The author makes an error while explaining Jevons’ Paradox. They say “the cheaper and faster we got at producing coal, the more coal we ended up using”. But that’s not paradoxical at all. That’s basic supply and demand curve from economics 101. It’s exactly what you’d expect.
Whereas if you take the extraordinarily difficult step of opening Wikipedia (https://en.wikipedia.org/wiki/Jevons_paradox), you understand the real paradox. Jevons observed that when steam engines get more efficient and individually use less coal, we end up deploying many more steam engines in many new industries and use cases, increasing coal consumption overall.
This seems like a basic oversight, maybe this guy hasn’t heard of Wikipedia. On the other hand, I think he wants to push the narrative that AI is seeing enormous productivity gains. He does this by using price per token, which has fallen. Very similar to Jevons supposedly observing “coal production”. But the author is struggling to point to new industries and use cases that were opened up by AI, like when we deployed steam engines back in the 1860s. So he misstates Jevons paradox, removes the paradoxical part, and makes it seem like his thesis makes sense.
Jevons’ Paradox could still apply here! I’m not saying it doesn’t. But we just haven’t seen the examples quite yet. A good example would be an observed surge in demand custom built software as software engineers become more efficient. But lower token costs ain’t it.
I think that statement is incomplete. It's cheaper because the AI providers are subsidizing the queries by burning cash in order to gain market share. So it's not a price, it's a subsidized and temporary price. Which will likely go up once burnable money runs out / the providers switch from "market acquisition" to "let's try to make this thing profitable at all".
Also, I'm sure I read something here in the last few weeks documenting that while every specific model shows falling price-per-token, "frontier models" which are the ones everybody wants to use - maintain steady prices (and are often significantly more expensive to use because they consume way more tokens than older models responding to prompt-fashion-de-jour).
a16z is heavily long on AI, so this article sounds very biased.
From the article: If you live in the United States today, and you accidentally knock a hole in your wall, it’s probably cheaper to buy a flatscreen TV and stick it in front of the hole, compared to hiring a handyman to fix your drywall.
Probably because the US has been focused on services for years rather than physical goods production. Everything else in US is focused on importing cheap(er) goods or materials.
> On the other hand, I think he wants to push the narrative that AI is seeing enormous productivity gains.
That is my impression as well. I would be thrilled to see this mythical 10x productivity. Even with 2x productivity, I would be highly pleased. This should mean developers (and everyone else) are producing 2x more quality, software (and general services) are 2x better? I see none of that, except 2x more junk. Did AWS, GCP, or anything else become 2x cheaper and 2x more stable? Maybe I'm living under a rock.
> From the article: If you live in the United States today, and you accidentally knock a hole in your wall, it’s probably cheaper to buy a flatscreen TV and stick it in front of the hole, compared to hiring a handyman to fix your drywall.
This also isn't true?
It costs almost nothing to patch drywall. You can also do this yourself. Unless the point they're making is "TVs are so cheap, you can mount a TV inside of the drywall for less money than it would cost to fix," which also isn't true.
Also the initial claim is just false — “if you live in the United States today, and you accidentally knock a hole in your wall, it’s probably cheaper to buy a flatscreen TV and stick it in front of the hole, compared to hiring a handyman to fix your drywall“. I can tell you that this isn’t true in San Diego. Unless they’re using flatscreen tvs that cost less than $300? Or perhaps making extremely difficult-to-patch holes somehow
This is true in pretty much every western country sadly.
TV's are really absurdly cheap (and awful) on the low end, we're not talking about your 60" LG OLED with AI TV here, we're talking: a screen with maybe 720p and a viewing angle of: dead centre.
Hiring a handyman is, what, $100/h in most countries, then there's a minimum call-out fee and materials cost- worse "I don't have the part". You're looking at about $300~ easy.
I live in the UK, and it’s basically £100 to get a tradesperson to show up to my front door regardless of what I want them to do. I can buy a flat screen TV from £100 new from one of the UK high street retailers [1] , or £85 [2] if I go on Amazon
Huh, i stand corrected, thanks! I think this brings both options to similar prices, so at the very least the spirit of the quote is true. Here’s my math:
- My handyman changes $50/hour, but if you find a new person maybe they charge $75-$100/hour
- materials are cheap, probably like $50 total for mud and drywall, or a repair kit
- with two hours labor, the total should be somewhere from $150-$250.
- if the handyman won’t accept a job less than 4 hours, the range is $250-$450.
You’re right it’s not a real paradox. It simply states that movements of a/b (efficiency of steam engines input coal / output work, price of coal per kilo of coal etc.) doesnt tell us anything about a and b. Easy maths. But for humans economics usually it means demand increasing so a and b increasing. It’s only a paradox for humans who focus on numbers too much
Oh that's a fine example of the Baader-Meinhof Phenomenon. Read this comment first, went to read one of the other top links on HN (https://news.ycombinator.com/item?id=45804377) and which Paradox shows up? Jevons'!
Was thoroughly confused which article this comment belonged to for a second.
> Jevons observed that when steam engines get more efficient and individually use less coal
And your basic supply and demand curve indicates that, at least until the supply is diminished in kind which isn't going to happen immediately given the all the constraints at play, using less coal leads to coal becoming cheaper and more readily available.
> This seems like a basic oversight
Not really. Improvements to the steam engine being why we got "cheaper and faster at producing coal" is immaterial. The added detail you've given is an interesting aside, I suppose, but doesn't change anything about the original premise. The only oversight is you not realizing this, perhaps?
Honestly I feel like Jevons Paradox is a distinctly unhelpful way of phrasing phenomena, designed to create a feeling of novelty where there's none.
It's completely obvious that if you need energy, and you have energy source A and B (lets say natural gas and oil), people will use the one that is cheaper.
Oil became cheaper because a new supplier entered the market, and people started using more of it! Jevons Paradox!
The other explanation is that people have an outsize demand for a resource, and are actively making efforts to make it cheaper so they can use more of it, then when it gets cheaper, they use more of it.
Transistors got cheaper and we are using more of them! Jevons Paradox!
That's a lot of economic theorizing, but to me, it doesn't seem to be substantiated all that well.
You can just as well argue that labor is getting more expensive in the West because of two non-market pressures. First, we have a multitude of government programs that seek to eliminate extreme poverty, so there are fewer people who are desperate to take any job for any money. Second, you have consumer protection policies that make it genuinely expensive to, say, be a HVAC repairman. Educational requirements, permits, licensing, business insurance, waste disposal, etc.
On my neighborhood Facebook group, every time someone asks for recommendations for menial, minimally-skilled backyard labor, they always insist the person needs to be licensed, bonded, and insured. And then, they're surprised that it costs $10,000 to paint a fence.
> (...) a multitude of government programs that seek to eliminate extreme poverty, (...) consumer protection policies that make it genuinely expensive (...)
I don't think this opinion holds a rational basis.
Extreme poverty being a factor in low job demand is an argument for coercing people into performing certain tasks even though they are not economically viable just because it benefits you personally. This is not a valid argument, neither indentured servitude or slavery. Isn't the US supposed to be a free market economy where free Enterprise reigned?
Complaining about regulation, including waste disposal, is also dumbfounding. Being required to dispose of air filters in a landfill is not the reason why you can't afford a repair. This opinion is also comical as HVAC also covers air quality because otherwise you can be cool in a room but literally sick.
This sort of opinion sounds completely irrational and unsubstantiated, and extremely ideological.
The main factors driving repair cost are things like device longevity, unit price, speed and ease of repair, parts availability, etc. That's mainly it. When you call someone to your house to repair something, the price tag covers that person's cost of living for the fraction of the time it takes them to deal with your problems. On top of that, you need to pay whatever parts they need to buy to get your things back to work. That's where the money goes.
> Being required to dispose of air filters in a landfill is not the reason why you can't afford a repair.
It is for someone - the market price is pretty much always going to be around the point where a small increase causes a noticeable drop-off in customers (otherwise the only sensible thing for the seller to do is charge more). If something causes even a relatively small increase in price will mean someone can't afford the thing any more.
Regulation, for example occupational licensing of jobs, has become much more prevalent. 60 years ago, only 5% of the workforce was occupationally licensed, now it is 25%.* There is a bipartisan push to reduce occupational licensing, and recognize licenses across state lines. So not irrational, unsubstantiated, partisan, or ideological.
> Complaining about regulation, [...] is also dumbfounding.
Regulation is often necessary, but it has a cost even though it's necessary.
If my country's regulations require nurseries to have one staff member for every three young children, there might be good safety reasons for that - but I'm going to have to spend a third of my salary to have one young child cared for.
Given that as soon as cloud computing happened we stopped bothering to debug VMs and just started deleting them and rebuilding, I don't know why people find the idea this applies to other industries surprising.
Repair involves establishing where in a very large state space an item is, and finding a path back to optimal.
Whereas building a new item simply involves traversing an already known path to optimal.
While in understand your reasoning it is a limited view perspective: The rebuild cost of a VM to society as a whole is marginal at best, an entire hvac system has a lot more crap it needs to dump somewhere when it is replaced.
A quick Google says it takes 4 years to get a license to be an HVAC technician in my state.
I understand large installs at businesses are a different problem, and granted I've only ever installed a mini split, but that was hardly rocket science. And home installs are likely what most people are thinking of here.
In Japan you can get minisplit's installed for $1k a unit, here you regularly find quotes over $10k. Something's gone wrong somewhere.
In Australia the going rate last time I got one installed when I was supplying the unit (this was back in 2019) was AU$540 (US$350) for a simple install (exterior wall of house etc.), because electricians just get an extra qualification and there’s a lot of competition.
I bought the 3.5KW unit online for $1080 including delivery (USD$702) so it was $1620 (US$1050) all up. I expect with the recent inflation it might cost more like $2000 (US$1300) installed for one that size and maybe $3500 (US$2275) for a bigger unit (8kW).
The splits themselves are mostly all Japanese brands that we have here (Mitsubishi Electric, Daiken, Panasonic, etc.) as well as some Korean (LG, Samsung), but Chinese ones are starting to appear in the market too. But they all seem very cheap compared to buying one in the US, before the installation there which just seems astronomically expensive to us.
> A quick Google says it takes 4 years to get a license to be an HVAC technician in my state.
