The problem with Bryan’s critiques is that they miss what we are trying to explain which is why some prices have risen while others have fallen. Immigration would indeed lower health care prices but it would also lower the price of automobiles leaving the net difference unexplained. Bryan, the armchair economist, has a simple syllogism, regulation increases prices, education is regulated, therefore regulation explains higher education prices. The problem is that most industries are regulated.
Suppose that there are two sectors, apples and string quartets. We observe that over the past 30 years, the relative price of string quartets has risen.
Using basic supply and demand analysis, we know that this could be a combination of four things:
1. A favorable shift in the supply curve for apples.
2. A downward shift in the demand curve for apples.
3. An unfavorable shift in the supply curve for string quartets.
4. An upward shift in the demand curve for string quartets.
The “Baumol effect” story says that it’s almost entirely (1). The main claim that Tabarrok and Helland make in support of this view is that the change in relative prices has been steady, so we need a steadily-changing factor to explain it. Regulatory changes are more herky-jerky. Bryan Caplan objected to this, and Tabarrok comes back with some new arguments.
Here, for example, are two figures which did not make the book. The first shows car prices versus car repair prices. The second shows shoe and clothing prices versus shoe repair, tailors, dry cleaners and hair styling. In both cases, the goods price is way down and the service price is up. The Baumol effect offers a unifying account of trends such as this across many different industries. Other theories tend to be ad hoc, false, or unfalsifiable.
Oh, please. In 1950, imports of shoes and cars were low. In later decades, they shot up. But car repair and shoe repair don’t face import competition.
In fact, it could be that the main reason that the prices are relatively high in health care and education is that they do not face import competition. That also would explain why a lot of us don’t feel richer. If the favorable supply curve shift were all due to domestic productivity gains, our incomes would be a lot higher. But a lot of the favorable supply curve shift comes from foreign supply added to the market. That is not a Baumol effect, It is a traded goods/non-traded goods effect.
I would like to expand Arnold’s list of mechanisms that explain “What Gets Expensive, and Why”:
1) The Baumol effect
2) Category creep
3) Demand de-linked from outcomes
4) Taxpayer subsidies for demand
5) Government restrictions on supply (Regulation)
6) Government production of the output
Arnold links no. 4 (Subsidies) and no. 5 (Regulation) in the phrase, “Subsidize demand, restrict supply,” because often the two go hand in hand. But these mechanisms are analytically distinct.
Bryan Caplan has penned a 2nd critique of Helland and Tabarrok:
https://www.econlib.org/gratis-and-extravagance/
Bryan’s 2nd critique is what prompts me to add mechanism no. 5, Government production of the output.
Bryan proposes what seems like a testable hypothesis:
“When government supplies free (or highly subsidized) products, we should expect private suppliers to supply gold-plated versions of the standard government-issue. As a result, their prices and costs will be closely synced.”
He concludes:
“when there is a dominant public sector with a private fringe, we should expect the Helland-Tabarrok pattern – [with] or without a Baumol effect. Given the pathologies of the public sector, it’s nice to have a private option, but don’t get too excited. The private options will generally be extravagant, because gratis government strangles budget-friendly alternatives. Tragically, this in turn sustains the illusion that in the absence of government, only the rich would be able to afford whatever the government now provides.”
Note: Government production plays a much greater role in education than in health care. Presumably, the empirical weights of the several mechanisms at work in d*mn high prices differ between education and health care.
Typo:
My comment should read: “Bryan’s 2nd critique is what prompts me to add mechanism no. 6, Government production of the output.”
In private K-12 education most of the affordable options (that cost under $10k/pupil/year) have vanished. Ultimately, either you are buying expensive prep schools or you are buying something not distinguishable enough from public school to be worth additional cost.
If you gave everyone a voucher for $10k/year (less I think then we spend today by a lot) I suspect there would be tons of budget friendly alternatives.
I don’t think it’s clear what would happen with vouchers.
To a large extent, what people are wanting to buy for their kids is status and the ability to select the quality of their peer group. That means selecting-out less desirable types, which is another way of saying discrimination which will necessarily have disparate impact, and the only legal way to do that is to screen via affordability via the price mechanism, which sets up a zero-sum scenario. So, the unaffordability and high prices that exclude large parts of the population is the point, and can’t be remedied.
