A less melodramatic post on vouchers

November 5, 2007 | 41 comments
By

So it’s vouchers time in Utah. Here are what I see as the relevant issues, minus the apocalyptic rhetoric:

Public funding and public production are two easily separable issues. We fund lots of health care and food purchases through the government, but mostly allow the production to be by the private sector. The voucher movement is an attempt to allow for some of the same in education. Whether or not this is a good idea depends on a few things.

I. There is a well grounded principle in the study of government finance that taxation and spending should try to treat equivalent people equivalently (doing so is, in the parlance, Pareto Efficient, meaning that it maximizes some agglomeration of social welfare). The big exception to this is when one person imposes a cost on those around them by their behavior, in which case you want to charge them for the cost they impose (dubbed a negative externality). Don’t let anybody tell you economists hate all taxes. There are some great reasons to impose taxes, and they typically fall into this “negative externality” camp.

The current system charges private school parents (withholds funding, the effects are identical) for not going with public education, which is probably priced at around $6000, give or take a thousand. That fee makes sense under one of two beliefs:

1. Putting your kid into private school imposes a $6000 “social” cost on your neighbors because you aren’t in the public school or
2. One finds it politically impossible to implement other ways to extract more money from the rich for redistribution (and one wishes to) so sticking it to all private school attendees is considered the next best thing.

In the first case, it is worth remembering that these kids aren’t leaving school, they are switching. Thus they will be part of a social network, just not the same one as their neighbors. They would probably be attending a pretty good school regardless, not propping up some failing system. The voucher proposition implicitly says that the social cost to the neighbors is more like $4-5000 (in the case of a 1-2 thousand dollar voucher). Not all that different.

So how do we decide what the social cost is to all your neighbors of having you leave?

One rough guess is to think about how much money those neighbors would be theoretically willing to pay to keep you in the school. On the other side, what is the least amount you could pay them to make them indifferent to you leaving?

Now, personally, I really doubt my neighbors would be willing to pool together and pay $1000 to keep my son in their public school, much less $6000 (despite his unquestioned brilliance). Thus I strongly doubt that the social cost is anywhere close to enough to justify the implicit tax. But I don’t know of any solid research on the question, so I suppose you are free to have your own beliefs about your children’s value. And remember that you must also weigh in the gains to the other private school students now blessed with the presence of your child.

Now in the second case, well, so few people in Utah go to private school, this is probably not a real good way to stick it to the rich as a group, but I guess some people will take what they can get.

Point I, above, is a really big deal, if you don’t understand it and tackle it, you are missing a fundamental issue in thinking about vouchers.

II. 3% of Utah is in private school. Thus this program really is not likely to alter the landscape too much. I think a very optimistic projection would be that after the reform, 5-6% of Utah would be private, in which case the program would more than pay for itself. Quite possibly, almost nothing would happen. In which case, the program will cost in the tens of millions, for a state educational budget in the billions. The effect would be, on an individual level, imperceptible either to increasing or decreasing total costs of state education spending.

III. Educational reform is incredibly difficult. Almost nothing has any significant, long term effect on student test outcomes. That is a skeptical take, but looking over the research, it is hard not to be skeptical. Vouchers tend to have modest positive gains. Ladd (hater) finds no positive gains in test scores among most students, Neal or Hoxby (lovers) find some positive gains. Both sides actually agree that there is one group that does gain, which is inner city blacks (of which, Utah has essentially none). The value of vouchers depends very much on the quality of the surrounding system. Utah does okay, so I would not expect large test score gains. Other outcomes are harder to track, so there is less research. What little I know tends to favor private schools as a good option.

IV. The typical use for low income vouchers is to move to the (cheap) Catholic private school system. Utah has some of these, so there are low income options for those who want them. Quite a few Utah private schools are under $5000/year. Thus the voucher could potentially make or break the deal for some families, especially when coupled with private financial aid.

V. Utah has a charter movement that is picking up steam. This is a far more likely place to see competition. They share many of the features of vouchers, but disallow state/church mixing and require the whole level of state testing and much of the regulation. A voucher law would let private schools compete on more equal footing with these schools, providing more options to parents.

VI. Public education can be justified on a lot of interesting grounds. In general, though, it is not particularly analytically useful to try to group it (or vouchers) with “welfare”.

VII. Public school teachers are normal people, so are private school teachers. They are, on average, neither saints nor villains (okay, actually, in Utah, many of them are Saints…). Teaching credentials tend to matter very little in the quality of a teacher after about three years. In fact, after about 3 years, additional experience or education shows basically no increase in the teacher’s ability to raise student test scores (the most easily measured result).

