Bryan Caplan  

Pritchett on Private vs. Government Schools

Chinese families move from nic... German President defends the m...
From The Rebirth of Education:
Whereas formerly only the elite may have gone to private schools, there has been a massive proliferation of private schools, especially in Asia and Africa.  These budget-level private schools are producing better learning outcomes, often substantially better, than publicly controlled schools - even for the same students - and often at much lower costs.
This isn't to say that across the board, private schooling is better than that available in government-run schools; in general, the evidence that private schools outperform government schools in well-functioning systems is weak.  In the United States, where there has been the opportunity to do the most rigorous experimental studies, most researchers agree that the private sector edge in learning is nothing like a full effect size [1 standard deviation], almost certainly not even a tenth of an effect size, and some legitimately dispute whether the private sector causal impact is even positive.
Yet the fact remains:
Broader than just the success of specific interventions inside government schools is the observation that even in low-performing government systems one finds excellent schools, but also, even nearby and even operating under apparently exactly the same conditions, terrible schools... The problem is not that government schools cannot succeed, for in nearly all developing countries some of the very best schools are government schools.  The problem is, as the LEAPS study authors emphasize, "when government schools fail, they fail completely"...
Case in point: In Mexico, "essentially all of the weakest-performing schools - those more than 100 points [2 standard deviations] below the average - are government schools."

My main objection: I strongly suspect that private schools have a big cost advantage over public schools even when they don't have much of a learning advantage.  This effect is easy to miss in the First World because there is relatively little demand for cheap adequate private education when there's free adequate government education.  But religious schools strongly suggest that private education for the masses can be provided at WalMart prices.  And if parents were paying their own money, WalMart pricing would probably dominate.

COMMENTS (12 to date)
Julien Couvreur writes:

This small-to-none difference on quality is quite surprising. How can it be explained, when voluntary (and therefore competitive) provision of other services usually leads to improvements in both quality and price?

Could it be a relative lack of demand (not much competitive pressure)?
Are customers of private schools not discerning enough?
Is schooling not such a big factor or the bottleneck, compared to genes, culture or environment?

I thought the main reason to oppose government schools is that they will indoctrinate everybody into the same mistakes.

MikeP writes:

Is schooling not such a big factor or the bottleneck, compared to genes, culture or environment?

These results are perfectly consistent with the model that individual children will learn at the rate that they are able to learn unless the conditions at school are so bad that they can't. If there is an environment of at least nominal safety and competence at a school, each kid will do as well at that school as that kid will do in most any school. If you have a 960 kid, he'll be a 960 kid at a 960 school or at a 720 school, but he may not be a 960 kid at a 400 school.

The reason that one finds only public schools at the low end is that conditions that are not at all conducive to learning are obvious to all and no one will pay out of pocket for them. They're better off doing anything else.

And, yes, this model also leads to the conclusion that the US pays far too much for its public schooling -- top, bottom, and middle.

Brian writes:
I strongly suspect that private schools have a big cost advantage over public schools even when they don't have much of a learning advantage.

A big price advantage, you mean. Very often there is no cost advantage at all. The government is spending a lot to provide free schools.

Case in point: In Mexico

Mexico doesn't have free public high schools available to everyone. You can test into the free public schools if you're of average or better intelligence and you study. Or you can pay for private schools. Most of the intelligent students are in the free public schools and the private schools are mostly filled by children of wealthy elites propagating their privilege. Wealthy elites usually have smart enough kids, though, so things are fine there academically.

That's not much like how high schools in most of the developed world work.

MG writes:

Besides the cost advantage, there is value for providing society something akin to "educational catastrophic insurance". To be sure, I would like to see how much of this effect is -- essentially -- the equivalent of student adverse selection, but I would not simply dismiss a system that could add value this way.

tom writes:

@ Julian

One of the biggest problems in comparing is deciding what to test for the comparisons. Private schools in the US often have a religious association (ie catholic schools)- in these schools the students get the same courses as public schools + religious studies, but of course there isn't a standarized test on religion so any learning in that area is ignored by the tests, despite the fact that mny parents value it, and many shools put resources into it.

You could potentially find a similar effect for non religious private schools if you took into account the portions of the curiculims that weren't covered by standardized testing- ie arts and sports (among others).

Chris H writes:


Tom already gave one good answer but I'd like to expand upon it some. Looking at test scores is a bad way of determining the overall quality of schooling. It's kind of like saying what kind of car is best based solely on how fast it can accelerate. Perhaps for a subsection of the car-buying audience that's all they care about, but the vast majority of people generally want to take a lot more factors into account.

Scott Sumner sometimes talks about this, and his point (which I fully agree with) is that the only test that really matters is the market test. Now, since people have to pay for public schools regardless of whether or not they actually send their kids there (and those schools are also subsidized by people without kids helping to pay for them)the US currently doesn't have a fair market test going on. Now given that a section of the populace still doesn't go to public schools despite their very large price advantage, I would be surprised if under purely free market or even voucher conditions a lot more people would didn't wind up sending their kids to private schools. And this would be a good thing regardless of what happens to the measured test scores.

Unless we assume a) parents are more likely to be irrational than rational on where their kids go to school (which is not totally solved under the current system given how parents decide where to live based on what public school their children will go to) and b) that current public schools are a better option than whatever irrational option the parents would choose (a debatable point), then the market test is the best test of school quality. Given that parents generally really care for their children and their well-being I doubt a) is true (and there's no obvious reason for a market failure here, even under a human capital story for education the students mostly end up capturing the gains that human capital provides) so I doubt the market test would be misleading.

Maniel writes:

Schools provide a service.
In the private sector, service providers generally gain and retain or lose customers based on the quality of their service. Over time, inferior providers either improve or go out of business.
Therefore, one would expect that, over time, the quality of education provided by private schools would improve.
Absent competition, public schools do not have the survival (customer retention) incentive to improve. Instead, the demands (or absence of demands) of neighborhood parents drive improvement (or not).
For these reasons, charter schools and vouchers for private schools tend to be of most interest to motivated parents in low-income neighborhoods. Snapshot comparisons of schools based on testing is of marginal interest.

Steve Sailer writes:

Mexico has perhaps the most notorious teacher's union in the world. Many teaching jobs in Mexico are hereditary sinecures. When a teacher dies, his or her heirs have dibs on the job. If none of them want it, they can auction the job off to the highest bidder.

The new president of Mexico had the head of the national teacher's union arrested on a visit back to Mexico from her luxury estates in San Diego.

In summary, Mexico is a bizarre example of how much can go wrong with public schools, but it's a polar case, not a representative one.

Floccina writes:

Another advantage of private schools is that they conform more to parents wishes. An example would be day care hours and activities for the time between 3:00 pm and 6:00 pm.

Slim934 writes:

"This small-to-none difference on quality is quite surprising. How can it be explained, when voluntary (and therefore competitive) provision of other services usually leads to improvements in both quality and price?"

3 likely answers.

1) Genetics

2) Regulation of the sector. If the state dictates the structure of the market, there's little the private sector can do except follow the regulations.

3) Lack of for-profit purveyors who would be the drivers of innovations which would make the experience better.

John Robbart writes:

It is important to know that voucher schools in the US serve a wide range of needs. Some are specifically for "problem" children, some focus on the arts, some on a classical education (including mandatory Latin). Lumping this diversity together and trying to generalize from it makes no sense.

The programs also, of course, vary enormously from state to state, some states not having vouchers at all.

Where they exist they put power in the hands of parents who are responsible enough to use it. I'm sure the parents are more involved in the schooling of their children than the average. But this is self selection, of course.

Comments for this entry have been closed
Return to top