A flat time mandate for HVAC tech certification seems really out of place. And a 4 year path of any sort seems excessive for a technician. I couldn't find anything like that. Most results I found were in the 6-12 mos range - which is often spent employed.
WV was an outlier with a 2000hr requirement. How I have seen (non-hvac) 2k requirements get satisfied are thru a HS VoTech (my son) or 18-24mos doing paid tech work toward the official certification (electrician techs do this).
I can't find a state that requires anything a like a 4yr college degree, where life is put on hold to focus on that. And then 4yrs of living and school expenses are investments that need to be earned back. Not for any trade tech.
> eliminate extreme poverty, so there are fewer people who are desperate to take any job for any money
One way to eliminate extreme poverty and increase skilled labour, is to ensure children have enough nutrition, health, schooling, and funds to pursue skilled worked in adult life, which usually involves making time outside of work, adequate health, some prerequisite education, and a sufficient financial buffer to actually upskill.
A second thing is Baumol's Cost Disease. If there are other industries that are more rewarding that require less effort, no-one is going to pursue those options. Why be a clever guy who makes a product (or becomes a HVAC repairer, electrician, etc.) when you can get a comfy job at a FAANG (or whatever the acronym is today). You could sub in benefits here, but I don't think people are thriving on benefits in the US. But I'm an outsider, so I wouldn't know.
But there's also a secret third thing that people don't often consider, which is the culture. If there's a culture that doesn't privilege working hard, or educational attainment, etc., people won't seek those things.
Cost of living goes up. Cost of services if they are not done more efficiently must as well go up. In the end cost of labour in my theory comes down to cost of housing and cost of health care. Cost of food and most other things should follow cost of labour. When both are profit generations or being inflated. Well there is no other options than to cost of labour go up.
> You can just as well argue that labor is getting more expensive in the West because of two non-market pressures. First, we have a multitude of government programs that seek to eliminate extreme poverty, so there are fewer people who are desperate to take any job for any money. Second, you have consumer protection policies that make it genuinely expensive to, say, be a HVAC repairman. Educational requirements, permits, licensing, business insurance, waste disposal, etc.
How's what you wrote substantiated at all?
To me it seems you are suggesting we let people starve so that labour you characterize as "minimally-skilled backyard" is cheap.
You also seem to be suggesting that consumer protection policies need to go as well. I assume we are going to trust the end consumer to do due diligence cause "they know whats better for them?"
If they get a quote of 10k and they cannot get a better one, they might as well start writing that check.
I don’t think that’s what they were saying, but instead, labor is expensive now because labor is scarce, which is true. And that regulation is expensive, which it is, but also necessary.
Perhaps the question is why wages are not high enough to support these prices (globalization, productivity wage gap over the last fifty years, etc). This will change over time due to structural demographics [1] making labor much more scarce (pushing up wages), we’re still in the early days. Software is not going to eat the trades and HVAC repair.
I didn't see anything in their comment where they added "and therefore we ought return to rampant extreme poverty", did I miss it? They don't appear to be suggesting anything you say they are, just describing the effects of what exist.
It can be true that consumer protection laws raise the price floor for certain goods and services without "and therefore consumer protection laws are bad".
> I didn't see anything in their comment where they added "and therefore we ought return to rampant extreme poverty", did I miss it?
The US is already experiencing rampant extreme poverty. There are people in the US holding multiple jobs and still can't afford to eat, let alone healthcare.
Again, this argument that things are expensive because the poor can't work and regulation somehow is suffocating businesses is purely ideological and not supported by facts.
Yes, because if you get a leak in your air conditioner, instead of just patching the hole and refilling it with coolant, you have to replace your entire system in places like California because they require a more green, less gas, in the name of science, and still completely ignoring the simple fact that it doesn’t make a damn difference because we all live in the same globe.
The amount of waste that is generated is 1000 times that of just refilling the coolant. When will people realize that you can use an existential threat that you can’t prove to justify anything? What could be more important than our existence?
If the same scientist came out with a study that said if you don’t pay me $1 million by tomorrow, we are gonna get hit by an asteroid. Would you believe them and pay me? Or has this become a political issue we’re no longer thinking rationally
> Yes, because if you get a leak in your air conditioner, instead of just patching the hole and refilling it with coolant, you have to replace your entire system in places like California because they require a more green, less gas, in the name of science, and still completely ignoring the simple fact that it doesn’t make a damn difference because we all live in the same globe.
I don't think you are holding an informed opinion. Ozone layer depletion was tied primarily with CFC emissions used in air conditioning units, and since the production of CFCs and other ozone-depleting chemicals were banned in the late 1980s the ozone layer started to regenerate. In practical terms this means lower incidence of health issues such as skin cancer or cataracts. That sounds pretty neat.
But being able to use a 40+ year air conditioning unit is worth it?
> First, we have a multitude of government programs that seek to eliminate extreme poverty,
Clarification: In some places we absolutely do. In others we absolutely do not.
> so there are fewer people who are desperate to take any job for any money.
I think is is better reflective: There is huge surplus of employers have that set up systems that insure they do not hire qualified people. Job portals that auto-trash applications from unwanted applicants (1st time, most minimal of crim rec, wrong zip codes) are one massive example.
source: me+kids spent a decade in red state, hunger-level poverty. kids who got zero replies during months/years of entry level job apps.
Why do you think people insist someone needs to be licensed, bonded, and insured? It's not really a quality question, it's CYA, a direct response to the rise of ambulance-chasers and increasingly-ridiculous jury awards. Morgan and Morgan are probably doing as much economic damage here as Baumol.
Yikes! There are so many corner cases and pitfalls in construction, it would blow your mind! I did it for 2 months way back and concluded I would certainly die.
Paint a fence? Pay the neighbor kid. Patch a roof? eh ... what could go wrong.
They insist on that because if someone does work on your property they might destroy your property
Yeah it's CYA: the ass in question is yours. The guy who burns your house down without insurance will just file bankruptcy. You're the one left with the ashes.
You might want to really do a careful read of your policy, because almost all of them have provisions about requiring work to be completed by licensed tradespeople.
it's not surprising that you've picked up some unsubstantiated protestant dogma disguised as economic theory, if you are hanging out on facebook neighborhood groups
I'm not sure why everyone picks on radiology as the 'obvious' field that will get automated. So far it has been the opposite:
> In 2025, American diagnostic radiology residency programs offered a record 1,208 positions across all radiology specialties, a four percent increase from 2024, and the field’s vacancy rates are at all-time highs. In 2025, radiology was the second-highest-paid medical specialty in the country, with an average income of $520,000, over 48 percent higher than the average salary in 2015.
Simply put, radiologists do a lot more than merely read scans:
> Radiologists are useful for more than reading scans; a study that followed staff radiologists in three different hospitals in 2012 found that only 36 percent of their time was dedicated to direct image interpretation.
In a similar vein, studies have found that software engineers spend only ~30% of their time coding on average. Yet we're similarly meant to believe that AI is going to replace human engineers? Ironically this belief is often held by software engineers themselves, as if they don't realize that they do so much more than generate characters. Bizarre.
> everyone picks on radiology as the 'obvious' field that will get automated.
Nah, pre-LLM, I think the obvious fields to pick on were lawyers, middle-management, and "email jobs" generally. That was a big miscalculation, since most people (especially engineers) do not understand the politics of power. Those jobs tend to jealously protect the power that they have and systematically dismantle what accountability they might be subjected to. Engineers in general are much more likely to democratize (and thus threaten) power, by creating things like accountability via metrics, at the same time as they mostly refuse to unionize. Radiologists have some unions, medicine generally enjoys a moat of credentials and certification. Things that SWEs in particular rejected while they said "come on over, anyone can code". I doubt radiologists ever suggested themselves that they should be measured on throughput, but SWEs actually did push ideas of 10x engineers and metrics like lines-of-code for years to argue they are productive enough to deserve raises.
A radiologist read my brother-in-law's MRI and contacted him and told him to go directly to the ER and actually wrote up a referral to have him immediately admitted. This happened in the middle of the night since the radiologist was reading them on the night shift. He was admitted and operated on two days later (as he had to be prepared for surgery and have additional pre-op testing done).
No AI replacement is going to be doing that anytime soon.
I'm as skeptical of AI as the next guy, but in this particular case, wouldn't the AI have read your brother-in-law's MRI right after it was taken, and sent him to the ER before he even left the MRI testing room?
I'm not arguing with the GP's point that radiologists don't do many other things that the AI maybe can't do, but it feels like your example is the opposite of that.
And not only that, your example demonstrates a failing of the human's limited amount of time to get all their work done.
The article addresses this. It explains that if large portion of your job can be automated (and hence be done at little marginal cost) then the remaining parts of your job become more valuable.
Radiology seems like a good field for AI because it's easier to see what automation would look like in practice, since an MRI, CT-Scan directly produces data and which can be fed to an AI. How well this actually works, I have no idea.
Just had our minisplit develop a leak after 7 years.
The shop recommended replacing it, as just the refrigerant is half the price of a new one if most has to be replaced, due to environmental taxes.
Basically a repair is guaranteed to be at least 50% of new price of NOK 23k ($2.3k) installed and can quickly approach 80+% if the guy has to spend time on it.
That made me wonder how on earth they make money on selling these things. And how effective that tax was, given it's pushing us to replace a unit that's probably fine for many more years with just a bit of TLC.
23k seems like a lot, especially so if the leak is somewhere other than a hidden part of the line. Equipment cost is likely only a couple of k, and labour to replace without running new lines is maybe a few hours.
Please get a second opinion, especially if you can find a non-shop to give you one.
> 23k seems like a lot, especially so if the leak is somewhere other than a hidden part of the line.
I am inclined to agree. The rear 2½T unit in our rental blew a line last year. Two companies quoted my landlord $10k for a new 2½T condenser and air handler (1 was another renter of his). Even that seemed high to me but I could be out of date
In the end, I had a friend come out and look it over. Leak was near the compressor and he charged well under $1k for the fix, mostly for refrigerant.
That is because it's wrong. Sorry, just woke up. I'm in Norway, and it's 23k NOK, or $2300.
The shop said filling my system back up with refrigerant if it was empty would be around $1.3k, and based on what I've heard I don't think that's off by much.
Oh I don’t know how I missed that unless you made an edit, but I also just woke up :) This price makes more sense.
