Why were libertarians wrong about school vouchers?

Here is Megan on that topic, I agree with much or maybe all of what she says, namely that parents are seeking quality peers for their kids rather than school effectiveness per se.  But I’d like to add another, neglected factor.

If you look at the benefits of opening up international trade, it’s now well-known that the decrease in deadweight loss may be fairly small, but the gains from “putting under” inefficient firms can be large.  You are winning rectangles instead of small triangles, and in the longer run innovation spreads more widely and prices can fall with higher overall productivity levels.

OK, so now consider schools.  Vouchers typically are applied to pre-existing schools, and often in a fairly limited geographic area, such as a single city.  Those schools already had a stable place in the market, and now the demand for their product goes up.  The experiments typically are designed so that no public school goes under.

So, post-vouchers, no schools go under, including no low productivity schools.  Thus the efficiency gains from vouchers of this kind can be quite small, possibly zero.

Of course you can debate whether we should ever let K-12 schools fail, or under what terms.  But the general principle remains that markets are most potent when exit is an available and indeed exercised option.


A true cost per pupil voucher value would be over $12000 per year in most cities. 2x that in DC.

But taxes are the price I pay for society!!!!

Ah, but what's the marginal cost per student?

Correct me if I am wrong, but wasn't one of the main points letting children from poor neighborhoods attend schools from a different area that were better in quality? If that happened, the place in the market for the poorer schools would be threatened. Why did that not happen?

Because failing schools are backed by unlimited public money. Closing these schools is tough. Chicago Public Schools have around 1/3 of their schools less than half full, IIRC, and closing a single one is a huge political fight.

Schools that have too few students left are routinely closed. My home town shuttered my middle school about ten years ago because the town no longer needed two middle schools. Schools with lots of students however are rarely closed barring some catastrophe-- it's too disruptive to too many families.

Your impression is out of date, CPS now has a mix of overcrowding in some areas and underused buildings in others. That's partly because they closed 50 schools a couple of years ago.

That episode actually seemed like a marker of change in the politics of closing public schools. The opposition was ferocious, public, well-funded by the teacher's union, came from both poor and better-off parts of the city...and failed. The original list of 54 was trimmed only slightly, the mayor and the school stuck to their guns.

And the mayor won re-election a year later. Not easily, but he did, despite the school closings being the rallying cry of opponents throughout the campaign. I dunno if that could have happened a generation or two earlier at least in Chicago...and CPS is working on a fresh list of closings right now. (Won't be as many as the last time.)

all you really need to do is compare US education outcomes with those of the Netherlands to realize US education policy is a farce at best

Do Dutch students in the US do as badly as you imply? Compare like to like here.

Would the results hold if you swapped populations?

We're doing a study of school choice in Jerusalem where all the choices are within the public school system. The great innovation here is that Jerusalem's mayor, Nir Barkat, KILLS SCHOOLS that parents vote against by exiting. Stay tuned.

What Megan misses are the supply side effects. The true upside of school choice won't be realized until there are enough signals that the demand and resources will be available for a long enough time that entrepreneurs are willing to commit their capital to starting new schools. If vouchers can disappear at any moment, why would anyone want to deploy capital there, especially if the vouchers are the only way a school can survive? Similarly, at small scales and short time frames, vouchers won't work. It would be as if Flint, MI offered their citizens water vouchers-- in the days before bottled water was widely commercially available.

In practice, you'll find that charter schools that expect to have great chances of underperforming (read: will take a bunch of poor, black kids), end up opening in crappy environments, and therefore need very little capital. If the school only lasts a few years, they don't really care. as they use a variety of mechanisms to treat the school not as a long term investment, but purely as an extraction operation: Shitty for profit universities that never attempted to provide good education used pretty much the same playbook.

Ultimately, the real secret is that great schools, at any level, seem to be unwilling to expand very much, because they know, whether they admit it or not, that they are social clubs in which people happen to be taught some things. If a school is designed for two sections per grade, they won't try to go up to six or seven like a large public school does, no matter how many vouchers are there.

