Special Features of the Baumol Effect

I explained the Baumol effect in an earlier post based on Why Are the Prices So D*mn High?. In this post, I want to point out some special features of the Baumol effect that help to explain the data. Namely:

  • The Baumol effect predicts that more spending will be accompanied by no increase in quality.
  • The Baumol effect predicts that the increase in the relative price of the low productivity sector will be fastest when the economy is booming. i.e. the cost “disease” will be at its worst when the economy is most healthy!
  • The Baumol effect cleanly resolves the mystery of higher prices accompanied by higher quantity demanded.

First, in the literature on rising prices it’s common to contrast massive increases in spending with little to no increases in quality, as for example, in contrasting education expenditures with mostly flat test scores (see at right). We have spent so much and gotten so little! Cui Bono? It must be teacher unions, administrators or the government!

All of that could be true but the Baumol effect predicts that more spending will be accompanied by no increase in quality. Go back to the classic example of the string quartet which becomes more expensive because labor in other industries increases in productivity over time. The price of the string quartet rises but does anyone expect that the the quality rises? Of course not. In the classic example the inputs to string quartet playing don’t change. The wages of the players rise because of productivity increases in other industries but we don’t invest any more real resources in string quartet playing and so we should not expect any increases in quality.

In just the same way, to the extent that greater spending on education, health care, or car repair is due to the rising opportunity costs of inputs we should not expect any increase in quality. (Note that increases in real resource use such as more teachers per student should result in increases in quality (and perhaps they do) but by eliminating the price increase portion of the higher spending we have eliminated a large portion of the mystery of higher spending with no increase in quality.)

Second, explanations of rising prices that focus on bad things such as monopoly power or rent seeking tend to imply that price increases should be largest when the economy is doing poorly. In contrast, the Baumol effect predicts that increases in relative prices will be largest when the economy is booming. Consider health care. From news reports you might think that health care costs have gotten more “out of control” over time. In fact, the fastest increases in health care costs were in the 1960s. The graph at left is on a ratio scale so slopes indicate rates of growth and what one sees is that the growth rate of health expenditures per person is slowing. That might seem good but remember, from the Baumol point of view, the decline in relative price growth reflects slowing growth elsewhere in the economy.

Third, holding all else equal, the only rational response to an ordinary cost increase is to substitute away from the good. But in many rising price sectors we see not only greater expenditures (driven by increased prices and inelastic demand) but also greater quantity demanded. As I showed earlier, for example, we have increased the number of doctors, nurses and teachers per capita even as prices have risen. John Cochrane correctly noted that this is puzzling but it’s a bigger puzzle for non-Baumol theories than for Baumol. For non-Baumol theories to explain increases in the quantity purchased, we need two theories. One theory to explain the increase in price (bloat/regulation etc.) and another theory to explain why, despite the increase in price, people are still purchasing more (e.g. income effect). The world is a messy place and maybe that is what is happening. But the Baumol effect offers a cleaner answer.

A Baumol increase in relative price is always accompanied by higher income so it’s much easier to explain how price increases can accompany increases in quantity as well as increases in expenditure. The Baumol story for increased purchase of medical care even as prices increase, for example, is no more mysterious than why people can take more leisure when wages increase–namely the higher wage means a higher income for any given hours and people choose to take some of this higher income in leisure. Similarly, higher productivity in say goods production increases income at any given production level and people choose to take some of this higher income in services.

Summing up, if we examine each sector–education, health care, the arts, etc.–on its own then there are always many possible explanations for why prices might be increasing. Many of these explanations have true premises–there are a lot of administrators in higher education, health care is highly regulated, lower education is government run. But, on closer inspection the arguments often don’t fit the data very well. Prices were increasing before administrators were important, health care is highly regulated but so is manufacturing, private education is also increasing in price, the arts are not highly regulated. It’s impossible to knock down each of these arguments in every industry, so there is always room for doubt. Indeed, the great difficult is that these factors often do result in higher costs and greater inefficiency but I believe those are predominantly level effects not effects that accumulate over time. Moreover, when one considers the rising price industries as a whole these explanations begin to look ad hoc. In contrast, the Baumol effect appears capable of explaining the pricing behavior of a wide variety of industries over a long period of time using a simple but powerful and unified theory.

Addendum: Other posts in this series.


So, basically, the market for labour is constrained by the supply of labour? As demand for labour in other industries increases, the supply of labour in health/education/etc. would decrease, unless there was a commensurate increase in price?