23k didn’t sound too far off what an American installer would try charge for a new system in a high cost of living area, so this wasn’t too unbelievable.
> 23k didn’t sound too far off what an American installer would try charge for a new system in a high cost of living area
I've heard these insane prices. I think 2.3k is pricey enough, you can get cheaper units here but SO wants the pretty indoor unit so pretty indoor unit she gets.
For a non-obvious leak it's pretty normal to recommend a new system at that age. The reason is that a tech's time is expensive, and finding a leak can take a lot of time. Then once it's found, if the leak is in the exchange you're looking at a fairly large replacement part cost.
At least in the US, refrigerant costs have been high because of a shortage, not taxes.
If you need R410A refrigerant, the high costs are due to government regulations causing the shortage. I very much hope that the heat pump I had installed 2 years ago does not leak. Theoretically, if the leak occurs under warranty, it should be covered. The central A/C unit I replaced with the heat pump had developed a leak right before the warranty expired, which resulted in a costly repair being covered. Unfortunately, we paid another company to refill it before learning that it would be covered as part of the warranty repair by the company that installed it. It then developed a second leak several years later, so I had a heat pump installed. Of course, that heat pump uses R410A refrigerant.
The leaky A/C unit had been made by Lennox while the new heat pump was made by Fujitsu. I very much hope that Fujitsu engineered its heat pump to last. The heat pump had also replaced an oil heating system that was around 25 years old and still could have been used for many more years. Expecting similar or better longevity out of a heat pump does not seem unreasonable.
Yeah labor costs also make a lot of repairs uneconomical. There's been talk here about removing the 25% VAT on repairs to make the value proposition a bit better, but doesn't seem to have much traction currently.
> At least in the US, refrigerant costs have been high because of a shortage, not taxes.
The guy said it was due to taxes, specifically that they had gone up so much in recent years. Seems it's because the tax is tied to CO2 tax[1], which has been going up since they introduced it in 2020. Not sure what refrigerant they use in my minisplit, but even if it's one of the cheaper one the tax is about $90 per kg, so adds up quick.
Here in Vietnam nearly all ACs are these minisplit type. It costs about 500,000VND ($19) to get one serviced, including replacement coolant. It's cheap enough that I get it done at least once or twice a year for all units (the coolant doesn't always need replacing but I get it cleaned and serviced).
I know the US and Vietnamese economies are very different, but something doesn't add up there.
Minisplit AC units are not very popular in the US. The US mainly uses either central air conditioning units or all in one AC units that are designed to be installed in a window. The window units are cheap, but inefficient. The central units are more efficient, but very expensive. Minisplit units are more expensive than window units (by probably a factor of 10), but are more efficient than either. I suspect most people in the US do not know that minisplit units are even an option.
In my home, I recently had a heat pump unit replace my central A/C with some minisplits that connected to the exterior unit installed in the basement. The entire setup cost as much as it originally cost to install the central A/C, despite parts of the central A/C being reused.
Note that in the US, what we call air conditioners only support cooling and not heating. When they support both, we call them heat pumps despite that being the scientific name which applies to the cooling only units too.
I don't think it's just "most Americans don't know."
When I installed AC in my former Bay Area home, I would have needed multiple mini split units to cover 1200sq ft, with questions about how many you could have on at once to still get the right performance. I went with a single central air unit instead.
It also (at the time at least) didn't come with a centralized thermostat, which meant managing each room individually (that would've been fine with me, personally, but it's a drawback for lots of people).
On top of that, many (if not most) mini split units are also somewhat aesthetically displeasing.
In my new home (on a different continent), I have mini splits. I'm somewhat satisfied with them, at best. I'd still prefer central air but it's not a thing for residential homes here.
If I were to ask most people I know offline, I suspect only a small number of them would know what a minisplit unit is. For most people, A/C is like plumbing. They do not think about it until it breaks. Learning that there are better options is just not something that people do.
That said, I was replying to someone from Vietnam. Assuming that things in Vietnam are similar to China and Japan, people will only heat or cool the specific rooms that they are using, rather than heating or cooling everything like how many Americans do things. Those in the US who cannot afford to heat or cool everything, who are likely very underrepresented here, would be those using window units, since they are cheap upfront. A minisplit would be cheaper over the long term, but the high upfront cost dissuades people in thing market from even looking at them.
Finally, I had Fujitsu minisplit units installed in my basement two years ago. They are far more aesthetically pleasing than window A/C units.
How long of a long term are we talking about? I suspect this is very dependent on many factors (not least of which is labor, but also insulation, climate, etc).
The energy (gas+electric) bill for my whole home would've taken well over a decade to add up the amount the mini split installation alone would've cost, so even if it brought down my energy cost to zero (which it wouldn't), it would've taken a decade to pay for itself (including the cost of window units).
Edit: and yes, they're prettier than window units for sure, but as you've pointed out, they're not competing with window units. They're competing with ducted central air that has almost no visible impact.
Its really expensive, its true. Prices spiked like mad after the coronavirus shortages, AC parts were back up forever. It sucked and the costs rose tremendously and seemingly never fell. Then we tarrif the imports and change the rate constantly. Nobody can sort out their supply chains(But it is not just the tariffs).
The refrigerant is often expensive, poisonous and flamable. We don't replace the refrigerant, we reclaim it and dispose of it properly. Its not without risks. Our building codes are pretty rigid as well, its a big pain point.
New HVAC install is gonna need an electrician, the lines pass through walls allowing the possibility of condensation to produce deadly mold. Every installation needs individual consideration. If it isnt %100 perfect, the customer will be riding the installer's ass.
The US has is a Safety Above All mindset on some things. We improve safety far beyond economic rationality because we don't want to systematically kill each other in this way, however few that is.
Unrelated example: a national EMT outfit operating here made all their techs wear plate carriers and ballistic plates. There is nearly no gun violence here. All the techs stopped carrying the plates after a few months because its dumb, and wont stop a baseball bat or a knife, but this is a national outfit, surely they did a cost benefit analysis. Someone signed the check and wrote a policy to make them wear em every day. Plates expire, also.
That said, we can definitely do better, and the cost is too high. Installers are in demand, and we tariff the imports to the tune of %15-%30.
Almost certainly. Although at least they all use the newer R32 coolant these days. A typical AC unit uses 0.5 to 1kg of R32 coolant, and releasing 1kg is about the same as 1 person taking a 1000km flight (from a quick search). If you replace the coolant every two years it'll pollute more from the electricity used than the coolant being released.
I think it could make sense if that $2.3k mostly only accounted for Western labor involved rather than representing entirety of "real" costs. Like - just entirely making up - 1k to ship it from middle of the Pacific to nearby ports to you, 0.5k from the port to location of installation, another 0.5k to install, 0.3k for documents preparations and extra wood screws, 2.3k total. And $25k worth of man-hours in CNY to manufacture, which effect is isolated and contained within their own economic bubble.
I can't just believe in the "them subsidization" theory anymore, international prices just don't seem like compensating for anything. It feels like paying for energy costs of delivery systems rather than the product. It just doesn't make sense.
The problem isn't tradework costing more, the problem is most people making less money thanks to wage stagnation and can't afford reasonable trade rates. Capital costs for the neccessary tools are also higher than ever.
I can build a house 4x faster today than my father could back in the day, or alternatively with 4x less labor, and yet he could buy a house, 2 cars, and support 3 kids on his single wage. Meanwhile I struggle to do more than just survive in my little shack house built from scrap on an empty plot doing the same thing.
The only jobs that earn me a decent wage is for corporations or for people near or in the top 1% of wage earners.
If the average person made more money I would have more work and more money. But as it is a lot of residential work would be of negative value to me at the prices people can afford, and with less work to do and farther and farther travel distances for it my prices must keep rising to stay solvent.
It's rules and legislation. The job of banging together reasonably serviceable form of shelter isn't that expensive. It's only expensive if you have to jump through a lot of hoops to use certified designs, materials, etc. deal with inspections, use expensive people for that have had training for wiring things together, etc.
Any slum in developing nations features lots of housing that people bang together out of trash. Very unsafe. But it works. We've been creating shelters for ourselves since prehistory. It's not that hard.
Millions of people live in slums. Upgrading that kind of housing to something slightly better isn't all that hard. Most people that a are a bit skilled can bang together a shed in their garden in no time at all. Putting a few solar panels on top isn't that hard either. And you can plug them into batteries easily. You can buy that kit on Amazon and run your power tools of that. All legal.
It's only when you want that in your house connected to the grid that cost suddenly balloons from a few thousand to many tens of thousands. Exact same technology. Maybe you'll use slightly more panels and a bit bigger battery. But now it's a lot of gate keeping by inspectors, electricians, certified equipment, etc. that come into play.
Same with houses. You can buy a recreational vehicle or caravan for a reasonable amount of money. Second hand these are very affordable. And some RVs can be quite nice to live in and even have AC. So, why are houses so expensive relative to RVs? Prefab housing has been a thing for decades. If you remove the wheels from an RV, it's basically a house. If you live in an RV, you are referred to as trailer trash. It has a stigma. But it's very cheap. Poor people do that. Because it's very affordable housing. That's why it's a popular option for people that would otherwise build slums.
> The job of banging together reasonably serviceable form of shelter isn't that expensive. It's only expensive if you have to jump through a lot of hoops to use certified designs, materials, etc. deal with inspections, use expensive people for that have had training for wiring things together, etc.
"Reasonably serviceable" — ay, there's the rub!
The U.S. Navy's nuclear propulsion program has a saying: You get what you INspect, not what you EXpect. Certifications, inspections, etc., are meant to try to keep lazy-, sloppy-, and/or dishonest providers from passing off shoddy (or even unacceptable) work product.
In many areas, most customers don't have — and shouldn't have to have — the expertise or the time needed to inspect and assess providers' work product. By requiring knowledgeable inspections in critical areas, we increase the cost a bit, but we improve safety for everyone — which lets us non-experts spend our time more usefully. (It's a form of division of labor.)
To be sure: Certifications, etc., can be captured by industry groups and used to limit competition and increase prices. That's a separate issue, one that can be dealt with in other ways.
