11 ms·

The whole article rests on this false economic claim, that traditional industries don’t overhire based on expectations. They absolutely do, ALL THE TIME. Manufa

The whole article rests on this false economic claim, that traditional industries don’t overhire based on expectations. They absolutely do, ALL THE TIME. Manufacturing, automotive, airlines, energy etc. all of them make demand bets and lay people off when those bets fail. Cheap money amplified this cycle, but this isn’t a tech specific "failure", it’s just how forecasting under uncertainty work. It’s incredible how some engineers assume they understand economics, then proceed to fail on some of its most basic premises. This tends to happen when engineering-style certainty is applied to systems that are driven by incentives and uncertainty.

Bayramovanar 8mo ago
I will avoid getting into a “who understands economics better” debate. But factory workers usually require specialized machinery, tooling, and physical capacity, which makes overhiring slower, harder and more constrained. Those investments force more deliberate planning. By contrast, engineers mostly require a laptop and company hoodie... That low marginal cost makes it far easier to hire aggressively on expectations and unwind just as aggressively when those expectations change.
- thunky 8mo ago
  This, and I would also assume it's easier to spot idle factory employees, where idle devs (especially remote ones) can be easier to miss. And, (assumption again) the factory boss doesn't have an incentive to increase idle worker numbers, where a dev manager often benefits from being in charge of a larger number of hardly working people.
  - silvestrov 8mo ago
    It is a lot easier for a software engineer to overcomplicate things. All of a sudden it is no longer enough with a few shell scripts. No, we need a full kubernetes cluster to run a service used by 10 secretaries. No, we can't just use PostgreSQL as a queue, we definitely need Apache Kafka for 1 msg per second.
- blenderob 8mo ago
  > By contrast, engineers mostly require a laptop and company hoodie. Alas, gone are the days when engineers too required specialized equipment like a desktop computer on the desk that you couldn't move with you. Every evening, you left it at office and went home to live a 100% home life. Alas, gone are those days.
  - zrail 8mo ago
    Maybe for some. I've worked from home for 15 years and a huge thing that I've learned is that I have to have a hard physical boundary. My work laptop stays at my desk unless I'm on call and actively fighting a fire. When I want to use my desk for non-work things the work laptop gets put away.
  - QuiEgo 8mo ago
    Crunch time in those days sucked. I remember mandatory nights and weekends and the managers ordering in pizza for everyone.
  - bitwize 8mo ago
    I was a contractor for a FAANG. My immediate cubicle neighbor liked to work out on lunch break, and to hang his sweaty, smelly gym things on the framework of his desk to air dry when he returned. I would've KILLED for a laptop I could take to the complimentary office café so I could get something done without holding my nose. If I were a full employee, I could just ask for one. Alas, as a mere contractor I was something less than a person, so I had to remain tethered to my desktop per company policy.
    - steve1977 8mo ago
      I mean, if killing was an option, getting rid of the cubicle neighbor would have solved the problem as well.
- brk 8mo ago
  Factory overhiring can happen easily by adding speculative 2nd and 3rd shifts. No need for extra equipment, tooling, or physical capacity. Lines with specialty equipment and tooling can also often be sped up. That can allow for other jobs to be added to all the functions that support the processes involved before and after the specialty equipment. New employees also often require training and some apprenticeship time, meaning they can get hired ahead of actual demand.
  - Bayramovanar 8mo ago
    I am really not debating whether over hiring is possible in factories. in tech cost of hiring is lower which makes headcount a much easier speculative bet and layoffs a much easier reset when the bet fails.
    - rpdillon 8mo ago
      The 90th percentile of factory workers makes $45,000. The 90th percentile of software engineers makes $230,000. Hiring in tech is insanely expensive; I'm not familiar with hiring in factories, but it can't possibly have the supply problems that tech workers have. I do on average 115 interviews for every person I hire in tech.
      - paulnpace 8mo ago
        That $45,000 factory worker is often operating a $1.5 million piece of equipment with $10-50,000/hr of materials going through it. It's an entirely different cost structure.
  - paulnpace 8mo ago
    There exists greater issues in hiring for manufacturing when it is for new shifts. Master level technicians and foremen willing to work those hours can be exceptionally difficult to find, and everything flows out of these people. While similar issues likely exist relating to discovering talent for software development, I speculate that the factory will, in practice, have a harder time finding people (for new shifts). My experience with seeing new shifts added is initially with only specific processes, and even with those it is with journeyman level technicians running a small crew to support relieving a bottleneck in production. Alternatively, manufacturers can outsource until they have enough volume to add a shift, but across the economy the net is just transferring production from one facility to another.
