Greg Forster (Ph.D., Yale University) is a Friedman Fellow with EdChoice. He has conducted numerous empirical studies on education issues including school choice, accountability testing, graduation rates, student demographics, and special education. The author of seven books and the co-editor of four books, Dr. Forster has also written numerous articles in peer-reviewed academic journals as well as in popular publications such as The Washington Post, The Wall Street Journal, and the Chronicle of Higher Education.

Executive Summary

Oklahoma has the opportunity to lead the nation into a new generation of education reform. Policy research and recent experience suggest a reform agenda focused on three goals, none of which requires significant increases in the education system’s spending levels. Oklahoma should expand parent choice; putting parents in charge is the real accountability system, with a long track record of helping schools (public and private alike) perform their best. It should create ongoing revision of academic standards to focus on clearly defining and measuring educational excellence, instead of using standards as a stalking-horse for ambitious political projects to remake society. And Oklahoma should consider reforms to the governance structure of the public-school system that would make it more responsive to the public (instead of special interests) as well as more efficient.

The Next Generation of Education Policy—More Spending or Real Reform?

Education reform is reaching a turning point on the national stage. Two major developments suggest the moment is right for policy entrepreneurs to set a new direction for education policy in the coming generation. One factor is the near-total burnout of the No Child Left Behind and Common Core efforts, which have consumed nearly all the oxygen in the educational room since 2001. The other is the continuing, and surprisingly little-noticed, growth of school choice policies in a diverse array of regions and forms nationwide. There are now 65 private school choice programs in 30 states and Washington, D.C., serving about 482,000 students.[1] These programs have produced a track record of success and important innovations in design, but also instances of failure from which we can learn valuable policy lessons.

In the late 1990s, state-level policy entrepreneurs like Governor Jeb Bush of Florida redefined the education reform agenda by enacting innovative state programs in standards and choice. Now, with no clear direction for education reform nationally, state-level policy entrepreneurs once again have a chance to set the agenda. With 20 years of lessons learned, and new circumstances offering new opportunities that couldn’t previously be pursued, the new agenda can both build on the old and seek innovations of its own.

One possible direction, which never lacks vocal advocates, is to spend more money on education. Special interest groups that profit from this spending, such as unions that represent teachers and school staff, perpetually hold themselves out as the voice of education expertise and the guardians of justice. They and their allies demand more money as the price of educational success and social peace.

Increased spending has no visible relationship to educational results. Unfortunately, the otherwise-banal interest peddling of the special interests is abetted by sloppy journalism and shoddy research that create the impression of a relationship between spending and results when there is none. For example, many widely used measurements and rankings of state educational quality include spending levels, or factors that are proxies for spending levels, as a key metric of “quality.” States that spend a lot are held up as educational successes for reasons that have nothing to do with student outcomes.[2] Consultants employed by special interests cook up manipulative and baseless formulas that purport to show how far short we are of what we “ought” to be spending.[3]

One of the most important—and consistent—lessons of education reform efforts in the past century is that policies focused on increasing monetary inputs do not produce results. Spending levels consistently went up nationwide (even in inflation-adjusted terms) for generations, yet outcomes such as high-school graduation rates and scores on the National Assessment of Educational Progress (NAEP) remained flat; empirical studies looking for a relationship between education spending and outcomes have consistently failed to find one.[4] 

Two often-hyped policy approaches are worth looking at in particular because they illustrate the failure of input-based strategies.

Teacher pay raises—even in “merit pay” programs—do not visibly affect student outcomes. Teacher quality is essential to a good education. In most kinds of organizations, one way to get better workers is to offer more pay. But that only works where hiring and firing are related to performance. Unfortunately, in public schools, union rules and other bureaucratic obstacles create barriers to entry that make it very difficult for new talent to enter the teaching profession. Gifted people, potentially attracted by higher pay, will have enormous difficulty becoming teachers unless they first spend years acquiring education credentials—credentials that have long since been shown to bear no relationship to classroom effectiveness.[5] The more talented they are, the less likely they are to want to squander years of their lives pursuing bogus credentials. The same union rules ensure that increased pay is distributed within the system without regard to performance. Even “merit pay” programs, which strive to tie pay increases directly to performance, have uniformly failed to affect student outcomes. A review of the research by leading researchers finds that merit pay programs have failed across the board.[6] 

Smaller classes do not visibly affect student outcomes. Policymakers often expect smaller classes to produce better results, because they would mean more individualized attention for each student. However, to make classes smaller, you have to hire more teachers; to hire more teachers, you have to hire lower-quality teachers who wouldn’t otherwise have made the cut. The critical importance of teacher quality thus cuts against class-size reduction. Also, the research suggests that even if teacher quality is constant, benefits from smaller classes don’t kick in until classes are very small.[7] And very small classes are prohibitively expensive to produce at scale. These factors help explain why class-size reduction policies have not succeeded when transferred from tiny pilot programs to statewide policies.[8]

If more inputs are not what is needed, where should policymakers look for promising ways to improve educational outcomes? The experiences of the past generation point to three promising areas: parent choice, professional academic standards, and clean systems of school governance.