And yes, the racism is there even in private schools dedicated to kids with parents with high intellectual achievement. My city has schools that are mostly about kids with lawyers and businessmen parents, others for doctors, and even some for kids of 1st and 2nd generation rich immigrants. There's the good black schools too, and little bits of overlap across schools: For instance, the school full of immigrants has some black folks with lower economic status, but that pick it because there's enough variations in skin color that the kids will face less racism than in the similarly priced, 95% white, 5% asian schools.

Better efforts around new school formation are happening. I would say the number of non-religious private schools in my area has tripled in the past 10 years. Like anything, these initiatives need to be made gradually.

The key change public schools could make to keep pace is to offer more selective admissions options, particularly at earlier grades. This gives you the freedom to create more programs that don't need to spend all of their time on discipline, forcing distracting kids into the classroom, and teaching kids that don't do homework.

"quality peers" is obviously East Coast code for "my kind of people". The fact that TC can write such stuff reminds me of the vast silent majority in Hollywood who stood by and did nothing about Weinstein. Where are his students protesting this kind of barely disguised intolerance? All I cared about for my kids schools was that they had a learning environment with competent enthusiastic teachers (hopefully, challenging them (unfortunately, a rare occurrence), few if any disruptive peers, and a consensus by the parents, students and teachers that getting an education was the most important task that the kids had. I'd give my kid's school district a B- on those. For me THAT'S what quality peers implies. Not money, race, ethnicity, accent, or IQ. Of course, if (Maslow's needs hierarchy's )basic needs aren't being met: physiological & security needs as well as (to some extent) love&belonging then education isn't going to be very effective.

Just to clarify:
1. When you use the term “quality peers,” we should assume that you mean peers that are not disruptive and who agree that getting an education was the most important task.
2. When TC and “East Coast” folks use the same term, we should assume that they are using it as code for money, race, ethnicity, accent and IQ.

Is there evidence that 1) TC and others have this racist definition of “quality peers,” and 2) that you have the non-racist definition of quality peers (e.g., by the choice of school district to live in or voucher into).

That's what every civic-minded American tells himself as he tacks on an extra $100,000 to the family's debt load to be in a majority white/Asian school district. In other contexts, we are told markets should be unfettered so they can provide accurate signaling.

I'm optimistic enough to think many problems could be solved just by allowing schools to kick out the small percentage of consistent troublemakers. But that's a whole flood of lawsuits waiting to happen as well.

"“quality peers” is obviously East Coast code for “my kind of people”. The fact that TC can write such stuff reminds me of the vast silent majority in Hollywood who stood by and did nothing about Weinstein. Where are his students protesting this kind of barely disguised intolerance?"

Trigger warning!

" few if any disruptive peers... For me THAT’S what quality peers implies. Not money, race, ethnicity, accent, or IQ."

You don't believe Tyler, why should we believe you?

Wrong, wrong, wrong. You make the presumption that "my kind of people" is determined by something like money, race, ethnicity, or accent. It's nothing to do with that. IQ- yes, you may be right about that, but that is certainly correlated with the admissions criteria for selective schools across the world. The schools and programs where my kids excelled had far fewer of "my kind of people" by your superficial measures than the public school they would have attended, but they were selective admissions, with a literal IQ test as an admissions criteria.

For me, it's all about getting advanced kids into a program that moves faster than other programs so that they have an advantaged position in future competitive scenarios, have the opportunity to more deeply explore subjects before they commit to one in college, have a safe environment, and don't get bored in class. These kinds of programs also tend to not tolerate ill-disciplined kids, ones that can't read very well, etc. This is where good teachers want to spend their time- instilling a love for the subjects they have expertise in, not in keeping kids from fighting each other. For kids that are struggling to read or exhibit self-control, they need a different kind of school, and we should work to establish those as well.

One problem comes in assuming that all kids need the same thing. Another comes in presuming that is some kind of racism. How brainwashed are you?

You: "for my kids schools... few if any disruptive peers"
Tyler: "quality peers."

You're saying the SAME THING for crying out loud.

This is such a bullshit statement. TC has posted in the past studies quantifying the impact of 'disruptive peers' in the classroom. One of the biggest impacts on school performance for the rest of the students. You're putting words into his mouth which is actually a statement he is making about the decision making process of other people. And rust me, what you care about just as easily codes for racism as what TC wrote. You don't want to get into this game - you won't win.