One interesting consequence of this is that your opportunity cost is now highest it's ever been. By choosing to post on Hacker News instead of spending my time better (perhaps even as simply as watching my daughter sleep) I am wasting more value than ever before. In days past, you were substituting something mindless for this. Today, by not spending your time with your family, or reading, or on your hobby or relaxing you are paying an ever more valuable slice of time on addictive social media like this.
There is a simpler explanation - HVAC jobs are directly influenced by the push to covert everything to heat-pumps. And good heat-pumps are expensive. So when you install a $15K to $25K, installation fee also goes up. I am surprised Solar isn't up there. Panels are cheap, install is what we pay for 10 years+
If it was so lucrative, wouldn’t more people set up shop undercutting the current offerings? Why not become an HVAC installer and make millions, if you’re really able to make $15k profit on a job you can turn out in a day.
The truth is probably more that the various money sinks in our society are starting to add up, things like healthcare, legal protection, licensure, compliance, rent (business or personal), even just having appropriate work vehicles, fueling them, compensating people for the time spent sitting in traffic to come across town to your house. Somehow you’re paying for all of that when someone’s livelihood is installing your mini split. A lot of those costs have grown faster than wages, if you try to point to a reason why it’s different today than 20 years ago. More people looking to make a quick buck without doing any work or providing any real value, and more people succeeding.
In my state (OR) it takes 4 years to become licensed to do the work for others but homeowners can do the work themselves.
My experience is that it’s not generally well understood how simple it is to install mini splits. The supply companies won’t sell to you directly outside of d2c web companies like hvacdirect
I acquired a minisplit for $450 or so (labelled "For export only - not for sale in US or Canada", because it contained R-410 and doesn't meet the current efficiency standards) and installed it myself.
Getting a similar system installed would have been north of $10,000, and before anyone says "well, that would be a licenced HVAC installer", no it wouldn't - it would be a barely-trained person who is simply "supervised" by a licenced HVAC technician.
It is expensive, but I think you're underestimating the costs.
If the AC catches fire because your electrician skills are bad, what happens? I guess you can rent a ladder if you need one, but they're at least $200 if your split is on the second floor and ladders can be deceptively tricky, and load ratings must be considered. Condensation can kill you and be an extreme cost with mold. Your first mini split is going to take a real long time to install, I promise, assuming you size it right. There is a non trivial risk to life and limb.
This is one of those "Reality has a surprising amount of detail" things.
I don't know what the deal is about people saying heat pumps are expensive. They used to be a little pricier than AC units, but it's just a 4-way valve in addition to one.
I just looked it up, and I can buy a heat pump for 200-400 euros (depending on desired output), installation is ~400 euros. Why are you paying 20-30x for something identical? This sounds like a price difference created by government behavior, like with solar panels and related hardware which seem to be significantly overpriced in north america.
> This sounds like a price difference created by government behavior
It's a price difference created by market segmentation of heat pumps as a luxury product in the US, and the relative lack of qualified installers due to our under-investment in education in the trades.
Is this some country specific terminology? At least in Australia I've never seen an air conditioner that didn't use heat pump technology. Aside from evaporative cooling that is.
Air conditioners (the things that can make a room colder, but not hotter), are indeed heat pumps, but in the US when we refer to a "heat pump" we mean the same technology, but with a reversing valve so that it can make rooms both colder and hotter.
Interesting. I’ve never seen one that couldn’t heat and cool before. Even crusty 30 year old window units can do both. Seems almost absurd to not utilise it both ways.
You have to specifically look for cooling only AC where I am. Most ACs come with heat-dehumidify-cool mode selection and therefore qualify as "heat pumps", as far as how the term is used. I think it's just quirks of regions that traditionally didn't have ACs by default.
Example of a solar install that was under $20k via DIY - maybe took this guy 1-2 weeks of full-time work. But he was quoted $90k and he did most of it himself - only hiring a backhoe operator to move some dirt around.
We're at the point where the trades are only going to cater to rich/desperate people because the margin they'd make on a job that is DIY-able (and charged a fair price) is not worth it to them. Why do 5-6 low margin jobs when you can do 1 high margin (rip-off) job? Your only competition out there is someone with the will to do it. Builders in CA are massive ripoffs as well.
I think the only way you bring costs from trades down is by having all your workers in-house - which is only doable for corporations. Your average homeowner is just fucked and is gonna have to youtube everything.
> It's that rich people are the only thing the trades are catering to now.
This is my gut feeling too. I've known so many rich people who just accept whatever number a tradesperson quotes. There is no way that hasn't had a mass effect
I'm not rich (or at least, don't feel rich with the enormous cost increases on everything combined with no promotions for 3 years) -- but how do you negotiate with a tradesperson exactly?
All of the major trade companies around where I live (HVAC, plumbing, electrical etc.) in the US have rates that they quote before the person will even show up. As a new homeowner who didn't grow up in the US, that's all I've ever dealt with.
If the answer is “Give them a number you're comfortable with, and just DIY it as an alternative” -- that's fine, and I do it for anything simple; but for the remaining ones, I have already made a determination that learning this skill would be way more in terms of time invested than the $500 or whatever absurd number they are quoting for a simple repair (this logic likely breaks down over time, and I'm trying to invest more time into learning more house repairs).
I have tried pre-purchasing some parts in the past; and asking them to use them for the install -- that one had some success and a guy told me how much his company marks up parts (n00%).
I do try to get multiple quotes for something, but the difference between them isn't usually appreciable; they're all absurdly high. I've tried to ask them for a parts v/s labor breakdown in the past; some won't even provide that.
If it is a company with more than 2 or 3 employees you are already going to be given a significant markup right off the bat for their marketting and commercial office expenses and have obfuscated costs, individual contractors have lower overhead. Location is also important, if I gotta drive an hour each way for a job that is going into the price. Getting legit parts, especially with short notice, also costs more. While I might buy a $8 ignitor on amazon for my own furnace or a friends, I cant risk crappy counterfit parts for regular customers that might blow up in a week so now its a $35+ part.
Buying your own parts can help but can also burn you if its wrong, or cost the same if something extra is needed that wasnt expected and requires a second visit or bought on short notice from a local parts dealer. A contractor often eats the costs of wrong parts they ordered and just hopes they can use it elsewhere later, but if you bought the parts that is just cost on you.
My recommendation, which is still probably of limited help and won't always be worth it, is to start by hiring a local handyman instead of a specialist and having atleast 2 weeks of lead time for parts. Of course finding a worthwhile handyman can have its own difficulties because so many tradesmen leave the industry after realizing corporate contracting pays workers like trash while taking a lot of the most valuable and worthwhile contracting work off the market from independent contractors.
You'd need to be negotiating with an actual tradesperson, not someone who is a glorified salesperson / unskilled installer who works for a company that has been acquired in a PE rollup.
The easiest way to find a good tradesman is to ask another tradesman. There is an HVAC person in my local area who will come out and do most jobs (such as, for example, moving an AC) for about $500. A PE rollup firm would quote $10k for such jobs.
What I'm seeing is trades workers basically don't want small one off jobs because you waste so much time on the overheads unrelated to the actual work. If someone has a hole in the wall they need patched, you're spending time answering the phone, estimating the price, driving there, billing the customer, etc. And a lot of jobs are very one off and difficult to evaluate the price.
Why even bother when you could just work for a mass build and plaster up hundreds of walls in a single job on a new apartment building or housing development.
So as an individual it's almost impossible to get someone over to do a small job, and your only realistic option is to do it yourself.
Most of the tradesmen around me won't even get out of bed for less than $1,000. They really don't want piddly little jobs from us peasants. I've started DIY'ing almost everything I need to do around the house because if you call an electrician, he'll quote you a $2,000 "go away" price because he's busy doing $50K new construction jobs.
I don't really know how this is any different now from the 1990s. I think there is a sense of bottom level on price they'll take for a job due to litigation/insurance concerns. But, handymen are what should be doing a lot of these jobs. I do think there is another concern about quality of workmanship as well. You take on these small jobs, one goes poorly - you take a huge hit on your reputation comparatively and have to fight to get small jobs even more than before. That wasn't as big of a concern before the internet made everyone's reputation so important. Losing one customer wasn't losing the whole market - whereas it is now. A few negative reviews can permanently tank your business.
The inflation in health care cost is not down to Jevons in my opinion. The likelihood of AI job displacement is low because the demand backlog is high.
Robotics will reduce back injury rates and so reduce Nurse shortages, for example. Misdiagnosis rates cause increases in compensation claims so improvement in diagnostics reduce risk so insurance premiums(for hospitals) should go down or at least not rise as quickly compared to inflation.
Note that the cost of health is high even in the economies without the broken US health system, and we have the baby boom moving through which is a huge bubble of cost associated with age.
Not that there's no component of Jevons in this, but it doesn't explain everything.
A thing said to me in Argentina resonated: They pay doctors, nurses, pharmacists slightly more if they will do out-calls to elderly people in the community at large. The increase in staff cost to have home calls offsets the massive increase in care costs if these people cannot be cared for in their home and move into a managed care facility. It's partly an externality if you privatise healthcare: who cares if the state has to pick up the tab, right? But in a more integrated view of the costs here, its better to pay more for people to help keep the elderly in their own home as long as possible.
(elderly, intending to stay in my own home, not in the US health system)
> if you automate 99% of a job, that last 1% the human has to do is incredibly valuable because it’s bottlenecking everything else.
The author is being (intentionally) naive here.
History and current research suggest that when technology automates the vast majority of a complex job, it can lead to the "deskilling" of the human worker.
This Lancet article (published Oct 2025) discovered that doctors were found to be less adept at finding precancerous growths during colonoscopies after just three months of using an AI tool designed to spot them: https://www.thelancet.com/journals/langas/article/PIIS2468-1...
And the thing is, I think the above finding is pretty intuitive.
If AI performs 99% of a radiologist's diagnostic work, the human's role will very likely be reduced to a skill that is more routine, that requires less expertise, like a final check or something - and thus commands lower wages.
Yet prices of these tests will not go down. AI and the companies that perform the test will see the rewards of eliminating these skilled jobs. The consumer will not see the savings.
It's cheaper to fix the hole, even if you screw up color matching on the paint a few times.