Parent Choice: The Real Accountability System

Putting parents in charge is the real accountability system that improves education.[9] It is the only education policy that has consistently worked to raise student outcomes, not just in pilot programs or special cases, but at scale and in a wide variety of cities and states over long periods. The benefits are sometimes moderate in size, and there are cases of failure. But the overall track record not only supports moving forward with choice, it suggests that better policy design would produce bigger improvements. Hence the opportunity for policy entrepreneurs to take the next step.

The research finds that private school choice policies have positive effects on student outcomes. This goes both for the students who use choice and for nearby public schools. For choice participants, 14 out of 19 U.S. studies using random assignment methods (the gold standard) found academic improvements from school choice. For studying the impact on nearby public schools, it is not possible to use random assignment methods, but 32 of 34 U.S. studies using all methods found that student outcomes were improved by school choice programs. The research on fiscal impacts, ethnic segregation in schools, and good civics is just as positive.[10]

Recently, attention has focused on a poorly designed choice program in Louisiana that produced negative results. The importance of one failure shouldn’t be exaggerated, but we also need to pay attention to it, to learn what not to do. Among other design flaws, the program imposed difficult burdens on participating private schools; as a result, most private schools stayed away from the program. Less than one-third of eligible private schools participated, in stark contrast to the norm for choice programs.[11] It appears that the program’s burdensome terms of participation caused it to inadvertently “select” the worst schools. This conclusion is additionally supported by the fact that participating schools tended to report significant enrollment declines prior to entering the program, suggesting these were underperforming schools that turned to the program for survival.[12] 

Universal choice should be the long-term goal. The Louisiana debacle and other experiences from the past twenty years point to the value of making school choice programs more widely and easily accessible. Programs created with strict limits on which students they can serve and how they can serve those students struggle, educationally and politically. Programs that create more freedom for more parents create more benefits for schools and students, and have a larger political base upon which they can draw to defend the program against invidious attacks of special interests.[13]

The whole logic of school choice—that it puts parents in charge, that it creates opportunities for schools to try diverse approaches and invent the education of the 21st century—points to the value of universal choice.[14] As long as we’re only putting some parents sort of in charge, school choice programs will be limited in what they can accomplish. Only putting all parents really in charge holds out the promise of more dramatic improvements, which has always been latent in the idea of school choice but has not yet been unleashed in a universal and unfettered school choice policy.

Policy options for Oklahoma to consider include:

  • Expand eligibility in its existing private-school voucher and tax-credit scholarship programs.
  • Enact Education Savings Accounts, which represent a design improvement from vouchers and scholarships.[15]
  • Move its pre-K programs to a school-choice basis.[16]
  • Resist efforts to impose burdensome regulations and government-controlled “accountability” in school choice programs; beyond reasonable health and safety regulations, parents are the real accountability system.

Professional Standards: Defining and Measuring Academic Success

Oklahoma has recently had a bad experience with standards reform. The Common Core initiative sounded good on paper—who doesn’t want to raise standards? In reality, Common Core turned out to be a top-down effort to transfer control of educational standards from states to a newly created central system, controlled by the federal government. Such a system would inevitably have been captured by education special interests, and was already in the process of being thus captured when the Common Core initiative lost steam and burned out.

Setting the right standards for public schools is an ongoing need. Systems can’t perform if they don’t have clear goals. Unfortunately, most states have inherited standards that are unclear or too low, as a result of historical factors ranging from ethnic discrimination to messy political battles. Oklahoma’s repeal of Common Core standards, followed by the national collapse of the Common Core project, provides an excellent opportunity for the state to revisit its standards on its own terms. Independent state-standards reforms adopted (for example) in Massachusetts in the 1990s demonstrate the enormous value that comes from states setting their own standards and doing it right.

Standards must define success clearly, not open the door for big political ambitions. The “standards movement” for the past 20 years has had two underlying problems.[17] One is pluralism. America is a big, diverse, pluralistic country, and we have different ideas about what kind of people we want to be. However sincerely reformers may want to stick to such limited goals as raising math scores, any major political effort to create a single set of national education standards must inevitably produce an irresolvable culture war over the total content of education. This is partly because the reformers themselves often didn’t stick to their script when it came to their allegedly limited goals; reformers kept describing Common Core in grand moral rhetoric as a historic movement to fundamentally transform American society. Parents didn’t want to be guinea pigs for elites with hidden political agendas promising to transform them for the better. 

