Granted it's a Wikipedia article and not the original research, but I have two quibbles. The first is that it still takes the same number of people to perform a Beethoven symphony, but 4 kids with electric guitars and barely any training can now perform music to a packed concert hall or dance floor. And those kids are under price pressure from someone who can now deliver the same productivity with a laptop. (I was one of those kids once).
The second is that college professors have been phased out of teaching for a long time, increasingly replaced by adjuncts who earn starvation wages.
Healthcare, I'm not sure about. At my most recent clinic visits, a doctor (i.e., MD) was nowhere in sight. A technician examined me, a physician's assistant interpreted the exam results, and ordered a visit to a MRI machine operated by technicians. (Thankfully, it turns out I'm OK). My healthcare costs twice as much as in most civilized countries for no good reason despite roughly equal labor inputs.
It would be useful to remove education and healthcare from the graph and see if the theory still makes sense.
>Granted it's a Wikipedia article and not the original research, but I have two quibbles. The first is that it still takes the same number of people to perform a Beethoven symphony, but 4 kids with electric guitars and barely any training can now perform music to a packed concert hall or dance floor. And those kids are under price pressure from someone who can now deliver the same productivity with a laptop. (I was one of those kids once).
I'd argue this is still a version of the Baumol effect. For example, some people look back at 1950s nightclubs and wonder why they can't have that now.
It's because, as you say, the productivity of automatic music increased but live orchestras didn't.
So if you want "music" you can get it cheap, if you want "live orchestra", you can't.
That's interesting, you could get a live orchestra when cost of music production was higher, but no more because big-band music was replaced by power-trios.
People were able to afford something earlier which in these days they no longer can.
I agree with education and healthcare tending to obscure the effects. In both cases, normal price sensitivity gets skewed because the person selecting which product they want (the patient, or student) isn't directly paying (because of insurance, or student loans/grants).
But there are plenty of service jobs that haven't seen big productivity gains. Masseuses, for example.
Almost all of them. Barbers, teachers, nurses... construction productivity in the US has collapsed. The US construction industry is literally less productive today than it was at any point in the last say 75 years.
Performance of symphonies has become massively more productive over the 20th century.
In 1900 a performance could entertain at most a few thousand people. Now, a single performance can be heard by tens of millions. A performance in the Concertgebouw Amsterdam can be heard in real time* in New Zealand, on the other side of the world, with better fidelity than all but a few of those present in the theatre experience.
Baumol's choice of example of zero-productivity-gain work was egregiously wrong in this case. Likely the same is true for his other examples.
Wider distribution doesn't actually matter here if it doesn't result in linearly increasing economic benefit. For most modern music the majority of income is made on tours doing live music. I imagine this effect is far more pronounced for orchestras because millions people bought Taylor Swift's last album, but not so many are paying money for the Berlin Philharmonic's release of Mendelssohn classics.
So if the orchestra's economic production is still only being realized when they perform live music, the baumol effect will absolutely still apply.
> For most modern music the majority of income is made on tours doing live music.
Now, yes: internet streaming killed recording as a viable profit center in the mid 2000s. When Baumol was writing, musical acts did live shows more or less at cost, or even at a loss (covered by their record companies), to promote their recordings, which were where the money was, then.
The declining popularity of classical music is beside the point. Baumol intended classical music to stand in for performing arts in general.
The size of the potential (and in many cases actual) audience and revenue has increased for performances of all kinds: popular music, stand-up comedy, baseball games, motor car races, spelling competitions, etc. At the same time, costs to an individual consumer have declined dramatically. Those things are the essence of productivity gain.
Full-time tenure-track faculty salaries haven't really beat inflation since the late 60s. And senior profs have been substituted by adjuncts and grad students. And yet cost of college keeps soaring.
Baumol's cost disease theory makes sense in general. But empirically, for a specific service, there's always a lot of other stuff going on, are they providing a superior good with rising demand as society gets richer, how easy is it to substitute lower quality adjuncts, etc.
> The first is that it still takes the same number of people to perform a Beethoven symphony, but 4 kids with electric guitars and barely any training can now perform music to a packed concert hall or dance floor.
