Sweden’s vouchers

Sweden gives all parents vouchers to use at public or private schools, explains Lance Izumi on the New York Times site.

About Joanne


  1. Ponderosa says:

    The piece gives zero evidence that the voucher experiment is working. In fact, one of the voucher-receiving schools shown in the piece is clearly the worst kind of progressivist school (one teacher says something like, “Knowledge must come from within the child.”) It looks to me as if a couple of Ayn Rand-types got into powerful positions at the Ministry of Education and convinced the politicians to drink the competition-is-a-panacea Kool Aid. I don’t expect it will do much, if anything, to boost Swedish kids’ overall achievement.

    I believe that a nation can and should engineer one great standard model for schools –a sort of Honda Civic of education –solid, dependable, feasible, sustainable, affordable, replicable, high-performing. Vouchers and charter-schools are just ways of avoiding the difficult and important task of designing that model.

  2. What Ponderosa said, mostly. I think it’d be nice if school reformers held their preferred reforms to the same standard they – reasonably – apply to status quo arrangements, namely, “Does this benefit student learning?”

    I’ll also note that U.S. students outperform their Swedish counterparts, at least in math, and probably in science.

  3. P-

    Maybe the voucher system is what the parents want. Other people’s opinions often get in the way of a good idea, don’t they?

  4. My “P-” was intended for Ponderosa. Of course, another P person managed to sneak in a post before I finished responding. Sorry.

  5. Andy Freeman says:

    > I believe that a nation can and should engineer one great standard model for schools

    I believe that every child should have a pony.

  6. The Swedes definitely like their voucher system. A lot of the more liberal politicians would like to scrap it, but it’s quite popular. But surely we can have this debate at a higher level than a popularity contest.

    The question is, what’s best for kids?

  7. Paul

    Vouchers are about individuals choosing what’s best for their own children.

    You sound as if you are uncomfortable with the idea of an education system shaped by the people who use it.

    I’ve never understood why the people we hire to educate our children then get to turn around and tell us we don’t get a say.

  8. Charles R. Williams says:

    “The piece gives zero evidence that the voucher experiment is working.”

    If parents are satisfied with the educational choices they have, vouchers are working.

    “a nation can and should engineer one great standard model for schools”

    The cultural consensus that is required to develop this standard model does not exist now and has not existed in this country since large Catholic immigration began in the 1840’s. Then there is the issue of whether a state monopoly could execute such a model effectively and efficiently.

    It is one thing for the state to fund education so that all parents can afford to educate their children properly. It is another thing entirely for the state to monopolize publicly funded schooling.

    NOBODY says that the sole purpose of schooling is to impart academic skills to children except when they are trying to finesse the issue of culture. In other contexts people routinely talk about educating the whole child. In the video we see evidence that the Swedish voucher system is in fact accommodating very different educational philosophies to the satisfaction of parents.

  9. Ponderosa says:

    I believe that we as a people ought to take responsibility for what kids learn; we shouldn’t leave it to chance or the whims of individual parents. I agree, Charles, that coming to a consensus will be hard, but just because it’s hard doesn’t mean it shouldn’t be done. I think it would be unhealthy if Johnny went to Video Game Academy and Suzie went to Afro-Centric Academy and Joe went to Bible=Truth Academy. It’s Tower of Babel. Democracy needs a citizenry that can really talk to each other –and that means shared cultural literacy.

    MT: should we let freeway users get to vote on road and bridge designs, even if some of these designs will be utterly unworkable? Should insurance companies pay for any procedure its customers demand, regardless of whether it’s a quack cure? My point is that, while there are lots of misguided teachers out there, being IN the profession does tend to give one somewhat better judgment about that field. Parents, as lay people, are even more vulnerable to making grave mistakes about education (just as many of them made grave mistakes about mortgages and investments recently).

