Baumol's Cost Disease
(en.wikipedia.org)127 points by drra 2 days ago
127 points by drra 2 days ago
> People rightfully complain about tech support, and I always think "Yeah they're bad because anyone who knows how to do it ... does not stick around."
The say "pay peanuts, get elephants" exists for a reason.
A related thought I heard was "pay police and army well, because you don't want them to start looking for ways of making more money...".
Outside the assembly line, you aren't just paying for the work themselves but for the character of the person providing. Drive the wage down low-enough and only the desperate apply. I know that locally, one of the first jobs the agency can get recovering homeless drug addicts is a call-centre job. That explains a lot.
>Yeah they're bad because anyone who knows how to do it ... does not stick around
Can confirm! I'd also like to add that the employer is happy to keep you in that position for a decade or more (as long as the client keeps paying) but one job over an extended period of time reads very poorly on a resume. Switching is a disruption necessary for growth.
Related. Others?
The Baumol Effect and Jevons paradox are related - https://news.ycombinator.com/item?id=45955879 - Nov 2025 (67 comments)
Baumol Effect - https://news.ycombinator.com/item?id=43065115 - Feb 2025 (1 comment)
The Baumol effect - https://news.ycombinator.com/item?id=35220758 - March 2023 (77 comments)
Why are the prices so damn high - https://news.ycombinator.com/item?id=33150094 - Oct 2022 (2 comments)
Baumol Effect - https://news.ycombinator.com/item?id=24812620 - Oct 2020 (99 comments)
Baumol Effect - https://news.ycombinator.com/item?id=20443675 - July 2019 (62 comments)
William Baumol, author of 'cost disease' theory, has died - https://news.ycombinator.com/item?id=14284466 - May 2017 (33 comments)
Baumol's Cost Disease - https://news.ycombinator.com/item?id=12679629 - Oct 2016 (1 comment)
Is productivity the victim of its own success? - https://news.ycombinator.com/item?id=11964673 - June 2016 (57 comments)
Baumol's Cost Disease: Why Artists are Always Poor - https://news.ycombinator.com/item?id=972082 - Dec 2009 (14 comments)
It seems relevant that the categories healthcare, housing, and higher education are not paid by wages/cash but rather through various means of financing (in the form of insurance for healthcare). (In addition to not being outsourcable).
New cars not increasing looks strange but that might be a US phenomena, where most cars might be imported and the median car have "shrunk" in size over the period.
It highly depends on what your definition of a price increase is. The cars people actually buy have definitely become more expensive with more and more people choosing to buy SUVs.
The sedan hasn't become more expensive, but people don't feel safe driving them anymore between all the large SUVs, this pushing more people towards buying more expensive cars.
Really interesting stuff. Am I correct in thinking that if productivity were to rapidly increase in the service sector (e.g. due to AI) the same way it did in the manufacturing sector in the 19th and 20th centuries, that the cost of services would decrease?
Also, a side note: I dislike a lot of the popular conversation around the Baumol effect because they’re usually along the lines of “this can’t be the only reason my healthcare or education is expensive”, which is true (there are other factors at play), but the Baumol effect still explains a lot of it.
The submission is titled with "Cost Disease", though the Wikipedia article has the more neutral term "Effect". But it is important to remember that money is a relative resource, not a real resource. If some sectors of the economy become drastically more efficient, and some do not, overall society has become wealthier, even if the prices in the latter sectors rise a lot.
The phrase “Baumol’s cost disease” is widely used, and well known. It’s also in the first sentence of the article.
> In economics, the Baumol effect, also known as Baumol's cost disease…
> If some sectors of the economy become drastically more efficient, and some do not, overall society has become wealthier, even if the prices in the latter sectors rise a lot.
That's assuming all sectors have become more efficient. Some, like construction, have become less efficient. And that's a big problem when it's relevant to necessities like housing.
Suppose people used to spend 20% of their income on housing and healthcare and 20% on apparel and electronics. Then housing and healthcare triple in price, apparel drops by two thirds, electronics drops by 98%, and everything else stays the same. Are they better off? No, because the most you can improve the cost efficiency of something is 100% (it becomes free), but the things that that cost more can increase in cost by more than 100% of the original cost, and some of them have.