VIII. Based on early October polling, it would take a miracle for the voucher bill to actually pass. On the other hand, our state bird is the seagull…

All in all, there in no reason to get too worked up about the effects of the voucher bill on Utah. Passing it would help out some families who, for whatever reason, find that their children do better in a private school and have very little effect on the rest of us. There are beliefs that would justify either a yes or a no vote. I think I’ve outlined the basic ones. Feel free to point out (with as little melodrama as possible) relevant issues I’ve missed.

Tags:

41 Responses to A less melodramatic post on vouchers

  1. Ardis Parshall on November 6, 2007 at 12:08 am

    It’s nice to read something on this issue that is calm and rational. Thank you, Frank.

    Plus, you’re very, very funny. (“On the other hand, our state bird is the seagull…”)

  2. Dan on November 6, 2007 at 12:16 am

    I haven’t been paying attention to the debate on the Utah vouchers, so I was wondering if someone could help me understand just how this thing will be paid for. What I understand is this: A family that has a son in school wishes to take this voucher to pay towards a private school. The public funds that normally account for his being in public school remains in the public school fund, right? OR does that money that would have gone to pay for his public school education go with him to the private school in the form of this voucher? If it stays in the public school system, then where does the state get the extra money to pay for the voucher?

  3. MLU on November 6, 2007 at 12:17 am

    Well, there’s the freedom thing. Oh, sorry. That was the 19th Century.

    Overall, though, your analysis seems accurate. I doubt vouchers as modest as those in this proposal would make much difference, mainly because teachers and students and parents would be drawn from the same pool as the public schools, with the same schemas and scripts in their heads and would duplicate the existing system in most ways.

    To do something really different–which would be necessary to get a different result–would require strong vision and strong leadership so that people were held to a much higher standard, which nobody is asking for and few would put up with. Strong methods just feel wrong to people today.

    Don’t worry. Be happy.

  4. Frank McIntyre on November 6, 2007 at 12:41 am

    Dan,

    Utah has been running a bit of a surplus, so they have earmarked some funds to try the voucher thing. But in the long run, it’s all one pot, so the state would pay about $2,000 for the child. If the child would otherwise have gone public, this saves the state maybe $6000. If the child would have done private anyway, this costs the state the $2000. Whether or not the program is a net cost depends on entirely on if enough people are switchers. The rationale in my first point does not depend on the number of switchers, but is about achieving an efficient economic system.

    Ardis, thanks!

    MLU, yes, there is a value to freedom which should be added in to deciding whether or not to do the program. This would be a benefit to count against the hypothetical social costs mentioned in I.

  5. Frank McIntyre on November 6, 2007 at 12:46 am

    As an addednum, let me re-emphasize what may not be clear. In order for the current system to be better than vouchers, you need the social gains per public school student, above and beyond those to the student themselves, to be around $6000 more than the gains to the students’ peers (and families) in the private school. That would be a pretty huge effect. If you think the net gains are positive but more modest, than the vouchers as proposed are likely a net societal gain.

  6. Clark on November 6, 2007 at 1:19 am

    One issue I’ve heard some Utah teachers griping about is the issue of when the accounting takes place. That is what happens when someone leaves public school halfway through the year or someone from private school enters public school halfway through the year. From what they were saying (and which I’m skeptical of) the bill doesn’t deal well with this making schools accountable when they shouldn’t be.

    Anyone know the details on this?

  7. Matt Evans on November 6, 2007 at 1:33 am

    Clark, you were right to be skeptical. It’s an issue whether or not the student is transferring to or from private schools, and one the schools already deal with it all the time: when familes move to Utah, out of Utah, change schools, change school districts, begin homeschooling, stop homeschooling, are in the hospital for months, etc.

  8. Clark on November 6, 2007 at 1:40 am

    Matt, the issue isn’t whether schools deal with it all the time. (Clearly they do) The issue is the accounting of the money. That is if someone transfers in from a private school what happens to the money? Are the calculations done at the beginning of the school year or are they pro-rated?

    As I said, I’m skeptical, but then I’m cynical about politicians enough to know they have a tendency to leave pot holes in their bills.

  9. Frank McIntyre on November 6, 2007 at 1:46 am

    clark,

    If enough people are switching to private schools that this is even remotely a concern then vouchers will easily pay for themselves and we can do whatever the heck we want because we’ll have surplus funds. The number of students switching mid-year is simply not going to be a big enough issue that it should change anyone’s vote. In fact, I doubt there is so much current excess capacity at private schools that it even could be an issue. These things take time. Lastly, if that many people are so desperate to get out of these schools that they are leaving mid-year, there is no doubt that we should let them. Surely that behavior indicates that their perceived losses outweigh any inconvenience to the public school system of losing a semester of funding.