    Charles: prejudice against government effectiveness, and blind faith in private enterprise’s effectiveness, should both be discredited now. Think AIG or Chrysler. FEMA sucked under Bush because he staffed it with hacks; under Clinton it was a tight ship. Look, both government and business are fallible, human institutions. But exposure to the pressures of the free market is not a panacea, nor is sheltering from the marketplace a sure-fire recipe for bungling. There are earnest and effective workers in both realms.

  10. Here is a study of the Swedish voucher program. It works.

  11. (Ponderosa): “I believe that we as a people ought to take responsibility for what kids learn; we shouldn’t leave it to chance or the whims of individual parents.”
    Inevitably, for each child, somebody makes educational choices. The issue then becomes: who best repesents the public’s interest? State control of education is a threat to democracy, just as State control of newspapers would be (is, in totalitarian countries).

    Pots and Kettles: Governance
    Practices of the Ontario Securities
    Joel Fried
    “Citizens of a country also face a principal – agent problem. Citizens “own” the machinery of government and employ bureaucrats to act as their agents in running this machinery. To reduce the costs of monitoring, the principals choose a legislature/board of directors to oversee the agents. Monitoring mechanisms are similar to those in the private sector: there are financial accounting standards that are met for each budgetary unit, and an external auditor checks these internal accounts. Transparency is maintained, in part, through freedom of information regulations. Compliance with procedures and other regulations are met both through internal monitoring and checks by units external to the bureau. Finally, contracts are structured, at least in a limited manner, to align the incentives for agents with those of the principals.

    There is, however, an additional problem in the public sector that does not exist for private firms. The firm has a well defined objective function – the maximization of profits – whereas the apparent objective for the government is the maximization of some
    index of a (weighted) level of welfare of the electorate. An unambiguous index of social welfare has been impossible to construct and, in its absence, monitoring the public sector
    is further complicated because data is generally lacking on whether or not the objective was actually approached and/or achieved and what the costs are that are linked to any specific objective. In effect, because of distribution issues and public goods, the cash
    flows measured with traditional accounting procedures will be, at best, only superficially correlated with that objective. Thus, looking at cash flows will provide the principals an extremely poor method of monitoring their public sector agents.”

  12. Malcolm – I’m open to being persuaded the Swedish voucher system has improved Swedish student learning, but it would have to be from a much more reliable source than the Heartland Institute.

    MTheads – So, just to be clear, you do not believe that student well-being should be the primary criterion by which we judge an educational system? It should be parental or familial satisfaction, instead?

  13. (Ponderosa): “…we as a people ought to take responsibility for what kids learn; we shouldn’t leave it to chance or the whims of individual parents…(C)oming to a consensus will be hard, but just because it’s hard doesn’t mean it shouldn’t be done.”

    I see no virtue in uniformity when it’s not obviously necessary. Children are not standard. If we all have to wear the same size shoes, somebody’s feet will hurt, whatever size shoes we wear, but why do we all have to wear the same size shoes? Seems to me the burden of proof is on advocates for coerced uniformity, not on the advocates of non-violence (i.e., freedom).

    If we disagree on a matter of taste, federalism (local control) or a competitive market allow satisfaction of multiple preferences, while a State-monopoly enterprise must create unhappy losers. If we disagree over a matter of fact, where “What works?” is an empirical question, federalism or a competitive market will provide more information than will a State-monopoly enterprise. A State-monopoly system is like an experiment with one treatment and no controls, a retarded experimental design.

    (Ponderosa): “…prejudice against government effectiveness, and blind faith in private enterprise’s effectiveness, should both be discredited now.”
    Call it “prejudice” if you like, and “blind faith” if you like. That hardly changes the facts. The government of a locality is the largest dealer in interpersonal violence in that locality (definition). The system of private property and markets unites local control with the incentive to use that control in benign and beneficial ways.

    I recommend to the socialists around here that they read…
    Von Mises, Socialism
    Havek, The Road to Serfdom
    Friedman, Capitalism and Freedom
    Solzhenitsyn, The Gulag Archepelago.