Economics feeds into this. I happen to like (among other things) "big band" music, which flourished in the 30s through the end of WWII. The reason was that wages were depressed (it was The Depression, after all) so it was cheap to hire a big band (and it was a way to spread the money around and keep your friends employed). The war ended, the depression ended, and jobs were plentiful (plus some new technology wsa developed) and that was the end of the Big Band Sound.
Symphonies evolved in a time of patronage and low wages. They are absurd today -- real money losers. I'm glad people still try, but I feel like my experience must be subsidized.
I am compulsively unable to attend a concert without estimating the amount the band is paid and dividing it by the players. The economics rarely make direct sense.
I play in a big band. These days, performing and composing are both labors of love. My band is by mutual agreement, non-commercial. The money we make goes into sheet music and band expenses.
There doesn't seem to be any compensation for the vastly greater spending on administration in colleges. Yale's non-instructional spending has grown 3 times faster than the cost of tuition in the last two decades. Is that sort of support structure really necessary or has it simply been taking advantage of the relatively easily obtained money that students now have available to them. If colleges have little skin in the game with a student's inability to ever pay, what economic incentive is there to run a lean operation?
The answer is it's both. On the one hand there isn't enough incentive to rein in costs. But on the other hand, administrative requirements really have been going way up:
> Perhaps most controversial is an increasing raft of federal and state regulations that universities must abide by: the Clery Act, which requires campuses to report their crime activity; new Title IX regulations that govern the handling of sexual assault; and Family Educational Rights and Privacy Act (FERPA) requirements for providing educational records.
> In 2013 and 2014 alone, the Department of Education released rules and directives on 10 new sets of issues, ranging from proposed rules on teacher preparation programs to Net Price Calculator requirements to specific regulations for FAFSA verification. Complying with all these rules requires additional staff and additional money. The resources required are not insignificant: a Vanderbilt study of 13 colleges and universities found that regulatory compliance comprises 3 to 11% of schools’ nonhospital operating expenses, taking up 4 to 15% of faculty and staff’s time.
> “It is pages and pages and pages of regulations that require more sophisticated professionals,” says Penny Rue, vice president for Campus Life at Wake Forest University and board chair-elect of the National Association of Student Affairs Professionals. Rue adds that incidents on college campuses, such as the 2007 shooting at Virginia Tech, contributed to a need for administrative spending that often goes unnoticed, from case management services to threat assessment teams.
Baumol’s law enables this growth in the administrative state - the money is available to pay for tuition, so it gets spent however the university wants.
The wages are what's relevant with Baumol. It's possible rising wages of parents helped generate the administrative burden but 90% of the time it was the flood of cheap student debt via gov, every kid thinking they need to go to college, the massive boom in demand via international students, etc.
The fact university administrators responded to boom in demand/capital by hiring way more administrators is probably not directly a consequence of Baumol either. But it is why the costs of these administrators is so expensive, t
which is reflected in tution costs.
Maybe you could blame it because being a uni admin is an easy job to get with little actual output demanded of you, but you still get paid like a high productivity sector employee, so people flocked to it. And if they were paid a wage connected to their individual output they wouldn't be nearly as many of them and tuition costs could go down.
But the universities could also just fire 2/3rds of them and still function while paying the rest normal high wages.
>There doesn't seem to be any compensation for the vastly greater spending on administration in colleges. Yale's non-instructional spending has grown 3 times faster than the cost of tuition in the last two decades.
Maybe Yale is an outlier here because there's another analysis[1] that looks at aggregate data for nonprofit and private colleges collected by NCES, and that analysis showed that support (ie. non-teaching) costs have outgrown teaching costs.
Sorry, I mistyped my original comment. I meant to say that non-teaching costs have not grown faster than teaching costs. From the article:
>Again, everything is per FTE per year. So support cost (student services, academic and institutional support) is roughly comparable to instruction cost (teaching), and the two have risen at similar rates in the 1999-2013 window. Research expenditures, meanwhile, have been pretty flat.
This is a very personal take, but one of the things I have always attributed to the Baumol effect is the following.
My partner works in the (semi) healthcare. She loves it but her main stress factor (nemesis even) has been lousy management. And she has a point. The managers in her organization are people who would be fired after a few weeks in my organization. I work in IT and we pay our managers serious money. I am the last one to say they are perfect or more than marginally competent, but they get the job done.