The big problem with the American K-12 educational system is we are running out of white children.

There are a lot of different systems that would work okay if K-12 students were 80% white/Asian (much like American 4-year colleges work pretty well under a lot of different systems for organizing them).

But That's Not Who We Are ... anymore.

"The big problem with the American K-12 educational system is we are running out of white children."
I have heard Moldavians and Ukrainians are cheap. Maybe you should buy a few. I would say white people are not worth the bother. We imported European colonists after emancipating the slaves and, except for the Italians, it was a disaster, almost worse than the Japanese.

A Nobel Laureate fled South Africa for refuge in Adelaide, and as far as I can tell, nobody except me ever asked: Why isn't J.M. Coetzee at, say, UC San Diego instead of in Australia?


Why doesn't the U.S. have a policy of welcoming talented Boer refugees?

"Why doesn’t the U.S. have a policy of welcoming talented Boer refugees?"
Is it so difficult for a Boer Nobel Prize winner to immigrate to America? Two white cousins of mine immigrated to America (don't say we are not helping you with your white people shortage - we pay all the rearing expenses and let you take the cream of the crop). None of them was, to the best of my knowledge, Nobel Prize winners. Coetzee's works are read in America, they are filmed by Americans. He probably could get a sinecure if he were savy enough.

If you ask me, the problem is the Chinese and the Indians. There are too many Indians and Chinese in America, sucking the air out of the other minorities and the whites. They are mostly disloyalto America and, as Fukuyama pointed out, clannish. It will end with rivers of blood. It will be the 1946 Japanese Uprising in Brazil multiplied by one million. Whatever the problems Black face, they are Americans, they are part of the American nation.

Foreigners from advanced countries have rarely been fans of American K-12 schooling. Many are not fans of American undergraduate education either. It's American postgraduate research schools that have often been highly rated, indeed very highly rated. I can think of exceptions: for example, the Californian system of undergraduate education had its fans, but more for its structure than for its academic standards.

Jerry Pournelle's blog often quoted this: "If a foreign government had imposed this system of education on the United States, we would rightfully consider it an act of war."
Glenn T. Seaborg, National Commission on Education, 1983

Does anyone here know the gist of Seaborg's criticism?

Lots of white kids in West Virginia and Kentucky.

I think you’re conflating race and socioeconomic status.

Today's episode of YMBAWSI:

You might be a white supremacist if....you think that the big problem with the American K-12 educational system is we are running out of white children.

"I think you’re conflating race and socioeconomic status."

Race as a variable is more important than (within-race) socioeconomic status. The children of low-income Whites outscore the children of high-income Blacks.

Yes, it's harder to be a white racist these days, with the government less willing to do the segregating for you.

The big problem with the American K-12 educational system is we are running out of white children.


And here's the list of state NAEP profiles, with Kansas slightly ahead of the national average:


Wow, California schools had deteriorated to a greater degree than I realized. I guess they can go with the slogan, "We're better than Alabama!".

Sadly, again it is the smaller wages and influx simple denizens brought by open borders. California school ratings are quite a shock to those Silicon Valley techies inhabiting the better socioeconomic bubbles of the Bay Area. Now they become nervous about their Tesla Model X and Model 3 sitting overnight in their driveway.

Why are they parked in the driveway overnight anyway?

Is that what America became? A big Kansas?! Why can't all states be number one like Massachusetts?

"Why can’t all states be number one like Massachusetts?"

So, are all states in Brazil number 1 ?

Some more so, maybe, but I doubt there are such discrepancies as those between Alabama and Massachussets. The Brazilian brain is a wondrous thing, I dare say.

Americans could put a man on moon, but they can't teach Johnny to read.

"Albion’s Seed."
So that is Alabama is fated to the outer darkness and the people of Massachussets is a master race...

No, it really doesn't matter if the kids are white, black or pink with purple polka dots. What matters is if the kids are raised in middle class homes with two parents and certain expectations of discipline and good behavior-- and a high value put on learning. That's what what we are running low on as families like that are having fewer kids.

Read up on adoption studies.

That must be why Polish, Romanian and other European white kids perform the worst here in UK schools, far behind Nigerians and Pakistanis. Lol.

So Johnny still can't read? Maybe America should create a commission to study the problems of American education.