    Especially given that the state has has committed to smooth the financial transition for up to 5 years.

    Mid year switching is almost surely a red herring.

  10. Bill on November 6, 2007 at 1:50 am

    I’m not sure I understand all this freedom talk. Are we talking about freedom from the economic consequences of our decisions? People are certainly free to go to private school now, provided they’re willing to pay. It seems more like an issue of priorities than freedom. Unless you’re talking about the 19th-century freedom to not go to school at all.

  11. MLU on November 6, 2007 at 2:06 am

    I’m not sure I understand all this freedom talk.

    Yes. It seems odd, doesn’t it?

  12. Bill on November 6, 2007 at 2:20 am

    I understand the meaning of freedom quite well. What I don’t understand is people who twist its meaning to support their ideological position (freedom talk).

  13. Mark D. on November 6, 2007 at 3:02 am

    Bill,

    Suppose Medicare only reimbursed expenses incurred at government operated medical clinics, and the proposition was to allow Medicare provider network to be expanded to private hospitals. Clearly the beneficiaries would have increased freedom if they had the effective (i.e. covered by insurance) ability to choose from a larger network of providers, both public and private.

  14. m&m on November 6, 2007 at 3:22 am

    Frank, thanks for this assessment. It’ll be interesting to see how the vote goes tomorrow.

  15. Peter LLC on November 6, 2007 at 8:43 am

    I have responded to gst’s cries for succor, written up a post on vouchers, and posted it on the best blog in the Universe.

    Excellent post.

    However, I am firm in my belief that the best blog in the Universe will be one that eases the html formatting burden on users by providing buttons.

  16. Ranbato on November 6, 2007 at 9:46 am

    Funding counts happen Oct. 1. Schools in Utah pray no one is sick that day because it is based on actual attendance.

    It is a red herring though.

  17. Timer on November 6, 2007 at 10:01 am

    If you ask a parent of a public school child in a major American metropolitan area, “Would you be willing to pay ten percent of the cost of public school tuition per year if doing so would prevent the top ten percent (in motivation/test-scores/connectedness/etc.) of students in your school from leaving?” I think that the answer, in many cases, would be an emphatic yes.

    There is a tremendous premium in these places placed on getting children into “hot” schools with wealthy, well-connected peers. There are streets in the LA area where virtually identical three bedroom houses, on opposite sides of the same street, will differ in price by $300,000 to $400,000 because one of them is in a “high test score” school district and the other in a school district (e.g., Pasadena) in which nearly all of the city’s many middle and upper middle class white families have fled the private school system. (See the wikpedia article on “Pasadena Unified School District” for some history.)

    The housing price difference does not reflect a difference in funding per pupil from one school to another (that’s covered by property taxes, which are comparable in different places). Rather, it reflect the tremendous premium people will pay for having a child in a program in which other academically strong children are present. Of course, there are lots of not so pleasant racial overtones to all of this. Once a school becomes more than fifty percent black/hispanic (which is usually also associated with elevated crime rates low test scores), the whites/Asians become willing to pay tremendous amount of money to remove their children from the school system.

    Many people say, “I would be happy to send my child to public school if all the other people in my neighborhood were doing so.” But now they’re stuck. The social costs to a badly failed public school system are enormous.

    Currently, Utah is pretty special. Most Utah schools are reasonably good (with a range of A.P. classes and the like for those who want to excel), and there has not been a big flight from the public school system. There are lots of people in CA, MA, NY who would give anything to have what Utah has — to be able to just go down the street and enroll your kid in Orem High and know that it’s a safe, decent place that virtually everyone (rich and poor) in your neighborhood attends.

    You’re probably right that Utah’s voucher program is too insignificant to matter; Utah is more racially homogenous and is not nearly on the verge of becoming Pasadena. People funnel their surplus income into additional children, not additional private education for the ones they have. But don’t downplay the social costs of a badly segregated school system. In principle, how much would Utah parents be willing to pay to prevent their public schools from becoming segregated the way they are in Los Angeles?

    Quite a lot.

    I hope.

  18. a spectator on November 6, 2007 at 10:25 am

    Interesting read. I would love to hear more about this:
    “Public education can be justified on a lot of interesting grounds.”

  19. Frank McIntyre on November 6, 2007 at 10:42 am

    Timer, You are welcome to believe that the negative externalities are more than $6000 in California or New York. But you need to wrap up a few loose ends.