> That's assuming all sectors have become more efficient. Some, like construction, have become less efficient. And that's a big problem when it's relevant to necessities like housing.
Housing prices aren't going up because of construction costs alone. The biggest increase is from the cost of land. For that the cost of a house on top has become less and less relevant. If construction became really cheap, prices would still trend upwards since there's always some billionaire's money to be parked somewhere.
The biggest increase is from the cost of zoning making land artificially scarce. But construction costs layer on top of that because of the nature of it: Instead of being able to build 20 new housing units on a lot that currently only has one and enough land to add more without destroying the existing building, adding more is now restricted to a small strip of the downtown where the lots already each contain 10 housing units. Which effectively doubles your construction costs to add 20 housing units because you still have to build 20 housing units but now you have to destroy 10 in order to make space, and then do that twice to actually add 20.
Which is a disaster if construction also got twice as expensive.
> If construction became really cheap, prices would still trend upwards since there's always some billionaire's money to be parked somewhere.
If construction became really cheap and there wasn't an artificial limit on how much housing you could build on a given lot then there would be tons of cheap housing and billionaires wouldn't find it a useful place to park money because it would have lower returns than competing investments.
Even if construction would be zero it wouldn't make more than a dent in the overall trend because the investments are done into land not houses. This is also why zoning doesn't matter. Look at the big picture. Pretty much every argument you made can be easily refuted by looking at any other jurisdiction that has completely different zoning laws and completely different construction prices and yet prices for real estate are skyrocketing the same, regardless of use. What those areas do have in common: general availability of real estate to international investors.
Properties aren't bought up anymore to develop them or have any returns from use. They are being bought as a non-depreciating asset. Want to effortlessly park money and not have the money rot? Buy land. Never mind what happens to stand on top of it. You can see more and more of those rotting real estate plots all over the western world. And there's always someone who wants to park money.
Your thinking isn't wrong on a smaller scale. All those aspects matter for a local housing market. But the overall trend is governed by something else.
> money is a relative resource
I think this is the better way to think of money and wealth:
Money is the unit of measure for wealth. It's not in itself wealth.
That doesn't quite make sense to me. A meter is a unit of measure for distance, but a meter is a distance.
Yeah, it's not the same as physics units.
Money does have real value, but only because it can be traded for valuable things.
But money in itself, as bills or numbers in a bank account, is useless until you trade it for something "real".
Your conclusion falls victim to the same conflation you’re calling out. If some sectors become drastically more efficient, society has become wealthier in terms of money, but not necessarily in terms of real resources.
For example, consider a case where finance becomes much more productive (in terms of $ per employee-hour) and raises wages to attract smart people, leading to fewer people becoming doctors because finance is much more attractive. Is society wealthier? The money says yes. The line goes up. But finance doesn’t set a broken bone or treat cancer. This may well have made society less wealthy in terms of what ordinary people actually care about.
The Baumol effect says wages for doctors will also have to go up. Society can afford this because it now has commensurately more resources due to increased efficiency. It’s a tide that raises all boats, precisely because of this effect. This is why a taxi in London costs and pays better than the same service in Cairo.
> Society can afford this because it now has commensurately more resources due to increased efficiency.
Does it though? Suppose that Wall St has discovered a strategy, like high frequency trading, that produces nothing but allows the one doing it to extract a margin that would otherwise have gone to the second-fastest trader. Many people are employed in a competition to be the fastest because being the fastest is rewarded but it's a zero-sum game where nothing useful is produced and the players each have to continuously spend resources to keep running faster in order to stay in the same place.
What benefit is the person now paying more for healthcare getting in exchange for this?
> It’s a tide that raises all boats, precisely because of this effect.
What if it's not all boats? Suppose it causes doctors to get paid more because people who have the wherewithal to become doctors could also work in finance, but it doesn't cause retail clerks to get paid more because Wall St isn't hiring them away from their existing jobs, and in the meantime they now have to pay more for healthcare.
Exactly. Even though Baumol himself used the phrase "Cost Disease", I think that framing distracts from the fact that it is a result of something desirable happening, namely increased efficiency in some sectors. You could also posit a case where some sectors become less efficient, due to badly conceived regulations, exhaustion of non-renewable resources, an unchecked monopoly, or some other factor, but you don't need a special mechanism to explain why prices rise in such a scenario.