  14. Paul, the Heartland Institute was not the source.

  15. From the pdf document…
    “1 Financial support from the Expert Group of Public Finance, Swedish Ministry of Finance, is gratefully acknowledged. We are grateful for valuable comments from Tore Ellingsen, Stefan Fölster, Sune Karlsson, Per Skedinger, Sören Wibe and seminar participants at the Stockholm School of Economics, the University of Uppsala, the IUI, the
    Institute of International Economics and the EEA Congress 2001.
    2 Corresponding author. Research Institute of Industrial Economics (IUI), P.O. Box 6501, SE-114 85 Stockholm, Sweden, +46-8-665 45 35, mikaels@iui.se. 3Swedish Research Institute of Trade (HUI).

  16. (Paul): “…just to be clear, you do not believe that student well-being should be the primary criterion by which we judge an educational system? It should be parental or familial satisfaction, instead?”

    May I play through? The above formulation posits a false dichotomy. Who is to assess “student (child) well-being”?

    I reason axiomatically, here.
    1) Most parents love their children and want their children to outlive them.
    2) If you live among people, there are basically three ways you can make a living: (i) you can beg, (ii) you can steal, (iii) you can trade goods and services for other people’s goods and services.
    3) Most parents accept proposition #2 and prefer 2(iii) for their children.
    4) Therefore, most parents want what taxpayers want from schools, that children be educated to make their way in the world.

    Bureaucrats, however, have systematically different interests. What we in the US call “the public school system” originated in anti-Catholic bigotry and survives on dedicated lobbying by current recipients of the taxpayers’ $500 billion+ K-12 education subsidy. The public (i.e., State-monopoly) school system has become an employment program for dues-paying members of the NEA/AFT/AFSCME cartel, a source of padded supply and construction contracts for politically-connected insiders, and a venue for State-worshipful indoctrination.

  17. A question for Paul…
    You suggest that the Heartland Institute is unreliable. As you say: “I’m open to being persuaded.” Do you have evidence?

  18. Paul

    As to judging the value of an education, I’m sure parents will rely heavily on the same things we do now in judging schools — test scores, teacher quality, security, curriculum, etc. Will data disappear if we institute school choice? Or are you afraid people won’t use the same criteria to judge schools as you yourself would use?

    The State has mandates and objectives that fluctuate over time as resources and politics change, but parents have obligations to their children that go much deeper. Hence, parents have every right to decide on the best education for their child.


    The State has had decades to prove they know best for our children. That experiment has failed. And that is all State-controlled public education is. We as a society are free to figure out other ways to educate our children without going against any natural laws.

    You seem to have a religious belief in the need of the State to bring order and meaning to the chaos and unpredictability of individual choices.

    I say this so many times, but the State serves us, not the other way around. You would stifle my freedoms and responsibilities to further your own vision of the world. Freedom is scary when other people have it…dang that Constitution.

  19. Malcolm – Whoops, you’re right. I didn’t notice that the host site on that study wasn’t the publisher. And the Journal of Public Economics is actually pretty reputable, as far as I can tell. Their measure of student achievement is a little too narrow for my taste – I’d like to know about more than a few elements of math performance – but not unreasonable. I also like that they checked for a grade inflation effect from competition, which I’ve always wondered about.

    So that report is suggestive, I think, of some value from vouchers. I don’t think it’s open-and-shut, though. Back in 1998 the Comparative Education Review published an article concluding that “The Chilean and Swedish cases suggest that national voucher reforms fail to do what their proponents claim, even when conditions demanded by the purest voucher advocates are met…”, although part of the motivation for that conclusion was a lack of publicly-available achievement data, something the authors of the Journal of Public Economics article had to grapple with as well.

    Also, just to be clear, I think parental satisfaction is good. I just don’t think it correlates very strongly with school quality, and that school quality should be the main policy goal. The evidence from voucher experiments in the U.S. suggests very strongly that it does not. Places like Wisconsin and D.C. have extremely satisfied participant families but little or no improvement in learning outcomes.