What my partner ends up with are the managers that cannot get a job in the better paying segment. And that is very much to the detriment of our healthcare.
The Baumol effect is not as much about, say, musicians who decide against their calling to become engineer instead, but about the job roles that can switch markets easily. Among those, the real talent follows the money. And to be honest, I cannot blame them. It is just human nature. But that does not make it less of a problem.
> What my partner ends up with are the managers that cannot get a job in the better paying segment.
This observation is sort of downstream from the Baumol effect rather than the effect itself or more accurately when the effect is countered explicitly.
The core loop of the effect is when you need to pay a healthcare manager a ballpark similar salary when putting out a "for hire" ad to get applicants or alternatively hide the salary filter till much later in the interview sunk cost.
The problem is that you end up having the managers first apply for other jobs before getting to the low-paying job, so the initiators of the interaction (i.e "apply for jobs") sieve out before getting to the job that pays almost the same but needs less competence.
So you pay about as much, but get even less value for money than paying more.
When you put Baumol effect, Dutch Disease[1], Gale & Shapley[2] and the Market for Lemons[3] together, you get to see the job market from a lot of different angles in my immediate neighbourhood.
Silicon valley has a Dutch disease for math teachers for instance, but also the Baumol effect for the English staff. Not complaining about them, I'd like my kids to learn history, math and english from great teachers & don't want to do Kumon or whatever else the other kids are doing after school.
The way the schools try to fix it is by making the schools initiate hiring through temps and do extensive adjunct periods before any concept of tenure to work around the market for lemons (you can't hire a temp managing director, which is what's different there).
This really sucks for the good teachers who want to have a happy late 20s in the career they prefer.
I think of a related effect every single time I'm at the DMV, or on the phone with a customer service representative, or dealing with insurance or some other job where few people love and I'm dealing with them clearly doing a shitty job.
I remember that people on the left side of the bell curve need jobs too. And, all else being equal, I'd probably prefer them manning the phones at a hotel and sometimes screwing up my reservation than wielding a scalpel when I'm under the knife.
I wonder if it's something else. I see constant complaints about the DMV on the Internet and popular culture. Dealing with Service NSW in Australia however has always been fast and painless for me.
These comments are the worst sort of Hacker News comments: a bunch of people who spend 45 seconds scrolling a Wikipedia article about a major concept from a field other than software and dismissing it as “obvious.”
Hi! I'm an economist, and I am genuinely enjoying reading everyone's perceptions. If we were to presume that this was the first time your representative HN commentator read about it, then we can congratulate the Wikipedia authors for sufficiently communicating the concept (a passing grade, as it were).
I thought you might like a counterpoint to your consternation with your fellow HN commentators.
you have to admit its kinda of a strange 'law', unless you really believe that supply and demand is the core inviolate law of collective human behavior..in which case such an exception might be interesting.
I had a similar contradiction with a friend yesterday, and I argued that the big winners of the AI revolution might actually be those in construction/electricians for instance, as their livelyhood won't be significantly affected in the next 10 years, as LLM can't be used to help or replace them in any way.
I'm very worried and currious about the future, regardless.
There are a number of interconnected things, not all of which are probably properly called the Baumol effect. However, it seems that given more efficient manufacturing, demographic trends, and other things, one of the results will be that we see more automation/AI, self-service, and things we decide are maybe occasional luxuries rather than something we partake of routinely.
If the max federal student lending is $5k or so per year, how does that explain college costs going past $50k a year? How is the balance being afforded?
> American economists William J. Baumol and William G. Bowen pointed out that the same number of musicians is needed to play a Beethoven string quartet today as was needed in the 19th century—in other words, the productivity of classical music performance has not increased. However, the real wages of musicians have increased substantially since the 19th century.
This raises the question of what counts as productivity, and by what metrics is it measured. Yes, the number of required performers for a given symphony is static, but that symphony is now able to reach far more people, be it in real-time (via live broadcast, and via larger auditorims (auditoria?) with capacity for more people) or after-the-fact (via recordings). Quality is also a consideration here: better instruments, better acoustic design in the venues, better recording equipment, better playback equipment, better storage media (allowing better lossless recording quality), possibly even better performers.