- Propelloni 8mo ago
  This only holds for companies that do not have to comply with some regulation or standard, e.g. ISO 27001, to do business. Especially infrastructure, banking, and defense tech have high compliance requirements that also, and sometimes esp. cover software development.
- AnimalMuppet 8mo ago
  What's the cost of a programmer-quality laptop, with two large monitors, plus licenses for all the software that they need? That's 2-3K. Does anyone know what the typical tool cost is for a factory worker? (And the tool cost for a factory worker can be zero if you're hiring for a second shift - they would use the same tools as the first shift, just at different times. In contrast, I don't think there are very many places that ask programmers to use the same laptop in shifts.)
  - arter45 8mo ago
    >Does anyone know what the typical tool cost is for a factory worker? It's probably lower. Apart from safety gear like helmets and gloves, everything else is most likely reused across shifts or shared across workers simultaneously (think industrial machines).
  - Bayramovanar 8mo ago
    they even cut costs by leasing the computers...
    - gdmka 8mo ago
      I'd add that companies write off their equipment investment over a 3-5 year depreciation period. After that, when the laptops reach EOL, employees can often buy them through the company's EPP (employee purchase program) usually for just the fair market value, which might be a couple hundred € or sometimes even less. So, eventually they recoup the equipment costs, wouldn't get surprised if they even make a profit out of it sometimes. First thing they do, they write off VAT (20-23% of laptop price). Then at the EOL, they just set the EPP price to the higher market value. And return the cost.
      - ThrowawayR2 8mo ago
        Is that common? I've never worked at a corporation that had an EPP for EoL computer equipment. It always all went to a specialist recycling/refurbishing business.
- re-thc 8mo ago
  > By contrast, engineers mostly require a laptop and company hoodie... > That low marginal cost Not true. That's how everything falls apart. Scaling software teams isn't simple. There's lots to account for. Do you just assume everyone commit to the same file and let it crash? The more people, the more work is required to standardize, organize, document and fix gaps. You can say you're already at scale and things "should" be done but they never are. You hire a team into a specific area to find it was a giant hack supported by a part-timer and as you try to fix it the problems keep escalating.
- aleph_minus_one 8mo ago
  > But factory workers usually require specialized machinery, tooling, and physical capacity, which makes overhiring slower, harder and more constrained. Those investments force more deliberate planning. > By contrast, engineers mostly require a laptop and company hoodie... That low marginal cost makes it far easier to hire aggressively on expectations and unwind just as aggressively when those expectations change. Software engineers also need - specialized machinery (at least when they have to upload to some computation cluster or cloud), think for example of the costs for GPU/TPU clusters for AIs at the moment - tooling: depending on the sector, the license costs for the sector-specific business software can be similar as expensive as specialized machinery - mental capacity (instead of physical capacity)
- _m_p 8mo ago
  The low marginal cost also means that the job market is easier for software engineers in times when expectations are optimistic.
Imustaskforhelp 8mo ago
> It’s incredible how some engineers assume they understand economics, then proceed to fail on some of its most basic premises. This tends to happen when engineering-style certainty is applied to systems that are driven by incentives and uncertainty. Dunning Kruger effect, am I right? We consider we are smart because we can make computers go beep boop so we know about the economy too. I mean, I am part of this too even though I (or we all) know the effect but I guess my point is that there should be an humility if we are wrong. I can be wrong, I usually am. If someone corrects me, I would try to hopefully learn from that. But let's see how the author of this post responds to the GP's (valid, from what I can tell) critique. Edit: Looks like they have already responded (before I wrote it but I forgot to see their comment where they said that its not at the scale or frequency we see in tech)
- thunky 8mo ago
  > We consider we are smart because we can make computers go beep boop so we know about the economy too Funny, the person you are agreeing with made the strongest "I know more than you" flex than anyone.
- boh144 8mo ago
  The scary part is when people who can make beep boop, and because the market has rewarded them for it, think they can skip the entire corpus of the humanities yet put forth opinions on how to shape society.
rcxdude 8mo ago
There's also this tech exceptionalism/grass-is-greener bias where software engineers especially tend to romanticize other industries and professions.
