I’ve long been critical of local government. Yes, local officials are “closer to the people.” And yes, moving to a new town is a lot cheaper than moving to a new state or a new country. Yet local governments are still far inferior to for-profit businesses.
Recently, however, I’ve realized that I’ve been too generous. The two main things that local governments do are:
1. Provide K-12 education.
2. Regulate construction.
And on reflection, local governments do both of these things terribly. Consider:
1. Voucher systems are clearly more efficient, yet virtually every locality continues to directly supply K-12 education. Nor is this a product of rising state and federal involvement. America’s local governments have been funding systems, not students, for centuries.
2. Local governments’ construction regulations are usually quite strict, especially in the most desirable locations. The resulting draconian system of height limits, zoning, minimum lot sizes, minimum parking requirements, and beyond roughly double the cost of housing and greatly retard national economic growth. While state and federal governments also regulate construction, local regulations are clearly vastly more important. That’s why we call them Not In My Backyard policies.
Some economists try to rationalize the status quo, but to no avail.
1. While voucher systems’ effect on test scores is debatable, the effect on customer satisfaction is not. How so? Because if you let parents take their money elsewhere, plenty will. Hence, they are not currently satisfied customers. You can deny that customer satisfaction is a good benchmark, but that’s the standard we use for almost every other business. Why should education be any different?
2. While you can argue that housing regulations curtail negative externalities, the leading examples are parking and traffic. The optimal response to both is not construction regs, but peakload pricing. In any case, the damage of housing regulation vastly outweighs the harm of any negative externalities that it plausibly prevents. And what about all the positive externalities of construction? More homes means more playmates for my kids.
The other route is to concede the inefficiencies, but insist that local government still “works.” After all, local governments are supposed to maximize the interests of their own citizens. And that’s what they do, right?
At least for housing regulation, this is superficially plausible. “Existing property owners benefit if you restrict supply to keep housing prices high” sounds right. Yet on reflection, this slogan is far less clear than it sounds. Most obviously, if housing prices in your region are high, and all of the other localities strictly regulate housing, deregulation allows your locality to sell out to developers and earn massive profits. Local deregulation is like violating your OPEC quota: In the absence of strident retaliation, it’s practically a license to print money.
How, though, can local government be so dysfunctional?
First, as I’ve argued before, non-profit competition is weaker than for-profit competition, even if the number of competitors is vast. Why? Because no one is trying very hard to win. As I’ve explained before:
Tiebout implicitly assumes that non-profit competition works the same way as for-profit competition. It doesn’t. If a business owner figures out how to produce the same good at a lower cost, he pockets all of the savings. If the CEO of a publicly-held corporation figures out how to produce the same good at a lower cost, he pockets a lot of the savings. But if the mayor of a city figures out how to deliver the same government services for lower taxes, he pockets none of the savings. That’s how non-profits “work.”
With non-profit incentives, neither the number of local governments nor the ease of exit lead to anything resembling perfectly competitive results. The “competitors” simply have little incentive to do a good job, so they all tend to perform poorly.
Second, voters are deeply irrational, even at the local level. Most people childishly refuse to grant that allowing more construction will reliably make housing more affordable.
Yes, you can point to my Myth of the Rational Voter and object, “How can voters be so irrational even though the expected cost of voter irrationality is especially high at the local level?” Reply: Even at the local level, the probability of voter decisiveness is so low that the expected cost of voter irrationality is approximately zero. If you have more than a hundred voters, “Your vote doesn’t count” is basically correct.
To reiterate, I am not arguing that local governments have two little blind spots. I am arguing that local governments have two main jobs – and they’re awful at both.
READER COMMENTS
Garrett
Nov 4 2021 at 4:25pm
Regarding vouchers, it seems like the data shows that private schools can’t improve outcomes over public schools after adjusting for student characteristics, but they can provide the same outcomes for way cheaper.
robc
Nov 4 2021 at 4:34pm
Dunbar’s number is pretty much a hard limit on the size of effective government.
