Henrik Jordahl on Swedish education and its privatization

Not long ago I linked to an article suggesting there were some burgeoning problems with the Swedish educational privatization model.  Henrik wrote me this email:

Dan, Tyler and others!

As the article claims there are really serious problems in Swedish schools – and it does not help that the minister of education from the liberal party does not seem to understand or be willing to admit what is going on.

Importantly, there is no evidence of the private (so called independent) schools or of the introduction of school choice contributing to worse performance. The two most relevant studies establishing this are:

1. Böhlmark, Anders & Lindahl, Mikael, 2013. “Independent Schools and Long-Run Educational Outcomes Evidence from Sweden´s Large Scale Voucher Reform,” Working Paper Series 6/2013, Swedish Institute for Social Research.
http://su.diva-portal.org/smash/get/diva2:675521/FULLTEXT01.pdf  Finds that an increase in the share of independent-school students improves average performance at the end of compulsory school as well as long-run educational outcomes. Also finds that international test results (TIMSS) deteriorated less in Swedish regions with a higher proportion of independent school students.

2. Wondratschek, Verena, Karin Edmark and Markus Frölich (2013). “The Short- and Long-term Effects of School Choice on Student Outcomes – Evidence from a School Choice Reform in Sweden”, IFN Working Paper No. 981. Forthcoming in Annals of Economics and Statistics. http://www.ifn.se/eng/publications/wp/2013/981 . Finds that  increased school choice had very small, but positive, effects on marks at the end of compulsory schooling, but virtually zero effects on longer term outcomes.

Hope this is useful and not too late.


YES! The forces of privatization strike back! I'm reading a book by journalist Michael Grabell "Money Well Spent?" on the stimulus, and it too says that magnet school funds funded by the Obama $5B "Race to the Top" program were rejected by schools who prefer the old ways. The truth is: none of this will work unless the students are motivated, by stick (Asia style) or carrot, to study.

Or by intrinsic motivation

Where can I get some of that?

Gene pool.

Which gene?

The gene pool has ubiquitous potential for enthusiasm for learning. The problem lies with the narrow homogenity of methodologies propegated by a monopolistic education system that only interfaces with a narrow band of learning styles offered by the gene pool. Industrial era mass production teaching must give way to information era individualized teaching. Diversification of methodologies is the key.

This is an interesting case of how to weigh empirical evidence. On the one hand, we have the evidence from the article describing mechanisms by which private ownership may result in innovations and changes that don't support student learning. We also have evidence from PISA of Sweden's continuing decline since 2000, ten years after the voucher reforms were first implemented http://www.thelocal.se/20131203/sweden-slides-in-global-education-rank-pisa-students-schools.

On the other hand, we have really just one study (so long as we don't consider a "very small, but positive effect on marks" evidence of quality) that finds a pair of positive results associated with increases in the share of students enrolled in independent schools. Given both Sweden's social norms against corruption, dishonesty, etc and its track-record of very effective governance, I don't see how continuing declines in both the level and equality educational outcomes for the past thirteen years in Sweden is anything less than a condemnation of privatization (not necessarily school choice) as a means of improving educational outcomes. One study, in this context, means very little, unless it points to some confounding reason why quality declined has been so associated with increased privatization nationally.

In general, northern Europe did well in the 2012 PISA scores, but the Scandinavian (i.e., non-Finland) countries lagged, and Sweden came in last in all of Western Europe (except for Greece). We are familiar with the Scandinavians scoring a little better on average than the rest of Europe at whatever requires good government and social solidarity, so Sweden scoring worse on PISA than, say, Portugal is, well, interesting.

Interestingly, the real high scorers in 2012 in Europe were a North-central axis of excellence: Finland, Estonia, and Poland. Of course, it helps for them to have negligible levels of diversity. We see in the U.S. that problems caused by diversity just soak up all the attention of elites, preventing much focus on helping the majority of students. Perhaps that's true in Sweden as well?


'Perhaps that’s true in Sweden as well?'

Not according to the Swedes - but then, they aren't interested in riding your hobbyhorse. Nor, it seems after looking at the results over a reasonable time span, are the Swedes thinking that the solution to their problems is more of the same reforms - the previous system having delivered better results, it is apparent that many Swedes believe it is time to get back to their proven arch-egalitarian ways.

prior talking about hobbyhorses. Got my laugh for the day. Thanks.

Really? There isn't a huge immigration debate going on in Sweden? Do you read tino.us? The majority in Sweden wants immigration restrictions.

If getting back to 10 years ago is their plan, good luck with their proposed invention of the time machine.

High levels of diversity provide students with the experiences they need to compete an a diverse world. Furthermore diversity is a chance to explore alternative life-styles and worldviews and thus strengthens the character of young people.
Northern Europe would greatly profit from more diversity as shown by many statistics and brilliantly argued by many intellectuals.

Agreed. I regularly take my kids down to the ghetto and drop them off for a few hours every night. For their own good.

It's an art to abuse every quote & draw bizarre conclusions from them.

I dunno Rahul, usually when people speak of diversity they do mean third world muds

Max, the public schools have deteriorated faster than the private. I would revisit your logic on this.