This libertarian would prefer school vouchers even if the education results were exactly the same. The freedom to choose your own children's school is a good thing in itself. It may be, as measured, that voucher schools have less impressive results since state schools are usually orientated towards the top 5% kids that will follow traditional professional careers, even when it is obvious that most children are not capable or interested in such subjects. But state schools are measured against these things so will do their best, via punishment and other incentives, to get the kids to do what they don't want or need to do. A voucher system would perhaps allow less academic schools with orientation to actual real life careers for the majority.

Simply put; we as tax payers put a lot of money into forcing kids to learn useless stuff. Vouchers can help change this.

Megan McArdle's column is an odd one.

The title: "We Libertarians Were Really Wrong About School Vouchers"

The conclusion: "My overall takeaway from the literature is that voucher programs probably do a little bit of good. But the emphasis is on the word “little”; they are not a cure-all, or even much of a cure for anything."

So, Vouchers don't cost any extra money (or shouldn't if properly set up) and they provide parents with choice and are a net positive.

I agree. Its a very statist approach, the theme is unless you can show clear gains versus having a state controlled industry, you should keep state control. Surely it should be the other way round, I way say only if there is absolutely no way to deliver the program privately should we opt for a state approach.

The other thing that irks me about this opposition to vouchers is that generally speaking the people opposing are fine with rich parents privately educating their kids at their own expense. But surely the logic against vouchers applies the same to private education at your own expense?

Yeah, I'm confused that not enough value is being put on simply giving parents a CHOICE. I pulled my kid out of his (large and well-regarded) public middle school this year to try a brand new STEM based micro school. He was not being failed by his current school (although we had some issues), we just felt he would do better in a smaller, more personalized environment. And he is thriving and genuinely loving school this year. We could make that choice because we can afford the tuition, but another parent who could not afford that choice would be stuck. Just having the ability to make a choice, and not be stuck with only one flavor, is a net good.

"Vouchers don't cost any extra money". This is questionable: the argument for it is generally made by noting that the per-pupil voucher is less than the per-pupil cost for public schools. But the latter figure includes the cost of special education, which various sources suggest is on the order of twice the per-pupil cost of non-special-ed students.

Second, a widespread voucher system would be a middle-class entitlement, and we'd see a great deal of political pressure to increase the amount, so that parents could spend less of their own money while sending their offspring to better and better schools—and, all too often, "better" would be defined in terms of athletic facilities rather than academic offerings. The situation would be very much like that produced in colleges and universities by the availability of federally-guaranteed student loans: schools would increase their costs in order to compete on amenities, and parents would demand larger and larger vouchers so that little Bobby could attend a school with a domed stadium.

Net positive, but not what she was expecting:

"Like many of my fellow libertarians, I genuinely believed that this would be an economic and social revolution that would, over time, lift millions out of poverty and alleviate all manner of social ills."

Libertarianism + HBD denial = constant failure.

Vouchers and charters serve conscientious people of all races in urban environments who have been zoned into districts with ghetto thugs because they can't afford better real estate. For this narrow segment it offers a chance to go to school in a stable environment even if it doesn't fundamentally change test scores (IQ) of the voucher users.

There is also a narrower set of magnet schools that focus on the super able (top 1%) who feel stifled at regular schools that don't have that much wrong with them but teach to the middle which is a real waste for people on the far end of the bell curve.

I wonder how different this would look if brains were applied to "let's improve schools" rather than "let's use schools to fight our favorite holy war."

"I wonder how different this would look if brains were applied to “let’s improve schools”"

The US has been doing that for 50 years. The quality of education for K-12 has declined during that period.

Which are the best proofs for each end of that claim?

1. That we have been trying pragmatic non-political things.

2. That median output has declined.

I think 50 years ago we just worried a lot less about the 55% of high school graduates who did not go to college (now just 30%). Not to mention the 15% who dropped out of high school (now 6%).

"I think 50 years ago we just worried a lot less about the 55% of high school graduates who did not go to college (now just 30%). Not to mention the 15% who dropped out of high school (now 6%)."

I think we did worry too much about this. That's how we got from there to here. We had to dumb down schools to get the number improvements. Now colleges give remedial math and reading courses to get them ready for .. colllege.