Wiki seems to agree with you:

Baumol's cost disease is often used to describe consequences of the lack of growth in productivity in the quaternary sector of the economy and public services, such as public hospitals and state colleges. Labor-intensive sectors that rely heavily on non-routine human interaction or activities, such as health care, education, or the performing arts, have had less growth in productivity over time. As with the string quartet example, it takes nurses the same amount of time to change a bandage or college professors the same amount of time to mark an essay today as it did in 1966, as those types of activities rely on the movements of the human body, which cannot be engineered to perform more quickly, accurately, or efficiently in the same way that a machine,

Economics is about indivisible humans. Enticing a musician o build cars takes time. But humans are not a regular market, we cannot trade humans in open auctions, too much private information gets released.

So, then. Why not call the Baumol a labor market model and quit confusing?

There is another effect, a consumer price rise causes prices to spread, temporarily, and that effect is easily confused with Baumol.

Except that it doesn't.

Nurses can now use precut, preformed bandages with vastly better adhesive management and change the average hospital dressing vastly faster than even 20 years ago.

College professors can now mark an essay by uploading it to software that can do the base tedium of checking spelling, grammar, etc. Even ignoring the completely automated grading options, software has made grading much easier. For instance, laboriously writing out comments on essays is vastly faster with a word processor than by hand. It gets faster still if you use software libraries to cut and paste recurring comments from class to class or year to year. Plagiarism detection and proof has gone from an arduous process to something all but automated. And of course software makes it faster to "hand back" assignments outside of class hours allowing for less wasted time in class.

We may not change the movements of the human body (which actually is highly debatable as modern Olympians demonstrate), but we can certainly adapt with pre-cut bandages using highly water resistant bandages or other technological adaptions.

Sort of. Better to say that the supply of labor in these industries follows the equilibrium of the labor market wages in industries that have productivity gains. So, for example, if first grade teachers and paralegals both had $3,000 wages in 1960, that would be about $25,000 in inflation adjusted dollars today. But if the paralegal wages rose with gains in legal research technology to $50,000 (as they have), then potential first grade teachers would be pulled from education to law (and other such sectors) if the wage of the first grade teachers didn't rise to follow the productivity and inflation gains in those other sectors.

Here's what I find confusing. Consider higher education. Baumol's argument implies that professor's salaries should rise in accordance with what they can get in other, higher-productivity industries that they could potentially enter. This implies that wage growth in higher education should rise at about the same pace as the wage growth of more highly educated people. And yet, the price of education rose significantly faster than the growth of wages of highly educated people.

So maybe Baumol's cost disease explains a portion of the increase, but I doubt the whole thing.

Didn't you switch from salaries to total costs there? My understanding of the data presented so far is that salaries rose with general wage inflation, but total costs rise more where many salaries are involved.

Teacher/Student ratio is higher than Smartphone-designer/Smartphone-user ratio.

"Teacher/Student ratio is higher than Smartphone-designer/Smartphone-user ratio."

The introduction of the subject of fractions is one of the first milestones where future scientists and engineers are separated from future social justice warriors.

Except, teach salaries have NOT gone up in university education by anything like what the tuition has done.

That's a good point. Aren't there also a lot of under-employed Ph D's out there muddling along doing post-doc fellowships? Why are these people not driving wages down? Is it because education is a superior good and having professors making 60k per year isn't good for public perception?

$60k isn’t an unusual salary for a professor, outside of big public universities. And even those big public universities have a lot of courses taught by “lecturers” and kinds of faculty.

We have a big Baumol effect in sports, but quality of the athletes does go up: e.g., look at field goal kicking percentages in the NFL over the decades.

NFL placekicker salaries rank from $500k to $5 mil:


So dads send their teen sons to Chris Sailer's kicking academy and kickers no longer get offseason jobs down on the docks but instead just practice their kicking.

LOL, brilliant.

It's likely that string quartet quality has gone up considerably over the centuries. A richer society can afford more instruction and practice for violinists.

Probably, but who cares? They are still playing the same music written centuries ago. Either you can play it, or you can't.

It's like being a teacher. Not everyone can be good at it, but the fact is, you only have to be "so good" at it before the task is fulfilled.

With regard to music we have a substitute good: recorded music, which nowadays is of very high quality. If attending a symphony performance costs too much you can buy a CD or download mp3 files. There's no equivalent to that in healthcare and insofar as education is about credentialing there isn't much that serves as a low cost substitute in education.

Unless you are into nostalgia, signaling, or the visual or participatory aspects of musical performance, the substitute (recorded music) is superior to the actual good (live music). (The participatory aspect is quite broad. It could mean anything from: you like clapping along with everybody else to your favorite tunes or you want to get in the musician's pants.)