    1. Private school students are not the top 10 percent, they are distributed through the range of students.
    2. The public school’s loss is the private school’s gain, thus you need to subtract off the benefit to them when figuring the social cost.
    3. My vague recollection is that students in many Catholic private schools have, on average, worse socioeconomic characteristics than the general population. Thus the cream skimming argument is suspect.
    4. Segregation in the public school system happens by people moving away from each other, but it still happens. It is possible that it is even worse that way, as it segregates not just schools but neighborhoods. People like being with people like them.

    “There are lots of people in CA, MA, NY who would give anything to have what Utah has”

    Anything? Because they could move to Utah for a heck of a lot less than “anything”.

    Now here’s the big one. Suppose I have the choice of paying interest on $300,000 or of paying for private school. We’ll ignore asset appreciation, although, in California, we probably shouldn’t. Assume a 7% interest rate and that a private school of comparable quality to the good public schools costs $10,000 a year. If you have 2 kids, you pay $20,000 for private school or $21,000 in interest. This suggests a net social benefit of… $500 a child. And that is ignoring asset appreciation which would, using these made-up numbers, actually drive the number negative. You can play with the numbers to get different possible outcomes– I don’t know what a decent public school costs in CA. But clearly this is no iron-clad argument for massive social benefits. Of course, what with the tendency to have bubbles, one must be careful about using CA housing prices period.

  20. Vada on November 6, 2007 at 12:14 pm

    Frank,

    While I appreciate your economic analysis of the vouchers, it seems to be quite lacking to me. (I am not an economist, by any means, so please feel free to further explain where I’m going wrong.) Your analysis seems to assume that everyone is paying the money for their children to attend public school ($6000 per child), and so the parents who have their children in private schools are being unfairly penalized by $6000 per child (or maybe a little less, since they are removing some social benefits to their neighbors). But very few of those parents are really paying the $6000 per child, because that assumes that the only ones paying for public schools are those who have children. There are many people whose children are grown and gone (and thus not utilizing the public schools) who are still paying for those schools. There are a number of people who don’t have kids who are paying for those schools. There are people who only have young children, or who homeschool their children, who are also paying for those schools. Are they being unfairly penalized as well? Because, like the parents of kids who go to private school, they are paying for a system that they don’t use.

    You ask if your neighbors would be willing to pay 6000 to keep your child in the public school system, and you think they wouldn’t. But my guess is there are at least 6000 houses in your school district (there certainly are in mine). So the question becomes, would they be willing to pay a dollar a year to keep your child in the school district? I think the answer would probably be yes, though I could be wrong.

    And just to clarify, I’m not for or against vouchers. I don’t live in UT, and I don’t have any kids in school yet, so I haven’t really looked into it. But I’m very interested in your economic analysis.

    And, on a complete tangent, Peter LLC, for the best blog in the universe (with buttons!) see http://zelophehadsdaughters.com.

  21. Vada on November 6, 2007 at 12:17 pm

    Um, I just posted a comment, but it didn’t show up. Am I in moderation?

  22. djinn on November 6, 2007 at 12:20 pm

    I’m having a difficult time with this argument. It posits no public education at all as the default best choice–perfect pareto efficiency.

    Frank McIntyre says: “The current system charges private school parents (withholds funding, the effects are identical) for not going with public education, which is probably priced at around $6000, give or take a thousand. ”

    The current system charges people more taxes the less their chance of using the public school system, as there is a child deduction. This is, pareto-wise, terribly unfair. Those with more children use more resources, because kids go to school free. This does not mean each child is allocated 6K from the not-bottomless state budget for school fees. Why should the family down the street with 8 kids get that 8 * 6K, while the childless couple up the street get nothing? Why does the mere act of having a child trigger some sort of unearned payout? How fair is that? It’s still only one or two taxpayers per family unit. Those without children, those whose children have grown, whose kids all have yet to enter school, are not receiving their 6K. Unfair, unfair, unfair.

    Why not do away with public schools altogether? It would be much fairer on everyone. Only those with appropriate-aged children will have to shoulder the burden of educating them, they can choose how much to spend, where to send them, etc. Free universal education, bah, so 20th century.

    Obviously, what Mr. McIntyre’s argument is missing is the idea that an educated populace is a public good worth expending resources on, as are children. We pay taxes for this privilege, free universal education. It is a blessing, not to be taken lightly. This is different than having a chunk of money passed out to each child during K-12; point 2? 3? above, is a really big deal, if you don’t understand it and tackle it, you are missing a fundamental issue in thinking about vouchers.