Where do you get the idea the number of dropouts is 6%? Real number is close to 17%:


If you count all people who drop-out of school as drop-outs, even if they passed the GED later, the high school graduation rate peaked in the 1970s:


I did a quick Google and found this:


I extrapolated a hair (down from 6.5 to 6.0) but 6 to 7 looks legit.

Another recent piece on the improving picture:


Maybe the trick is that "dropout" rate is not the mirror of "graduation" rate?

Could a higher fraction "complete" but not graduate?

This analysis implies not that libertarians were wrong but that there ideas were never really tried, a remarkably common state of affairs.

The program Betty DeVos ran in MI had very good results, especially for minorities.

You mean like True Communism was never tried?

Vouchers are at an all time peak in the US and yet still less that 0.5% of students are using Vouchers and that's up from around 0.3% in 2010. And even among those numbers, most of those vouchers have significant restrictions on usage.

However, all that being said, the consensus is that Vouchers are a net positive.

Aren't all parents "seeking quality peers for their kids"; indeed, aren't all adults seeking quality peers for themselves. Of course, it's a fallacy that the quality peers will cause the rest to rise up, while the reality is often the opposite. Why kids gravitate to certain peers is a mystery to me. The same can be said about adults. Like water seeking its own level, kids do too. One can only hope that the leveling is up rather than down, but the forces that determine it are often beyond the parents' control.

"Why kids gravitate to certain peers is a mystery to me. The same can be said about adults." - RaywardThurstonHowell the Third, talking at the country club to a group of fellow lawyers.

Why kids do what they do is a mystery. Two brothers who were my own son's boyhood close friends went on to entirely different careers, one a medical researcher at a prestigious college the other a comedian specializing in blue humor. My own son and his college roommate spent their weekends successfully doing what college boys do, the roommate eventually obtaining a Ph.D. while my son got his degree in the life skills that he and his roommate learned on weekends. Go figure.

Libertarian Vouchertopia must never fail but we can it fail it!

1) Probably the main aspect that is forgotten is it is pain in the ass to transport kids to school versus having the kids walk to school.

2) Most public schools are fairly good already and probably work for 75%+ of the students.

3) My guess is the higher grades would benefit more than lower grades. The benefits of local elementary school and like peers are much larger in 2nd grade than 10th grade. So we should focus only on upper grades.

4) I still think our society will not benefit long run with parents and vouchers as I still dont see how vouchertopia will create decent schools for average and below average students. (Also I still don't see how it will cost less in the long run.) So now we will have more helicopter parents competing the for the best middle schools and paying more above the vouchers. Why would vouchertopia not become similar to our college system? (And drive parents cost up more which will decrease birth rates.)

4) I always assumed one primary reason why Libertarians hated the school systems so much was the teacher's unions not the results.

5) I find it ironic that the mix of teacher union charter schools do best and I have noticed the best teachers are the most pro-union teachers. So if you dont like the teacher union than find ways to have the best teachers stop supporting the union. Additionally, I would suggest with good but not great teacher salaries/benefits the pool of 'good' teachers is very limited.

6) Finally, any education study runs into the same two realities: Schools can only do so much for each student and all school comparison studies suffer from extreme self selection problems. I assume the union & charter school do best because of this reality more than union charter schools are best.

If you compare states, the ones with the strongest unions --like Mass -- have the best results while the states with the weakest unions have the weakest schools.


How do you measure "the strongest unions?" Doesn't look to me like you have any data backing your claimed correlation, which would be pretty worthless anyway.(No accounting for race, unions in high-performing states should be expected to be under less pressure, ect.)

So, not interesting.

"If you compare states, the ones with the strongest unions –like Mass — have the best results while the states with the weakest unions have the weakest schools."

That's not true.

California has some of the strongest unions and yet is at the bottom. New York, Rhode Island, Michigan and Hawaii also do poorly, and those are all states with large, strong unions.


A reminder that these things don't work neatly for big states (a national map with county-level detail):


I suspect the reality here are:

1) When all else fails, assume the best results are the best student self selection process.

2) Again, the best teachers are also the most pro-union teachers.

"2) Again, the best teachers are also the most pro-union teachers."