The best things about recorded music are that: 1) you can skip over the sections of music that you are disinterested in, 2A) you don't have to sit still for 1.5 hours (in the case of orchestral music), 2B) you don't have to get pushed around (in the case of popular music), 3) other people's noises don't interfere with your listening, 4) you can almost always find a recording where the musical arrangement is better and the musicians make fewer mistakes, and 5) the cost is much lower

In the case of popular music where composers work directly with the musicians, if there are fewer live shows they can actually produce more music. Spending less time touring allows for less burn-out plus more time to learn more music.

IDK. The quality of painting/sculpture has gone way down. The great stuff was all done circa 1700. The modern stuff seems comparable to circa 1200 works.

No, we don't have a Baumol effect in sports. It's the same fallacy as with the string quartet. The measure of productivity for entertainment is not the number of performances per performer but rather the number of audience members reached. Big-league sports salaries exploded because the productivity increase associated with TV contracts brought an ocean of money into that end of the industry. That also explains (which Baumol can't) why minor league players, who are exactly as productive as big leaguers in terms of games played per unit of time, make subsistence wages.

Baumol probably applies more if you try to hire a generic, fungible, string quartet, and less so if you try to hire Yo-Yo Ma.

I think you're confusing Baumol effects and winner-take-all games. The productivity of a string quartet has not improved, you still need a quartet of approximately the same level of training as ever.

What *has* improved is the efficiency of the delivery system, such that the quartet deemed even slightly better than all the others might get a lot of exposure and payoff via recordings, streaming, etc.

But Baumol still applies: if I want to hire a run-of-the-mill quartet for a wedding, they will still only be on offer at a seemingly inflated price (though not as inflated as the *best* quartet would be.)

The wedding example fits my point that productivity is properly measured by audience size.

And "winner take all effect" is just another way of saying that the best players are far more productive than ordinary ones, in terms of the economic activity that they are capable of generating. Baumol thus doesn't apply to big league sports. And again, it fails to explain minor league ball, whose players make next to nothing.

Second, explanations of rising prices that focus on bad things such as monopoly power or rent seeking tend to imply that price increases should be largest when the economy is doing poorly.

OK -- somebody is going to have to explain that one to me. Have the unions (both public and private sector) been able to negotiate their richest contracts with the greatest bonuses, wage-increases, and job protections when the economy is doing poorly?

He wasn't talking about unions. Lefties like Alex do not consider unions to be "bad things."

When he talks about the evils of monopolies, you can be absolutely certain he is not talking about Labor Monopolies.

'Lefties like Alex do not consider unions to be "bad things."'

Probably the winner of most hilariously uninformed comment of 2019. Enjoy reading this post of Prof. Tabarroks, which opens as follows - 'Union violence against non-unionized workers and their employers is an under-reported story. Everyone knows it happens but they look the other way. It’s hard to look the other way, however, when the violence and vandalism is videotaped and put on the web–that’s what two Philadelphia developers did and the result is making waves far beyond the workplace.' https://marginalrevolution.com/marginalrevolution/2013/01/uncovering-union-violence.html

And it ends this way - 'Read the whole excellent piece. Hat tip to EconJeff who notes that the article is flawed only by a bit of lazy history suggesting that the 40 hour work week and good working conditions are primarily due to unions (there is also a bit of weak economics in a suggestion that the wages of builders and city rents should be closely related). Still it’s a very astute article that connects the dots in the iron triangle.'

Prof. Tabarrok a union supporting lefty - truly laughable, and a fine example of why the MR comment section is so entertaining.

Libertarians: regulation and moral hazards / bad incentives are why the productivity growth is so bad.

Alex: no no no, Baumol effect, not government interference, explains why prices rise in low productivity growth sectors.

There is obviously an opportunity here for public spending to improve productivity, to reduce public spending in the long run.

But perhaps that's a hard place to get to in today's politics.

Any specifics? Or just your usual platitudes and trolling ?

So .. on this page there is a thing called Marginal Revolution University .. but you think *I'm* trolling.

We mostly just think you're an idiot. Trolling or otherwise.

Right, the ankle-biters who come through with content free and context free insults are the real geniuses.

You deserve nothing less and nothing more.

And you show yourself for what you are. Not one who has any constructive comment on productivity improvement in education or medicine. Just someone wanting to add stink to a serious discussion.

I love you, man, warts and all.

I mean, if I have to feed it to you by the spoon-full,


but I really expect MR readers to know these things.

A link to EdTech.

So you were, in fact, trolling the entire time.

The efficacy of online programs will be entirely determined by the signaling to human capital formation ratio.