Even if this is so, a the theory of a voucher system is not that private schools will do better than public, necessarily. It's that private schools will force public schools to compete/improve or lose students. All schools should get better, at least in the long-term. Public school deterioration is therefore still a sign of a failure of policy.

I see that as a beneficial side effect, but blaming private schools, as they outperform, for public school deterioration does not seem helpful. Should the public maybe fix the actual problem? Or least investigate it?
This debate's purpose seems to be more for distraction than improvement.

So yes, certainly the problem is not just that there are some evil private-equity-funded private schools. But there's a lot of reason public schools could be deteriorating that can be traced to privatization: Schools may be competing by offering sub-standard education, which may force public schools to actually focus on nonacademic things to hold onto students (Sumner may not think this is a bad thing, but then this whole debate becomes groundless http://www.oecd.org/edu/Untapped%20Skills.pdf); the cost to students of instability resulting from school closures, voting with your feet, enrollment shifts, may outweigh the benefits from competition. I think the only way to examine this problem is to look at the system as a whole, not at one kind of school or another.

wrong link, sorry:

"Schools may be competing by offering sub-standard education"

And yet they score better.

"evil private-equity-funded private schools" ?

I'm more worried about the evil public schools.
At least the private one don't force you to go there as the public ones wish to.

Bottom line is the private schools are performing better. Fix the public ones and the private ones will go out of business for lack of customers,

I don't think you are understanding the point. I'm not saying private schools are competing by offering non-academic benefits, I'm saying schools are competing in this way, all schools. That's what happens in a voucher system, ALL schools become competitive. Public schools often are more competitive, because they are operating in buildings meant for 600 and with three new schools in the district they are fighting an uphill battle for their life to keep up enrollment.

Further, while perhaps, studies show marginal benefits to private schools over public in Sweden, and perhaps they are able to adequately control for unobservable differences between those who seek out private schools and those who do not (though this is unlikely, since I haven't heard of any lottery-based studies), these schools overall are still doing much, much worse than they did thirteen years ago.

Where are you getting that from?

It's funny that the New York Times is expectedly attributing America's mediocre PISA test performance to "income inequality," when arch-egalitarian Sweden did the worst of any country in Western Europe.

The 2012 PISA test found about a 57 point gap in math between Swedish and immigrant background students in Sweden. (Same scoring system as the SAT with 500 supposed to be the OECD mean and 100 the standard deviation.) With immigrants making up about 15% of the student population, that means the Swedes would score about 9 or 10 points higher, which still isn't very good. That would still leave Sweden behind white Americans, and the white folks of Minnesota (the most Swedish-like American state) generally score above the white American average.

You can see the PISA math scores for immigrants and indigenes on P. 36 of


Indigenes, Steve? I am looking forward to the moment when you start referring to métèques.

And here is a perspective from the UK, an anything but arch-egalitarian country -

'Sweden was once a shining example for Michael Gove's reforms.

Sweden's education system has often been cited by Michael Gove as a role model, especially for its policy of state-sponsored free schools providing increased choice for parents. In 2008 Gove told the Conservative party conference that Sweden's school reforms would be introduced if he was in government – and in 2010 promptly did so, with the advent of free schools.

A few years later and Sweden's star has dimmed. The 2012 Pisa results show Sweden's exam results falling abruptly across all three measures of reading, maths and science – with the country recording the largest drop in maths performance over 10 years. Anna Ekström, head of Sweden's National Education Agency, said in response: "The bleak picture has become bleaker with the Pisa review that was presented today."

Dr Susanne Wiborg of London's Institute of Education said: "The Swedish free schools have played an indirect role in the decline of the Pisa scores over the last decade. However, the question still remains to what extent these schools actually can be blamed for this."

In recent months a number of for-profit companies running free schools in Sweden have been in financial difficulties, while a recent TV exposé revealed that the state-funded privately-run schools were prepared to bend selection rules to admit bright pupils.

Sweden's education minister, Jan Björklund, said the Pisa results were "the final nail in the coffin for the old school reform," and speculated that the central government could take over running schools from Sweden's municipalities.' http://www.theguardian.com/world/2013/dec/03/swedish-results-fall-free-schools-pisa-oecd

Sounds like the Swedes want to get back to their old arch-egalitarian ways, actually. And for a more Swedish perspective -

' The Pisa report prompted scathing criticism from the Swedish Teachers’ Union (Lärarförbundet).

"We're losing ground on all fronts and find ourselves in a very precarious position," union head Eva-Lis Sirén said in a statement, adding that Sweden's results had "sunk like a stone".

"We're losing not only those who are having a hard time, but also high-performing students."

Sirén accused politicians of pulling Sweden in the wrong direction when it comes to education policy, arguing increasing differences between Swedish schools is to blame.

"A lack of equality is the price Sweden has had to pay as a result of free school choice. That's a price we can never accept," she said.

The union head was also critical of recent reforms, claiming they left teachers overwhelmed with administrative tasks. She expressed fears that the teachinng profession had "lost its status" at a time when tens of thousands of new teachers need to be recruited.