1. True, but indicates they are worth the extra effort to the parents.
2. I's say for the middle 60%. The top and bottom 15% aren't served nearly as well as they could be.
3. "The benefits of local elementary school and like peers are much larger in 2nd grade than 10th grade. " Wouldn't this suggest that it's more important in the lower grades?
4a. Voucher schools often benefit minority students more than white kids. Looks at the good results DeVos got in her program, good for white kids, better for minority.
4b. Libertarians hate the unions for the poor effect on results. Better to run the schools for the benefit of the students rather than the benefit of the teachers.
5. I'd like to see any evidence that teacher union schools do better. That would be new to me, but interesting
6. You shouldn't assume that. Private schools almost always do better than public, without teacher unions. In my area, Catholic schools dominate private schools and they outperform the public ones, even the very good public ones. This is true for high schools too. And no more free labor, I don't remember the last time I saw a nun teaching.

1. True, but indicates they are worth the extra effort to the parents.

Well for average income parents, there might be two incomes and might not have the time. Also I suspect the parents most dedicated to education are also the helicopter parents that libertarians don't like. And I do believe these education helicopter parents would decide the optimal number of children is one.

2) I’s say for the middle 60%.

I almost used 60 - 80% here so we are agreeing.

3) Wouldn’t this suggest that it’s more important in the lower grades?

No because there a lot benefits of local schooling as kids get to know their neighbors, create friendships, etc. Note most test scores on math and reading have improved in 4th grade the last decade where as 11th grade is fairly stagnant. I assume that in High School teaching students at different levels has more value. Also learning is more basic as my brother-in-law questioned spending $10K to teach finger painting.

4) I was exaggerating my point, but teacher unions are too powerful. However, I am not sure how we are going to increase teacher quality without a raise in pay to teachers as to get more good teachers means taking from the private sector.

5) I think these realities are a problem of self-selection of students here. So special charter schools in urban areas with lots potential students probably benefit but these also tend to more union districts.



6) I should have been more clear that most school studies are too dominated by student self selection so any study comparison to prove the right school approach.

Two other point, I am guessing the two issues of modern education compared to 1950/1960 education:

1) Most people have good memories of their school even if the school was poorly run.

2) We forget that public schools probably really benefited from the sex discrimination job market in the private sector. So an reasonably smart and amibition women college graduate in 1950, say Megan Mcardle, would have been pushed into teaching public schools as opposed to her chosen career path in 1990s.

1) Few kids walk to school. 3x as many take a car. 3x as many take the bus. In many jurisdictions is illegal for children under 9 to walk to school.
2) Working for 75% of the students isn't true and isn't the problem being solved.
3) Lower grades is where there is some of the greatest divergence.
4) Vouchers don't create schools, people create schools to capture voucher funds. If they don't get used, no loss. It would be different than the system the colleges are using to exploit the taxpayers, because the voucher is for a fixed amount. If the the voucher is $15k, you aren't getting a federally subsidized loan for the rest of the tuition.
4 [sic]) The results are the primary reason, the unions and their exploitative contracts are a secondary reason.
5) Teachers' unions are generally not in charter schools.
6) Self selection is the point here, allow schools to select students, while parents select schools. It aligns a lot of incentives.

Here is the simple fact that very few people want to face on any side of the debate. Schools aren't the problem and debating how to fix them is largely a waste of time. The problem is poverty and schools can't fix it. Poor kids who live in unstable households aren't going to do well in school. If you have enough of those kids in a school, it's going to perform poorly and there will probably be additional negative impacts on the other students. There is no school focused policy that will allow us to solve that problem.

That's exactly my thought on the matter. This country makes a fetish out of education. There's a consensus that the right school can fix broken people, a consensus that I'm skeptical of at best.

It cannot be said enough: Good schools do not make good students. Good students make good schools.

Move all of Harvard College's students to Bridgewater State and all Bridgewater State's undergraduates to Harvard. The former Harvard students will still do much better than the former Bridgewater State students.

And there are very few bad schools in the USA or the rest of the developed world but many schools with bad students.

Good students should be able to go to good schools, which means allowing schools to select good students.

MLK says to judge people by the content of their characters. Thus, schools should discriminate by the quality of student's character, not intelligence. Good students are not necessary the most intelligence students, but there is a correlation between content of character and intelligence, a minimum intelligence threshold is necessary to have a good content of character.