BC
Nov 4 2021 at 5:06pm
Agree with your two points regarding K-12 education and construction regulation. However, arguably, the most important function of local government is police protection. It’s not obvious that most local governments have done a poor job in that — compared to what, private security? — other than perhaps the ones that have recently Defunded Police or refused to prosecute “petty” crimes. Most communities seem pretty safe. Similarly, it’s not clear that national government in the US has done such a poor job at national security. Again, compared to what? At most, perhaps one could argue that similar results could have been achieved for far less cost.
Not coincidentally, police and security are the quintessential public goods for which we might expect government to do a decent job relative to alternatives. K-12 education is far less of a public good as the non-excludable external benefits to the general public are far smaller than the excludable internal benefits to the student.
Maybe, better to say that government does a terrible job at almost everything other than *pure* public goods, defined in the narrowest terms.
Floccina
Nov 5 2021 at 3:56pm
Would you feel safe walking in almost any neighborhood in the USA?
BC
Nov 7 2021 at 12:07am
Most neighborhoods, yes, but not all. Many more neighborhoods two years ago than twenty years before that. Fewer neighborhoods now than two years ago.
Josh S
Nov 4 2021 at 5:12pm
I don’t think that customer satisfaction is a good benchmark for schools. This is obviously a complex area with multiple objectives, and depends greatly on how you define the purpose of education and the purpose of schools. I tend to side with people who say the purpose is to develop intelligence and character so that people can live better, fuller lives and contribute to society. Customer satisfaction is both too short-term and too focused on the individual aspects to account for this full purpose.
I’m definitely not claiming the existing system does well against my stated purpose, but I do think there’s a kind of “commons” that should be a supported by any proposed alternative and wouldn’t naturally emerge if you simply aggregate individual preferences.
Jose Pablo
Nov 5 2021 at 1:52pm
“I do think there’s a kind of “commons” that (…) wouldn’t naturally emerge if you simply aggregate individual preferences.”
Then, how this “kind of commons” would emerge? (If not by “aggregating individual preferences”). Out of the mind of a benevolent ruler who can see what’s better and what’s worse for “ignorant aggregated individuals”?
And, by the way, even to “aggregate individual preferences” is not possible (see Arrow’s Theorem). Much less possible should be to “overrule” an impossible “aggregation of the individual preference of so mistaken individuals”. Poor guys! so lost without a leader of the commons” that can force into them the “preferences” that they don’t seem to have but so clearly need.
But you are right, there are “multiple objectives” in K-12 public schools and some of those explain the rejection of vouchers. The main “objective” of the public school’s system is to provide jobs to people that could be barely employed in any other place (what’s a “Global Perspectives” or even an “English” teacher good for out of public schools?).
By this account public schools are pretty successful: they employ a lot of, otherwise, unemployable people.
The fact of this being the main “objective” of public schools explains a lot of the “observable dynamics” in public schools: teacher’s behavior in the classroom, rejection of vouchers and the downplaying of “customer satisfaction” relevance, as just but some examples.
Matthias
Nov 6 2021 at 6:17am
I agree with the gist of what you are saying, but Arrow’s theorem isn’t particularly relevant here.
That theorem is way too narrow to constrain what you can do in practice. Eg it doesn’t even cover using probalistic election methods, or allowing people to compensate each other.
Jose Pablo
Nov 7 2021 at 3:15pm
“It doesn’t even cover using probabilistic election methods, or allowing people to compensate each other.”
My point was not that it does. But what Arrow’s Theorem does show is the impossibility of transforming individual preferences into “collective” preferences.
And this (impossible) “collective preferences aggregation” is a necessary first step for the ulterior “overriding” of this (impossible) aggregated preference that Josh was advocating.