People don’t go to Harvard to learn things.

You haven’t demonstrated what you’re purporting to have demonstrated.

If your theory was correct, wage increases should be directly proportional to years of college.

Instead, it’s a step function upon graduation.

Three years of college give essentially no wage gains compared to a high school degree.

If it were human capital formation, it would approximate a linear function as human capital accrues.

You’re way out of your depth. Stick to random trump trolling.

And this is why he's to worth responding *to*.

Does anyone see "my" theory above? No, I posited no new and unique theory of my own, I just linked to research and progress around the world.

You googled a gated paper. That you haven’t read.

A paper using IV to try to tease out private and social returns to education in Norway.

Try again?

"You’re way out of your depth. Stick to random trump trolling."


"That's how you fight the cost disease."

The demand is for machinists, plumbers, welders, manufacturing engineers, etc, so where have MOOCs demonstrated the low cost production of skilled machinists, welders, plumbers, manufacturing enginerpers and technicians?

Any education that requires even a few thousand in capital plus spending on consumption can not be done "online".

I know teaching the principles of welding, machining, etc can be done online in an hour, but its easier to combine it with the half hour of time in a classroom before starting the 6 hours in the lab with each student using one of the $5000-$50,000 machines in the $500,000 real estate building with the electrical service, the HVAC, the floor load bearing, capacity needed to support teaching manufacturing science.

Anonymous has been mulp all along.

In spirit.

These are complimentary explanations. The libertarian critique explains why it is low productivity and the Baumol effect explains why the prices are high.

That's my point.

Libertarians know how to manufacture 9nm feature size electronic in a baseement shop with sodering iron and a few issues of popular electronics or a few issues of CQ from the 60s?

The cost of electronics production facilities have increased from the cost of a garage and soldering iron and wire cutters to something in the range of $1B-$10B.

What was missing from Ayn Rand was the tens of thousands of workers required to produce just the things she handwaved as the production of one man. The entire value chain of producing just the steel rail was ignored as if god created steel rail to be dug up from the ground, a god which did not exist, because real men aka libertarians, created the universe.

I accept Tabarrok's explanation of the Baumol effect, but the usual suspects, higher education and health care, have other influences that increase demand notwithstanding price increases.

For higher education, the conventional wisdom is that one can't live without it (or cannot live very well), so families make all manner of sacrifices to consume higher education. At the societal/political level, higher education cannot be only for the rich, so subsidies and funding sources are created so that the children of the not rich can consume higher education and experience the illusion that upward mobility in America is possible. But the rich still consume higher education at the elite colleges, while the not rich are sent to the not elite universities. As for productivity, many of the consumers of higher education today are not well prepared for it, so the productivity of the teachers necessarily suffers: it takes more time to teach the same course materials to a student population that is trending downward in preparation.

For health care, the conventional wisdom is that one can't live without it (or can't live very long without it), so families make all manner of sacrifices to consume health care. At the societal/political level, health care cannot be only for the rich, so subsidies and funding sources are created so that the most important constituents (e.g., seniors who vote in large numbers) can consume health care and experience the illusion that the same life expectancy is possible for every America. But the rich still consume health care at the elite health care facilities with the elite health care providers, while the not rich consume health care at the clinic and the public hospital. As for productivity, many of the consumers of health care today are not in very good health, whether due to lifestyle or genes, so the productivity of the health care providers necessarily suffers: it takes more time and effort to maintain the health of the unhealthy in a population with health status that continues to trend downward.

Good one, very clever. I like it.

rayward's batting average is low, but he crushed this one.

Try a problem without labor cost effects, as in:
Assume a general rise in goods prices from an exogenous event. Labor markets do not react fast enough, nor does productivity react, the shopkeeper must assume he now has a larger set of poorer customers. He will need to buy cheap stuff in higher quantities and expensive stuff in lower; losing economies of scale for the one and gaining economies for the other.

The result is a skew of store flow that happens even before he sets prices. His prices are set to cover the economies, the intended price hike will be lower fro cheap items and higher for expensive items. The customer basket size increases, poor customers need bigger baskets to buy less often and minimize their share of the general price rise.

Then the shopkeeper sets prices, he sets prices to cover the extra inventory needs for customer jitter, and he knows how to transform the flow distribution. Queue management first, setting prices second. Or, in other terms, PSST from Kling first then Econ 10 from Mankiw second.

The Baumol effect cleanly resolves the mystery of higher prices accompanied by higher quantity demanded.
My PSST solution solves it better. The price spread increases but the economies of scale favor poor people when we suddenly create more poor people.