In short: attempting to measure the productivity of an artistic endeavor is unlikely to result in anything coherent.
This is such bogus malarky. How is it possible to seriously write an article like this without once mentioning money, from what I saw.
The value of the money commodity today is not tied to productivity. How much effort does it take to make a dollar? And how does this effect its relative value to all other commodities.
And so of course the amount of the "money" commodity needed for a given piece of work can change without the productively of a given piece of work changing based on the independent value change of said money commodity.
Seems incredibly shortsighted, even in the limited examples provided; It says "the number of musicians required for a quartet is the same as in Beethoven's day" which is true, but even by 1960 recording technology provided those musicians the chance to be more "productive" by all but the most myopic of measures (i.e. "times they drew their bow across the strings").
It's a poor example to use for sure but it's also just the wiki. Not all of them are designed to be persuasive or necessarily have gotten much attention to be well written/supported.
You're free to contribute a better example to the Wiki. There's plenty of low productivity jobs with high wages being propped up by the general productivity of the other sectors (corporate middle management, administrators everywhere, tons of consulting gigs, etc). Or someone could expand the critique section if you can find some better sources.
The market for recorded music overlaps but is not the same as the market for live music, as proven by the existence of chartered live performances of already recorded music.
> By most measures, productivity growth in the education sector over the last several decades has been low or even negative;[40][41] the average student-teacher ratio in American universities, for instance, was sixteen to one in 2011, just as it was in 1981.[37] Yet, over this period, tuition costs have risen substantially.
I'm just a layman, but this method of measuring "productivity" in education is absolutely bonkers and seems to treat the actual education received by each student as a fungible commodity. Thus a student in a "Rock Music History" class is roughly equivalent to a student in Quantum Physics 401.
Is this how economists actually think of education?
If the Baumol Effect is true for college tuition, are college professors receiving better than inflation pay raises commensurate with the increases in tuition?
I am sure at the time it was a novel insight, but its punchiness is largely undercut by our contemporary economics which recognizes (if somewhat begrudgingly) that for consumers to continue contributing to the economy, they need to get paid sufficient wages to afford goods and services locally. It is ultimately an ancillary effect resulting from the holism of economies using common currencies.
If you look from the POV of low-level worker in a rich region it's nice that you earn 5 times more than a person doing the same work in a poor country.
If you look from the POV of low-level worker in a poor region it sucks that you earn 20% as much as a person doing the same work you do exactly as efficiently in a rich region. And it sucks that it doesn't depend on your effort and efficiency. It mostly depends on how well the big buck industries work in your region.
This is the main reason I'm frustrated every time people post maps of "labor productivity per hour". With the implication that people in poor countries are inefficient and that's why they are poor.
>If you look from the POV of low-level worker in a poor region it sucks that you earn 20% as much
And if you pit a slave from an even poorer region against them then it sucks even more. They earn nothing at all while that worker from a poorer region earns money for work.
Some people think that because of this their privilege they need knocking down a peg.
>This is the main reason I'm frustrated every time people post maps of "labor productivity per hour". With the implication that people in poor countries are inefficient and that's why they are poor.
I dont even see why this is controversial. Your efficiency would drop considerably if you moved to a 3rd world country and vice versa.
> Your efficiency would drop considerably if you moved to a 3rd world country and vice versa.
Not necessarily. A barber in Bździszewo Kolonia earning 5 USD per haircut and a barber in New York earning 50 USD per haircut can work exactly as efficiently and do exactly as good a job. The main difference is in how much money their customers have.
Baumol's effect doesn't make things more expensive, it makes them get cheaper at a slower rate than other things, so they only get more expensive relatively.
I think either Elon Musk or Peter Thiel was talking about a similar effect, where the cost of things in regulated industries continues to increase as non-regulated industries increase in productivity, such that regulated industries eat a larger portion of spending over time. Could be wrong and it might have been some other person that said this...
It was Marc Andreessen. He has pointed out that cost of a 100 inch high definition television is on the way to $100 while a four year college degree is eventually going to hit $1,000,000
This effect is distorted by our soft-money standard which enables unlimited money printing and which allows the economy to be detached from the process of value creation.