lumost 8mo ago
Tech companies also benefit from over hiring. It gives them slack to absorb crises, and Fast Fuel to move quickly on new growth. Eliminating the fear of layoffs allows employees to take more risks and explore. The current crop of tech companies cutting staff is going to lead to a large number of dead giants. The staff who services the layoffs will be risk averse, and defensible in a job cut situation. You see this in legacy firms where it takes 10 people to make a change because each person has a small slice of permissions required to effect the change. This pattern is by design as laying of any of the ten people on different teams would kill dozens of critical business processes. This is not how you make high growth firms.
lenerdenator 8mo ago
> Cheap money amplified this cycle, This is more-or-less the rub. Fortunately we had a kinda-sorta sane monetary policy under the Biden administration once the pandemic started to ebb, but now we've got Mr. Appearances trying to push Jerry Powell to make stupid mistakes.
WalterBright 8mo ago
> It’s incredible how some engineers assume they understand economics I would say most engineers. The reason is simple - basic economics is not taught in the public schools, and economics/business is not a required course for an engineering degree. One of the best classes I ever took was a summer class in accounting.
JKCalhoun 8mo ago
My take is that the traditional industries you mentioned (manufacturing, automotive, etc.) are making expectations on the market for what it is they're selling. With tech though, it makes expectations on markets that are brand new, or don't exist, and/or have no proven track record whatsoever. And with that, VR would like to take the microphone and have a few words…
cyberge99 8mo ago
They typically have an elastic workforce padded with several contractors or temp workers that can expand or contract based on market conditions.
queenkjuul 8mo ago
It's incredible how engineers of all stripes assume they know all sorts of things they don't actually understand
- dang 8mo ago
  That's a property of human beings in general. It's also a nice example of how sometimes a conundrum goes away when you make it more general. That is, it seems incredible (going back to the GP) that engineers assume they understand economics when they don't. It's somewhat less incredible (applying your formulation) that engineers assume they understand things they don't. While it may still be incredible that engineers are so assumey, at least we no longer need to marvel at their assuminess about economics. If we go one more step and state it as everyone assuming they understand things they don't, it's no longer incredible at all. What would be incredible is if anyone didn't.
palmotea 8mo ago
> It’s incredible how some engineers assume they understand economics, then proceed to fail on some of its most basic premises. This tends to happen when engineering-style certainty is applied to systems that are driven by incentives and uncertainty. Software engineers are stupid, often because they think they are very smart (which breeds arrogance). And on top of that, they often have many biases that they often fail to account for (e.g. a preference for neat-and tidy-systems, leading them to get seduced by oversimple neat-and-tidy explanations, e.g. Econ 101).
fuzzfactor 8mo ago
The title could be re-stated: The economy is screwed, tech is just the scapegoat. People who already had plenty absorbed the cheap money, until there was basically not enough value left to go around for consumers to be able to carry on the way that was expected. >this isn’t a tech specific "failure", Nope, looks like every "market" has become more nonideal, and the more "financialized" they are could be what's making certain markets worse. Not every job is actually on a "market", some are deep in and others not so much. In the old days people held on to their jobs for so much longer that it's a really big difference when the actual "job market" in so many areas was virtually insignificant compared to 21st century rambling as the norm. Regardless of pay, the more your job is part of a market, the more you would be subject to market forces beyond your control. Markets are "always" irrational except for those times when they are neither in your favor nor against you. If a neutral situation holds for a period of time, that can be negotiated with a more "straightforward" approach than when volatility is the dominating factor. People might think with almost "engineering-style certainty". Once there are wild swings, it can be like a sine wave where half the curve is positive and the lower halves are negative. It still balances out to neutral but you may not know where you are at any one time. Every positive excursion will not last very long and be followed sooner rather than later by a negative equivalent in some way. With a significant cycle occurring, then the only truly rational, neutral times are those brief zero-crossing events when the market reverses from favorable to unfavorable or back again. If your only real opportunities occur when the market is in your favor, then you are dependent on the market being irrational which you need, but is about as uncertain as it gets. Opportunities like this can reverse more easily than the mainstream where they more often merely fade.