    MTheads – The criteria you suggest families will use to choose schools are demonstrably not the ones that are used. Georgetown has a series of reports on the DC voucher program finding, among other things, that “a very small number of families” use test score information in choosing a school. Whether you’re “sure” families will behave in a certain way is irrelevant – the data suggest that they do not behave that way. (Security might be an exception, although here the issue is that while families in DC care about security, it’s not clear what their evidence is that a given school is more or less safe than any other.)

  20. My point is that, while there are lots of misguided teachers out there, being IN the profession does tend to give one somewhat better judgment about that field.

    I don’t see how being a teacher means you are better qualified to judge what children should be learning. Being a teacher may mean that you are better qualified to judge how kids should be taught than the average person. But teaching is just one of many possible careers that an adult may pursue, I don’t see how teachers are going to be particularly well-informed about what kids should learn.

    And then there are all the non-job purposes of education, such as making well-informed voting decisions, or dealing with falling in love, or grief. How do teachers get a legs-up on those things?

    As for your comparisons with road and bridge designs, and insurance companies not paying for quackery – engineers have always had to face the problem that their failures are bloody obvious – you can’t hide a collapsed bridge. Meanwhile medicine for centuries provided really really bad advice, so bad that the average patient would have been better off just ignoring it entirely, such as bleeding. Medicine only started to improve once attention really started being paid to the effectiveness of what doctors did, I don’t think the school system is there yet.

    Parents, as lay people, are even more vulnerable to making grave mistakes about education (just as many of them made grave mistakes about mortgages and investments recently).

    You are aware that many of the big banks have lost their shirts in the recent financial crisis? The mortgage and investment affair is not exactly a great argument for handing over important decisions to professionals in that field.

    prejudice against government effectiveness, and blind faith in private enterprise’s effectiveness, should both be discredited now.

    Nice strawman – no one advocates blind faith (well, outside actual religions). How about intelligent, considered faith in private enterprises’ effectiveness? Market economies are still doing far better than the non-market ones.

    There are earnest and effective workers in both realms.

    What matters is the relative proportion of the effective ones. And not just effective workers, but effective systems. A teacher may be incredibly hard-working, earnest and effective individually, but if she keeps getting fourth grade classes with 1/3 of the class reading only at second grade, and the school administration doesn’t back her up in maintaining classroom discipline, and the curriculum is badly designed, how effective is the school as a whole going to be?

  21. Jay Greene compiled this list of US voucher studies. Abstract arguments (since kids are not standard, a State-monopoly structure of the edeucation industry is counter-indicated) and empirical evidence support policies which give to individual parents the power to determine for their own children which institution (if any) shall receive whatever support taxpayers provide for education services.

    Gerard Lassibile and Lucia Navarro Gomez
    “Organization and Efficiency of Educational Systems: some empirical findings”, pg. 16,
    Comparative Education , Vol. 36 #1, 2000, Feb.
    “Furthermore, the regression results indicate that countries where private education is more widespread perform significantly better than countries where it is more limited. The result showing the private sector to be more efficient is similar to those found in other contexts with individual data (see, for example, Psucharopoulos, 1987; Jiminez, et. al, 1991).
    This finding should convince countries to reconsider policies that reduce the role of the private sector in the field of education”.

    Joshua Angrist,
    “Randomized Trials and Quasi-Experiments in Education Research”
    NBER Reporter, summer, 2003.
    “One of the most controversial innovations highlighted by NCLB is school choice. In a recently published paper,(5) my collaborators and I studied what appears to be the largest school voucher program to date. This program provided over 125,000 pupils from poor neighborhoods in the country of Colombia with vouchers that covered approximately half the cost of private secondary school. Colombia is an especially interesting setting for testing the voucher concept because private secondary schooling in Colombia is a widely available and often inexpensive alternative to crowded public schools. (In Bogota, over half of secondary school students are in private schools.) Moreover, governments in many poor countries are increasingly likely to experiment with demand-side education finance programs, including vouchers.

