with the remedies spelled out in the law. The law is frequently misunderstood as demanding student academic proficiency. In fact, it only requires that states and districts comply with its guidelines regarding the reporting of data, spending, planning, and adoption of interventions. So long as officials do those things, whatever their progress or nonprogress on reading and math achievement, they are home free. While new data show growing compliance with nclb in urban districts, state and district analyses document how unenthusiastic and formulaic most such activity appears to be. This is particularly perilous for the remedy provisions, as the impact of public choice, ses, or school restructuring depends almost entirely on earnest and imaginative execution.

After strong initial promises, the Department of Education accepted half-measures and backed away from fights.

Federal officials bear much responsibility here. Lawmakers assembled their bipartisan majority by agreeing to vague statutory language, and the executive branch has generally been unwilling to stand firm on implementation. After strong initial promises — that “lack of capacity,” for example, would not be an acceptable excuse for failing to provide school choice — the Department of Education accepted half-measures and backed away from fights with states or districts. Some of this may be due to the particular challenges of nclb and political decisions by the administration, but much can be attributed to decades of inertia in a bureaucracy built around regulatory compliance.

Third, nclb’s remedies (notably school choice and ses) are not, in fact, being much used or are being deployed in their mildest forms (e.g., school and district restructuring). Little nclb-inspired choice is occurring; ses participation rates, though higher than before, remain laughably low in most places; and we see scant evidence of systematic school restructuring. Nor do states and districts appear to have the capacity  to restructure more than a handful of schools, certainly not the hundreds, soon to be thousands, that nclb has flagged as warranting such interventions.

Yet separate and apart from nclb, we find a lot of standards-based and choice-based reforms occurring. Most of this action is state-initiated, some locally inspired, belying the view that nclb is the only game in town or that tempering its aspirations would slow the nation’s education renewal.

In practice, nclb may have caught a reform wave more than causing one.

Fourth, on the plus side, there are places where nclb has created political cover for state, district or school officials to take bolder actions than they otherwise might have taken. Scholars David Plank and Chris Dunbar, for example, suggest that the imagined threat of nclb restructuring in Michigan has fostered a sense of urgency at low-performing schools. In some locales and schools, it has brought exigency and focus that were previously lacking. It is possible that the actual design and operation of the remedies aren ’t as important as their mere existence — and the mythology that envelops them. We may be witnessing a “Wizard of Oz” phenomenon in which nclb matters not for what it actually does but because it creates a scary presence “behind the curtain” that can be used to prompt otherwise painful changes and blamed for difficult decisions.

But can such an illusion last? As the nclb wizard is revealed to wield little real power, the “political cover” he supplies is sure to fade. As it becomes apparent to all that neither federal nor state nor local officials have much idea how to “fix” schools nor any desire to sanction perpetual low-performers, educators and local officials will see the little man behind the drapes struggling futilely to scare them, and the law ’s remedies will lose credibility. This may also undermine non- nclb accountability in the process.

NCLB sows confusion where there are discrepancies in school ratings between state and federal models.

Fifth, nclb works very differently from state to state. In some, its prescriptiveness impedes the state ’s own approach to standards-based reform, as in Florida, where it mandates the restructuring of some schools that simultaneously earn honors grades from the state ’s own accountability system. Certainly nclb  sows confusion where, as in California, there are discrepancies in school ratings between state and federal models. In particular, the law ’s crude pass-fail grading system and its mandated restructuring cascade are complicating homegrown improvement strategies in such leading reform states as Florida and Massachusetts.

Less is known about nclb’s potential stimulus effect in those states — such as Oklahoma, Mississippi, South Dakota, Nebraska, and Idaho — that had done little by way of standards-based, accountability-style education reform as of 2002. The task for nclb proponents is to demonstrate that the benefits in such states are substantial enough to outweigh the difficulties wrought in states that had already forged coherent accountability systems of their own.

Sixth, ses poses a particular challenge. Districts don’t like losing money (which happens when parents opt for outside providers to deliver ses) but they don’t mind their teachers earning extra dollars on the side. Providers of tutoring are loath to be regulated by hostile state authorities and may not teach in ways aligned to district curricula or state tests — but definitely want to maximize their enrollments. Parents get information through the school, which has little reason to steer them to outsiders who will take district funds and not necessarily help schools make ayp.

Across the land, evaluation and quality control have received scant attention, parent outreach has been half-hearted, and the ses machinery has proven balky. This development is not simply a consequence of lackluster implementation and perverse incentives; it ’s also a result of the hodgepodge of theories and compromises baked into the ses program itself. One faction views ses as prelude to full-fledged vouchers and, believing in competition as a tool of school improvement, holds that allowing private tutoring providers to compete for Title 1 dollars will press low-performing districts to improve. A second group regards ses  chiefly as a way to provide extra instruction to youngsters in low-performing schools, crafted to avoid harming either those students or their schools. A third cadre sees it as a mechanism for nurturing an industry of for-profit and nonprofit providers and innovative approaches to serving at-risk students.

