The history and future of workplace automation (2015)
aeaweb.org
aeaweb.org
A significant number of floor staff only have highschool degrees and are being speedrun through an associate's to stay with the company on a higher staff grade. Others are doing months of on the job training to be brought up to the level that they can still be employed. I know some just can't hack it and are quietly being let go.
I know we are already asking our youth to mortgage their lives to get higher education, but what about a 50 year old woman who has been working for 30 years and still has 15 to go, but their job is completely gone? Do we send them back to college? This is already happening en mass in Japan, elderly are working entry level jobs because their old jobs are replaced. I think the college and retraining argument are only going to become more entwined in the next 20 years.
There's not enough rich people to do that in the first place. You will end up taxing the whole middle class, which is effectively what politicians are increasingly doing, which brings everyone down
I’m not sure that’s true. There are more rich people than ever and taxes are at their lowest point since ~1940
>I’m not sure that’s true. There are more rich people than ever and taxes are at their lowest point since ~1940
There are 2,640 billionaires in the world, 28 fewer than in 2022 (according to Forbes). Their total wealth is $12.2 trillion. Some portion (how much?) could certainly help a sizable number of people, but it is difficult to know how many and for how long.
> but it is difficult to know how many and for how long.
Permanently? Finding out how many is the job of the IRS after the top marginal tax rates are increased by an extra 10-20%. The top capital gains tax rate is also much too low.
Globalization and the removal of capital control resulted in a race to the bottom between countries offering ever lower tax rates. That’s a huge problem, it’s something only rich people can utilize and leads to continually increasing inequality.
According to the Distribution of Wealth page on Wikipedia (https://en.wikipedia.org/wiki/Distribution_of_wealth), millionaires (> US $1M, i.e. incl. billionaires) own 46% of property, while the middle class (US $100 - $1M) own 39%. https://en.wikipedia.org/wiki/Distribution_of_wealth#/media/...
Excluding primary residences would likely significantly shift this figure, but the figures seem reasonable..
A lot of people can become a millionaire throughout their lifetime just by having a normal job or small business. That's nowhere the level where I would describe as "rich".
There are also more people than ever, in case you haven't noticed
Bezos himself could personally pay to educate a generation of US citizens. Most of us can’t process the amount of wealth that one individual has, let alone that whole demographic of people.
There’s also a question of what fair taxation even looks like when you have such a small number of people in the US economy controlling the lion’s share of the capitol.
Let’s not fool ourselves into thinking that the capitol isn’t there. It most definitely is.
Let’s assume there’s a little over 3 generations in the US at any given time. Call it 100M people per generation.
Bezos has around $150B.
$150B divided equally across 100M people, there would be $1500 per person for all of their education. I think that’s obviously not enough to cover the education of an entire generation.
> Most of us can’t process the amount of wealth that one individual has
My experience is that many also can’t comprehend just how many people there are out there and how quickly program costs multiply out as a result.
> Globally, we counted 2,640 ten-figure fortunes, down from 2,668 last year. Altogether, the planet’s billionaires are now worth $12.2 trillion, a drop of $500 billion from $12.7 trillion in March 2022.
For comparison (via various wikipedia pages reachable from[1]), the US federal government budget in 2022 was $6.3 trillion (against an income of $4.9 trillion).
If 100% of the $12.2 trillion owned by billionaires was seized and annually disbursed at 3% annually for social programs, this would provide $366 billion dollars annually. In 2022, the US spent $747 billion on Medicare alone.
Increasing the Medicare budget by %50 seems nice, but it also doesn't seem like it would be a transformational change to society. It may be a better idea to spend a large chunk of the $12.2 trillion immediately on high-upfront-cost things, but I don't have the imagination to come up with good ideas.
[0] https://www.forbes.com/billionaires/
[1] https://en.wikipedia.org/wiki/United_States_federal_budget
It doesn't have anything to do with how many there are, but how much wealth they hold. And the 0.1% hold an enormous amount of wealth which absolutely would help fund effective services for the poor.
You could have one guy with a hundred trillion dollars and the fact that he's just one guy would be irrelevant.
We just need to enforce the tax laws we already have on the books.
[1]https://www.cnbc.com/2023/06/20/hunter-biden-to-plead-guilty...