> ...consider a case where finance becomes much more productive... leading to fewer people becoming doctors because finance is much more attractive.
This is the opposite of what one would expect from a sector whose efficiency increases, as modeled by Baumol. See the first bullet in the article: "The share of total employment in sectors with high productivity growth decreases, while that of low productivity sectors increases" (also see the detailed analysis in the Technical Description section). It might be theoretically possible that induced demand could still increase overall employment in a sector as its efficiency increases, but I think you have to make an argument why that would be true. During the industrial revolution, automation eliminated 98% of the labor required to produce a yard of cotton cloth, but between 1830 and 1900 the number of weavers in the US increased by a factor of 4, because demand increased due to lower prices [0]... although the US population also increased by a factor of 6, so as a percentage of the workforce weavers still declined, even as people consumed much more cloth per capita.
[0] James Bessen, Learning by Doing - The Real Connection between Innovation, Wages, and Wealth (2015), pp. 96–97.
I picked finance for my example because demand is practically unlimited. People only need so many clothes, but when your business comes from making money directly, there’s a lot of room for growth.
Imagine some new math allows HFT to make more money. HFT firms wouldn’t start laying off quants. They’d probably hire more to try to capture more of that new money, and they’d have more money available for hiring.
Finance funds hospitals and cancer research institutes - or at least, it enables the gathering and concentrating of resources to do so.
Now, advertising...
Some finance is needed and beneficial. The ability to form corporations and raise money through the stock market enhances many other fields of endeavour.
But this can go too far. In London during 2000-2008, finance consumed every spare IT worker, as well as mathematicians and physicists. Salaries were far higher working for a bank than working in any other IT-related industry or start-up. Did this produce great works? Is London now better off because of this? In a word, no.
Advertising enables innovation-producing firms to drive awareness of their services in a cost effective manner, and for less informed consumers to understand what is available on the market. Your typical physician might not be fully caught up on what is the state of the art in arthritis treatments, but advertising enables this to happen.
> If some sectors of the economy become drastically more efficient... overall society has become wealthier
That's a weird one - what's your metric for the "wealth of overall society"? Stock market indexes can't be it because those are subject to extreme levels of unreported inflation and gaming.
How can you measure something that is subject to extreme inflation when that inflation is not only unmeasured but not even acknowledged as a phenomenon?
At present, the "wealth of overall society" is a unicorn metric as opposed to the perfectly measurable and extreme levels of income and wealth inequality. In other words, the overall losses from skewed distribution dwarf the gains from higher efficiencies.
If the orchestra performs less often because the violists have better paying jobs in a factory making the latest and greatest TVs, more homes will have the latest and greatest TVs.
Of course, this relies on the assumption most work - and hence most productivity - is a net social good. If the violinists have instead got jobs operating an orphan-crushing machine, that would be a bad thing. But hopefully your society is structured in such a way that the average worker is contributing to the prosperity of their local community.
GDP produced divided by costs required measures intensity. These will be typically normalized (inflation removed) or, if a ratio, can be nominal since both have the inflation ratioed out.
GDP is known to be an imperfect measure, especially for capturing cottage industry and due to the distribution effect you described, but it's not horrible to start with.
GDP is how much wealth is produced in society at a moment in time.
The total accumulated wealth in a society is a related but entirely different number.
Suppose that every single person in society receives the same compensation and has the same wealth, i.e. there is perfect equality.
Society produces housing and other necessities and people consume them in some amount and then have what's left as disposable income to spend on whatever they like, e.g. for going on dates.
Then a law is passed prohibiting the construction of housing with more than two stories. Building ten housing units on ten lots with ten foundations requires more labor than constructing ten units all on the same lot, so the price of housing increases, people have to spend more on housing and have less money left to buy flowers and some people quit their jobs at the flower shop to go pour concrete (while still getting paid exactly the same amount as before).
There is zero change in inequality but the cost of something has gone up and people have to eat it by getting less of something else.
I can’t help but notice the items of increasing price / cost are not optional / discretionary purchases. Price can increase and not affect demand. Whereas the items decreasing in price are all subject intense competition and price sensitivity.