The Baumol effect predicts that the increase in the relative price of the low productivity sector will be fastest when the economy is booming. i.e. the cost “disease” will be at its worst when the economy is most healthy!
This effect needs a labor market. I can explain the other two effects with a neural transformation of the queueing structures. This part:

"relative price of the low productivity sector" means labor prices for musicians which are their product prices. This is a different effect, it is not a distribution change from exogenous monetary inputs effecting prices and involves labor pricing with respect to booming labor demands. Pick Baumol for this labor problem, then debate the labor market until you are blue in the face.

It is just a hypothetical change in the money measure, stick with number theory, it should predict a new configuration from the old and the price change, a probability transformation.

Take Walmart, for example. Tell Walmart to compensate the soap inventory for a 10% hypothetical, and mysterious price hike everywhere on goods.

The Walmart manager goes to the soap bin and adds more space for plain ivory soap and less for scented dove. She has done that so many repeated times, she computes, right away, the change in soap variance in the bin. On the spot, and they have just enough clerk flexibility to adapt to change of scale at the checkout counter. Once they know inventory variance, they can count the current cost of inventory space, and set prices accurately before the fist customer comes in.

The real supply demand competition is queues size, agents don't like to wait in line, and that is the fundamental law of finite things aggregating.

>The Baumol effect predicts that the increase in the relative price of the low productivity sector will be fastest when the economy is booming. i.e. the cost “disease” will be at its worst when the economy is most healthy!

This seems like a key point of difference with what we observe. In times of plenty, there are ways to get ahead that do not require extensive credentialing, thus there should be a demand shock for education. The mirror image of this can be seen in the commonplace that during recessions, a lot more people sign up for grad school.

This series of posts has been fascinating and really helped me adjust my thinking, so thank you!

That said, I have one question. The Baumol effect tells us that sectors that have low productivity growth will increase in costs and eat up more spending relative to high productivity growth sectors. That said, low productivity growth is not always (and perhaps rarely is) baked into the cake like it is with a live-performance string quarter. For example, to observe that the information revolution hasn't increased labor productivity in education and medicine the way it has in other areas is a mechanical explanation of why costs are going up for those services, but it doesn't banish the question of why productivity isn't improving. Some of your comments suggest that the Baumol effect is the ultimate explanation and we don't need to probe into the details of an industry to figure out why costs are rising, which is perhaps a misreading of your argument. What seems correct is that rising costs still need an explanation, but we should expect that the explanation is a description of the factors that are holding back productivity growth. If it's inherent to the service (string quartets), then c'est la vie. But it could well be that industries are overregulated or mismanaged (medicine? education?) and that the high cost is a signal of that those effects are raising prices by holding back productivity.

Alex is forgetting first principles.

Economies are zero sum.

Assume an economy producing only food.

Gdp = price of labor per unit of food × quantity of food.

Alex argues for

Gdp'= 2×gdp

Gdp' = .05 × price of labor per unit of food × ??? × quantity of food.

Is ??? 4 times the quantity of food paid for with the same household income, prices tracking labor costs, 2 times the household income plus an equal amount of debt with prices of food unchanged but profits now equal to labor cost of production.

Workers are consumers, which means cutting worker incomes, labor costs, cuts consumer spending which cuts gdp.

Alex sees falling gdp as virtue, or does he?

If food is the entire economy except for the king spending on a motor car, then the king builds a car factory and every autoworker buys a car until one person in every household is an autoworker and thus every household has a car..

In a population of one million, this nation has suffered massively from Baumol effect in a hundreds times increase in household spending on cars due to hundreds times increase in labor costs making autos.

Is this nation with such massive cost increases of autos worse off?

The Baumol effect also predicts that labor-intensive parts of the healthcare sectors should be increasing in expenditure faster than non-labor-intensive parts, which is stunningly untrue. It would also imply parity in spending growth across specialties (unless high-growth specialties somehow have better outside options), which is equally untrue!

As in your article, you manage to sidestep the explanation that a long set of actual health economists have given: Technological change, reallocation to high-cost technologies, and generally increasing inputs used per patient. This would be obvious if you spent much time in the literature rather than knocking down straw men. (Also, monopoly explanations are not at all inconsistent with spending rising during booms. This is a bizarre assertion.)

Also, the Baumol effect only matters if labor markets are sufficiently monopsonistic, because otherwise wages are only determined by marginal product. For it to explain the trajectory of health care expenditures requires substantial monopsony. How can you possibly reconcile this with this blog’s repeated denial of monopsony effects when discussing minimum wages? You really think doctors have less bargaining power than fast-food workers?