    By the way, here’s a study showing that when factors such as parental involvement are factored in, students do equally well in public and private schools:
    http://www.boston.com/news/education/k_12/articles/2007/10/10/study_examines_public_private_schools/

    As an aside, what is the major difference between “public” and “private?” We’re the public in public schools, after all.

  23. Frank McIntyre on November 6, 2007 at 12:36 pm

    djinn,

    “Why should the family down the street with 8 kids get that 8 * 6K, while the childless couple up the street get nothing? ”

    The argument would be that the payouts are to the child, not the parent. Children are each worth something separate from their parents.

    “Why not do away with public schools altogether? It would be much fairer on everyone.”

    Once again, the benefits accrue to the child, and everyone is once a child, so the system is not as skewed as you present it.

    “It posits no public education at all as the default best choice–perfect pareto efficiency.”

    1. You can still have public _funding_ in either system.
    2. Your claim is simply not true, some people like publicly funded schools, just as some like public hospitals or government cheese. In no way does the setup posit no public education as the default. It lets people choose what they prefer. If one posits that no one prefers public schools, then I would be positing no public schools as the default. But I didn’t.
    3. I am not sure you understand what is going on in point I.

    “Obviously, what Mr. McIntyre’s argument is missing is the idea that an educated populace is a public good worth expending resources on, as are children.”

    No, Mr. McIntyre specifically understands that exact point, which is why he discussed it both when he talked about how public funding is not the same as public provision, and in point 1 where he discussed the public good aspects of public schools. On the other hand, private schools are also a public good (if we are going by the actual definition of the term) if by “public good” you mean providing education to children. Thus your (perfectly reasonable) arguments about the social benefits of educating a child apply equally to them. To prefer public schooling, you need arguments specific to public schooling. Reasons that are separate from public funding and do not accrue to students in private schools.

    “By the way, here’s a study showing that when factors such as parental involvement are factored in, students do equally well in public and private schools:”

    That is an interesting and useful study, but is one of many, each with their own flaws and limits. I would discourage you from cherry-picking studies that reach conclusions you already like. Especially when so many are done poorly. But the results there are, I think, in line with what I discussed above. And even in that study, as I’m sure you noticed, some of the private Catholic schools do significantly outperform the public schools.

    “As an aside, what is the major difference between “public” and “private?” We’re the public in public schools, after all.”

    We are also the public in private schools. (not me, per se, just keeping with your theme).

  24. Frank McIntyre on November 6, 2007 at 12:40 pm

    Vada,
    I just salvaged your comment out of the spam filter, sorry for the delay.

  25. Frank McIntyre on November 6, 2007 at 12:57 pm

    Vada,

    That is a good question. What you are talking about, though, is funding, whereas vouchers are about provision (who produces the good). Thus if I present a person with two choices, I want the cost (to the decision-maker) of those choices to represent their social cost. For most goods, this happens naturally because markets tend to this price.

    For schooling, free education could be justified if one thinks the cost of the education is roughly equal to the social gain (_beyond_ the gain to the child directly internalized by the parents) of having people educated. But if the argument is about being educated then it applies equally to private and public schools, and so there is no reason on that basis to have different prices.

    If, on the other hand, you think there are unique gains to _public_ education that make it $6000 more valuable, then one should maintain the price difference between the two.

    Now you argue that maybe people are willing to pay that much to keep my son around, but you are aggregating the payers and not the people being paid.
    Thus, the right thought experiment is would taxpayers be willing to pay twice as much for a public school system than for a private one that is publicly funded, because they would be willing to pay an extra $6000 to keep every student? I know of no evidence that people are willing to pay that much, but perhaps you are right. If so, the optimal thing to do is to vote against vouchers. If you think people would only be willing to pay an extra $4-5,000 then you should vote for vouchers. And remember, these are social gains after netting out the social gains to the community formed by the private school.

  26. djinn on November 6, 2007 at 2:35 pm

    OK, so, in your estimation, society grants a $6k benefit per year for each child to educate said child. If a parent doesn’t use that $6k, in your argument they are being effectively taxed–but the benefit is not being withheld, it is simply not being used. There’s a major difference. You don’t have to cash your tax refund either, accept Social security, use the water from your tap, walk on the sidewalk, or use any other government benefit.

    Look, this is simply not a Pareto-efficient choice and cannot be made such. People determine which schools to send their children to based on a huge variety of non-economic factors (the exclusionary nature of many wealthy private schools, other schools teach a specific, uh, worldview that the parents are interested in, etc.) with the caveat that only those in privileged positions–those with sufficient money (for private schools) and free time (a stay at home mom with the right characteristics to teach her children, the ability to drive the kids to the school, etc.) can even enter the marketplace.