“Don’t nationalize standards,” while it is clearly the pons asinorum, is not the only lesson here. Any state-level movement promising to use educational standards as a political tool to transform society would devolve into the same kind of irresolvable culture war. The more important point is to get the job of standards right. The main purpose of standards is not to change society or advance any other political agenda, but to provide a clear definition of success for the ordinary, ongoing job of educating children. If people think standards are being politicized, they will oppose them regardless of who is politicizing them and what they are politicizing them for. Big, transformational aspirations are great in some policy areas, but not in this one.

Standards need measurements that provide transparency without driving change. The other big problem in the standards movement has been technocracy. National standards are meaningless in practice if they’re not tied to a national system of quantitative metrics that test whether students are living up to the standards. This forces schools to divert attention from educational priorities that aren’t being measured quantitatively (and which in many cases can’t be measured quantitatively) to the ones that are. Education would be whittled down more and more ruthlessly to serve narrower and narrower goals—goals defined not by parents, but by a technocratic class of “experts” and the politicians who control them. Parents saw where the logic of this system leads much better than most of the reformers in the standards movement did, which is a major reason the movement failed. 

Transparency is essential to any kind of accountability. To hold schools accountable, parents and other stakeholders need to know what’s going on. So having educational metrics is essential. However, a key lesson of the Common Core disaster is that there is a tradeoff between metrics that provide transparency and metrics that drive change. When metrics are used as a tool to change schools, everything gets myopically distorted to fit the metrics. You can use metrics to find out what is going on or to change what is going on, but not both. 

Policy options for Oklahoma to consider include:

  • Create a stable, recurring process to review state educational standards for possible reforms—say, every 10 years—so standards reform will be less tied to short-term political passions. The legislature would of course remain free to take action outside this process, but its existence would discourage irresponsible policy adventurism.
  • Give a primary role in this process to well-established subject-matter experts, but require a high degree of transparency, including opportunity for public comment, at all stages. The best way to convince the public you don’t have a hidden agenda is to remove the tools that are used to hide agendas.
  • Harmonize public standards with private school choice by including in the standards process a consideration of alternative approaches being used in private schools. This builds in a positive valuation of the educational diversity made possible by choice, rather than a one-size-fits-all “accountability” that hates and fears difference.
  • Invest in the development of better metrics of success that are designed to provide information rather than drive change. Recent research has revealed the ineffectiveness of annual test-score changes as a metric, for example.[18] New metrics of non-cognitive outcomes such as honesty and perseverance are emerging. Oklahoma could lead the way in these promising developments.

Clean Systems: Responsive and Efficient Governance Structures

It has been widely recognized for a long time that public school systems in the U.S. are not well governed, and that this is not because of mere accident or carelessness but because special interests have colonized the systems for their own benefit. Indeed, an awareness of this reality has been one of the primary motivators behind both the school choice and standards-based education reform movements. School choice is, as we have seen above, the most promising and best-proven approach to circumventing the special-interest colonization of schools. However, other policy reforms can also help.

Governance reforms to the public-school system should be ambitious, but aim for clearly defined, non-comprehensive objectives. An effort at comprehensive overhaul is unlikely to be successful; the special interests that have colonized the system are very powerful, with large resources of money and volunteer labor to deploy in political battle, and they would go all out to resist anything that threatens their power at its source (as the ferocity of their opposition to school choice attests). At the same time, small reforms that merely tinker around the edges are not likely to be worth the effort. Governance reforms should be big in terms of ambition, but narrow in terms of scope.

Reforms can make the system more responsive to the public, instead of to special interests. Oklahoma, like all states, has adopted a number of school governance policies whose primary if not sole purpose is to protect special interests. These policies produce no educational benefit, and often make education actively worse. Changing these policies would not only be an inherently good thing, it would also harmonize with the expansion of school choice—when parents are put in charge, they push schools to improve, so it’s important to give schools more freedom to improve in response.

Reforms can make the system more efficient. Special interests have an incentive to avoid clear lines of authority and efficient decision-making. The more complex and ambiguous the system is, the more opportunities these interests have to extract resources from it. Fixing all the inefficient systems would be unrealistic, but specific systems can be identified for reform.

Policy options for Oklahoma to consider include:

  • Require school board elections to occur at the same time as general elections. Holding school board elections at odd times helps ensure low turnout from everyone except the special-interest voters. This may be the most important governance reform, because not much else is possible while school boards are colonized by special interests.
  • Reform teacher hiring, firing, and certification. Teacher quality is essential, and the primary obstacle to attracting better teachers is the arbitrary and educationally worthless barriers to entry that keep talented people out of the profession.[19] Let principals decide whom to hire and fire.
  • Cut back on overreaching regulations. The 1889 Institute maintains a database of state regulations that shows Oklahoma has gone far beyond the kind of reasonable health and safety regulations it ought to impose—right down to monitoring the calorie content of diet soda.[20]
  • Oklahoma has a separate vocational-ed system, CareerTech, with its own property tax. That’s a recipe for bloat and abuse, as funds will come in regardless of how many students choose these schools—or are pushed into them. A single public-school system, plus private school choice for parents seeking options, would make these systems more transparent, responsive, and efficient.
  • School-employee benefits can be a locus of special-interest extraction. After Wisconsin required its schools to put health insurance contracts out for bid instead of making sweetheart deals with favored providers, it saved over $3 billion in a seven-year period.[21] Oklahoma could audit its school-employee benefits processes.
  • A huge portion of public school employees provide non-educational functions such as cafeteria and bus services. There is no reason for these services to be provided by government employees; as with health insurance, bidding out to private providers would create savings.

Conclusion

Education reform is ripe for a new, next-generation agenda. It is now widely understood that a lack of money is not the systemic problem. Top-down, test-based technocratic standards systems have burned out politically. So the field is clear for new directions. Meanwhile, private school choice is reaching maturity, and the movement is ready for the next big step toward universal educational freedom—to establish school choice as the real school accountability system. Academic standards, disrupted by the dalliance with Common Core, are ready for a careful revision. Clear opportunities to make public schools more responsive and efficient through governance reform have been identified. State-level policy entrepreneurs who take the lead at this moment will be in a position to shape not just their own states, but the national educational agenda in the coming generation. 

References

[1] The ABCs of School Choice 2019 Edition, EdChoice, 2019, p. 7-12.

[2] Stan J. Liebowitz and Matthew L. Kelly, “Fixing the Bias in Current K-12 Education Rankings,” Social Science Research Network, June 15, 2018.

[3] Eric A. Hanushek, “Science Violated: Spending Projections and the ‘Costing Out’ of an Adequate Education,” in Courting Failure: How School Finance Lawsuits Exploit Judges’ Good Intentions and Harm Our Children, ed. Eric A. Hanushek, Hoover institution, 2006.

[4] Jay P. Greene, Greg Forster and Marcus A. Winters, Education Myths, Rowman & Littlefield, 2006, p. 7-19.

[5] Douglas O. Staiger, Robert Gordon and Thomas J. Kane, “Identifying Effective Teachers Using Performance on the Job,” Brookings Institution, 2006.

[6] Stuart Buck and Jay P. Greene, “Blocked, Diluted and Co-Opted,” Education Next, Spring 2011; shortly after this report was published, a major merit-pay experiment in Texas was found to have failed; see Matthew G. Springer, et. al., “No Evidence that Incentive Pay for Teacher Teams Improves Student Outcomes,” Rand Corporation, 2012.

[7] Matthew M. Chingos and Grover J. Whitehurst, “Class Size: What Research Says and What It Means for State Policy,” Brookings Institution, 2011.

[8] Greg Forster, “Do Smaller Classes Help?” Oklahoma Council of Public Affairs, January 28, 2019.

[9] Greg Forster, “The Next Accountability,” EdChoice, July 6, 2016.

[10] Greg Forster, “What Does the Research Show on School Choice?” Oklahoma Council of Public Affairs, April 29, 2019.

[11] Jonathan N. Mills and Patrick J. Wolf, “The Effects of the Louisiana Scholarship Program on Student Achievement after Two Years,” School Choice Demonstration Project, 2016, p. 38.

[12] Atila Abdulkadiroglu, Parag A. Pathak, and Christopher R. Walters, “School Vouchers and Student Achievement: First-Year Evidence from the Louisiana Scholarship Program,” NBER Working Paper 21839, 2015, abstract.

[13] Greg Forster, “The Greenfield School Revolution and School Choice,” EdChoice, 2012.

[14] Greg Forster, “Universal Choice or Bust!” in Freedom and School Choice in American Education, ed. Greg Forster and C. Bradley Thompson, Palgrave, 2011.

[15] Matthew Ladner, “The Next Step in School Choice” Education Next, Summer 2016.

[16] Greg Forster, “How Beneficial Is Pre-K?” Oklahoma Council of Public Affairs, June 15, 2018.

[17] Greg Forster, “The Next Accountability,” EdChoice, July 6, 2016.

[18] For up-to-date citations to this literature, see Jay P. Greene, “The Achievement-Attainment Disconnect Strikes Again!” Jay P. Greene’s Blog, April 2, 2019.

[19] Douglas O. Staiger, Robert Gordon and Thomas J. Kane, “Identifying Effective Teachers Using Performance on the Job,” Brookings Institution, 2006.

[20] Greg Forster, “Relax School Regulations,” Oklahoma Council of Public Affairs, January 29, 2018.

[21] Ola Lisowski, “Since Act 10, School Districts Have Saved $3.2 Billion in Benefits Costs,” MacIver Institute, August 6, 2018.

Join Our Mailing List