That's not how it works: Most things that people call non-discretionary still have opportunities for market effects. Take food: You might have spikes for, say, beef, but what is necessary is just enough calories, not necessarily having them come from beef. Therefore, you can switch preferences, eat something else, and not be affected by the price increases. Poeple rarely do though, because the amount of money spent on food is significantly lower than historical, precisely due to agricultural productivity improvements. In the US, we also have to consider the effects of recent tariffs: When supply gets far more expensive, prices go up in a market, regardless of whether people must eat or not.
For housing, there's also significant location effects: One doesn't have to live in, say, Manhattan. People trade time for location, and then select the space they want. A whole lot of the space Americans use is completely optional. Go look at cities in Asia, or in Spain: You can have a city with an average density similar to NYC's Upper East side, but not even NYC comes close. That's not about a limitation of supply, but very specific policy choices.
It's similar in other mandatory things: In healthcare, the amount of things that are actually mandatory isn't that large, training to become a doctor is offered to far fewer people that would want the job, drugs can be handed long monopolies... It's not about non-discretionary, but mostly a regulatory problem. Same with American colleges, which waste an order of magnitude more money in what is shaped like an old luxury good. Anyone that has gone to a public university in continental Europe and to a US college can tell you it's a completely different good, and the American approach isn't all that focused on efficient education, as it's still shaped like a finishing school. And again, it's not necessary.
So I'd argue it's almost always regulation written to help certain incumbents, instead of inability of market forces to keep prices low even when it appears that a good is non-discretionary.
Yeah you’re right, that’s fair. I just think people don’t behave that rationally. I’m not moving away from my kid’s grandparents because my local costs have gone up, for example.
There exists greater friction with many of the items in red than the highly automated ones.
My question is how linked is this friction to the lack of automation?
With text books and meat packing there are few players due to consolidation. This means they can avoid investing in automation and keep prices high because they face less resistance from consumers and virtually none from competitors.
In short I’m asking if market forces are to blame for lower automation. And therefore automation is not the root cause of price increases.
At some price level you would move. Of course there are probably many more people that would move before you so that situation may not arise in practice. Economics works on the margin.
Sure but there’s a significant change in price that I can withstand before moving. My point is there’s less price sensitivity for some of these items. And because of that I wonder if that affects the amount of automation suppliers invest in.
College is optional. Actually, if you're just looking for "a skilled job" trade school is a better bet than college now. But the only people actually saying that are conservative nut-jobs trying to fight a culture war against a balanced education. People send their kids to colleges for the same reason why they demand more car lanes instead of better buses and trains: it's a status symbol.
The thing about status symbols is that you are buying them to feel better than someone else. So they almost have to be scarce - and therefore expensive. That's the basic idea behind cost disease; scarce things in an economy of abundance become more expensive, not less.
Perhaps it is optional for some. The point isn’t if it’s technically required. It’s that these items have less price sensitivity, which could be a greater factor in what industries invest in automation.
> Actually, if you're just looking for "a skilled job" trade school is a better bet than college now.
Better for whom? And better in what sense?
Long-term, on average, post-college careers still blow the trades out of the water in earnings.
In my case certainly, if I had bought into the “trades are better!!” online rhetoric I would be making far less money than I am now, and I get to work remote.
> Long-term, on average, post-college careers still blow the trades out of the water in earnings.
That average has a lot of outliers. There are a handful of degrees which almost guarantee you gainful employment. Like, someone getting a law degree or prepping for hospital residency will make waaay more money than maths, liberal arts, or anything on PhD track. The latter do not have anywhere close to the same job prospects.
Furthermore, some degrees are extremely expensive to get. My guess is you got an engineering or CS degree, which in terms of "degrees with job prospects" are still reasonably priced. You can graduate and go into the work force with little debt (or at least, I did, YMMV). Less so for the lawyers and doctors pushing up the college average, who have to go to more expensive schools and even more expensive post-graduate programs. They rack up lots of student debt in the process. Even if it gives you a higher salary, you might not be comfortable with a decade and change of debt slavery.
The clearest example I’ve seen of this effect is with dental hygienists in the U.S.. Big labor crunch as lots of industry veterans left the labor force during COVID, wage and career growth prospects are weak vs. alternative options for newcomers, very difficult to automate in practical terms, and the way they produce revenue is usually per-procedure, with a ceiling largely fixed by insurance reimbursement rates, so the offices that employ them see a profitability issue when contemplating a raise.