On a hard-money system, monetary rewards can only be attained by providing value to others who also provide value to others. The extent of your reward for servicing your customers is proportional to the extent of the economic value provided by those customers in aggregate (in terms of how they service their own customers). It's more profitable to service people/companies who are effective at servicing others.
In a soft-money system, there is no guarantee that a specific entity earned their money through value creation; e.g. the money could have been sourced from a loan from a now-bankrupt company which provided no economic value (just spent the money into the economy and went bankrupt soon after)... Or it could come from huge, overpaid government contracts. It could come from corrupt dictators who are laundering government-issued money between each other, etc... With money printing, it's too easy to spend money which you can just create out of thin air. There are too many weak points which can be exploited and which can undermine the entire premise of free market capitalism.
If an industry experiences productivity gains, it does not entail that they would need more workers; the opposite should be true. The entire premise of productivity gains is that you need fewer workers to provide the same amount of value. If the productivity gains within an industry were real, that industry would not need more workers and it would not be pulling workers from other industries and thus driving up the salaries of remaining workers (due to labor shortages in those unproductive industries). Specialized skills are required to deliver productivity gains; it's not like you can hire a petrol station attendant to build your AI with Tensorflow...
What's really going on (a much more powerful force) is that some unproductive industries are subsidized by money printing so they grow regardless of the value they bring to society and they just create bullshit jobs. Workers just move from industries which aren't well subsidized by money printing to industries which are. This movement of workers does drive up increases in salaries in other industries, but it has nothing to do with productivity. If anything, the fact that some industries benefit more from money printing than others may explain why there is such inequality in productivity growth across industries. Money printing creates incentives which, for example, keep tech workers out of certain industries (since they can get paid more to work a bullshit job). Why are so many tech people working for Big Tech (where they are clearly not needed given the recent rounds of mass firings) and so few working on farming, building/construction automation and robotics (for example).
This is nonsensical. The Baumol effect is just supply and demand for labor. The monetary system in use does not change whether or not supply and demand work this way.
You also appear to be conflating the notion of "hard" vs "soft" money and the notion that one should be paid based on the economic gain of your output to others. Those are orthogonal concepts. Their only commonality is that the economic consensus on both notions is that they are wrong.
I just don't see how productivity gains (e.g. automation) in an industry would lead to increased demand for workers in that industry. I feel like it should be the opposite.
Maybe except for brand new industries in the early stages but then I don't think productivity is measurable in such young industries because there would be no meaningful productivity metric to use to make any such analysis.
In theory, productivity gains on automobiles leads to cheaper cars and increased car sales. So an increase in workers is possible.
But I agree that looks like a questionable assumption. If the productivity improvement changes car prices from $250K to $25K, sure the assumption will work out. But not sure $25K to $24K would work out.
I think the only consistent and reasonable measure for productivity of labor is how much the economy is willing to pay for the labor[1]. So if the society is willing to pay nurses n times higher salaries than 20 years before, the value of nurses work has become n times higher, and the productivity has increased n-fold. (minus inflation). And the baumol effect vanishes as a puff of smoke into air.
[1] But surely a worker creating only one thingie at the same time than other produce five thingies must be less productive, you say? Well, if the first one is able to sell his thingies at more than 5 times higher price than others, quite obviously s/he is more productive than the others.
Why do you think it is inconsistent and unreasonable to choose a measure of productivity that says y=5x is more productive than y=x in terms of real output of y given real input of x?
> So if the society is willing to pay nurses n times higher salaries than 20 years before, the value of nurses work has become n times higher, and the productivity has increased n-fold. (minus inflation).
How do you disentangle value and inflation? I think that makes your definition incoherent as you just loop back to real output and real productivity to measure those things.
This isn’t just about productivity of labor, it’s about productivity in terms of end result. That kind of or productivity is deflationary, not inflationary.
Eg a TV costs 10% of what it did 20 years ago. Fewer person-hours of work are involved in making one TV. Less of the average home budget goes to TVs than it used to, even while people are buying better tvs.
That’s a massive productivity increase in the lingo of economics.
You could replace TV with solar panels, cars, computers, etc. Less human involvement per good produced => productivity increase.