Unfortunately, these disparate theories and strategies work at cross-purposes. In the end, there has not been competition, demonstrably effective remediation, or much evidence of innovation. We conclude that the ses  provision is unworkable as presently designed.

Seventh, parts of nclb are providing a wealth of data, sunlight, and transparency with regard to school-, district-, and state-level performance. This has empowered public officials and parents to lean on schools and districts and has equipped educators with new tools. We welcome the flood of additional information and suspect that, in the long run, nclb’s greatest accomplishment may be the school-performance data it furnishes to parents, educators, and state and local officials.

Yet even as it generates reams of information on pupil achievement and school performance, nclb does a woeful job of gathering data pertaining to its own implementation, the deployment of its “remedies,” and their effectiveness. In other words, nclb is good at telling America how students and schools are faring today in math and reading but bad at telling anyone what they are doing, how they are changing, what remedies are being used in which ways where, and what difference (if any) those remedies are making. This means that very little can be learned about what to do more — and less — of by way of altering  the present achievement picture.

In the private sector, outmoded competitors can be rendered irrelevant. In the public sector, this almost never happens.

Eighth and finally, the challenges posed by nclb’s remedies raise fundamental issues of federalism and whether the circa- 1965 architecture, so reliant on state education agencies and local school districts for implementation, is even suitable for a “reform” regimen where it’s the behavior of those very agencies that is to be altered. Such change-forcing steps require somebody  to drop the hammer. Today, that somebody is the district in the case of schools and the state in the case of districts; yet neither hammer-wielder has shown much inclination to take politically tough steps. Meanwhile, colleges of education have done an abysmal job of providing school or district leaders with the skills to turn around troubled schools or districts, while state licensure regimes ensure that nearly all principals and superintendents get trained in those institutions.

In the private sector, outmoded competitors can ultimately be bypassed, bankrupted, restructured, bought out, or otherwise rendered irrelevant. In the public sector, this almost never happens. Yet nclb’s remedies seem to take for granted that it will. That is a strategic error that threatens to undo the promise of the restructuring provision.

Thus, we find a truly mixed picture. Other parts of nclb  may or may not be working, but its “remedies” are at best iffy when it comes to catalyzing changes that boost school effectiveness and student achievement. At the same time, those remedies may impede school effectiveness, confound promising practices and functioning reform programs, and ask for such unrealistic acts that they breed cynicism.


There are two ways to draw guidance from what’s been learned to date about the nclb remedies. One is narrowly to ask how Congress might retool them to make them work better within the law ’s current structure and basic design. This is the path that Washington is most apt to follow at reauthorization time — the path of least resistance. It is the path charted by President Bush in his 2007 nclb proposals, by most of the congressional leaders who have signaled their intentions, and by the Gates-funded, Aspen-housed National Commission on No Child Left Behind, which offered 60-plus separate recommendations in mid-February. It is also consistent with a long tradition of refining major laws on the basis of experience rather than fundamentally reconstructing them. And it places boundaries around the political tussles ahead.

A very different approach is to rethink the law’s basic design. Though politically less likely, it would do children more good. Our eight recommendations draw upon both the incremental and more radical perspectives.

First, federal policymakers ought to be more realistic regarding what they undertake in k–12 education, acknowledging that Uncle Sam isn’t good at finely calibrated, escalating sanctions of the kind that nclb currently expects states and districts to execute. His capabilities are better aligned with a simpler model that leaves more room to state and local policy entrepreneurs to devise creative solutions. This may be an unwelcome message to advocates frustrated by the glacial pace of education change and eager to make a big difference in one giant leap, but it ’s a message they should heed.

Rather than imposing an incremental cascade of remedies, Washington should insist that states label schools that need help; that they act to strengthen such schools; and that they shut, replace or turn inside out those schools that resist improvement. That kind of mission is better attuned to Washington ’s strengths — and more closely resembles the recipe that Uncle Sam has used to excellent effect in reforming welfare.

In education, the key is to make clear that states — working through districts as they see fit — must assist schools that they identify as low-performing and help to clear away the barriers that stymie change agents from improving them. Simultaneously, choice programs should provide decent options to students — but for the kids’ sake, not with any expectation that they will improve malfunctioning schools. After a certain point is reached, however — when schools have gone several years without showing sufficient improvement — the interventions that follow should be truly draconian. Such schools should be closed (with their buildings recycled to house new schools). In other words, there ought to be a presumption of good intention in the initial years, with the law geared to providing essential political cover and local muscle to clear-eyed reformers. Thereafter, the law should be designed to replace persistently ineffective schools. We believe this sequence — presumption of competence followed by swift and sure action — can provide the clarity of purpose and design that is currently lost amid nclb’s scattered components and disparate theories of action.