    Although not a randomized trial, a key feature of our Colombia study is the exploitation of voucher lotteries as the basis for a quasi-experimental research design. Because demand for vouchers exceeded supply, the available vouchers were allocated by lottery in large cities. Our study compares voucher applicants who won a voucher in the lottery to those who lost. Since the lotteries used random assignment, losers provide a good control group for winners. A comparison of voucher winners and losers shows that three years after the lotteries were held, winners were 15 percentage points more likely to have attended private school and were about 10 percentage points more likely to have finished eighth grade, primarily because they were less likely to repeat grades. Lottery winners also scored 0.2 standard deviations higher on standardized tests. A follow-up study in progress shows that voucher winners also were more likely to apply to college. On balance, our study provides some of the strongest evidence to date for the possible benefits of demand-side financing of secondary schooling, at least in a developing country setting.”

    It’s charming to encounter a Secondary Science (Biology) Education major who believes in Intelligent Design.

  22. I read a study of the Chilean voucher program by Martin Carnoy. Seemed to me it did not support his conclusion. Independent schools got less money than State (government, generally) schools and did as well. Church-operated schools got as much money (per pupil) as State schools and did better. Looks like a win for vouchers to me.

    I heard Martin Carnoy speak. He made a dismissive claim about the evidence for vouchers, so during the Q&A, I asked about the evidence and analysis in Herman Brutsaert’s study of vouchers in Belgium. He answered that religious battles for control over the curriculum had driven Belgian officials to the policy of subsidized options outside the State school system. Like, that’s an answer?

    Sorry, I do not trust Dr. Carnoy.

    Tracy makes an important point. Teachers may be good at developing in students some fluency in their particular area of expertise, but the choice of which area of expertise to develop is a normative judgment, a value judgment. The education industry is not a natural monopoly. Beyond a very low level, there are no economies of scale at the delivery end of the education industry as it currently operates. Education only marginally qualifies as a public good as economists use the term, and the “public goods” argument implies subsidy and regulation, at most, not State operation of schools.

    Even if one supposes that the political process can reliably summarize some “democratic” collective value judgment, why suppose that aggregation of resources and decision-making authority will yield a better result that a process which leaves resources and authority in the hands of individual parents? Do we vote onnext week’s lunch menu?

  23. Paul

    School choice isn’t about the best quality education. For one thing, “quality” is itself a whole different debate.

    School choice is about parental rights, which the State has no right to usurp except under the most stressing circumstances. We don’t live in Sparta. Our State doesn’t get to treat us as a means to some idealized end. This isn’t Brave New World where we bring up children with the benefit of the State in mind.

    The point is, we as human beings have the natural right to rely on our own judgement and not on a Government’s, even in the case of our children.

    Study groups, experts, policies, and data can help, but underneath it all runs ideology. The best way to defeat ideological based education is to diffuse it with a hundred choices.

    I’ve enjoyed the debate. Thanks

  24. Paul, if TIMSS 8th grade Math or Science scores or NAEP 8th grade Math or Reading scores provide too narrow a measure of school system performance for your taste, consider this chart, which a statistician in the Office of the Attorney general, State of Hawaii, gave to me. During the 10-year span which the chart summarizes, most schools in the Hawaii DOE operated on a September through June schedule.

  25. MTheads – I think it’s fine to say that school choice isn’t about school quality. I disagree, but at least the disagreement is clear. But note that you’re changing the argument, which had previously been about how sure you were parents would use good metrics of school quality to make their decisions.

    Malcolm – I think you are looking at the “working paper” version of the paper on Sweden’s vouchers. I’m looking at the one that was actually published. It’s not clear to me that they used TIMSS, since they say they used two small parts of “a standardized national achievement test”. (pg. 361 of the published version) TIMSS is a standardized international achievement test. I haven’t read the paper cover to cover, though, so maybe they clarify at some point.

    As for the voucher research, let’s not do too much cherry picking. There’s quite a lot of research of voucher experiments in the U.S. that show either no gains or only very slight gains after voucher implementation:

    On Milwaukee
    On Cleveland
    On vouchers generally
    I’d mentioned D.C. research previously.