"If we're concerned about the results today, we have every reason to be worried about 2020," Síren said.' http://www.thelocal.se/20131203/sweden-slides-in-global-education-rank-pisa-students-schools

PISA 2012 scores, average for all three subjects:

Finland 529, 3% immigrant
Estonia 526, 1% immigrant
Poland 521, 0% immigrant

Denmark 498, 9% immigrant
Norway 496, 9% immigrant
Sweden 482, 15% immigrant

Norway has all the money in the world for social services but scores mediocre on the PISA tests compared to the much poorer Poles and Estonians.

Perhaps the Scandinavians' problem is that they really worry about why the diverse aren't doing so well. It sure takes up a lot of space in the brain of American elites. For example, the New York Times just featured a story on how the autism rate nationally is 1 out of 88 students, but in Minnesota it's 1 out 36 white students and 1 out of 32 Somali students. The NYT's headline: "Study Links Autism and Somalis in Minnesota." Sure, there are two orders of magnitude more white students in Minnesota, but who cares about them when we can obsess over how the Somalis have an autism rate that is 1/8th higher than the white rate?


What percentages are these? I'm confused because Wikipedia reports Estonia as having 16% immigrant?

What gives?

C'mon Rahul, you know the answer. White immigrants don't count.

He usually doesn't cheat on the raw numbers though.....

First and second generation immigration to Sweden makes up 27% of the population. Some of that is other Europeans, but that's still a massive amount.

"And for a more Swedish perspective" Er, you mean And for more form a perspective of those who would financially benefit from...

My kids attend public schools in Singapore. All government run but very competitive, both within the class and between schools.

I think educational ambition and attainment bubbles up - from the community to the student and then to the school.

If Sweden is starting to decline - as is my own country of Canada - it because social norms have changed. Privatize your schools or hand them over to the unions, by if parents aren't demanding and pushing for results you're not going to achieve much.

I don't think educational "ambition" is a big driver of basic math and reading scores. This sounds like a fantasy of blank-slaters.

No, then who reads to young children if not parents who want their children to read?

Who pushes the math homework when math is hard and boring?

Do you have kids? Ever lived in Asia?

Don't twin and adoption studies deny the causality here? Good parents don't raise more successful kids; they give birth to more successful kids. There's a high correlation between being a high-IQ, ambitious, conscientious parent and having high-IQ, ambitious, conscientious kids, but the causal mechanism isn't bedtime stories and pressure over math homework.

Finch, not sure about your studies, but I have kids and trust me, parental pressure matters a lot. Probably ALL that matters.

You're not curious enough. I'd like to think that to, as a parent of two, but it aint so.

Parent and teacher expectations are a huge driver of all academic success, not sure what you mean by "ambition". The literature is very clear on this. Your kid gets a "B", do you congratulate him? If you do, he will feel better right then, but will also have learned that your expectations of him were lower than that. Parents who say "A 'B'? You better study harder, son" will have hurt his feelings, but he will have learned that the people who know him best consider him capable of more.

Educational ambition will push you into a Kumon cram course (or whatever the kids do nowadays). When I was in year 4-5, those things were turning C+ students into an A- students.

@Tarrou, the other things is the parents might have twigged to grade inflation in their school. "Better study harder" is probably a more rational thing to say to a B-grade in say England than in Singapore.

"those things were turning C+ students into an A- students"

Today we use grade inflation. Always works and is a lot cheaper.

Yeah, grade inflation is another beast entirely. Quite frankly, if the kids had anything like a reasonable education, no normal-IQ child should ever get less than an A in American public schools. Public school is a fucking joke. I tested graduate on the Iowa exam when I was eight (and no, that's not a brag, that's a knock on how desperately retrograde our schools are). Three years of homeschool > twelve of public.

This sounds a lot like Arnold Kling's null hypothesis that changes in educational format have little effect on overall academic performance. I'm curious about the rate of spending between Sweden and other countries, did the privatization help keep expenditures under control?

Its funny that after all this time of being told that rich peoples children are getting an unfair advantage because they are attending private schools that can give a better education, the left wing now are arguing that a private school system is probably worse than a public one.

I am in the place that even if on some measures private schools are worse than public ones, they are still better, because they are going to be more customer focused. My son attends a wonderful private school that is probably less academic than the public school I attended, but so what? He is very happy there and the teachers are really nice and supportive. There is only a very small section of the population (probably less than 0.5%) that needs academic skills like advanced mathematics anyway. As long as you can read and write, and do simple maths and know something about science, then that's all that is really needed. The remainder can be handled by employers who can then give people the specific training they need for their specific jobs. As all the literature shows, learning general things is absolutely no help in learning specific skills.

Yeah. The problem with your logic is, your son needs to get into a half way decent college - and it may be harder if he doesn't have a rigorous enough high school curriculum or doesn't score high enough on standardized tests which include math

1. IMO those international tests do not measure anything important.

but virtually zero effects on longer term outcomes.

Almost nothing beyond the basics in schooling seems to effect the long run.

Comments for this post are closed