It seems odd to discuss labor income as the primary source of cost disease without a full breakdown of what comprises labor. Physician pay accounts for 7.5% of all health care costs (essentially 1 year of growth). There are a lot more people employed in healthcare now, however. That shows up as labor driving costs, but doesn't explain why it takes 10 people to do the job it used to take 1. It also shows up as low productivity in the model if you think quality hasn't changed. Baumol effect says nothing about why we need to hire a lot more people to do the same job, it just says something about how much we have to pay all these extra people.

According to free lunch economics, the faster labor costs are cut in all sectors, the lower household income, the more consumer spending results driving higher gdp growth because profits create wealth, ie, asset price inflation, and its price inflation that pays for GDP, not wage income.

Alex is argung that,



2 × gdp = 0.5 × a +0.5 × b +0.5 × c

Alex is outraged that instead

1.5 × gdp = 0.5 × a + 0.9 × b + 4.0 × c

His anguished cry "Why? Why?"

"Why does the price of c rise so much when gdp goes up?"

Hes failing to ask the right question:

Why don't Americans want the falling gdp and falling prices of goods consumed and household spending seen in Somalia, Syria, Afghanistan from 1975 to 1995, etc, driven by falling wages and wage incomes?

It seems odd to discuss labor income as the primary source of cost disease without a full breakdown of what comprises labor. Physician pay accounts for 7.5% of all health care costs (essentially 1 year of growth). There are a lot more people employed in healthcare now, however. That shows up as labor driving costs, but doesn't explain why it takes 10 people to do the job it used to take 1. It also shows up as low productivity in the model if you think quality hasn't changed. Baumol effect says nothing about why we need to hire a lot more people to do the same job, it just says something about how much we have to pay all these extra people.

This is part of the story (really a tautology, so not that super-interesting) but if anything it is more of an indictment of health care and education, not less of an indictment. It's also patently untrue that we have not invested real resources in trying to improve education and health care. One large differecen in those industries is lack of consumer choice and the fact that the consumer does not pay directly for the services (ass opposed to say cars, or cell phones.) Again, it is part of the story but the indictment over lack of quality improvments should be more severe, not less, if we have no choice but to pay more in real terms just to tread water.

Costs and spending on veterinary services has increased just as much as health care costs for humans. That inclines me to believe Baumol cost disease is driving the increases in spending, not something else.

Again, why are *these* sectors low productivity sectors?

Given evidence of productivity gains in sub sectors with less 3rd party (mostly government) interference, eg, plastic surgery, lasik, it still seems to me that you’re not addressing the root cause for why substitutions cannot be made and why productivity is not growing.

Per the Christensen link above, I think a lot of it is cultural inertia. We are used to doing things a certain way, and resistant to change.

HMOs have a higher productivity than see-your-GP-first medicine, but most people want to see their doctor (dragging down his productivity and that of his office).

I don't think we should constrain it to an American political model, because many of these factors are true all over the world (education is especially similar everywhere).

Though certainly mandated HMO would improve much.

>I think a lot of it is cultural inertia. We are used to doing things a certain way, and resistant to change.

I tend to agree, but now were just back to the usual suspects, along with teacher unions, administrators and/or the government.

"The price of the string quartet rises but does anyone expect that the the quality rises? Of course not." As a musician, music presenter and music reviewer I cringe when I see nonsense like that. In point of fact the quality of string quartet performance, as with classical performance generally, has increased dramatically over the last 50 years. The golden age of string quartet performance is right now. But with graduates pouring out of conservatories and competing fiercely with each other for the crummiest jobs in the field, I doubt that classical musicians' incomes in real terms have gone up significantly in that period.

As an aside, it also amuses me that Baumol looks very much like bémol, which is French (with cognates in other languages) for "flat" (in the designation of a note, like E-flat).

Cost disease might be the root problem but administrative bloat makes it worse.

If admin costs in my public school district were what they are in a comparable sized catholic school system, IE 10% instead of 50%, a comparable percentage growth would be 40% less bad.

Ad. Figure 3 The Relative Growth in Expenditure and Math Scores - there clearly was a growth in Math Scores - but the author chose a scale that flattened it relatively to the Expenditure growth - this is a classic misleading technique.

I believe that the Baumol hypothesis fails for the case of education if we take a broader perspective. The government owned, operated, and funded (GOOF) schools control 90% of the K-12 education market. They offer a zero-tuition option versus parents who must pay tuition for their children to attend private schools.