Would a voucher magnet school that uses test scores to discriminate (again, as an imperfect proxy for peer quality) like Thomas Jefferson or Stuyvesant be legal for long?
The two ‘eliteness’ proxies correlate with what people want, but only roughly. Mostly people aren’t really concerned that their kids peers are the richest and brightest: they’re more concerned about character, behavior, values, compatibility, and lack of bad influence. Also mate-worthiness.
The way people try to discriminate on that basis is to apply a religious test and impose requirements that make the organization unattractive to most people without the target qualities, which again operate as a kind of costly screening mechanism.
“Would a voucher magnet school that uses test scores to discriminate (again, as an imperfect proxy for peer quality) like Thomas Jefferson or Stuyvesant be legal for long?”
I don’t know. It is somewhat beyond my ability to speculate. My own did but with difficulty. Charters remain popular in many places, though not all are “magnets.”
I see two drivers for non-elite vouchers.
1) Urban schools are dangerous. This can be solved by removing the bottom 10% and having better discipline methods.
2) Public schooling is ideologically dangerous, and elite private schooling isn’t any better.
Is there any way to help regular people solve these problems in a middle class budgetary conscious way (either free or as low as 1-2k per kid per year)?
Obviously, we know the impediments that exist today. We also know the successes (charters, limited) and failures (many non-elite religious schools had zero idea how to differentiate themselves or respond to progressivism).
I guess I’m just talking my book, what I’d like to have available.
One of the most important ways we get richer (or have higher welfare, well-being, utility, civilizational capital – however you want to put it) is to overcome constraints leading to zero-sum tournaments and Malthusian scenarios. Mostly, economists tend to focus on the role of innovation of physical technology in reducing the relative scarcity of commodities and goods, e.g., agricultural tools, chemicals, and techniques that increase the productivity of land so that calories per acre rise faster than the population. Or fracking making the supply of petroleum more elastic and thus affordable.
But what has been less studied and emphasized is the role of certain cultural and social technologies (including, but definitely not limited to, state policies) in mitigating certain zero-sum problems. I think “Social Technology” is superior to “Culture” or “Formal Institutions and Informal Folkways” (of which it is a subset) since it makes the utility of this resource-expansion analogy more explicit.
Yes, a few people like Elinor Ostrom have studied things like coercive cultural practices to limit externalities of over-extraction that do not rely on the state and which deal with physical “tragedy of the commons” scenarios with constraints deriving from scarcity of limited resources.
But there are also “Tragedies of Social Rat Races” where people are competing for limited “Social Resources”. Status is an obvious example – critically important for success in a sexual free marketplace – but “quality peer environments” / “quality neighborhoods” are another. If everybody is trying to appear better-than-average, or competing with each other to select better-than-average people to shape their social environment, there are only ever so many better-than-average folks to go around. And that leads to a rat race which causes people to expend all their surplus resources in an effort to not be a loser.
Let’s say that most parents are terrified of sending their kids to a school where they are likely to be physically assaulted. Well, if they can’t rely on the schools to create a reliably secure environment by, for instance, imposing strict discipline, with harsh punishments and swift removals of all bad apples (regardless of how the identity statistics come out), then those parents will try to find whatever legal proxy is available that correlates however roughly with “propensity of kids to commit assault” (e.g., household income), and then will clamber over each other to bid up the price of more selective environments which tend to screen out more dangerous elements, which sets up an obvious “Tragedy of the Social Rat Race” situation: the one is which we all now live.
But the “Social Technology” (a “traditional” or “conservative” one at that) of imposing sufficiently strict and effective discipline and security policies on everyone without fear of liability for disparate impact substantially lowers the legitimate concern, and thus significantly mitigates the wastefulness and harm of the Social Rat Race.
Same goes for neighborhood security. If the police are ineffective, then “safe, quality neighborhoods” become a scarce resource, and, again, a neighborhood is defined by the people who live, work, and transit through it, and there are only so many reliably pro-social potential neighbors to go around. On the other hand, the Social Technology of policing that is sufficiently effective to ensure all neighborhoods are reliably safe increases the supply, alleviates the scarcity, and neutralizes the surplus-draining scramble for the few better bubbles. If you don’t have to worry about the rabble because the police effectively deter them, then it also means that there is less political support for consequential coping-mechanism policies that keep prices high to keep that rabble out. For example, requiring large lots specifically to waste scarce land resources and raise prices, so the neighborhood stays unaffordable.