Alternatively, make the Government waste WAY LESS money is a better answer. have you ever seen a mafia going bankrupt every year despite collecting protection money from everyone? only the government is so utterly incapable.
Who is making people do pointless work?
What does this mean?
Money only has value when they are spend. If you lock money up in an account and never use it, it is practically taken out of the economy.
Do, do we let governments circulate money for education, health, care, etc. or do we let corporations circulate money in the pursuit of more money?
For me, it's a balance. Right now it does seem like the inequality has been raising (in particular under the pandemic) and we probably should try to redistribute a bit more. But the increasing interest rates are already leading us in this direction. And the social unrest will probably also accelerate salary increases over the next years – which in turn needs to be financed.
I think the biggest issue we face are reaching multi-lateral agreements like the global minimum tax that can hand the power to tax companies back in the hands of governments (as it becomes unattractive to move to low-tax jurisdictions).
This is a problem because most people thought that they would have a career or a job forever without learning anything for 30 years. If you ask me, this is a crazy thought that we have given them in the first place. We should tell kids that learning is a lifelong journey and that it does not stop once you leave school - if you sleep on your skills you WILL be replaced.
I'd rather work towards a more stable future myself, but politically I have no idea how we can get there.
Honestly it was quite nice as by that point most high schoolers are bored out of their minds anyway, sliding down into destructive patterns, and this kept things challenging and interesting.
And also gave me the foundation to more easily learn the uni stuff that I needed on my own.
With the current availability of anything on youtube and such it should be trivial or at least a lot cheaper to learn anything you have the desire to learn. I think schools should be there to instill into us that desire.
Not saying unis are a bad idea - plenty of good things about them too, just I don’t think in our world they are a particular requirement - just make the schools system a few notches more demanding and we should be golden.
Simply put I could know everything, the problem is you now have to develop your own series of tests to ensure that I do and that I'm just not lying that I do.
Making schools more demanding on a single path likely doesn't solve this either. You just stress out people that are not going to be academically great no how hard they try.
Education does not mean you have to sit in some school. There are many ways to educate yourself.
As for your actual question, the internet is full of free knowledge. Many employers give employees a budget for self development and learning as well, but honestly, learning new things is an investment made by individuals for individuals.
The post GP is replying to reads as satire unless you hold a Hobbesian view of the social contract. It could be a line from Silicon Valley.
My point is that:
> the internet is full of free knowledge
Most of which is not useful for doing a career 180°. Unless you're going into software, putting theory into practice requires access to tools, material, and access to instructors, all of which cost money on an ongoing basis.
> Many employers give employees a budget for self development and learning as well
In tech. And few other career paths where this is something that makes sense. Most people in most jobs don't have this.
> honestly, learning new things is an investment made by individuals for individuals.
Again, only possible in tech and few other career paths. It's particularly easy with software, given that the job is usually... not consistently demanding, and you can get away with a lot of self-investment while on the clock before your boss starts to suspect something. Or you and your team can just start or re-write a project in a technology stack that is a great choice to put on your resume, but not so much for the problem being solved. Again, few other jobs give such opportunities.
The point I'm trying to make is, tech jobs are an outlier. Most other careers don't leave space to learn new trade when your boss isn't looking. Most adults come home exhausted after 8+ hours of work, some amount of commute, and have plenty of time-consuming responsibilities to deal with, which they can't outsource because they're not earning tech salaries. Sure, there are always some people that still manage to re-skill and switch careers. There are also some people who'll risk everything on a new venture and not end up homeless in the process. But it's not what most people can do - not adults, with dependents and costs of living at the level adequate for their current job, not when that job is suddenly taken away from them, while they're being ridiculed by the same people who said and keep saying to everyone else "nah, automation isn't a big deal, it creates new jobs, better jobs, more jobs than it takes away".
EDIT: Also, this is the kind of a problem that we can keep ignoring at our own peril. People who end up having their lives derailed by the "system" or "progress" don't disappear. They're still there, still a part of society - and when there's enough of them and when they all feel that society has screwed them over and left to wither and die, they will rise up and tear the society down. As it happened many times in history. And us tech workers with decent salaries and little responsibility - we aren't the top elite that always manages to buy or talk their way out of trouble. We're the very class that is first to be hanging on trees or lubricating guillotine blades.