Sounds like some other places use capitation to break the tight coupling between hourly productivity and profitability. Sounds interesting but politically very challenging. Would be interested to hear some perspective from consumers in e.g. the Nordics with experience.
Computer software is on the chart as getting cheaper with time but that can’t be right? I remember Photoshop used to be a disc and you owned it for a reasonable price and now you subscribe to the Adobe protection racket. Software as a service would be a service which should go in the top of the graph with other extortionists like colleges and hospitals.
The reasonable price was $600 in 2000 which is over $1000 adjusted for inflation. That's four years of subscription, the useful life of a perpetual license may not be much more than that.
Your perpetual license on a fixed version of Photoshop would not have been transferrable across OS/compute generations for 25 years.
In practice most longtime Photoshop users paid the $600 once and some $200~ upgrade cost every 2-4 years. Adjusted for inflation its same or more than what you pay now.
If you think you'd be fine now with 25 year old Photoshop features you maybe forget how basic the product was compared to today. Further besides OS compatibility there were file format / camera raw version additions made over time that you'd have wanted.
You can get 2000-level Photoshop for free from many sources now, so you shouldn't be complaining about cost increases. Paid software has moved upmarket and bundled. The work done by people who need paid software has become more complex and vertically integrated (productivity increase.)
Are you also happy running Windows XP or Mac OS X 10.1 (Puma) which seem to be the last versions officially supported? (I guess on Windows it may have worked in some form or another longer than that but surely it doesn't work anymore on Windows 11 which is the only version currently supported.)
aren't there open source programs that have caught up to early 2000s era photoshop?
Well there are other examples, eg Google Docs is free, no need to pay for Office.
I don’t know about the aggregate data tbf
This is what came to mind for me. Certainly there are free versions of everything that used to cost money a decade or two ago. The paid software now is much more advanced.
Productivity has always imposed an upper limit on wages.
Those wages are far below the upper limit imposed by the value of college education...
That it's possibly signaling rather than something more concrete doesn't really matter to the economic analysis.
1. This is certainly a real effect that has some effect on relative wages at the margins in some cases.
2. In 2025 if you hear someone talking about it in the context of the US economy you are most likely hearing propaganda, designed to provide a dodge for the real driver of higher costs which is mostly concentrated corporate power, consolidation, and collusion.
Evidence for claim #2? The sectors where the Baumol effect has been most painful (housing, childcare, education, with the exception of healthcare) are ones that have much higher levels of competition and distribution than areas where prices have rapidly dropped. Construction Physics, for example, did an analysis [1] that showed that the top multifamily housing developer has 2% marketshare; the top residential housing developer (DR Horton) has 8.4% and subs out almost all the work, and the top 4 together have only 20% of the market. Compare with tech markets like browsers, search engines, or operating systems where the top firm alone often has 80% market share.
[1] https://www.construction-physics.com/p/why-are-there-so-few-...
I'm not sure what the answer is for housing, but there are tons of factors that go in to the growth of cost there. For one, the people making the buying decision aren't comparing DR Horton to Lennar. Usually, they're thinking along two lines: monthly mortgage cost and location.
Still, that doesn't rule out other types of consolidation (that are not necessarily corporate in nature.) There are no new "cities" being built, and even if you want to live in a small suburban community, chances are that you want or need to live near a city for economic reasons. I bet a lot of people on this forum wouldn't even consider living outside of 15-mile radius of SFO or NYC.
For individual families, the choices are often even more constrained. Assuming a dual income household, it's unlikely both earners will be able to geographically relocate at the same time. So you end up with situations where new housing outside of economic centers is pointless to build, and new housing in economic centers is expensive or impossible to build due to regulations and existing suburban street layouts.
Bringing it back to Baumol, we can think of an invisible "land value tax" as rising much like a wage rises without an increase in productivity. Since we're not making new economically productive regions, the cost of living near one of the existing ones has to rise (and we're not doing anything to counteract those trends.)
Housing is all messed up because land is a limited resource and regulation artificially limits it even further.