So, the only way of fully understanding the scope of what Josh seems to be advocating is:
a) we are in a case in which Arrow’s Theorem does not apply (Unanimity, less than 3 alternatives, …) and so an “aggregated preference” can exist. Let’s say unanimity
b) Even if there was “unanimity” in the preferences of all the individuals, somebody or something, could still “overrule” this unanimity to “defend” the interest of some “kind of commons”.
rick
Dec 3 2021 at 4:52am
This idea that people who study humanities aren’t employable is a joke. A degree in a English Lit, Classics or Philosophy program is ideal for law school. English degree in particular is excellent for all manner of white collar work that requires communication and persuasion.
My partners mentor was the head of a large traditional trust bank and only hired from humanities and social sciences, because they showed better aptitude and sales with clients and other skills could be easily taught.
When I was in college less than I decade ago, plenty of English majors were talking about finance jobs as a backup.
Global studies has applications in corporations as well as military. These should be obvious.
Many young people begin with either ideals or confusion that would put them in a low paying public teaching job. That these jobs are low paid is a problem of funding as private teachers with same outcomes get paid MORE but educate students for LESS $. So where is the public school money going because it’s not going to teachers. Who are not the main problem. The administrative bureaucracy is much moreso.
Greg Esres
Nov 4 2021 at 5:12pm
Most parents like their kids’ schools, even while they think poorly of the education system as a whole. So, by your metric, government schools are successful.
Many of these parents are wrong, of course, but the feedback on their error is delayed long after they could use it to make better decisions.
Other countries have a much better K-12 education system, and the kids’ parents are both more satisfied and more justified in that feeling than US parents. This result shows that governments aren’t inherently deficient in providing high quality products to their citizens.
The reason that US K-12 schools are often bad is that our education establishment is ideologically committed to defective pedagaogical approaches, such as Whole Language/Balanced literacy for reading instruction, inquiry-based instruction, and focus on skills rather than knowledge. And they have a reluctance to do what’s necessary to control the behavior of kids to create a proper learning environment. Vouchers won’t fix that.
Jose Pablo
Nov 5 2021 at 2:03pm
“but the feedback on their error is delayed long after they could use it to make better decisions.”
That’s simply not true. Most of the parents in my K-12 public schools have, at the beginning of every school year, a crystal-clear idea of which teachers they want to have for their kids and which ones they want to avoid at all costs.
The “knowledge” about the “quality” of the teachers is out there. Loud and clear. No need to standard test, teachers’ evaluation or any other useless administrative teacher’ evaluation tool. There is simple no available mechanism for the parents to express these opinions and preferences that they do have about the teachers at their local school.
Failing to see the wonders that the existence of this mechanisms would do to the overall quality of public teachers is a very surprising kind of denial.
Floccina
Nov 5 2021 at 4:04pm
I do not see that in the data. I think almost all schools in the developed countries are about the same.
Mark
Nov 5 2021 at 11:14pm
German kids still show better results (TIMMS) than US, and 97% attend government schools (home-schooling is banned). Am I happy with the schooling my kid gets (first year, same primary school I attended in 1976)? Nope. They could easily get better results without some silly “reforms” (nowadays kids of the 2nd + 1st class – 12 each – have to be taught in one room by one teacher – I have seen that happen in poor 1-teacher-village-schools in Turkey, I did never expect to see that in my place! – Here it is not about saving money! – It will pass. The decade before the fashion was: “writing before reading”. Madness.) –
Would I buy at a supermarket with all products 10% more expensive? Eat at BK if I found it noticeable less yummy than McD? Rarely, if I had the choice. And thus McD has more outlets – and the less efficient “real”-supermarkets got taken over now by the better “kaufland”-markets.
Would I put my kid into a school with a less silly approach? Tomorrow, if I could. I have no choice. There is a (kinda private) school not far, seems better, but their waiting list is long – no chance they could just “take over” the less-well-run schools. (schools here are mostly state gov. ruled – local gov. here does … other silly things in silly ways … not much really into “decision making”, mostly just applying the rulz. )
Laron
Nov 4 2021 at 5:24pm
Except in Hawaii, where public education is provided at the state level for reasons unknown to me. That leaves it with one job, construction regulation, which it fails at so spectacularly that Honolulu was the most restrictive area studied in the Wharton Land Use index.