I've seen a lot of people blow a lot of money on really basic stuff like this (myself included). The lack of basic awareness around hvac, landscaping, drainage, drywall, plumbing, electrical, et. al. has me wondering if I'm still in the right business. ChatGPT can't carry a bag of sand up a hill or dig a ditch. It can tell you about these things and make you feel like a hypothetical god over them, but it can never do the actual work on site. I don't feel like there's a lot of competition around being in Texas crawl spaces during the summer.
This was pretty interesting. But the part I didn't get is where its stated that things got more expensive, but we consume more of them because we got richer.
Is this not just inflation? If everyone got paid more and everything got more expensive, are we not essentially level?
> If you live in the United States today, and you accidentally knock a hole in your wall, it’s probably cheaper to buy a flatscreen TV and stick it in front of the hole, compared to hiring a handyman to fix your drywall.
While this is true, the costs are inflated because you need to repaint the entire room to get the original look, rather than only pay the cost of merely replacing the drywall. Of course, some handymen are much more expensive than others, so it is possible that is more expensive too.
If you are one of the few using wallpaper and have extra wallpaper for just such emergencies, using the extra wallpaper to paper over it should be cheap.
> With widespread AI adoption we plausibly could consume 10x or more of the service: Legal services, for example, plausibly fit this bill.
A ten-fold productivity gain in legal services sounds simply awful for society. Imagine the time and money sink if everyone can sue you for every frivolous thing because AI can prepare and file the paperwork instantly without needing a lawyer. You'll need your own AI to defend against the onslaught of legal disputes.
Every contract for jobs and every terms & conditions for services will be 10x longer because AI has a much higher complexity threshold compared to a human. My belief is that one reason tax returns became much more complicated in the last ~30 years is because of tax preparation software. In the era of paper tax returns, there was a limit to the complexity that an individual or even an accountant could handle, so there was a limit on how complicated the government could make it.
Most normal people rarely need a lawyer in their lives. With AI's productivity explosion in the legal services, you're going to need legal services every day. Your neighbor wants to borrow your chainsaw? Your AI legal agent will negotiate a liability waiver with his AI agent.
Id blame the complexity of tax returns more on lobbying. Everyone wants their own industry custom exemption and that in turn creates holes that need to be patched etc
> How did this happen? 100 years after Jevons published his observation on coal, William Baumol published a short paper investigating why so many orchestras, theaters, and opera companies were running out of money. He provocatively asserted that the String Quartet had become less productive, in “real economy” terms, because the rest of the economy had become more productive, while the musicians’ job stayed exactly the same. The paper struck a nerve, and became a branded concept: “Baumol’s Cost Disease”.
I find the term “cost disease” very negative. It actually describes positive social progress.
Instead of a “disease,” it’s really a sign of a healthy, advanced economy.
This is a very interesting piece. I'd never heard of "Baumol’s Cost Disease" but I think the thing it is appliesd to most often here in Australia is coffee.
Everyone complains that it is $7 for a cup of coffee, and yet demand for barrista coffee keeps increasing and demand for barristas keeps increasing.
Fortunately the massive amount of information available online, continued search-engine-degradation notwithstanding, often means finding out how to do many things yourself is much easier and cheaper than hiring someone else to do it.
A friend works for accounting firm servicing decent number of HVAC shops they are not as profitable as it would appear from outside from crazy markups.
Lots of people do all kinds of things that are not explicitly written down at their jobs. Those things cannot be tokenized, they cannot be taken in by AI no matter how intelligent and sophisticated it gets. They are judgement calls, made by real people, who think harder and bring more to the table than any machine, no matter how many FLOPS.
But, nonetheless, those jobs will probably disappear, and the machine will reorganize itself to maximize its self-legibility, making tons more money but becoming yet more shitty and inhumane in the process. Discarding the whole reason the jobs exist in the first place (stuff like Value and Service — oh excuse me, sorry, meaningless cost centers to be optimized).
Sort of like how Radiologists do all kinds of important shit, that gets entirely hand waved away in this think-piece as 99% automatable. Yeah, sure guys, the radiologist — actual doctors for fucks sake — I’m sure he is nothing but a warm body between the patient and the computer, signing paperwork and collecting cheques.
> “Baumol-type effects” means that everything that didn’t get more productive got more expensive anyway, but we consume more of it all the same because society as a whole got richer.
It depends how you measure. The cost in goods or purchasing power goes up, but the cost in hours stays the same.
There is probably an equivalent of Amdahl's law for GDP - overall productivity will be bottlenecked by the least productive sectors.
Until AI becomes physically embodied, that would mean all high-mix, low-volume physical labor is likely going to become a lot more valuable in the mid-term.
> We are at the point in the technology curve with AI where every day someone figures out something new to do with them, meaning users will take any chip they can get, and use it productively.
Sure, that's an assertion.
But (with just as many citations), mine would be:
This boom is absolutely, 100%, fueled by the combined factors of:
1) employees outsourcing the cognitive load of their jobs to models that are, impressively "close", but not quite _as_ good as a well-trained human.
ie, we're replacing google with a fun, but terribly energy-wasteful (and _very often_ factually wrong) "make up an answer" tech.
and 2), AI "app developers" who are having fun with the previously "impossible" (*cf. https://xkcd.com/1425/) APIs of multi-modal natural language, and "didn't sci-fi warn us about this?" simulations.
Neither of which are good for productivity, if we measure productivity as "improving circumstances for the mutual commonwealth of all life". Which is the goal.
* oh, I _did_ use a citation after all.
It is an interesting article, but _far_ too sure of itself in all the wrong areas.
This is mostly down to people being afraid of anything even remotely trades-like. Learn to do some basic home repair, it will save you thousands.
> This graph can mean different things to different people: it can mean “what’s regulated versus what isn’t” to some, “where technology makes a difference”
Cars are pretty heavily regulated...
What I see is what is necessary to live and what isn't. Elastic vs inelastic demand.
> the average American middle-class household can comfortably manage a new car lease every two years
Huh, no, the average American middle-class household cannot do this.
> If one sector becomes hugely productive, and creates tons of well-paying jobs, then every other sector’s wages eventually have to rise, in order for their jobs to remain attractive for anyone.
I'm sorry, but anyone who has lived in the lower income brackets knows this just isn't true.
This is hard to read. Whoever wrote this is extremely out of touch and thinks they're eminently intelligent. It reminds me of the "smug San Francisco" South Park episode. The world is going down a road of hurt and you've got elites who are so busy "winning" over the past 50 years running around sniffing their own farts.
> This is mostly down to people being afraid of anything even remotely trades-like. Learn to do some basic home repair, it will save you thousands.
Installing a new HVAC system is not "basic home repair".
Yes, there are HVAC-related repairs that qualify as basic, but we're also talking about the big things.
And while yes, many homeowners could learn how to install a new heat pump, run refrigerant lines, make sure every connection is torqued properly, etc., most would not want to or have the time to do so, and that's fine, normal, and expected.
Seems like most uses for hallucinating bullshitters that steal intellectual property are going to require significant human oversight, likely often multiple levels of human review, with almost everything they do that has value.
Since AI is the future, here is what AI says (which is basically sourced from a couple Reddit posts):
"The cost to repair an AC in Hong Kong varies by the problem, but basic repairs can range from approximately HKD $500 to $1,000, while major repairs like a compressor or refrigerant leak can cost HKD $1,500 to $4,000 or more. Inspection fees are often around HKD $500, with the repair costs and any needed parts being quoted separately. "
Whereas if you take the extraordinarily difficult step of opening Wikipedia (https://en.wikipedia.org/wiki/Jevons_paradox), you understand the real paradox. Jevons observed that when steam engines get more efficient and individually use less coal, we end up deploying many more steam engines in many new industries and use cases, increasing coal consumption overall.
This seems like a basic oversight, maybe this guy hasn’t heard of Wikipedia. On the other hand, I think he wants to push the narrative that AI is seeing enormous productivity gains. He does this by using price per token, which has fallen. Very similar to Jevons supposedly observing “coal production”. But the author is struggling to point to new industries and use cases that were opened up by AI, like when we deployed steam engines back in the 1860s. So he misstates Jevons paradox, removes the paradoxical part, and makes it seem like his thesis makes sense.
Jevons’ Paradox could still apply here! I’m not saying it doesn’t. But we just haven’t seen the examples quite yet. A good example would be an observed surge in demand custom built software as software engineers become more efficient. But lower token costs ain’t it.
I think that statement is incomplete. It's cheaper because the AI providers are subsidizing the queries by burning cash in order to gain market share. So it's not a price, it's a subsidized and temporary price. Which will likely go up once burnable money runs out / the providers switch from "market acquisition" to "let's try to make this thing profitable at all".
From the article: If you live in the United States today, and you accidentally knock a hole in your wall, it’s probably cheaper to buy a flatscreen TV and stick it in front of the hole, compared to hiring a handyman to fix your drywall.
Probably because the US has been focused on services for years rather than physical goods production. Everything else in US is focused on importing cheap(er) goods or materials.
> On the other hand, I think he wants to push the narrative that AI is seeing enormous productivity gains.
That is my impression as well. I would be thrilled to see this mythical 10x productivity. Even with 2x productivity, I would be highly pleased. This should mean developers (and everyone else) are producing 2x more quality, software (and general services) are 2x better? I see none of that, except 2x more junk. Did AWS, GCP, or anything else become 2x cheaper and 2x more stable? Maybe I'm living under a rock.
This also isn't true?
It costs almost nothing to patch drywall. You can also do this yourself. Unless the point they're making is "TVs are so cheap, you can mount a TV inside of the drywall for less money than it would cost to fix," which also isn't true.
I had to buy the sanding pole, the joint compound, the putty knife, and the paint the other day. A TV would definitely have been cheaper.
TV's are really absurdly cheap (and awful) on the low end, we're not talking about your 60" LG OLED with AI TV here, we're talking: a screen with maybe 720p and a viewing angle of: dead centre.
Hiring a handyman is, what, $100/h in most countries, then there's a minimum call-out fee and materials cost- worse "I don't have the part". You're looking at about $300~ easy.
But for $129 you can get this; https://a.co/d/7cdztf8
[1] https://www.argos.co.uk/product/7623909?clickPR=plp:6:323
[2] https://amzn.eu/d/bVCBLv3
- My handyman changes $50/hour, but if you find a new person maybe they charge $75-$100/hour
- materials are cheap, probably like $50 total for mud and drywall, or a repair kit
- with two hours labor, the total should be somewhere from $150-$250.