Education is a hard problem.

poverty isn't a hard problem. just give poor people money.

Give people money does not make them not poor. Poor is not by the amount of money that they have, since it has to do with intelligence and content of character. A poor character would not be effective at using money; thus, they will not leave poverty. Giving them money only increases inflation. Giving money to the rich does not affect inflation for the poor. Giving money for the poor affects inflation for the poor. Remember that people buy different things based upon their background, culture, etc.

Care to show your data on kids from unstable poor houses doing worse at school? Of course in your analysis please correct for the heritage of the kids. I think you perhaps are going to find a problem understanding cause and effect. Poor kids can, and absolutely do well at school. Everyone was poor 50 years ago compared with today, but somehow people still managed to get themselves educated.

Having read this blog and its links for a couple years, what I remember is that "poor" kids do about as well as the rest, but a minority in truly "dire" circumstances (kids without a quiet place to sleep, no one to wash their school clothes) do much worse.


Funny thing to laugh at, buddy.


Before washing machines, nobody would go to school.

Washing machines, lol. I'd like to see the results after the novelty wears off. Maybe the parents are making the kids go so they can get the laundry done.

It would be cool if it were something as stupid as this.

"..."but a minority in truly “dire” circumstances (kids without a quiet place to sleep, no one to wash their school clothes) do much worse.

I believe the evidence indicates that it's poor attendance (and parents who don't care about education) that is the biggest contributing factor to abysmal performance. Indeed, the article you post is about using washing machines to boost attendance.

However, I'm doubtful the idea will work in the long run. The issue is less related to clean clothes and more related to a general attitude. If a parent can't care enough to wash your children's clothes (I've used a kitchen sink and hung them to dry in the bathroom before) , then they probably don't care enough to make sure the students get fed, go to bed on time or actually go to school. That kind of parent certainly doesn't care enough to help them with their homework.

Sure, I'm sure "dire" conditions often exist in that gap between "bad" parents and "so bad" that child services takes the kids away.

Is it the relatively low income or the unstable households? IE can you solve it by just adding money? http://www.nytimes.com/2011/04/21/nyregion/kiryas-joel-a-village-with-the-numbers-not-the-image-of-the-poorest-place.html

Poverty is not the problem. Culture is the problem.

If there is actual failure here, it would be in the compromise. It isn't free market- it is a voucher system based on government which people hope would act like the free market.

But then, if I actually had sufficient resource allocation, I'd check Megan's work. What literature is she reading? Research done by a bunch of Phd's in education is likely to be very badly biased.

NH has school choice for small towns that don't have high schools... it works beautifully because the small towns don't care about subsidizing bad public schools, and simply abandon the bad ones completely.

...and where are you guys getting $12,000 per pupil public-school costs? It's over $18,000 in NH.

NH is way up there.


Per that, an amazing difference between neighboring Mississippi and Louisiana.

Cool source. It really doesn't seem to be a big correlation between spending and results. NY and DC lead the pack with mediocre and poor results.

To me, the problem with our big school districts isn't that exit isn't possible. It's that voice is so ineffective. In big cities, school board members represent tens of thousands of parents and have to manage hundreds of schools. There's no way these people are spending time trying to understand particulars of individual schools. If you're a parent who wants to try and improve your child's school, your options to use voice are thus extremely limited.

If big city school districts were broken up into dozens of manageable districts with a few elementary and middle schools and a couple of high schools, parents and community members would have far stronger voice and be able to push their schools in a positive direction. This would be much more effective than a quasi-market voucher system; information is so bad in these systems that it's hard not for parents to use bad heuristics like "peer quality". School effectiveness metrics are just too poor for parents to understand the product they are buying. In contrast, it's much easier to determine what might make a community school marginally better versus worse (and these outcomes probably differ by community!) Tiebout sorting provides an exit option for parents who aren't satisfied by their local marginal improvements.

In short - the next wave of school reform should emphasize the ability for local communities to self-govern. If we aren't willing to let schools go under, let's let the people who live near them give a go at making them better.