You need to be able to read to access most knowledge.
One solution is that which the Nordic countries provide: a dole that is sufficient to cover basic cost of living, and enough community-organized events and cultural and educational infrastructure that the now terminally unemployed can (or can at least choose to) spend their days doing something besides drinking and getting soft.
About five years ago I got fired/quit from my second dev job and realized nobody was hiring node.js developers. So I looked at the local dev listings and saw the majority were .NET or Java. I spent the next month while unemployed reading an 800 page book on .NET (to be honest I got to page 400 or so before I was like “okay I get it”) and snagged a job as a mid level dev at a .NET shop in a Fortune 500 company. I didn’t even buy the book, I pirated it (apologies to Phil Japikse!).
I think we need to teach people that learning stuff yourself is free if you know where to look. We teach everyone school is the only path to knowledge which is, quite frankly, the most expensive, slowest, and overall worst way to learn almost everything. Outside some limited domains most concepts necessary for gainful employment can be self taught.
Isn’t that what the internet was supposed to do? What we all imagined when we’d talk, wide-eyed and endlessly optimistic, about the internet to people back in the 90s and early 2000s? Wasn’t it supposed to make information free? Which it has, for those of us who know how to look.
While I agree with what you are saying and am also 100% self taught myself, I think a large portion of people (especially outside the tech world) just aren't able to pick up a book, read it, pick up the knowledge and put it into practice efficiently.
Especially with no accountability or social pressure that sitting in a classroom provides...
I am not arguing for college or university, but I think that just giving people (especially someone in their 40s or 50s) a book and expecting them to pick up new skills may be asking a bit too much...
In software. Where there are almost no costs to practice or even to start making money off your knowledge. Where the market is growing so fast that everyone and their dog can half-ass some learning for a few months and then land a job - at least if they're young enough, and not operating at peak capacity like most adults. There's literally no other career path like this right now. No one else gets to pretend they've learned something, and get a shot at catching up to minimum competence levels on the job before the employer realizes what's going on and fires them. And that's just a reflection of the shitty state the industry is in.
No, giving people advice based on tech career paths is just a thinly veiled insult.
> I don’t really have much sympathy for these people. It’s important to stay plastic and keep learning through your life.
Remember that 5 years from now, when your job and all your career prospects disappear thanks to GPT-6.
I also think it's quite the opposite - knowledge jobs can be self-taught with a spectrum of degree within that statement as well. Computer programming - absolutely. Electronics - once you start working with real devices rather than sims you now have to invest not insignificant capital if you want to use testing equipment and measurement tools. Accounting - I'd guess you have to invest capital to take the CPA exam (United States) and you may even have to have interned before that.
However, if we look at other 'gainful employment': mechanical work - sure you just need to have the engines available and ready to be worked on; not necessarily a free endeavor. CNC programming - maybe there are simulators but again you will want to eventually learn the feel of real equipment. Woodworking, metal-working, other-fabrication - you are going to burn through a decent amount of source material before you learn what you're doing. Commercial driving - some places will pay for you to go through the whole program, you can't just self-teach yourself and apply how to drive a commercial truck.
The last sentence just reads as condescending (there are some individuals that are morally opposed to the idea of pirating, you have indicated this is not the side you fall on).
Maybe it takes you a decade to get that degree, maybe you end up not needing it, but taking classes is better than sitting around in meetings all day. Besides, the security of not being pigeonholed in a single career is going to make people less fearful and more likely to take risks like starting new businesses.
Here in Kentucky, there are lots of people who worked in coal mines for several generations ("grandpa was a coal miner, dad was a coal miner, I'm a coal miner and my son will be a coal miner"). Coal is going the way of buggy whips. It is a dying industry. Automation is replacing some, and a collapse of customers wanting coal are getting rid of more. There are state and federal programs to help retrain workers. Yet the men affected frequently refuse the retraining. People still complain about the times in 2016 when Hillary campaigned and promised more retraining with stuff like "how dare that carpetbagger come here and tell us how to live"
I have yet to come across an organization that has the opportunity to eliminate a skilled role - that is a white collar, certified expert position (think accountant or financial analyst) - because I have automated one of their business processes. I have worked for small, medium, and very large enterprise organizations. In every instance there has been at least two or three other business processes that were and will be out of our automation capabilities that the team member simply could not do because they were spending the vast majority of their time doing a business process that could be automated. Automation simply frees that team member up to do what was previously seen as "value added" work but now can be considered business critical. The author of this study probably even touches on that by addressing how the meaning of work and a job role simply shifts upon the automation of previously executed business processes by moving to what are still considered "in role" processes.
https://docs.google.com/viewerng/viewer?url=https://www.aeaw...