- if the handyman won’t accept a job less than 4 hours, the range is $250-$450.
The key part is the "demand more than eclipsed the cost savings" bit.
The cotton gin is another well known example, labor per unit down, labor for the whole industry up.
Was thoroughly confused which article this comment belonged to for a second.
And your basic supply and demand curve indicates that, at least until the supply is diminished in kind which isn't going to happen immediately given the all the constraints at play, using less coal leads to coal becoming cheaper and more readily available.
> This seems like a basic oversight
Not really. Improvements to the steam engine being why we got "cheaper and faster at producing coal" is immaterial. The added detail you've given is an interesting aside, I suppose, but doesn't change anything about the original premise. The only oversight is you not realizing this, perhaps?
It's completely obvious that if you need energy, and you have energy source A and B (lets say natural gas and oil), people will use the one that is cheaper.
Oil became cheaper because a new supplier entered the market, and people started using more of it! Jevons Paradox!
The other explanation is that people have an outsize demand for a resource, and are actively making efforts to make it cheaper so they can use more of it, then when it gets cheaper, they use more of it.
Transistors got cheaper and we are using more of them! Jevons Paradox!
You can just as well argue that labor is getting more expensive in the West because of two non-market pressures. First, we have a multitude of government programs that seek to eliminate extreme poverty, so there are fewer people who are desperate to take any job for any money. Second, you have consumer protection policies that make it genuinely expensive to, say, be a HVAC repairman. Educational requirements, permits, licensing, business insurance, waste disposal, etc.
On my neighborhood Facebook group, every time someone asks for recommendations for menial, minimally-skilled backyard labor, they always insist the person needs to be licensed, bonded, and insured. And then, they're surprised that it costs $10,000 to paint a fence.
I don't think this opinion holds a rational basis.
Extreme poverty being a factor in low job demand is an argument for coercing people into performing certain tasks even though they are not economically viable just because it benefits you personally. This is not a valid argument, neither indentured servitude or slavery. Isn't the US supposed to be a free market economy where free Enterprise reigned?
Complaining about regulation, including waste disposal, is also dumbfounding. Being required to dispose of air filters in a landfill is not the reason why you can't afford a repair. This opinion is also comical as HVAC also covers air quality because otherwise you can be cool in a room but literally sick.
This sort of opinion sounds completely irrational and unsubstantiated, and extremely ideological.
The main factors driving repair cost are things like device longevity, unit price, speed and ease of repair, parts availability, etc. That's mainly it. When you call someone to your house to repair something, the price tag covers that person's cost of living for the fraction of the time it takes them to deal with your problems. On top of that, you need to pay whatever parts they need to buy to get your things back to work. That's where the money goes.
It is for someone - the market price is pretty much always going to be around the point where a small increase causes a noticeable drop-off in customers (otherwise the only sensible thing for the seller to do is charge more). If something causes even a relatively small increase in price will mean someone can't afford the thing any more.
*https://documents.ncsl.org/wwwncsl/Labor/NCSL_DOL_Report_05_...
Regulation is often necessary, but it has a cost even though it's necessary.
If my country's regulations require nurseries to have one staff member for every three young children, there might be good safety reasons for that - but I'm going to have to spend a third of my salary to have one young child cared for.
Given that as soon as cloud computing happened we stopped bothering to debug VMs and just started deleting them and rebuilding, I don't know why people find the idea this applies to other industries surprising.
Repair involves establishing where in a very large state space an item is, and finding a path back to optimal.
Whereas building a new item simply involves traversing an already known path to optimal.
clarification: Of the things that make it expensive to run an HVAC repair company, consumer protection related expenses are super far down the list.
source: 2 decades (on/off) supporting a close friend's regional hvac biz, national geothermal interests.
I understand large installs at businesses are a different problem, and granted I've only ever installed a mini split, but that was hardly rocket science. And home installs are likely what most people are thinking of here.
In Japan you can get minisplit's installed for $1k a unit, here you regularly find quotes over $10k. Something's gone wrong somewhere.
I bought the 3.5KW unit online for $1080 including delivery (USD$702) so it was $1620 (US$1050) all up. I expect with the recent inflation it might cost more like $2000 (US$1300) installed for one that size and maybe $3500 (US$2275) for a bigger unit (8kW).
The splits themselves are mostly all Japanese brands that we have here (Mitsubishi Electric, Daiken, Panasonic, etc.) as well as some Korean (LG, Samsung), but Chinese ones are starting to appear in the market too. But they all seem very cheap compared to buying one in the US, before the installation there which just seems astronomically expensive to us.
A flat time mandate for HVAC tech certification seems really out of place. And a 4 year path of any sort seems excessive for a technician. I couldn't find anything like that. Most results I found were in the 6-12 mos range - which is often spent employed.
WV was an outlier with a 2000hr requirement. How I have seen (non-hvac) 2k requirements get satisfied are thru a HS VoTech (my son) or 18-24mos doing paid tech work toward the official certification (electrician techs do this).
I can't find a state that requires anything a like a 4yr college degree, where life is put on hold to focus on that. And then 4yrs of living and school expenses are investments that need to be earned back. Not for any trade tech.
Its not excessive if the purpose is job protection.
See also any career that involves interning (law, accounting, …)
How much does the unit cost? What work does it take to install it? How large does it need to be to support your home?
What are the energy needs of your typical home in Japan vs your home town?
Those are the key factors, not how many years someone spends in tradeschool.
One way to eliminate extreme poverty and increase skilled labour, is to ensure children have enough nutrition, health, schooling, and funds to pursue skilled worked in adult life, which usually involves making time outside of work, adequate health, some prerequisite education, and a sufficient financial buffer to actually upskill.
A second thing is Baumol's Cost Disease. If there are other industries that are more rewarding that require less effort, no-one is going to pursue those options. Why be a clever guy who makes a product (or becomes a HVAC repairer, electrician, etc.) when you can get a comfy job at a FAANG (or whatever the acronym is today). You could sub in benefits here, but I don't think people are thriving on benefits in the US. But I'm an outsider, so I wouldn't know.
But there's also a secret third thing that people don't often consider, which is the culture. If there's a culture that doesn't privilege working hard, or educational attainment, etc., people won't seek those things.
How's what you wrote substantiated at all?
To me it seems you are suggesting we let people starve so that labour you characterize as "minimally-skilled backyard" is cheap.
You also seem to be suggesting that consumer protection policies need to go as well. I assume we are going to trust the end consumer to do due diligence cause "they know whats better for them?"
If they get a quote of 10k and they cannot get a better one, they might as well start writing that check.
Perhaps the question is why wages are not high enough to support these prices (globalization, productivity wage gap over the last fifty years, etc). This will change over time due to structural demographics [1] making labor much more scarce (pushing up wages), we’re still in the early days. Software is not going to eat the trades and HVAC repair.
[1] https://www.sas.upenn.edu/~jesusfv/Slides_London.pdf
It can be true that consumer protection laws raise the price floor for certain goods and services without "and therefore consumer protection laws are bad".
The US is already experiencing rampant extreme poverty. There are people in the US holding multiple jobs and still can't afford to eat, let alone healthcare.
Again, this argument that things are expensive because the poor can't work and regulation somehow is suffocating businesses is purely ideological and not supported by facts.
The amount of waste that is generated is 1000 times that of just refilling the coolant. When will people realize that you can use an existential threat that you can’t prove to justify anything? What could be more important than our existence?
If the same scientist came out with a study that said if you don’t pay me $1 million by tomorrow, we are gonna get hit by an asteroid. Would you believe them and pay me? Or has this become a political issue we’re no longer thinking rationally
I don't think you are holding an informed opinion. Ozone layer depletion was tied primarily with CFC emissions used in air conditioning units, and since the production of CFCs and other ozone-depleting chemicals were banned in the late 1980s the ozone layer started to regenerate. In practical terms this means lower incidence of health issues such as skin cancer or cataracts. That sounds pretty neat.
But being able to use a 40+ year air conditioning unit is worth it?
Clarification: In some places we absolutely do. In others we absolutely do not.
> so there are fewer people who are desperate to take any job for any money.
I think is is better reflective: There is huge surplus of employers have that set up systems that insure they do not hire qualified people. Job portals that auto-trash applications from unwanted applicants (1st time, most minimal of crim rec, wrong zip codes) are one massive example.
source: me+kids spent a decade in red state, hunger-level poverty. kids who got zero replies during months/years of entry level job apps.
Paint a fence? Pay the neighbor kid. Patch a roof? eh ... what could go wrong.
Yeah it's CYA: the ass in question is yours. The guy who burns your house down without insurance will just file bankruptcy. You're the one left with the ashes.
> In 2025, American diagnostic radiology residency programs offered a record 1,208 positions across all radiology specialties, a four percent increase from 2024, and the field’s vacancy rates are at all-time highs. In 2025, radiology was the second-highest-paid medical specialty in the country, with an average income of $520,000, over 48 percent higher than the average salary in 2015.
Simply put, radiologists do a lot more than merely read scans:
> Radiologists are useful for more than reading scans; a study that followed staff radiologists in three different hospitals in 2012 found that only 36 percent of their time was dedicated to direct image interpretation.
Source: https://worksinprogress.co/issue/the-algorithm-will-see-you-...
Nah, pre-LLM, I think the obvious fields to pick on were lawyers, middle-management, and "email jobs" generally. That was a big miscalculation, since most people (especially engineers) do not understand the politics of power. Those jobs tend to jealously protect the power that they have and systematically dismantle what accountability they might be subjected to. Engineers in general are much more likely to democratize (and thus threaten) power, by creating things like accountability via metrics, at the same time as they mostly refuse to unionize. Radiologists have some unions, medicine generally enjoys a moat of credentials and certification. Things that SWEs in particular rejected while they said "come on over, anyone can code". I doubt radiologists ever suggested themselves that they should be measured on throughput, but SWEs actually did push ideas of 10x engineers and metrics like lines-of-code for years to argue they are productive enough to deserve raises.
No AI replacement is going to be doing that anytime soon.
I'm not arguing with the GP's point that radiologists don't do many other things that the AI maybe can't do, but it feels like your example is the opposite of that.
And not only that, your example demonstrates a failing of the human's limited amount of time to get all their work done.