That sort of differentiation is naturally suspect. Here in [large red state], "property-wealthy" districts' budgets are frozen - or running a deficit - as they must send most of their school tax revenue away, to poorer, or simply more rural, school districts. (That's how it began, anyway, and so for awhile; but now, I believe, the $ mostly just goes straight to the state's general fund, because that's how it goes here - the legislators here never saw a tax they couldn't undedicate. Or a lottery. :-))

Understand, the students of "property-wealthy districts" can be almost entirely disadvantaged - we're not just talking Fancy-pants ISD. But there are routinely calls to make it illegal for parents in such districts to raise money privately, as through foundations, or even at the band parent or booster club level, where lots of moms do a lot of work for small returns. It hasn't happened yet, but the other rule of politics here is, if somebody thought of it, no matter how novel it sounds, if it involves a grievance, it's as good as done. Watch the bathroom thing play out, and get back to me.

Peri - I'm not sure I understand your claim, so I will try to restate it and ask you if I am getting the gist of it.

Basically, big school districts do a lot of the revenue-raising legwork for states. The districts are rich as a whole (though many district students are in poverty) and thus send a lot of tax dollars to the states funding pot. The state redistributes more of that money out to rural areas than it sends back to the city. State legislatures tend to over-represent rural areas (certainly true in my purple Midwestern state), so they like this arrangement. Breaking up big districts is thus politically unlikely because it would result in a lot of money being redistributed from rural areas back to the new poor districts within the city.

Is this an accurate understanding of your claim?

If so, you may well be correct, but in my opinion that's more or less the right thing to do. So I'm not dissuaded from my advocacy of this policy.

I didn't read your comment very well. I apologize. I am not sure in my state that the broken-up districts would be property-poor enough, but in any case I can't see the big "rich" urban districts ever wanting to break up their empires. Even if they did, a lot of our school districts that are takers rather than donors are by no means what you'd call rural districts. Many are urban sprawl districts (as are many legislators); in fact, the district that was chosen here to be the face of the lawsuit back in the nineties, that led to this transfer scheme, was just such. [Despite the victory and all the years of transfers, incidentally, the state took over the running of that district a couple years ago.] Plus we have a border region that is pretty much all sprawl minus the rich metropolis. They are politically powerful and used to sucking up money. Your plan would have too few advocates here.

And still the moms put on their fall festivals, and the kids go around selling their coupon books - I admit, it bothers me that even this money is coveted by the equalizers. However, I'm actually not bothered that 54% of my school property tax bill leaves my school district. I live in a distinctly baja part of a small property-wealthy district, and it's like an annual demonstration that we didn't really need to be taxed that much. The kids are alright.

"If we aren’t willing to let schools go under, let’s let the people who live near them give a go at making them better."

Both of those are going to be uphill battles due to an entrenched bureaucracy and teacher's unions. Vouchers are often an attempt to side step entrenched bureaucracies and teacher's unions

Re: Vouchers typically are applied to pre-existing schools, and often in a fairly limited geographic area, such as a single city.

As a practical matter, unless we invent teleportation, it will always be the case that there's a very limited numbers of schools to which it is practical to send your kids.

Can someone explain the phrase "You are winning rectangles instead of small triangles" to me?

Google "Deadweight Loss Graph" and look at the Wikipedia entry....

After reading research pro and con on this for probably 30 years, I think somehow we should try to get to the point where the median income and above family pays directly for all their children's schooling. That is so that costs are reduced to what the median income family values schooling relative to other things. I am convinced that you cannot subsidize people with above median income so they currently pay every cent of their children' schooling one way or another. I am also convinced that the Government schools are very inefficient, back 10+ years ago I paid about $3,500 to send my boys to a Christian school. I think the evidence is that private schools do not raise PISA scores much, and maybe even lower them, but that PISA is not so useful a measure and gains/loses will be too small to worry about , so our main push should be to reduce costs.

My mother told me that her parents went to school for just 1 year and yet were educated and ran a successful barber shop and sent their son to Brown University. They also speculated in real-estate, they were not superstitious and were quite knowledge about the world and could read write and do arithmetic. I think school is oversold as a cure all.

Most international intelligence measurement are Western and liberal bias. Thus, non-whites have hard time. Of course, religious schools teach better content of character to their students from bottom up.

Comments for this post are closed