Parents, educators, and officials should be able to see clearly how their school, district, or state is doing.

Second, though Washington should trust states to turn around their own schools, all should be measured against a single set of national standards and uniform national tests, at least in the core subjects of math and reading. (This presupposes that such standards and tests can be competently and coherently designed — and not by politicians.) This would permit parents, educators, and officials to see clearly how their school, district, or state is doing. That strategy has the great merit of sorting out roles and responsibilities in the school reform domain, though we recognize that “national testing” will prove unpalatable to many Republicans and that “trusting states” will appall many Democrats. It is urgent, however, to distinguish the actions that the federal government can do well from those that must be entrusted to others.

Note, too, that the “others” do not necessarily have to be state and district education agencies, as assumed by the 40-year-old federal aid-to-education hierarchy that nclb  inherited. It may be that governors, mayors, nonprofit organizations, for-profit firms, regional authorities, and other such entities are better suited to handle the task of school reform. Washington might envision a contracting and multiple-provider regime in which a host of nontraditional players are empowered to provide and monitor schooling.

Because states differ so dramatically in their approaches (and commitment) to education reform, Washington should provide greater freedom for states that earn it. States that can document success in providing options to eligible kids or delivering effective school improvement deserve greater flexibility and the opportunity to serve as models and labs for other jurisdictions.

Third, instead of mandating “one step per annum” over a seven-year sequence, nclb should afford states and districts the option of interventions that span several years. So, for example, if a school fails to make ayp (properly calculated), even for a single year, it would go into “corrective action” and  its students would have access to ses and  they would have the right to leave for other schools. But that phase would last four or five years, during which time the state or district can do pretty much whatever it likes to improve the school ’s effectiveness — and any federal rules, mandates, and spending restrictions that get in the way should be waived. The goal is to empower hard-charging superintendents and principals — and encourage others to charge harder, mindful that Washington will abet rather than impede them. States must do the same. For example, they should set aside provisions of collective bargaining contracts that impede the reform of faltering schools. (The Bush administration recently made a similar proposal.)

If the school does not begin to make ayp during that 4–5 year “corrective action” period, the hammer comes down. Now there are no loopholes and scant flexibility. The failed school must be turned over to someone else to run or else closed and its students accommodated at different schools. The key point is that if district efforts fail to strengthen a school, the district loses “ownership” of that school. (Similar logic should apply to state efforts to strengthen low-performing districts.)

This law identifies competent schools as “failing” and pushes states to dumb down standards.

Fourth, for any of this to work as intended, both parents and educators need to have confidence in the reliability of ayp  as an identifying mechanism; any version of this scheme goes to pieces if states or districts are ordered to shutter schools that fair-minded observers view as reasonably effective. Right now, the law identifies hundreds of generally competent schools as “failing” and pushes states either to set unrealistic targets that ensure that this designation will apply to many more schools or else dumb down their standards. So long as this law ensnares borderline-effective schools in a confusing web of remedies, it will prove difficult for even the best-intentioned implementers to make work.

ayp determinations and reporting must be better attuned to schools’ effectiveness (i.e., their pupils’ academic gains as well as absolute performance) and better at distinguishing between schools (and districts) in serious trouble and those that succeed with most of their students. (The administration ’s 2007 nclb reform recommendations point toward the possibility of ayp “growth models” for all states; four are piloting them today.)

What’s more, the annual identification of school status needs to happen far faster so that educators, policymakers, and parents know what ’s up well before the next year commences. The incapacities of the testing industry must not be allowed to perpetuate the dysfunctional practice of delaying such information until August, September, or October. All of the nclb remedies currently require that a school’s status be determined annually. Ensuring the accuracy of such determinations has combined with the failings of overburdened, underaccountable testing firms and balky data systems to produce an unworkable timetable. Sixteen states failed to finish identifying their “need improvement” schools from the 2005–06 school year before September, when the 2006–07 school year had already begun. For the remedies to work, states must radically alter their testing and data processes so that school identification is executed — and publicly reported — in weeks instead of months.

Fifth, parents, in particular, need better, faster, clearer information regarding their ses and school-choice options. These remedies also demand monitoring of ses providers, focused on actual delivery of services, better templates for communication and evaluation, and support for districts that do their best to make them work. States should conduct regular audits to encourage districts and schools to pay attention to customer service. It also makes sense to provide both ses  and school choice simultaneously to students whose schools “need improvement.”

A long-standing concern is the conflicted role that districts play as both ses providers and “gatekeepers” charged with negotiating agreements with alternative providers. This asks them to work against their own interests, manage responsibilities they ’re not equipped for, and engage in activities they regard as peripheral. The cleanest solution is for districts to cease controlling access to ses. Instead, states should explore how they could provide for other public or private entities to assume those responsibilities. Meanwhile, states must be required to monitor and report on the effectiveness of all their ses providers, districts included.