Bayramovanar 8mo ago

I will avoid getting into a “who understands economics better” debate. But factory workers usually require specialized machinery, tooling, and physical capacity, which makes overhiring slower, harder and more constrained. Those investments force more deliberate planning. By contrast, engineers mostly require a laptop and company hoodie... That low marginal cost makes it far easier to hire aggressively on expectations and unwind just as aggressively when those expectations change.

thunky 8mo ago
This, and I would also assume it's easier to spot idle factory employees, where idle devs (especially remote ones) can be easier to miss. And, (assumption again) the factory boss doesn't have an incentive to increase idle worker numbers, where a dev manager often benefits from being in charge of a larger number of hardly working people.
- silvestrov 8mo ago
  It is a lot easier for a software engineer to overcomplicate things. All of a sudden it is no longer enough with a few shell scripts. No, we need a full kubernetes cluster to run a service used by 10 secretaries. No, we can't just use PostgreSQL as a queue, we definitely need Apache Kafka for 1 msg per second.
blenderob 8mo ago
> By contrast, engineers mostly require a laptop and company hoodie. Alas, gone are the days when engineers too required specialized equipment like a desktop computer on the desk that you couldn't move with you. Every evening, you left it at office and went home to live a 100% home life. Alas, gone are those days.
- zrail 8mo ago
  Maybe for some. I've worked from home for 15 years and a huge thing that I've learned is that I have to have a hard physical boundary. My work laptop stays at my desk unless I'm on call and actively fighting a fire. When I want to use my desk for non-work things the work laptop gets put away.
- QuiEgo 8mo ago
  Crunch time in those days sucked. I remember mandatory nights and weekends and the managers ordering in pizza for everyone.
- bitwize 8mo ago
  I was a contractor for a FAANG. My immediate cubicle neighbor liked to work out on lunch break, and to hang his sweaty, smelly gym things on the framework of his desk to air dry when he returned. I would've KILLED for a laptop I could take to the complimentary office café so I could get something done without holding my nose. If I were a full employee, I could just ask for one. Alas, as a mere contractor I was something less than a person, so I had to remain tethered to my desktop per company policy.
  - steve1977 8mo ago
    I mean, if killing was an option, getting rid of the cubicle neighbor would have solved the problem as well.
brk 8mo ago
Factory overhiring can happen easily by adding speculative 2nd and 3rd shifts. No need for extra equipment, tooling, or physical capacity. Lines with specialty equipment and tooling can also often be sped up. That can allow for other jobs to be added to all the functions that support the processes involved before and after the specialty equipment. New employees also often require training and some apprenticeship time, meaning they can get hired ahead of actual demand.
- Bayramovanar 8mo ago
  I am really not debating whether over hiring is possible in factories. in tech cost of hiring is lower which makes headcount a much easier speculative bet and layoffs a much easier reset when the bet fails.
  - rpdillon 8mo ago
    The 90th percentile of factory workers makes $45,000. The 90th percentile of software engineers makes $230,000. Hiring in tech is insanely expensive; I'm not familiar with hiring in factories, but it can't possibly have the supply problems that tech workers have. I do on average 115 interviews for every person I hire in tech.
    - paulnpace 8mo ago
      That $45,000 factory worker is often operating a $1.5 million piece of equipment with $10-50,000/hr of materials going through it. It's an entirely different cost structure.
- paulnpace 8mo ago
  There exists greater issues in hiring for manufacturing when it is for new shifts. Master level technicians and foremen willing to work those hours can be exceptionally difficult to find, and everything flows out of these people. While similar issues likely exist relating to discovering talent for software development, I speculate that the factory will, in practice, have a harder time finding people (for new shifts). My experience with seeing new shifts added is initially with only specific processes, and even with those it is with journeyman level technicians running a small crew to support relieving a bottleneck in production. Alternatively, manufacturers can outsource until they have enough volume to add a shift, but across the economy the net is just transferring production from one facility to another.
Propelloni 8mo ago