These private options have to appeal to niches in the educational market to justify a non-zero tuition. We observe the 2 main niches as religious instruction and sorting for high achieving scholars headed toward top colleges. At least in California, the market share of private school attendance is under 10% of the under 18 population, and these private schools are probably exhibiting the same or higher costs as 50 years ago.

This is the world in which Alex's observations are made. This world is controlled by government regulations, unions, etc. where GOOFs have a 90% market share.

But now imagine a regime where K-12 education is supported by taxpayer dollars for children attending non-GOOF institutions. Suppose a policy was in place that paid parents a grant if their child was not enrolled in a GOOF school, who passed a criterion referenced exam based upon what students his age in GOOF schools were expected to learn to qualify for grade level promotion.

Call this the Scholastic Achievement Grant for Education (SAGE) program.

Now imagine how the cost structure of this market would evolve. For illustration sake, I'll stipulate the average fully-loaded annual per pupil expenditure, including pensions, health insurance, and capital expenditure often omitted from cost statistics, to be $18,000. Suppose that the amount of the SAGE is $10,000.

It's not hard to imagine that a lot of college-educated, stay-at-home Moms would opt to use the Khan Academy's online course to home school their kids. Or these parents would form co-operatives with other parents (using internet community resources) to supplement home-schooling for young students under 10 with paying for tutors in subjects like mathematics and science that demand expertise beyond the typical soccer Mom or Dad's capabilities.

Likewise, you'd see an explosion of for-profit schools cropping up similar to Crossfit gyms have rocked the world of the big-box fitness centers like 24-hour Fitness. Now we have niche providers for Pilates, Yoga, Spinning, Orange Theory Boot Camp style bare bones facilities that focus on niche markets.

I think that we could expect similar innovations in the market for delivery of education. Of course, this assumes that the public school teachers unions will sit still and not try to impede this radical competition that would eviscerate their professional perks.

But I guarantee that the Baumol Effect would go out the window under a SAGE regime.

I’m usually to blogging and i actually appreciate your content. The article has genuinely peaks my interest. I’m going to bookmark your web page and maintain checking for new details. I am looking for some good blog sites for studying. I was searching over search engines and found your blog site. Well i like your high quality blog site design plus your posting abilities. Keep doing it

It seems odd to be measuring productivity and the opportunity costs of inputs only on the labor side.

Healthcare is a process whereby we attempt to take sick people and turn them into healthier people. The single biggest impact on how healthy the patient becomes after experiencing healthcare is how healthy they were to start. For instance I had two patients not too long ago. One was a former nurse who noted that she had diffuse abdominal pain one night and then next day she was exquisitely tender in the lower right abdomen; she self-diagnosed with appendicitis. We did an ultrasound and sent her back to surgery. Another patient had toughed out her abdominal pain for about 5 days. She had air in the belly on x-ray, was febrile, and had begun to have some kidney trouble. First patient was fixed and gone after spending one night on obs in the hospital while the second required an ileocecostomy and IV antibiotics after going septic.

In both instances, we were equally "productive". The second patient had objectively worse end outcome, but she started worse. Taking a 7 to a 9 (i.e. the first patient) is much easier than taking a 3 to a 7 (i.e. the second patient).

And it is not just the acute complaint. If you give me a healthy young 20 something, I can do many more things to maintain their health. Surgery, meds, PT, and everything else works better when you have more physiologic reserve. When you are sick I often can't use the best meds (kidneys are shot), can't send you off to surgery (skin won't heal), and cannot manage PT (joints are too stiff from arthritis).

At a population level, Americans (and most of the developed world) have become less healthy. In 1960 the average American woman had a BMI of 22. Today it is over 28. If you want to quibble over BMI, I can easily rattle off other statistics like diabetes incidence, exercise amounts, or drug use. As a nation, our healthcare system has gone from working on the first patient above to working on the second. If we managed to keep the population at a "6", we are doing far more today than when we hit a "6" in 1960.

Likewise, education has vastly changed inputs. In 1960 you had far more children with married parents. This, for whatever reason, makes them easier to teach. Likewise, you had far more intangible support from stay-at-home-moms, community support, and the like. We have gone from a school system that had exceedingly low external restraints (crime, multi-partner fertility, drugs) to ones that do. Teaching Beaver and Wally is necessarily going to be easier than teaching the Gallagher clan.

If Baumol predicts flatlined productivity, then this clearly cannot be Baumol. The most important inputs into both healthcare and education have dramatically worsened over the years. If we are seeing flat lined gross outcomes, then productivity MUST be increasing.