Jose Pablo
Nov 5 2021 at 2:12pm
That was predictable!
If they are left with just one job to do, local governments will devote more resources to this job and will do “more” on this area. Since the only thing they do is “regulation” their full devotion to this, have to result in “more regulation” (there are incentives in place for “doing more regulation” but not, as Bryan suggest, for “doing better regulation”). And that can only mean more restrictions.
Laron
Nov 6 2021 at 3:16am
I hadn’t thought about that, but it makes perfect sense!
Gene
Nov 6 2021 at 1:41pm
Great point. This reminds me of a story I read years ago about how the North Korean government suspected that a certain kind of anti-socialist subversive activity was happening, and so it created a special unit of police/army (I don’t recall which) to investigate this one problem. I remember thinking, god help those investigators if they DON’T discover any of that anti-socialist subversive activity in every place they look!
The consequences of failure among Hawaiian government officials, of course, don’t include all of their relatives being sent to labor camps, but the incentives in both cases echo each other.
robc
Nov 4 2021 at 5:25pm
Those are the exact things that vouchers can fix. (Not sure if they do)
Choose a private school that doesn’t use defective pedagogical approaches. Choose a school that will punish/kick out poor behaving kids. Etc, etc.
Vivian Darkbloom
Nov 5 2021 at 9:42am
Vouchers might be a good solution in urban America with sufficient population density; but, I’m not sure how much good vouchers would do in a town of, say, 1,500 residents. Realistically, what percent of the US population can support more than one local K-12 school? Even if possible, would it be economically efficient to do so?
robc
Nov 5 2021 at 2:06pm
Parents could use the voucher to send the kid to school in the next county over. In some parts of the US, it could be a few hour drive to the nearest decent school in some cases, but parents do that kind of commute for figure skaters and etc.
robc
Nov 5 2021 at 2:08pm
A single classroom school might be MORE economically efficient than a large public school, due to the lack of bureaucratic overhead.
Jose Pablo
Nov 5 2021 at 2:20pm
So, the reasoning goes: “since, maybe, vouchers are not a great idea in towns of less than, say, 1,500 residents they should not be implemented anywhere in the whole country”?
It seems like a poor reasoning to me, akin to say that “since four lines highways are not a great idea in towns of less than 1,500 residents they should not be built in LA or New York”
Vivian Darkbloom
Nov 5 2021 at 5:46pm
Perhaps the very first clause of my comment went right over your head?(!):
“Vouchers might be a good solution in urban America with sufficient population density;…”
Jose Pablo
Nov 5 2021 at 9:53pm
So, the second (and longer) part of your reasoning does not “qualify” the “might be a good solution” part, in any meaningful way, right?
Thanks for the clarification
Felix
Nov 6 2021 at 9:13pm
(IANAE) Your argument seems to be that vouchers wouldn’t work in rural areas because the population density is too small to support the multiple schools that provide the competition necessary for vouchers to work.
I’d suggest that any area with that few students will, at worst, have one school with or without vouchers, but the existence of vouchers means the threat of competition is present, and that is all that matters. If a school only has 10 students in all grades, parents will be very invested in how well it teaches their kids, and the threat of some parents forming their own school for, say, half those students, is very real, and will keep that one school on its toes.
Frank
Nov 4 2021 at 6:10pm
“Second, voters are deeply irrational, even at the local level.”
My neighbors — here in a smug, but not the smuggest, county near DC — just love their local schools! Change could only make things worse.
Therefore, things must be fine in DC, too.
That’s locally rational, if under-informed, and globally irrational.
Steve
Nov 4 2021 at 8:14pm
3. Police
(is police so obviously not in the same category as 1 & 2? i don’t know what to compare to, but maybe local governments do ok with police? maybe they’re terrible too?)