This only holds for companies that do not have to comply with some regulation or standard, e.g. ISO 27001, to do business. Especially infrastructure, banking, and defense tech have high compliance requirements that also, and sometimes esp. cover software development.
AnimalMuppet 8mo ago
What's the cost of a programmer-quality laptop, with two large monitors, plus licenses for all the software that they need? That's 2-3K. Does anyone know what the typical tool cost is for a factory worker? (And the tool cost for a factory worker can be zero if you're hiring for a second shift - they would use the same tools as the first shift, just at different times. In contrast, I don't think there are very many places that ask programmers to use the same laptop in shifts.)
- arter45 8mo ago
  >Does anyone know what the typical tool cost is for a factory worker? It's probably lower. Apart from safety gear like helmets and gloves, everything else is most likely reused across shifts or shared across workers simultaneously (think industrial machines).
- Bayramovanar 8mo ago
  they even cut costs by leasing the computers...
  - gdmka 8mo ago
    I'd add that companies write off their equipment investment over a 3-5 year depreciation period. After that, when the laptops reach EOL, employees can often buy them through the company's EPP (employee purchase program) usually for just the fair market value, which might be a couple hundred € or sometimes even less. So, eventually they recoup the equipment costs, wouldn't get surprised if they even make a profit out of it sometimes. First thing they do, they write off VAT (20-23% of laptop price). Then at the EOL, they just set the EPP price to the higher market value. And return the cost.
    - ThrowawayR2 8mo ago
      Is that common? I've never worked at a corporation that had an EPP for EoL computer equipment. It always all went to a specialist recycling/refurbishing business.
re-thc 8mo ago
> By contrast, engineers mostly require a laptop and company hoodie... > That low marginal cost Not true. That's how everything falls apart. Scaling software teams isn't simple. There's lots to account for. Do you just assume everyone commit to the same file and let it crash? The more people, the more work is required to standardize, organize, document and fix gaps. You can say you're already at scale and things "should" be done but they never are. You hire a team into a specific area to find it was a giant hack supported by a part-timer and as you try to fix it the problems keep escalating.
aleph_minus_one 8mo ago
> But factory workers usually require specialized machinery, tooling, and physical capacity, which makes overhiring slower, harder and more constrained. Those investments force more deliberate planning. > By contrast, engineers mostly require a laptop and company hoodie... That low marginal cost makes it far easier to hire aggressively on expectations and unwind just as aggressively when those expectations change. Software engineers also need - specialized machinery (at least when they have to upload to some computation cluster or cloud), think for example of the costs for GPU/TPU clusters for AIs at the moment - tooling: depending on the sector, the license costs for the sector-specific business software can be similar as expensive as specialized machinery - mental capacity (instead of physical capacity)
_m_p 8mo ago
The low marginal cost also means that the job market is easier for software engineers in times when expectations are optimistic.

Imustaskforhelp 8mo ago

> It’s incredible how some engineers assume they understand economics, then proceed to fail on some of its most basic premises. This tends to happen when engineering-style certainty is applied to systems that are driven by incentives and uncertainty. Dunning Kruger effect, am I right? We consider we are smart because we can make computers go beep boop so we know about the economy too. I mean, I am part of this too even though I (or we all) know the effect but I guess my point is that there should be an humility if we are wrong. I can be wrong, I usually am. If someone corrects me, I would try to hopefully learn from that. But let's see how the author of this post responds to the GP's (valid, from what I can tell) critique. Edit: Looks like they have already responded (before I wrote it but I forgot to see their comment where they said that its not at the scale or frequency we see in tech)

thunky 8mo ago
> We consider we are smart because we can make computers go beep boop so we know about the economy too Funny, the person you are agreeing with made the strongest "I know more than you" flex than anyone.
boh144 8mo ago
The scary part is when people who can make beep boop, and because the market has rewarded them for it, think they can skip the entire corpus of the humanities yet put forth opinions on how to shape society.