So because one set of chinese are able to make bunch of counterfeit casio watches way cheaper than another set of chinese can make apple watches, the first one are more productive because both make watches that show time? Maybe there is a reason why another watch or another nursing service or another tv (at another time) is more expensive? Maybe the reason is that in the more expensive case they produce more value i.e are more productive?
By the way, just started wondering why nobody is worried about the atrocious productivity development of CEOs over the last decades? Should shareholders maybe hire some consultants to advice the CEOs how they could make their work more efficiently that they could become cheaper?
So if a billion robot nurses suddenly appear, doing the same work as human nurses, driving prices down due to competitive and lower costs for the robots, how has value of the product decreased?
Because “value” in the economic sense is not synonymous with “value” in the ethical sense and neither is it a complete overlap with “value” in the way we use it in everyday speech.
In economics value is about choice and indifference. If you have two options and you are indifferent between them they have the same value. Everything about value in economics flows from there. It’s a very useful concept. But it’s not the same as “value” that a philosopher, a poet, or anyone else might talk about. Similarly “utility” in economics is fairly much equivalent to value in economics — but “utility” in economics doesn’t quite map to the common term “useful”.
How do you measure (and compensate workers for) compassion? Making ethical decisions? Maintaining confidentiality? Acting on a combination of training and instinct? I mean, okay, we can conceive of robot nurses who do the latter, but how does a robot nurse demonstrate compassion and empathy for patients? I think that stuff is priceless.
Economics is basically the ideology of the upper class.. it is not a science, it is ideology.. I would suspect that this cost disease thing is driven in large part by modern monetary theory and increased liquidity through printing money, driving up asset prices and causing inflation by financial speculation.. the increase in asset prices such as housing causes the increase in wages.. the upper class would have us believe that the wage increase causes inflation. Maybe in the past but in the last 100 years or so and increasingly in the last 20 or so inflation and the cost disease that is basically described in this linked article above is probably driven mainly by inflation caused by asset price increases caused by financial speculation fueled by hyper liquidity...
Edit.. those who vote me down take a look at inflation before the rise of central banks.. yes of course there were periods of inflation but then they were followed by periods of deflation.. this evidence of that take a look at the wages for an average man from 1700s all the way through the early 1900s it was basically the same.. whoever wants to central banks came into play inflation crept upward and upward
Hello, economist here! You are half correct. It's not a science, it is a social science. As far as methods go, it's quite imperialistic and has been the source of a lot of methodology improvements in many fields over the last several decades (especially in mechanism design and econometrics, IMHO).
Though, there are certainly ideological branches of economics. A good example is the "Austrian school," which claims to be the "mainline" of economics instead of mainstream per Pete Boettke.[0] There are also pretty dead arguments around Institutional Economics and similar that aren't really active fields anymore. However, this notion that standard economics is somehow lost is more reactionary and prosaic than it is sublime -- mainstream economics tends to do quite well in the marketplace of ideas and in evaluation, though subject to the same replication concerns all social sciences face.
Recent work in causal inference (which ML enthusiasts celebrate) has absolutely upped the bar for economists' output.
As for your hunches, they aren't really on point or based in reality. Modern Monetary Theory came a long time after Baumol's elucidation of sectoral cost disease. Recall that macroeconomic theory is only a small part of economics as a whole, even when considering its associated fields of inquiry.
Academia is fraught with fraud... Once the establishment has settled on a narrative regarding some subject, professors will be falling over themselves to present studies based on Cherry picked data that will support the establishment narrative.. very few studies are actually reproducible..
You're only saying this because you don't like the outcome of economic research. So called "soft" sciences are the best way to get close to the truth in those fields even if it's harder to get undeniable answers.
The second is that college professors have been phased out of teaching for a long time, increasingly replaced by adjuncts who earn starvation wages.
Healthcare, I'm not sure about. At my most recent clinic visits, a doctor (i.e., MD) was nowhere in sight. A technician examined me, a physician's assistant interpreted the exam results, and ordered a visit to a MRI machine operated by technicians. (Thankfully, it turns out I'm OK). My healthcare costs twice as much as in most civilized countries for no good reason despite roughly equal labor inputs.
It would be useful to remove education and healthcare from the graph and see if the theory still makes sense.