I live in a high demand area. A perfectly cromulent house on a particularly good lot will sell for $1.5 million as a teardown. The new house will be 6,000+ sqft and be inhabited by a family of four. Builders won’t build smaller because the land price sets a hard floor. The most profitable and economically productive thing would be to split the lot and build several smaller houses, or build a small apartment building, housing several times more people for the same cost. But this isn’t legal. Construction costs don’t make a difference. If construction costs doubled, the new houses would just get smaller. Some of these teardowns would stop being torn down. The cost of living in the area would stay about the same.
My mental model for high/low inflation goods/services are really just what % of their cost is driven by local on-shore labor which isn't / can't be automated to be more efficient. The "can't" is sometimes regulatory or cultural.
So for example childcare & education both fall into high inflation because we almost demand that it be inefficient. Customers demand to know the worker:customer ratios, and expect them to be low. It's held up by universities as a measure of quality!
Similarly with medical care, you don't see a lot of efficiency-increasing changes over time. The process of going to the doctor when you are sick, getting a prescription, and picking it up at the pharmacy is about 90% the same as it was in the 1980s. Maybe Amazon's efforts with telehealth&pharmacy can help here, tbd.
Housing is partially land use / zoning, increased regulatory burdens with time on multi-family housing, and that home construction itself is still something of an artisan craft than an industrial automated process.
I’d speculate that the cause of higher costs is excess government spending over the past few years, creating a lot more dollars chasing fewer resources.
One implication of this is that we need regulatory improvements (ie improvement via negativa, less) for healthcare, childcare, education, and building new housing. It’s non-ideal when government policies restrict supply and restrict competition, and entrench existing players.
I'm skeptical that anything actually is cheaper when normalized by offset wages. like sure you can buy a TV for cheaper, but isn't that just because now you have cheap labor and arbitrage? software is the main exception here. what physical product is actually cheaper when you remove offset labor arbitrage?
this would also explain why things that are not subject to said arbitrage do not actually get cheaper, e.g. anything that must be done locally.
Don’t automation technologies improve the productivity of labor?
If I can make one widget per hour, and some new tool lets me make 10 widgets per hour?
Conventional economic theory suggests the gain will be split between the widget-maker and the widget-consumer, in proportions determined by the relative slopes of the supply-demand curves, but definitely the product will become somewhat cheaper.
sure, but take furniture - is high quality furniture cheaper today than 50 years ago, normalized by inflation? from my investigation the answer is no.
It depends on what you mean by "high quality", I suspect. Above a relatively low floor, price of furniture seems unrelated to (e.g.) sturdiness or expected lifespan. It's more like fashion, in that you are paying for names or decoration.
Probably a difficult question to answer. My sense is that high quality new hardwood furniture hasn't gotten especially more expensive over recent time (adjusted for inflation) but you'd probably have to ask someone in the industry who is more plugged into various cost inputs.
>seems unrelated to (e.g.) sturdiness or expected lifespan
I'll disagree somewhat. At least in New England, there are smallish manufacturers who make high quality products that you're not going to find in most, if any, of the large retailers.
This is exactly Baumol: If by “high quality” you mean “difficult to mass produce” then yes, the lack of productivity gains in hand-made furniture makes real costs go up.
Of course, it’s easy to mass produce sturdy furniture, such as office furniture. But it’s not what consumers think of as “high quality”.
In a way, I thought about this when I changed my career the last time.
I had worked in what I will call "high end" tech support for some proprietary (and some less proprietary) networking equipment.
My job generally paid great and customers paid big support contracts, good deal for a good 20 years. But Tech Support is never glamorous, executives eventually think of them as just problems (even if they're solving problems) because that's all they hear. Quality management jumps to other more glamours departments and so on.
I was not so sad when a layoff occurred (company sold for parts and most of support was cut because more people on balance sheet looks costy). eventually and I learned to code late in life and got a new job / career.
Amusingly while I was learning to code a former coworker. (one of the people developing the products) at a company who bought some of the products I supported for a good 20 years reached out and said I should apply for a remote support job. I wasn't enthused but I did thinking they might make a good offer ... I never heard anything back. I was maybe one of 100 people who worked on those products in that capacity, I could have gone to work and done the job fabulously in an instant. Former coworker asked around was told "he doesn't have masters degree in CS". I wonder how those CS masters guys cost?
I got a lot of stories from that coworker how support was a complete disaster for a long long time at that company.
People rightfully complain about tech support, and I always think "Yeah they're bad because anyone who knows how to do it ... does not stick around."