    This will have the absolute effect of transferring money from the poor, who can’t even participate, to the rich, who can. It will be a loss for society; it will hurt those who need help the most while helping those who need it least.

    As to the studies about public versus private, as far as I can tell, it’s a tie. I remember reading a bit of a book awhile back that actually looked at real public and private schools and made the astonishing discovery that the public schools were actually more accountable to their communities than the private ones. I’ll see if I can provide a link….

  27. Frank McIntyre on November 6, 2007 at 2:52 pm

    Djinn,

    This is not “in my estimation”. This is run-of-the-mill public policy methodology. Saying that one cannot use the $6000 in private school means that it is not, in fact, “available” for private school use. Thus the public/private decision gets distorted from the actual social costs (unless there actually are $6000 in social costs to bring it back to equilibrium.)

    “People determine which schools to send their children to based on a huge variety of non-economic factors”

    sure, which is why we don’t just point at a study that claims there is no difference in test scores and then tell parents they are just fine in public schools. We let them decide, based on the societal costs of their actions and their estimation of the gains, where to put their child.

    “This will have the absolute effect of transferring money from the poor, who can’t even participate, to the rich, who can.”

    I certainly agree that rich people use more vouchers than poor people, but this argument is muddled. If your goal is to increase income redistribution, this is a glaringly inefficient way to do it (meaning it is a very costly way to get the money). There is a large literature on how best to do income redistribution with the least distortion and this looks like a pretty lousy approach.

    A better approach would probably be to raise property or income taxes, or to institute a larger standard deduction (which is progressive) or don’t lower taxes when you run a surplus, which is what has happened in Utah in order to fund the voucher proposal. But, I suppose if you don’t think tax changes are politically possible (even though most education bonds seem to pass easily around here), then you should oppose vouchers because you view that as the only available way to do income redistribution. You should also oppose the new light rail as it will disproportionately subsidize the well off. :)

    Personally, if what I wanted was income redistribution, I would be going after the other 90% of rich people who go to public schools as well, rather than trying to clobber the 10% who go private.

  28. Frank McIntyre on November 6, 2007 at 2:54 pm

    As for research on voucher schools and test scores, a great place to look for a starter would be the Journal of Economic Perspectives a few years back which did pro and con pieces by Derek Neal and Helen Ladd. These two researchers cover both ideological sides while not being two insanely dogmatic.

  29. Timer on November 6, 2007 at 3:01 pm

    Frank,

    “1. Private school students are not the top 10 percent, they are distributed through the range of students.”

    Yes, of course. And certainly not in Utah. But there are areas where the quality gap between private and public schools has grown to the point that most people above a certain income/motivation level have abandoned public schools. Of course, the demographics of who goes to private school will be different in Utah (The ultra-religious? The people whose children have special needs that public schools can’t meet?), and people may not be willing to pay so much to be with them.

    “2. The public school’s loss is the private school’s gain, thus you need to subtract off the benefit to them when figuring the social cost.”

    Absolutely. I was only pointing out that the value people assign to having smart kids near their own kids is high (and people might be willing to pay a lot more than you think to keep your kid in their class). As for overall economic efficiency… well, that’s a tougher question. Economically speaking, the amount of money the rich will pay to avoid being with poor kids is undoubtedly more (for obvious reasons) than the amount of money that the poor can pay for the privilege of being with the rich kids. But the rich also appreciate some amount of diversity, provided it can be kept at manageable levels and the level of opportunity a school provides for overachievers (AP classes, etc.) is kept high. They also like being able to live in the house they like best (rather than having to huddle together in little “good school” enclaves) and in having good schools nearby. They’re always saying, “Oh, if only the public schools could get back to quality X. Then I would participate again…”

    “4. Segregation in the public school system happens by people moving away from each other, but it still happens. It is possible that it is even worse that way, as it segregates not just schools but neighborhoods. People like being with people like them.”

    Oh, absolutely. In the LA area, the conventional wisdom is that you just either
    1. Pay an extra $21K/year in housing interest for a fancy school district.
    2. Pay an extra $20K/year per child for private school.
    3. Accept that your child will be in a heavily minority, poorly performing, rather scary school.

    If you declared private schooling illegal, people with the means would still choose 2 over 3.

    “Anything? Because they could move to Utah for a heck of a lot less than “anything”. ”

    No, I really don’t think “anything” (not even CA real estate prices) could get a true Californian to move to Utah. :) But I’m glad you like it. (And I think CA has sent a few hundred thousand people you way, hasn’ t it?)