Likewise, policies, education focused on traditional virutes and character formation, folkways, and other elements of Social and Cultural Capital (again, “Social Technologies”) can go a long way to raise the general trustworthiness of members of one’s community – even strangers – and lower the amount of resources one has to allocate into various forms of protection and insurance from less scrupulous counterparties. Capitalism, private propety, free markets, and the institutions which support and maintain their functioning, are all Social Technologies too.
Some of the major resource-limitation constraints we face are technologically determined, like the zero-sum problem with central real estate in Winner Cities in an era of continuing centralization and the increasing significance of economies of agglomeration. That’s something like a “natural disaster” in terms of it being caused by large, impersonal forces and trends.
But other problems are more like “man-made disasters”, a result of the progressives intentionally breaking the old Social Technogies in the name of “Social Justice” deriving from the false and insane doctrine of human sameness and statistical uniformity. Suppressing such destructive tendencies – akin to luddite sabotage, except throwing sand into the gears of old traditions instead of new machines – was itself a Social Technology – indeed, perhaps the most important one, a meta-Social Technology protecting the others. And once that defensive tech was disabled, it was only a matter of time before all the others were exposed to dismantling.
Robin Hanson proposes a mechanism, “the increasing-focus-on-prestige effect,” which I would like to add to Arnold’s list of mechanisms that explain “What Gets Expensive, and Why:”
1) The Baumol effect
2) Category creep
3) Demand de-linked from outcomes (Cf. Kling)
4) Taxpayer subsidies for demand
5) Government restrictions on supply (Regulation)
6) Government production of the output. (Effect = Gold-plated private alternative. Cf. Caplan.)
7) Prosperity effect on demand for prestigious services. (Cf. Hanson.)
My intuition is that no. 3 and no. 7 are related, but distinct.
Here is a link to Robin Hanson’s blogpost about d*mn high prices:
http://www.overcomingbias.com/2019/06/our-prestige-obsession.html
Bryan Caplan rebuts Alex Tabarrok’s rejoinder:
https://www.econlib.org/why-the-prices-are-so-damn-high-reply-to-alex/
Here Bryan emphasizes taxpayer subsidies and immigrations restrictions (i.e., Arnold’s “Subsidize demand, restrict supply”), rather than government production of the output.
Without delving deeply into the analyses of North, Wallis &Weingast, we might observe that as social structures and economic activities have become more “complex,” the relationships in transactions and interactions have become more and more impersonal, creating the requirements for “intermediaries” (or intermediary services), often at many stages, in the quests for objectives (including the objective of avoiding or shifting costs). Needs for intermediation have given rise to “opportunities” for “interventions” (which have many names).
Intermediations and interventions carry with them costs. All costs must be covered; and prices, rather than diminution of goods or services (including their distribution) become an appropriate allocator.
To repeat, the more labor-intensive the sector, the more regulated it effectively is, because immigration restrictions are restrictions on the supply of labor.
I think this is kind of cheap on Bryan’s part. Okay, sure, maybe this is true in a narrow sense, but the obvious rejoinder here is that healthcare and education were a lot cheaper back before the US liberalized its immigration laws in the 1960’s, so I don’t think this actually rules out Baumol as a big driver of higher costs.
Alex misses my most undeniable point: the education industry, unlike the auto industry, is funded almost entirely by government – and this funding has swelled over time. If government had not increased funding for education, education prices would have risen far less.
Did Bryan even make that point the first time around?
…healthcare and education were a lot cheaper back before the US liberalized its immigration laws in the 1960’s
Immigration laws aren’t the issue. It doesn’t matter how easy it is for foreign MDs to come the the U.S., what matters is how easy is it for them to practice medicine. And it’s actually very hard. They need to repeat residency, which means they need to grab a residency slot away from somebody else. Even if they’re successful in this, there’s no increase in supply of MDs. Many foreign MDs who come here are never able to practice medicine:
https://www.pri.org/stories/2018-03-26/highly-trained-and-educated-some-foreign-born-doctors-still-can-t-practice
In addition, Tabarrok claims federal regulation can be demonstrated to be a constant across all industries, which even if true, ignores the much more substantial role of state regulation in education and health care relative to other industries.