    I actually don’t feel particularly strongly about vouchers one way or the other, and wouldn’t strongly object to their implementation, but there’s just no evidence they’d improve educational outcomes much. It’s a big mistake for voucher advocates to premise their advocacy on major improvements to education quality, because vouchers won’t deliver on those promises. There are much stronger arguments to be made about cost effectiveness, which actually *are* supported by the bulk of the data. Once you look at the totality of the research, however, it’s clear that vouchers in the U.S. either do not improve educational outcomes, or improve them only very slightly. I don’t see the point of hyping them up.

  26. Paul,

    We disagree about the evidence. Read the ppost in Jay Greene’s blog to which I linked (above). The preponderance of evidence supports statistically significant improvement in standardized test performance with vouchers. Further, the same level of standardized test performance at lower cost is an improvement. This was not “cherry picking” (like “disingenuous”, “cherry picking” is a gone-to-college way to call someone “liar”). Greene collected all the random-assignment studies he could locate.

    One large cost of the US State-monopoly school system which appears on no balance sheet is the opportunity cost to taxpayers of the foregone evolutionary improvement in educational systems improvement.

  27. Sorry: “in educational system performance”.

  28. Andy Freeman says:

    > I believe that we as a people ought to take responsibility for what kids learn; we shouldn’t leave it to chance or the whims of individual parents.

    And the pony should come with a prince or princess.

  29. Paul

    Are you arguing that most parents are too stupid to make good school choices?

    Yes, I think most parents who would bother to make a choice would take into consideration the perceived quality of the school. It might not be the only consideration. It might not be the one which in the end determines which school. Perhaps the parent wants a school near their place of employment or a school with a great football program.

    The point is they will bring some sort of judgement to bear on the choice. and I’m taking an educated guess that many parents would turn to the same data we use now when deciding where to buy our homes.

    And no, that is not why I’m for school choice. I consider school choice to be a constitutional right. Whether I like it or not. I just happen to like it.

  30. Andy,

    You write: (Ponderosa): ” ‘I believe that we as a people ought to take responsibility for what kids learn; we shouldn’t leave it to chance or the whims of individual parents.’
    And the pony should come with a prince or princess.”

    You again raise an issue which four participants (Tracy, MThead, me, and now you) have tried to get defenders of the NEA/AFT/AFSCME cartel’s exclusive position in receipt of the taxpayers’ K-12 education subsidy to address. Their silence is deafening. It does not take 12 years to teach a normal child to read and compute. Much vocational training occurs more effectively on the job than in a classroom. State provision of History and Civics instruction is a threat to democracy, just as State operation of newspapers would be (is, in totalitarian countries). People do not become more intelligent, more altruistic, better-informed, or more capable (except in their enhanced access to tools of violence) when they enter the State’s employ. Quite the contrary; guns attract thugs. So why imagine that State operation of school will enhance school system performance?
    In The Open Society andits Enemies, Karl Popper observed that academics are often “dictators in piocket edition”.

    Eduardo Zambrano
    Formal Models of Authority: Introduction and Political Economy Applications
    Rationality and Society, May 1999; 11: 115 – 138.

    “Aside from the important issue of how it is that a ruler may economize on communication, contracting and coercion costs, this leads to an interpretation of the state that cannot be contractarian in nature: citizens would not empower a ruler to solve collective action problems in any of the models discussed, for the ruler would always be redundant and costly. The results support a view of the state that is eminently predatory, (the ? MK.) case in which whether the collective actions problems are solved by the state or not depends on upon whether this is consistent with the objectives and opportunities of those with the (natural) monopoly of violence in society. This conclusion is also reached in a model of a predatory state by Moselle and Polak (1997). How the theory of economic policy changes in light of this interpretation is an important question left for further work.”

  31. I think the world would be a better place if more kids had ponies, personally.

  32. Government mandated ponies?

  33. Lightly Seasoned says:

    Yes! I think a predatory pony cartel would be just the thing. Barn sour little Shetlands. Perfect.

  34. Andy Freeman says:

    > Barn sour little Shetlands.