    “Now here’s the big one. Suppose I have the choice of paying interest on $300,000 or of paying for private school. We’ll ignore asset appreciation, although, in California, we probably shouldn’t. Assume a 7% interest rate and that a private school of comparable quality to the good public schools costs $10,000 a year. If you have 2 kids, you pay $20,000 for private school or $21,000 in interest. This suggests a net social benefit of… $500 a child. And that is ignoring asset appreciation which would, using these made-up numbers, actually drive the number negative.”

    You’re drastically understating the price of private schools, but you’re right that is a tough tradeoff. I’ve been told that the conventional wisdom in Pasadena is that if you have two or more children, it is worthwhile to pay the extra $300,000 to $400,000 for a housing in a place with fancy public schools (say San Marino). If you have only one child then you are economically better off living in Pasadena proper paying for private school. But obviously the housing prices are going to be in equilibrium with the private school prices, so I don’t know what you are trying to establish by pointing out that these costs are similar. Of course they are.

    But CA shows that in a mixed community (maybe one third poor minority), it is possible have two equilibriums: one in which at least 80% of upper middle class whites send their kids to public schools (and the overall quality of public schools is kept high is kept high — think South Pasadena) and one in which at least 80% of upper middle class whites flee and send their kids to private schools (and the public schools fall apart — think Pasadena proper).

    People in the latter equilibrium often claim to wish they could be in the former equilibrium (which is what Utah has), and public leaders spend a lot of money and energy trying to move from the latter equilibrium to the former. (It’s not easy; you get towns where 3 bedroom houses cost over a million dollars — with all kinds of property tax money — that can’t manage to convince the owners of these homes to return the public schools.) Utah is already in the former equilibrium — and should be happy about that!

  30. fMhArtemis on November 6, 2007 at 3:28 pm

    Nice post, Frank. I just got back from voting–I’m For–and I was thinking on the way home how dismal the chances of this thing to pass are. But it was nice to hear your take, instead of all the spin I’ve been hearing from both sides.

    Praying for the seagulls….

  31. Kevinf on November 6, 2007 at 4:00 pm

    Frank,

    Pretty well reasoned post. I would only take issue with one comment:

    “Teaching credentials tend to matter very little in the quality of a teacher after about three years. In fact, after about 3 years, additional experience or education shows basically no increase in the teacher’s ability to raise student test scores (the most easily measured result).”

    Not so. There are a couple of national teacher certifications, the most recognized be the National Board Certification. A recent study from Arizona State University, among others, actually shows that kids with teachers that are National Board Certified actually do improve on their results in standardized tests. The NBC requirements take into account both teaching skills, and subject matter expertise, appropriate for the grade levels involved, so that a high school math teacher has to meet requirements relating to the curriculum at the HS level, and how kids at that age learn, where an elementary school teacher is judged on more general subject matter knowledge, and the skills required to teach pre-adolescents.

    Teachers with less than 3 years experience are discouraged from seeking the National Board Certification. It takes about a year to prepare and submit the work, including video taped sessions from actual classroom environments, a lot of self examination about current teaching practices, many pages of written documentation about experience, leadership, and educational philosophies.

    Most states reward nationally certified teachers with substantial bonuses, up to $10,000 annually in some states. Utah, unfortunately, does not.

    What is true is that by the third year, you’ve eliminated a lot of the learning curve of how to exist and succeed at a base level in a classroom.

  32. Frank McIntyre on November 6, 2007 at 4:56 pm

    timer,

    It sounds to me like we are pretty much on the same page then.

    I thought you were using the Pasadena example as showing the net social benefits to public over private school, if not then I don’t see that much disagreement as we both understand the numbers the same way.

    In which case, you still need to quantify the negative externality. As you know, knowing that there are two equilibria does not mean that one of them should be imposed. I like your point about the two equilibria, this is an interesting version of the same phenomena as I discuss with negative externalities. Your argument that people verbalize a preference for public equilibria over private has two problems– one is that it does not quantify the benefit, it just suggests that there is one. The other is that of course people paying for private schools would prefer, very much, to live in a place where they did not have to pay for school!

    Also, are these two equilibria really equally available to one city with a fixed demographic, or are they rather unique equilibria based on the demographics?

    “Utah is already in the former equilibrium — and should be happy about that!”

    I don’t see any realistic way that $3000 vouchers would move us to the other equilibrium. And if they did we would save so much money that I would find it very hard to believe it was a net loss.

    “And I think CA has sent a few hundred thousand people you way, hasn’ t it?”
    Depending on what you count, I’m one of those.