Paul A Sand
Nov 5 2021 at 7:25am
Also missing: fire department. In many places: water, sewer, waste disposal.
Scott Ullery
Nov 6 2021 at 4:25pm
…street maintenance, traffic engineering, parks, stormwater management…
Jose Pablo
Nov 5 2021 at 3:46pm
(… maybe local governments do ok with police? maybe they’re terrible too?)
https://ucr.fbi.gov/crime-in-the-u.s/2017/crime-in-the-u.s.-2017/tables/table-25
They clear around 60% of murders, one third of rapes, 30% of robberies, 17% of property crimes, 12% of motor vehicle theft …
I would not say this is “doing ok”. Closer to “terrible”, I guess. At least by this account. Maybe they prevent a lot of crimes from happening in the first place.
Anyway, relying on being good at activities that cannot be “seen” when you are pretty bad at observable activities, seems a little bit a “desperate” claim to me.
Kaveh pourvand
Nov 4 2021 at 8:30pm
In a completely privatised society, what ensures that the local governments are profit-seeking rather than NGOs more akin to local governments today?
Niko Davor
Nov 4 2021 at 10:46pm
Great post.
Why doesn’t some governor and mayor really push forward with school privatization and school choice initiatives? I suspect that’s not realistic for Democrats. Democratic voters wouldn’t mind, but the teacher unions have lots of strength and influence, and I can’t imagine a Democratic candidate would upset them and remain viable. But Republicans have nothing to lose. Why can’t some Republican local leaders charge forward with school choice and privatization and vouchers?
Betsy DeVos was a pro school choice cabinet pick, she seemed sincere, the tiny change she made were good, but overall, she seemed to have little impact. I suspect she had sincere intentions but didn’t have the skills to make big changes to established bureaucracy.
robc
Nov 5 2021 at 10:29am
Plenty have, and they have passed in many places.
Colorado has a large charter school enrollment. It isn’t full privitization/school choice, but it is a big step in the right direction.
My 5 year old goes to a charter Montessori school. Before we moved to CO this summer, she was in a private Montessori that cost $900 a month. The charter in CO costs $130 per year. The rest of the school is paid for thru the public school system, based on enrollment, so basically a voucher system.
Ken P
Nov 4 2021 at 11:49pm
A good paradigm for many things is read only vs read write. If I build something in tech it’s read write. If I want to build a shed in my own yard, in most locales, I will need permission because the physical world is read only. You need the approval of an admin (via permits) to make changes in the physical world. That slows progress considerably.
Phil H
Nov 5 2021 at 12:11am
There must be some locales that have high use of charter schools. In fact, there are… I remember reading about research on them on this very blog. So, it exists, and yet it is not spreading. Surely worth looking at the factors that are stopping the spread of charter schools, because it’s not all irrationality.
robc
Nov 5 2021 at 10:32am
I believe KY just recently passed a charter school law. So it is still spreading. Maybe not rapidly.
I think someone above hit it right…they may not be able to deliver better results than public schools, but they can do it for a fraction of the price (I think they can deliver better results, but not always, you have to choose carefully).
Since parents aren’t paying directly for the school, that makes the process slower.
robc
Nov 5 2021 at 2:11pm
KY’s law passed in 2017. They became the 44th state to pass a charter school law.
Everett
Nov 5 2021 at 1:00am
Health codes, parks, and road signage/maintenance are the things I notice the most.
Oh yeah, libraries too.
There is a school down the street.
Henri Hein
Nov 5 2021 at 1:56pm
I wonder how much of the problems with K-12 education is due to State and Federal involvement. It seems to me the dissatisfaction with K-12 schools in the US started growing around the time the Department of Education was established.