rcxdude 8mo ago

There's also this tech exceptionalism/grass-is-greener bias where software engineers especially tend to romanticize other industries and professions.

lumost 8mo ago

Tech companies also benefit from over hiring. It gives them slack to absorb crises, and Fast Fuel to move quickly on new growth. Eliminating the fear of layoffs allows employees to take more risks and explore. The current crop of tech companies cutting staff is going to lead to a large number of dead giants. The staff who services the layoffs will be risk averse, and defensible in a job cut situation. You see this in legacy firms where it takes 10 people to make a change because each person has a small slice of permissions required to effect the change. This pattern is by design as laying of any of the ten people on different teams would kill dozens of critical business processes. This is not how you make high growth firms.

lenerdenator 8mo ago

> Cheap money amplified this cycle, This is more-or-less the rub. Fortunately we had a kinda-sorta sane monetary policy under the Biden administration once the pandemic started to ebb, but now we've got Mr. Appearances trying to push Jerry Powell to make stupid mistakes.

WalterBright 8mo ago

> It’s incredible how some engineers assume they understand economics I would say most engineers. The reason is simple - basic economics is not taught in the public schools, and economics/business is not a required course for an engineering degree. One of the best classes I ever took was a summer class in accounting.

JKCalhoun 8mo ago

My take is that the traditional industries you mentioned (manufacturing, automotive, etc.) are making expectations on the market for what it is they're selling. With tech though, it makes expectations on markets that are brand new, or don't exist, and/or have no proven track record whatsoever. And with that, VR would like to take the microphone and have a few words…

cyberge99 8mo ago

They typically have an elastic workforce padded with several contractors or temp workers that can expand or contract based on market conditions.

queenkjuul 8mo ago

It's incredible how engineers of all stripes assume they know all sorts of things they don't actually understand

dang 8mo ago
That's a property of human beings in general. It's also a nice example of how sometimes a conundrum goes away when you make it more general. That is, it seems incredible (going back to the GP) that engineers assume they understand economics when they don't. It's somewhat less incredible (applying your formulation) that engineers assume they understand things they don't. While it may still be incredible that engineers are so assumey, at least we no longer need to marvel at their assuminess about economics. If we go one more step and state it as everyone assuming they understand things they don't, it's no longer incredible at all. What would be incredible is if anyone didn't.

palmotea 8mo ago

fuzzfactor 8mo ago

The title could be re-stated: The economy is screwed, tech is just the scapegoat. People who already had plenty absorbed the cheap money, until there was basically not enough value left to go around for consumers to be able to carry on the way that was expected. >this isn’t a tech specific "failure", Nope, looks like every "market" has become more nonideal, and the more "financialized" they are could be what's making certain markets worse. Not every job is actually on a "market", some are deep in and others not so much. In the old days people held on to their jobs for so much longer that it's a really big difference when the actual "job market" in so many areas was virtually insignificant compared to 21st century rambling as the norm. Regardless of pay, the more your job is part of a market, the more you would be subject to market forces beyond your control. Markets are "always" irrational except for those times when they are neither in your favor nor against you. If a neutral situation holds for a period of time, that can be negotiated with a more "straightforward" approach than when volatility is the dominating factor. People might think with almost "engineering-style certainty". Once there are wild swings, it can be like a sine wave where half the curve is positive and the lower halves are negative. It still balances out to neutral but you may not know where you are at any one time. Every positive excursion will not last very long and be followed sooner rather than later by a negative equivalent in some way. With a significant cycle occurring, then the only truly rational, neutral times are those brief zero-crossing events when the market reverses from favorable to unfavorable or back again. If your only real opportunities occur when the market is in your favor, then you are dependent on the market being irrational which you need, but is about as uncertain as it gets. Opportunities like this can reverse more easily than the mainstream where they more often merely fade.