Think of it like 'commoditize your complements'.
The shop recommended replacing it, as just the refrigerant is half the price of a new one if most has to be replaced, due to environmental taxes.
Basically a repair is guaranteed to be at least 50% of new price of NOK 23k ($2.3k) installed and can quickly approach 80+% if the guy has to spend time on it.
That made me wonder how on earth they make money on selling these things. And how effective that tax was, given it's pushing us to replace a unit that's probably fine for many more years with just a bit of TLC.
edit: fixed pricing brain fart
Please get a second opinion, especially if you can find a non-shop to give you one.
I am inclined to agree. The rear 2½T unit in our rental blew a line last year. Two companies quoted my landlord $10k for a new 2½T condenser and air handler (1 was another renter of his). Even that seemed high to me but I could be out of date
In the end, I had a friend come out and look it over. Leak was near the compressor and he charged well under $1k for the fix, mostly for refrigerant.
That is because it's wrong. Sorry, just woke up. I'm in Norway, and it's 23k NOK, or $2300.
The shop said filling my system back up with refrigerant if it was empty would be around $1.3k, and based on what I've heard I don't think that's off by much.
23k didn’t sound too far off what an American installer would try charge for a new system in a high cost of living area, so this wasn’t too unbelievable.
> 23k didn’t sound too far off what an American installer would try charge for a new system in a high cost of living area
I've heard these insane prices. I think 2.3k is pricey enough, you can get cheaper units here but SO wants the pretty indoor unit so pretty indoor unit she gets.
At least in the US, refrigerant costs have been high because of a shortage, not taxes.
https://www.coolingpost.com/world-news/us-ac-companies-move-...
The leaky A/C unit had been made by Lennox while the new heat pump was made by Fujitsu. I very much hope that Fujitsu engineered its heat pump to last. The heat pump had also replaced an oil heating system that was around 25 years old and still could have been used for many more years. Expecting similar or better longevity out of a heat pump does not seem unreasonable.
Yeah labor costs also make a lot of repairs uneconomical. There's been talk here about removing the 25% VAT on repairs to make the value proposition a bit better, but doesn't seem to have much traction currently.
> At least in the US, refrigerant costs have been high because of a shortage, not taxes.
The guy said it was due to taxes, specifically that they had gone up so much in recent years. Seems it's because the tax is tied to CO2 tax[1], which has been going up since they introduced it in 2020. Not sure what refrigerant they use in my minisplit, but even if it's one of the cheaper one the tax is about $90 per kg, so adds up quick.
edit: mine uses R-32, so yeah about $90/kg.
[1]: https://www.vke.no/artikler/2024/okning-avgift-hfk/
I know the US and Vietnamese economies are very different, but something doesn't add up there.
In my home, I recently had a heat pump unit replace my central A/C with some minisplits that connected to the exterior unit installed in the basement. The entire setup cost as much as it originally cost to install the central A/C, despite parts of the central A/C being reused.
Note that in the US, what we call air conditioners only support cooling and not heating. When they support both, we call them heat pumps despite that being the scientific name which applies to the cooling only units too.
When I installed AC in my former Bay Area home, I would have needed multiple mini split units to cover 1200sq ft, with questions about how many you could have on at once to still get the right performance. I went with a single central air unit instead.
It also (at the time at least) didn't come with a centralized thermostat, which meant managing each room individually (that would've been fine with me, personally, but it's a drawback for lots of people).
On top of that, many (if not most) mini split units are also somewhat aesthetically displeasing.
In my new home (on a different continent), I have mini splits. I'm somewhat satisfied with them, at best. I'd still prefer central air but it's not a thing for residential homes here.
That said, I was replying to someone from Vietnam. Assuming that things in Vietnam are similar to China and Japan, people will only heat or cool the specific rooms that they are using, rather than heating or cooling everything like how many Americans do things. Those in the US who cannot afford to heat or cool everything, who are likely very underrepresented here, would be those using window units, since they are cheap upfront. A minisplit would be cheaper over the long term, but the high upfront cost dissuades people in thing market from even looking at them.
Finally, I had Fujitsu minisplit units installed in my basement two years ago. They are far more aesthetically pleasing than window A/C units.
The energy (gas+electric) bill for my whole home would've taken well over a decade to add up the amount the mini split installation alone would've cost, so even if it brought down my energy cost to zero (which it wouldn't), it would've taken a decade to pay for itself (including the cost of window units).
Edit: and yes, they're prettier than window units for sure, but as you've pointed out, they're not competing with window units. They're competing with ducted central air that has almost no visible impact.
The refrigerant is often expensive, poisonous and flamable. We don't replace the refrigerant, we reclaim it and dispose of it properly. Its not without risks. Our building codes are pretty rigid as well, its a big pain point.
New HVAC install is gonna need an electrician, the lines pass through walls allowing the possibility of condensation to produce deadly mold. Every installation needs individual consideration. If it isnt %100 perfect, the customer will be riding the installer's ass.
The US has is a Safety Above All mindset on some things. We improve safety far beyond economic rationality because we don't want to systematically kill each other in this way, however few that is.
Unrelated example: a national EMT outfit operating here made all their techs wear plate carriers and ballistic plates. There is nearly no gun violence here. All the techs stopped carrying the plates after a few months because its dumb, and wont stop a baseball bat or a knife, but this is a national outfit, surely they did a cost benefit analysis. Someone signed the check and wrote a policy to make them wear em every day. Plates expire, also.
That said, we can definitely do better, and the cost is too high. Installers are in demand, and we tariff the imports to the tune of %15-%30.
... Why does someone in Norway need a split AC unit? I don't think they're very efficient for heating.
EDIT: I researched this and it seems the type designed for heating in negative zero temperatures are much more expensive. So maybe that explains it?
Or making those things.
I think it could make sense if that $2.3k mostly only accounted for Western labor involved rather than representing entirety of "real" costs. Like - just entirely making up - 1k to ship it from middle of the Pacific to nearby ports to you, 0.5k from the port to location of installation, another 0.5k to install, 0.3k for documents preparations and extra wood screws, 2.3k total. And $25k worth of man-hours in CNY to manufacture, which effect is isolated and contained within their own economic bubble.
I can't just believe in the "them subsidization" theory anymore, international prices just don't seem like compensating for anything. It feels like paying for energy costs of delivery systems rather than the product. It just doesn't make sense.
I can build a house 4x faster today than my father could back in the day, or alternatively with 4x less labor, and yet he could buy a house, 2 cars, and support 3 kids on his single wage. Meanwhile I struggle to do more than just survive in my little shack house built from scrap on an empty plot doing the same thing.
The only jobs that earn me a decent wage is for corporations or for people near or in the top 1% of wage earners.
If the average person made more money I would have more work and more money. But as it is a lot of residential work would be of negative value to me at the prices people can afford, and with less work to do and farther and farther travel distances for it my prices must keep rising to stay solvent.
Any slum in developing nations features lots of housing that people bang together out of trash. Very unsafe. But it works. We've been creating shelters for ourselves since prehistory. It's not that hard.
Millions of people live in slums. Upgrading that kind of housing to something slightly better isn't all that hard. Most people that a are a bit skilled can bang together a shed in their garden in no time at all. Putting a few solar panels on top isn't that hard either. And you can plug them into batteries easily. You can buy that kit on Amazon and run your power tools of that. All legal.
It's only when you want that in your house connected to the grid that cost suddenly balloons from a few thousand to many tens of thousands. Exact same technology. Maybe you'll use slightly more panels and a bit bigger battery. But now it's a lot of gate keeping by inspectors, electricians, certified equipment, etc. that come into play.
Same with houses. You can buy a recreational vehicle or caravan for a reasonable amount of money. Second hand these are very affordable. And some RVs can be quite nice to live in and even have AC. So, why are houses so expensive relative to RVs? Prefab housing has been a thing for decades. If you remove the wheels from an RV, it's basically a house. If you live in an RV, you are referred to as trailer trash. It has a stigma. But it's very cheap. Poor people do that. Because it's very affordable housing. That's why it's a popular option for people that would otherwise build slums.
"Reasonably serviceable" — ay, there's the rub!
The U.S. Navy's nuclear propulsion program has a saying: You get what you INspect, not what you EXpect. Certifications, inspections, etc., are meant to try to keep lazy-, sloppy-, and/or dishonest providers from passing off shoddy (or even unacceptable) work product.
In many areas, most customers don't have — and shouldn't have to have — the expertise or the time needed to inspect and assess providers' work product. By requiring knowledgeable inspections in critical areas, we increase the cost a bit, but we improve safety for everyone — which lets us non-experts spend our time more usefully. (It's a form of division of labor.)
To be sure: Certifications, etc., can be captured by industry groups and used to limit competition and increase prices. That's a separate issue, one that can be dealt with in other ways.
Quality — and quality assurance — aren't free.
A majority of mini splits are made in China and are inexpensive and reliable because they're so pervasive in Asia. Most are rebadged Midea or Gree.
You can get a decent mini split for <1.5K and install it yourself for $200 in tools.
The truth is probably more that the various money sinks in our society are starting to add up, things like healthcare, legal protection, licensure, compliance, rent (business or personal), even just having appropriate work vehicles, fueling them, compensating people for the time spent sitting in traffic to come across town to your house. Somehow you’re paying for all of that when someone’s livelihood is installing your mini split. A lot of those costs have grown faster than wages, if you try to point to a reason why it’s different today than 20 years ago. More people looking to make a quick buck without doing any work or providing any real value, and more people succeeding.
My experience is that it’s not generally well understood how simple it is to install mini splits. The supply companies won’t sell to you directly outside of d2c web companies like hvacdirect
Getting a similar system installed would have been north of $10,000, and before anyone says "well, that would be a licenced HVAC installer", no it wouldn't - it would be a barely-trained person who is simply "supervised" by a licenced HVAC technician.
If the AC catches fire because your electrician skills are bad, what happens? I guess you can rent a ladder if you need one, but they're at least $200 if your split is on the second floor and ladders can be deceptively tricky, and load ratings must be considered. Condensation can kill you and be an extreme cost with mold. Your first mini split is going to take a real long time to install, I promise, assuming you size it right. There is a non trivial risk to life and limb.
This is one of those "Reality has a surprising amount of detail" things.
The skill involved is that of tightening screws on screw terminals.