Many states need help in fixing their troubled schools. Most lack the capacity. But that’s not just an education problem.

Sixth, if choice is to be a serious element of nclb, as we believe it should be, more options are essential, including inter-district choice, the creation of greater capacity via a flood of high-performing charter schools, and the inclusion of academically effective private schools. (The Bush administration ’s 2007 nclb  reform proposals move in this direction.) Part of this entails the Department of Education ’s living by its stated position that lack of capacity is no valid excuse for failing to provide choices. Washington might require that districts find ways to offer feasible options — e.g., virtual schooling, expanding the capacity of successful schools, or raising state charter-school caps. States should lose federal dollars if their own statute books restrict (via charter school “caps,” for example) the provision of new options for children who are entitled to choice but currently can ’t find acceptable options. Washington should make clear that — at the very least — such states need to get out of the way and allow entrepreneurs to try to meet the need in various ways, including through virtual schooling.

Seventh, many states and districts need expert help in fixing their troubled schools. Most lack the capacity. But that ’s not just an education problem. We know of no sector, public or private, in which thousands of entities can separately assemble the know-how, talent, and organizational machinery to “turn around” troubled operations. Instead, such capabilities tend to be concentrated in a handful of specialized outfits such as “turnaround specialists” and niche consultants. If revitalizing low-performing schools (and districts) is to occur with any consistency at scale, primary-secondary education must develop that capability. The nation would benefit from a set of effective operators capable of contracting with multiple districts or states to provide the oversight, leadership and acumen to drive restructuring. Operating on that scale will permit specialization and cooperation and allow providers to build deep expertise. Washington cannot create this capacity, but it can provide resources, underwrite research, and encourage states to embrace non-profit and for-profit entities that show a record of success.

Eighth and last, in reauthorizing nclb, Congress should introduce real consequences for failure — and incentives for success. nclb’s current rewards and sanctions create little urgency for individual educators or school or district leaders. Collective bargaining agreements still protect the jobs of educators even in schools that pass beyond the restructuring horizon. (This one is the result of political timidity on Capitol Hill and at the Education Department, though in early 2007 the White House proposed to change course.) Failure to comply with the law or raise achievement has no clear results for the pay or job security of teachers or principals. The threat of federal dollars being withheld is all but toothless, mainly because Congress restricted this penalty to “administrative dollars” — and didn’t apply it to weak academic performance, but rather to failure to submit acceptable “plans” to Washington.

We recommend that federal policymakers encourage states and districts to adopt personal consequences for inadequate performance and failure to improve. Superintendents and principals should be held responsible for their schools ’ outcomes, rewarded when those outcomes are good — and penalized when they’re not. While many of today’s calls for performance pay focus on rewarding teachers for test score results, the most fruitful place to begin is by ensuring that the executives have skin in the game.

Some closing observations

Nclb began with the noble yet naïve promise that every U.S. schoolchild will attain “proficiency” in reading and math by 2014. While there is no doubt that the number of “proficient” students can and should increase dramatically from today’s 30-ish percent (using the National Assessment definition of proficiency), no educator in America believes that universal proficiency in 2014 is a serious goal. Only politicians promise such things. The inevitable result is cynicism among educators and a “compliance” mentality among state and local officials.

In critical ways, today’s nclb amounts to a civil rights manifesto dressed up as an accountability system. That ’s an untenable basis for serious reform, rather as if Congress declared that every last molecule of water or air pollution would vanish by 2014 or that all American cities would then be crime-free. The law is also pushing states to move aggressively in too many schools at once, ensuring that capacity won ’t match the challenges at hand. In this light, revamping the remedies begins with the need to refashion nclb as a functional accountability system rather than as an aspirational one. The failure to do so portends an eventual public backlash that will not only threaten nclb but may also discredit the years of clear thinking and coalition-building that have characterized educational accountability since the release of A Nation at Risk  nearly a quarter-century ago.

Whatever the political value of promising to “leave no child behind,” the results thus far threaten to undermine two decades of hard-won gains on educational accountability. nclb’s dogmatic aspirations and fractured design are producing a compliance-driven regimen that re-creates the very pathologies it was intended to solve. It ’s time to re-learn the hard lessons of the Great Society, when ambitious programs designed to promote justice and opportunity were undone by utopian formulations, unworkable implementation structures, and the stubborn unwillingness of supporters to acknowledge the limits of federal action in the American system.

Frederick M. Hess is director of education policy studies at the American Enterprise Institute.

Chester E. Finn, Jr. is a senior fellow at Stanford University’s Hoover Institution and president of the Thomas B. Fordham Foundation. This article is adapted from No Remedy Left Behind: Lessons from a Half-Decade of NCLB (AEI Press, August 2007).