"At a population level, Americans (and most of the developed world) have become less healthy. In 1960 the average American woman had a BMI of 22. Today it is over 28. If you want to quibble over BMI, I can easily rattle off other statistics like diabetes incidence, exercise amounts, or drug use. "

45% of Americans smoked in 1960, 15% are smokers today.
13% of Americans were obese in 190, 35% are obese today.

Y'all realize smoking cigarettes is one of the oldest appetite suppressants known to man, right?

I was going to mention that so thanks.

Yes. And?

Smoking is a cumulative lifetime risk for health impacts. Smoking rates peaked in ~1980, but cumulative pack-years did not peak until around 2000. Excess mortality from smoking is greatest in the sixth and seventh decade of life so I would say our risk of smoking related mortality peaked back around 2010. In another decade or so we will reach parity with Western Europe (if present trends continue) and maybe in a decade after that we will get back to the level from 1960.

It is almost as though slow consumption of carcinogens takes time to works it way into and out of the death curves.

"Excess mortality from smoking is greatest in the sixth and seventh decade of life"

The same with obesity.

excess morbidity with obesity starts much earlier than age 60
morbidity is more expensive than mortality

Nope, excess mortality from obesity comes sooner. You have vastly higher rates of maternal mortality during pregnancy, you have increased complications to every surgery, and you are vastly more likely to have ventilation problems with anything that mucks around with you. Oh and it is an independent risk factor for suicide.

The leading causes of death from 20-44 are unintentional injury (which is vastly worse in the morbidly obese and worse in the obese) and suicide. Smoking causes of death tend to make up more excess mortality a decade later.

Thanks for playing, but unless you have a really good epidemiological citation, the numbers on which I base my claims are going to end up being very robust.

Weight does cause excess death later in life as well, but long story short is that the bell curve of excess death is wider for obesity than for smoking. It already has more area under the curve and will shortly cause more gross premature deaths.

This still begs the question, why is education a low productivity sector, particularly higher education? The major tasks e.g., the large survey lecture, grading exams, answering student questions, etc. are all highly standardize-able (i.e., don't require phds) and digitize-able (don't require people).

What Alex is actually measuring is "college schooling" which is a proxy for "higher education." There are various technologies available for delivering education (seated in classrooms, online, self-study), but the vast majority of college students receive schooling utilizing 1950's technology.

My conjecture is that this is driven my the education authentication process. Right now, a schooling-diploma from a brand-name institution is the easiest way to demonstrate likelihood that you have received an education.

Suppose instead that we witnessed the development of institutions similar to associations that develop the Bar Exams or Medical Licensing Exams. These associations offered rigorous exams and long-form projects to measure educational attainment. They wouldn't care about your schooling, only with the outputs.

I could imagine this would completely revolutionize the delivery of education by separating authentication functions from educational delivery functions.
It seems that a competitive market would evolve toward this solution so what are the barriers? This is not the same as accreditation organizations which aren't equipped to handle alternative educational delivery systems that I envision.

For one thing, nearly every college is either government-owned, operated, and financed (GOOF) or a non-profit that is partially funded by the government. Regulations pervade this world, and lacking profit-motives, they have little to no incentive to embrace risky innovations to compete.

So the Baumol Effect might explain the status quo cost curve, but it doesn't explain why the status quo has remained intact for 50 years. Education is not the same as a string quartet playing a fixed piece of music. The quartet analogy is defined as a fixed technology. Unless you wish to introduce the notion of recorded music perfectly reproducing the sound quality of a live quartet coupled with holograms of the musicians, it makes no sense to discuss innovations and competition in that field of endeavor.

Well, Alex has yet to admit that in higher ed admins/student have been rising faster than faculty/student and their salaries have been rising much faster than the faculty ones. And while the explosion of assistants to deputy vice provosts is probably more important, there is also the explosion of pay going to athletics, with 40 states having the highest paid public official an athletic coach at a state university, even though there are only about 20 universities in the US where the athletic programs pay for themselves.

NCAA schools collude to keep player compensation down, they should at least agree to limit coach compensation to something like $300,000/year.

Nevertheless have athletic department taken a bigger chunk of school funds net of the income that they bring in?

At the average mediocre Big State U, the athletic departments are critical for alumni donations.

Professor Rosser knows this, of course. But it’s an easier target (football bad!) than the Title IX and Diversity armies of administrators (good!).

IDK. You can have athletic departments w/o multi-million dollar per year coaches.

Particularly, if you also price-fix coach salaries, like the NCAA and pro leagues currently do w.r.t. the athletes.

As the wages of string quartet players rise, would that not induce more string quartets to form? And, if so, would not competition among them drive out the "least good" of them?

Comments for this post are closed