I just looked it up, and I can buy a heat pump for 200-400 euros (depending on desired output), installation is ~400 euros. Why are you paying 20-30x for something identical? This sounds like a price difference created by government behavior, like with solar panels and related hardware which seem to be significantly overpriced in north america.
It's a price difference created by market segmentation of heat pumps as a luxury product in the US, and the relative lack of qualified installers due to our under-investment in education in the trades.
Example of a solar install that was under $20k via DIY - maybe took this guy 1-2 weeks of full-time work. But he was quoted $90k and he did most of it himself - only hiring a backhoe operator to move some dirt around.
We're at the point where the trades are only going to cater to rich/desperate people because the margin they'd make on a job that is DIY-able (and charged a fair price) is not worth it to them. Why do 5-6 low margin jobs when you can do 1 high margin (rip-off) job? Your only competition out there is someone with the will to do it. Builders in CA are massive ripoffs as well.
I think the only way you bring costs from trades down is by having all your workers in-house - which is only doable for corporations. Your average homeowner is just fucked and is gonna have to youtube everything.
This is my gut feeling too. I've known so many rich people who just accept whatever number a tradesperson quotes. There is no way that hasn't had a mass effect
All of the major trade companies around where I live (HVAC, plumbing, electrical etc.) in the US have rates that they quote before the person will even show up. As a new homeowner who didn't grow up in the US, that's all I've ever dealt with.
If the answer is “Give them a number you're comfortable with, and just DIY it as an alternative” -- that's fine, and I do it for anything simple; but for the remaining ones, I have already made a determination that learning this skill would be way more in terms of time invested than the $500 or whatever absurd number they are quoting for a simple repair (this logic likely breaks down over time, and I'm trying to invest more time into learning more house repairs).
I have tried pre-purchasing some parts in the past; and asking them to use them for the install -- that one had some success and a guy told me how much his company marks up parts (n00%).
I do try to get multiple quotes for something, but the difference between them isn't usually appreciable; they're all absurdly high. I've tried to ask them for a parts v/s labor breakdown in the past; some won't even provide that.
Buying your own parts can help but can also burn you if its wrong, or cost the same if something extra is needed that wasnt expected and requires a second visit or bought on short notice from a local parts dealer. A contractor often eats the costs of wrong parts they ordered and just hopes they can use it elsewhere later, but if you bought the parts that is just cost on you.
My recommendation, which is still probably of limited help and won't always be worth it, is to start by hiring a local handyman instead of a specialist and having atleast 2 weeks of lead time for parts. Of course finding a worthwhile handyman can have its own difficulties because so many tradesmen leave the industry after realizing corporate contracting pays workers like trash while taking a lot of the most valuable and worthwhile contracting work off the market from independent contractors.
The easiest way to find a good tradesman is to ask another tradesman. There is an HVAC person in my local area who will come out and do most jobs (such as, for example, moving an AC) for about $500. A PE rollup firm would quote $10k for such jobs.
Why even bother when you could just work for a mass build and plaster up hundreds of walls in a single job on a new apartment building or housing development.
So as an individual it's almost impossible to get someone over to do a small job, and your only realistic option is to do it yourself.
Robotics will reduce back injury rates and so reduce Nurse shortages, for example. Misdiagnosis rates cause increases in compensation claims so improvement in diagnostics reduce risk so insurance premiums(for hospitals) should go down or at least not rise as quickly compared to inflation.
Note that the cost of health is high even in the economies without the broken US health system, and we have the baby boom moving through which is a huge bubble of cost associated with age.
Not that there's no component of Jevons in this, but it doesn't explain everything.
A thing said to me in Argentina resonated: They pay doctors, nurses, pharmacists slightly more if they will do out-calls to elderly people in the community at large. The increase in staff cost to have home calls offsets the massive increase in care costs if these people cannot be cared for in their home and move into a managed care facility. It's partly an externality if you privatise healthcare: who cares if the state has to pick up the tab, right? But in a more integrated view of the costs here, its better to pay more for people to help keep the elderly in their own home as long as possible.
(elderly, intending to stay in my own home, not in the US health system)
The author is being (intentionally) naive here.
History and current research suggest that when technology automates the vast majority of a complex job, it can lead to the "deskilling" of the human worker.
This Lancet article (published Oct 2025) discovered that doctors were found to be less adept at finding precancerous growths during colonoscopies after just three months of using an AI tool designed to spot them: https://www.thelancet.com/journals/langas/article/PIIS2468-1...
And the thing is, I think the above finding is pretty intuitive.
If AI performs 99% of a radiologist's diagnostic work, the human's role will very likely be reduced to a skill that is more routine, that requires less expertise, like a final check or something - and thus commands lower wages.
I've seen a lot of people blow a lot of money on really basic stuff like this (myself included). The lack of basic awareness around hvac, landscaping, drainage, drywall, plumbing, electrical, et. al. has me wondering if I'm still in the right business. ChatGPT can't carry a bag of sand up a hill or dig a ditch. It can tell you about these things and make you feel like a hypothetical god over them, but it can never do the actual work on site. I don't feel like there's a lot of competition around being in Texas crawl spaces during the summer.
Is this not just inflation? If everyone got paid more and everything got more expensive, are we not essentially level?
Many things can get cheaper, some things get more expensive, and median person gets more wealthy and buys more of both.
In reality, yes, inflation plays a role in this, but the article is pointing to other patterns layered on top of it.
On average, over long term, despite inflation, people can afford way more good and services.
While this is true, the costs are inflated because you need to repaint the entire room to get the original look, rather than only pay the cost of merely replacing the drywall. Of course, some handymen are much more expensive than others, so it is possible that is more expensive too.
If you are one of the few using wallpaper and have extra wallpaper for just such emergencies, using the extra wallpaper to paper over it should be cheap.
A ten-fold productivity gain in legal services sounds simply awful for society. Imagine the time and money sink if everyone can sue you for every frivolous thing because AI can prepare and file the paperwork instantly without needing a lawyer. You'll need your own AI to defend against the onslaught of legal disputes.
Every contract for jobs and every terms & conditions for services will be 10x longer because AI has a much higher complexity threshold compared to a human. My belief is that one reason tax returns became much more complicated in the last ~30 years is because of tax preparation software. In the era of paper tax returns, there was a limit to the complexity that an individual or even an accountant could handle, so there was a limit on how complicated the government could make it.
Most normal people rarely need a lawyer in their lives. With AI's productivity explosion in the legal services, you're going to need legal services every day. Your neighbor wants to borrow your chainsaw? Your AI legal agent will negotiate a liability waiver with his AI agent.
I find the term “cost disease” very negative. It actually describes positive social progress.
Instead of a “disease,” it’s really a sign of a healthy, advanced economy.
Everyone complains that it is $7 for a cup of coffee, and yet demand for barrista coffee keeps increasing and demand for barristas keeps increasing.
But, nonetheless, those jobs will probably disappear, and the machine will reorganize itself to maximize its self-legibility, making tons more money but becoming yet more shitty and inhumane in the process. Discarding the whole reason the jobs exist in the first place (stuff like Value and Service — oh excuse me, sorry, meaningless cost centers to be optimized).
Sort of like how Radiologists do all kinds of important shit, that gets entirely hand waved away in this think-piece as 99% automatable. Yeah, sure guys, the radiologist — actual doctors for fucks sake — I’m sure he is nothing but a warm body between the patient and the computer, signing paperwork and collecting cheques.
It depends how you measure. The cost in goods or purchasing power goes up, but the cost in hours stays the same.
Until AI becomes physically embodied, that would mean all high-mix, low-volume physical labor is likely going to become a lot more valuable in the mid-term.
Cost of goods goes down - think factory automation improving line rate.
There is less human intervention, but that intervention requires more expertise.
Sure, that's an assertion.
But (with just as many citations), mine would be:
This boom is absolutely, 100%, fueled by the combined factors of: 1) employees outsourcing the cognitive load of their jobs to models that are, impressively "close", but not quite _as_ good as a well-trained human.
ie, we're replacing google with a fun, but terribly energy-wasteful (and _very often_ factually wrong) "make up an answer" tech.
and 2), AI "app developers" who are having fun with the previously "impossible" (*cf. https://xkcd.com/1425/) APIs of multi-modal natural language, and "didn't sci-fi warn us about this?" simulations.
Neither of which are good for productivity, if we measure productivity as "improving circumstances for the mutual commonwealth of all life". Which is the goal.
* oh, I _did_ use a citation after all.
It is an interesting article, but _far_ too sure of itself in all the wrong areas.
This is mostly down to people being afraid of anything even remotely trades-like. Learn to do some basic home repair, it will save you thousands.
> This graph can mean different things to different people: it can mean “what’s regulated versus what isn’t” to some, “where technology makes a difference”
Cars are pretty heavily regulated...
What I see is what is necessary to live and what isn't. Elastic vs inelastic demand.
> the average American middle-class household can comfortably manage a new car lease every two years
Huh, no, the average American middle-class household cannot do this.
> If one sector becomes hugely productive, and creates tons of well-paying jobs, then every other sector’s wages eventually have to rise, in order for their jobs to remain attractive for anyone.
I'm sorry, but anyone who has lived in the lower income brackets knows this just isn't true.
This is hard to read. Whoever wrote this is extremely out of touch and thinks they're eminently intelligent. It reminds me of the "smug San Francisco" South Park episode. The world is going down a road of hurt and you've got elites who are so busy "winning" over the past 50 years running around sniffing their own farts.
Installing a new HVAC system is not "basic home repair".
Yes, there are HVAC-related repairs that qualify as basic, but we're also talking about the big things.
And while yes, many homeowners could learn how to install a new heat pump, run refrigerant lines, make sure every connection is torqued properly, etc., most would not want to or have the time to do so, and that's fine, normal, and expected.
That’s funny I thought the exact same thing reading your comment.
"The cost to repair an AC in Hong Kong varies by the problem, but basic repairs can range from approximately HKD $500 to $1,000, while major repairs like a compressor or refrigerant leak can cost HKD $1,500 to $4,000 or more. Inspection fees are often around HKD $500, with the repair costs and any needed parts being quoted separately. "
Basic repair US$64 - $129 Major repair $193 - $515 Inspection $64
Somehow, American HVAC prices are orders of magnitude higher than that.