A lot of the construction regulations are beneficial. Having codes for important things like electricity and foundations improves the quality of housing, at least at the cheaper end. It also sometimes makes it easier for contractors, as they know what to expect. Inspecting the foundations of new construction and recording the result is an important device for establishing some level of trust when houses are bought and sold, since foundations cannot be easily evaluated after the house is put on top. Given the rare case of property disputes, surveying seems like it’s efficiently done. The Public Records department is never in the news. I count that in their favor.
Everett
Nov 6 2021 at 2:02pm
+1!
It would also be nice if districts were rewarded for passing kids through quicker. Not enough to bias them to pumping kids out faster than is beneficial to the kids, but maybe just not yanking their funds if a kid graduates early. Let them keep the funds for the full 13 years and suddenly the HS classes can be smaller, thanks to the earlier graduates, and more focused on the same ability level.
Capt. J Parker
Nov 5 2021 at 2:41pm
Assume a local town government is really acting like a business. It’s product is housing. It will keep producing housing until the marginal cost of producing one more house exceeds the marginal revenue it receives from that house. From the town government’s standpoint the marginal “costs” of producing that last house are:
1) The additional services it needs to provide (police, fire, schools, roads)
2) The loss in tax revenue it will incur by exposing the rest of the community to the negative externality of additional construction and congestion. (the mechanism for this is: negative externality=lower home values for all=lower tax base)
The marginal revenue the town government receives is:
1) The tax revenue from the additional house.
2) Any positive externality the additional house may create (as in: that weed filled eyesore lot next door became a nice new house with nice new neighbors.)
The way the town government can “decide” to produce one more house is through the building regulations. Less regulation promotes construction. More regulation stops construction. (The way to think about this, of course, is that housing developers are making their own decisions on the margin and more regulation adds to their marginal cost)
Now, consider some area like Greater Boston. Many communities to live in and many different balances between housing costs, tax level, services provided and land use and construction regulations. Why isn’t it really the case that our current housing costs are the result of market mechanisms? Alternatively, what is the evidence that the current regulatory environment is not an economically efficient mechanism of dealing with construction and development externalities?
Craig
Nov 5 2021 at 6:37pm
Another thing about vouchers that people often forget is that vouchers will allow many parents to choose to live in communities they would refuse to live in without vouchers because they simply won’t send their kids to the schools in that town and while they might be middle class or even upper middle class, private school might be financially burdensome.
Matthias
Nov 6 2021 at 4:19am
Your article is quite US centric.
In any case, don’t local governments there also run the local monopolies on eg supplying water and taking away sewage? Or waste disposal?
Those are pretty important jobs, and they seem to be done a bit more competently than the two jobs the article mentioned.
(Of course, the private sector might run those better yet. I don’t know, and it’s beside the narrow point.)
Slightly tangentially:
Governments all around the world and at different levels and different points in history provide lots of different services.
Some of those are horrendous, some of them are run quite well, many are run so-so. I wonder whether we can find some simple patterns.
Matthias
Nov 6 2021 at 4:20am
I assume part of why vouchers are so unpopular are because they are only one step away from just giving people money.
Jason S.
Nov 8 2021 at 1:38pm
It’s not fair to blame local governments for the policy of funding systems not students. All 50 states require local governments by statute to provide public schooling free of charge. Local governments merely manage public schools. Do they do that well? Perhaps not, but it does seem that states with a large number of competing local governments do better on school quality than states with few competing local governments (or Hawaii, where the state runs the schools). https://wallethub.com/edu/e/states-with-the-best-schools/5335
Niko Davor
Nov 11 2021 at 11:09am
Even there, there are Republican state governments very sympathetic to school choice, vouchers, and privatization. And during the Trump Administration, there was support at the federal level too. It seems that the opportunity exists if some well coordinated groups were to take advantage of it.
nobody.really
Nov 9 2021 at 9:47am
While Caplan overstates his case, his main points are that we could improve the ways we finance schools and do zoning.
Relatedly, the NYT offers a 15-min. video discussing how Democratic-controlled jurisdictions are among the worst regarding zoning restrictions, education funding disparities, and progressive taxation.
Comments are closed.