    “But the rich also appreciate some amount of diversity, provided it can be kept at manageable levels and the level of opportunity a school provides for overachievers (AP classes, etc.) is kept high.”

    I saw a paper that suggested everybody liked some diversity as long as they were the majority.

  33. Frank McIntyre on November 6, 2007 at 4:59 pm

    Kevinf,

    I am thinking of how teaching credentials (such as extra education) causally make you a better teacher. And the answer appears to be zero to not much. What you have pointed out is a program that sounds, to me, like it identifies which ones are already good teachers. I don’t think we disagree, although I appreciate your clarification.

  34. Kevinf on November 6, 2007 at 5:47 pm

    Frank,

    To some extent, the national certifications do identify good teachers, but most teachers that I have known that have gone through the process have made significant changes in their teaching techniques. It also costs in the neighborhood of $2,500 to apply for the certification, and if you don’t make it the first time, you have another year to reapply in the areas that you did not meet standard, and pay a portion for the second application. Some school districts may help with the application costs, but usually in the form of a promissory note that is taken out of the bonus the next year.

    Where Utah does not currently reward nationally certified teachers, I would suspect that fewer Utah teachers apply. Don’t have those numbers on hand, though.

  35. Frank McIntyre on November 6, 2007 at 6:57 pm

    Kevinf,

    It would be very interesting to know how teacher’s students improved before and after the certification process. Like I said, most things have very small or no effects, but this might be, for all we know, one of the better ones.

  36. Kevinf on November 6, 2007 at 7:27 pm

    Frank,

    See here: http://www.cna.org/documents/CavaluzzoStudy.pdf and here: http://epaa.asu.edu/epaa/v12n46/.

    A somewhat more critical review that has mixed praise for these programs is here: http://www.edletter.org/past/issues/2002-ma/boardcert.shtml.

    It does seem to hold some promise.

  37. Mark B. on November 6, 2007 at 7:55 pm

    ” when one person imposes a cost on those around them by their behavior,”

    Clauses like this, where “their” makes sense grammatically as a reference back to “them” which refers back to “one person”??, makes one wonder if we’re trinitarians after all. How else account for the inexplicable and confusing shifts in number?

    If schools would go back to teaching grammar, maybe we could at least talk intelligibly about who should pay for them.

  38. Frank McIntyre on November 6, 2007 at 8:14 pm

    Mark, it’s been downhill ever since the Nicene creed.

  39. MLU on November 6, 2007 at 10:59 pm

    Having reduced teaching to, well, teaching, whether in public or private schools, the numbers compare weak schools with other weak schools and find little difference. But it keeps people busy and off the streets.

    These are questions that I think are more interesting:

    Do intelligent and committed teachers with the freedom to teach get better results than bureaucrats in oversized and overmanaged schools? If they do, how do you manage to get your child into such a class? Where are the black swans?

    Neither Jesus nor Socrates would get tenure in these humming and buzzing airport-like institutions. Even worse, both could easily be accused of melodrama.

  40. Amy on November 7, 2007 at 12:11 am

    I live in FL and the voucher program is a big fat mess that has caused a lot of grief. I ‘m really tired over here on the east coast and i havent read all the comments so forgive me if I’ve missed something. Things I can think of off the top of my head (and keep in mind, UT’s demographics may be quite different):
    The students who qualify for vouchers here are typically from low socioeconomic classes. They often qualify for special programs such as OT, speech, EH, etc. Private schools generally do not offer these programs. Not to mention many schools here ask families to leave if their children do not meet their top academic standards…in other words anyone who has a learning disability is not welcome.
    To be frank, these families pay money to go to these schools, these schools attract these families by keeping up with top academics. Low socioeconomic kids are not wanted. I’m really condensing this whole thing here. But the only private schools left are these rinky dinkies where I’m not so sure they are getting a better education.
    It has also made problems in ways you don’t always think of. Transportation–bussing these kids. Lunch…there are no free lunch programs at private schools.
    I think I am making this too complicated. Sorry. I’m not sure if this helps the discussion, but I’ve typed this whole darn thing so I’m just going to post it.
    Charter schools have worked the best so far here. My kid goes to a phenomenal magnet arts school. But that is due to the choice proram here and not vouchers. So that’s a whooooole ‘nother mess our fabulous Florida education system has created. Good luck in Utah!

  41. Jeremy on November 7, 2007 at 1:57 am

    Amy — thanks for chiming in with the FL perspective. Probably a moot point anyway: with 80% of precincts reporting, vouchers are losing, 38/62.