Taborrok has previously decried the role of occupational licensing, which is primarily a state function, approvingly citing a study attributing $200 billion annually in increased costs, on a study using a methodology “exploiting “state variation in licensing law.” See Marginal Revolution December 7, 2018. In doing so, he rebuts both his Baumol’s Cost Disease theory of everything and his regulation is an irrelevant constant claims.
Both Taborrok and Caplan agree that increased immigration would solve all of our problems. And possibly dumping ebola-exposed illegals across the country will indeed accomplish that, but they don’t explain how increased immigration will change state occupational licensure regimes and their impact.
Too little, too late.
One of the founder’s worst conjectures was that free speech and a free press would promote political accountability.
It didn’t work during the age of newspapers and it is not working in the age of social media.
Meanwhile the electoral process becomes ever more corrupt. With no effective oversight of state electoral systems, and increasing numbers of state secretaries, attorneys general, and prosecutors on the Soros payroll, domestic terror groups like antifa are increasingly free to violently intimidate voters and suppress dissent.
And a sufficient number of the states have now turned their presidential electors over to California to ensure that in the next six years the US will become a one-party dictatorship. California voter rolls are not subject to any oversight, meaning that using vote harvesting techniques refined in 2018, California will be able to supply whatever number of votes are necessary to have the Democrat candidate elected in 2020.
The USA is over.
Nothing can be done to avert it.
I highly recommend Liao Yiwu’s “Bullets and Opium” to get a taste for what dissenters are in for. Undoubtedly the best book of the year and deserves the Nobel in literature this year.
I so not think their data on auto prices is correct, and I mean massively.
An upward shift in the demand curve for string quartets.
On some level the Baumol effect would also predict this, because anything you can possibly spend money on is (however indirectly) a substitute for anything else you can spend money on. If apples get cheap enough, it frees up cash for string quartets, which raises demand for them.
When apple production gains productivity, we store fewer of them at home because we can get them easier. Definition of productivity, consumer loans on apples drop in rate, shorten in term.
The is no extra cash in the deal (all things being equal) , so string quartets get less cash and thus respond by upping scale, going to larger auditoriums with more customers.
Change of scale is automatic, that is the Baumol effect. All thing being equal, a change in productivity in one sector leads it to move left and all other sectors move right on the yield curve. That is the Baumol effect. It is fundamental, it is fundamental to avalanches, bee swarms, quarks, prime numbers and humans doing transactions. It is fundamental to first grade arithmetic.
It was misleading to use “service prices” when discussing a Baumol phenomenon: prevailing sectoral wages (also corrected for local cost of living) make a lot more sense, which means you have to disaggregate the cost into component inputs. If hair stylists are more expensive mostly because the rent for hair salons has shot up while the wages of stylists has stagnated in real terms, then “Baumol Effect” is not a good explanation after all. Hair styling – like all the jobs I call “proximate services” – is inherently proteced from import competition, but if real wages are stagnant, then import competition isn’t a sufficient answer either. You could start adding a variable about specialized labor scarcity when there is a high human capital threshold, and while that is probably true, once you start adding variables, you can ‘prove’ anything.
(And of course how to correct nominal to real and the fundamental and intractable problem of index choice and measuring price level inflation makes the entire exercise fraught to begin with.)
Think about the other price that has shot way up, the one that orginally earned the “Too Damn High” complaint: Rent. Rent rise is obviously not a Baumol effect. Ironically, here Tabarrok would immediately pin the blame on government regulation, restricting supply and subsidizing demand – but the rent rise has been smooth-ish, so his own ‘herky-jerky’ rejoinder about instances involcing regulation would apply. Also, the prices for these TDH are too damn high on a global basis, in a way that seems orthogonal to local regulation. Indeed, the places where the prices for these things seems stable-ish and under control are precisely those countries in which the government has regulated prices and supply in order to keep the costs down!
You know what else is “too damn high”? Taxes and, let me assure you from my experience working in government – pretty much anything taxes buy, which we call “cost disease”.
Instead, let’s step back and look at all the “Too Damn High” sectors, and see if we can derive any more general and abstract “common thread”.
What do Rent, Health Care, Education, and Taxes have in common in a developed economy?
They are Neo-Ricardian. You can’t achieve anything approaching your best-available opportunity for utility without paying those pipers. The gap between the “pay the pipers” lifestyle and the “don’t pay the pipers” lifestyle is what they can extract, and normal market mechanisms will naturally lead in the direction of lowering that gap down to zero such that agents are indifferent between the alternatives.