It's not just a Berlin thing. It's happening all over the place. I bought some regional wine in the Netherlands when I visited my family there. In a small village from a tiny business that produces wine locally and sells it on local markets. I should add that the Netherlands is not known for its wine production. That's a completely new thing (thanks to global warming). This is far away from the big cities. The same market featured all sorts of local produce. In the US you'd call it a farmer's market. Expensive produce made by people with decent educations aimed at people with disposable income.
What's an artisanal business, it's a business model where the added value is in something being done with a personal touch and craft. Mostly without a lot of automation. People pay extra for that sort of thing and the people doing this stuff tend to enjoy what they are doing. The whole point is quality, craft, and tradition. As volume production for highly automated/processed stuff scales, prices for that drop and demand for the artisanal stuff increases.
Just an observation: automation is a race to the bottom and it merely serves to enhance the value of stuff that is harder to automate. We automate the boring, low value stuff first. That creates more capacity for doing the high value stuff. Which as the wording implies is great for economic growth.
Related to this is the re-shoring of manufacturing jobs. Now that automation has lowered the price of doing things locally with fewer workers, it no longer makes economic sense to ship stuff half way across the planet. Plus people like the idea of locally produced stuff. Americans want US made cars and phones. Germans like their German cars. Etc. Yes, a lot of that manufacturing uses less people to produce vastly more goods. But that's just causing them to grow faster and employ more people.
That's why there is a labor shortage instead of mass unemployment. Despite macro economic circumstances, a few years of lock down, etc. This stuff is just fueling economic growth. The whole economy is shifting to emphasize qualities such as being locally produced, sustainable, etc.
I suspect classical economists don't see value in this theory. But that said, the underlying point is that cost, price and value are really odd concepts and in some ways, "all made by machines" confuses the hell out of people who thought they had a sense of value in a product which is now made by .. whom exactly? I mean why do I even pay more than the capital depreciation, the input costs and some low margin?
Objects like resistors or pills reflect the sunk costs much more than the labour costs. Are electronics production lines and pill factories totally devoid of humans? No. They moved roles. They supervise machines, or they perform tasks where the costs of (mobile) robots or robots adaptive to contextual demand is still too high.
I have a friend who supervised a totally automated clarified butter (Ghee) factory in Queensland Australia. It operates in a very narrow temperature band: too high and the Ghee tastes burnt. Too low and it gums up the flow. He's like the man in Kurt Vonnegut's "Player Piano" keeping the robot working, which makes the Ghee. There's a driver or two, and some payroll staff, and cleaners, and finance/economics and advertising staff. It's hardly "nobody works here" but they don't employ people on the factory floor as much as they did.
Robot slaughter houses (there's a thought...) are using smart cutting devices to move highly skilled labour to the places they can add value. There are 4 important cuts on a side of beef which have to be made "just right" to extract maximum value in all the subsequent jointing. That demands a skill which goes beyond the normal, you have to sense where the muscle and bone points are on this specific carcase right then and there. 4 cuts, which ruin the value if you get them wrong.
Sheep shearers are highly paid. Robot sheep shearing has been a long time "in development"
It is also worth noting that if you don't have a workforce, you don't have people earning money, and thus don't have people buying your goods. This is not a problem for a Bottega making handbags to sell to Melania Trump and her friends, because you aren't selling them to your local workforce. It's a huge problem if you make milk, or tin pineapples, and expect to sell produce locally: if you don't pay people, they can't buy your product.
If the neoclassical theory of price converging to marginal cost actually holds, then this is exactly what we should be paying!
> Objects like resistors or pills reflect the sunk costs much more than the labour costs.
You mean up-front capital expenditures? Sounds about right, but then there's the problem that somewhere back in that chain, high-skilled professional labor was expended to design the machines (that design the machines?). So you do then have present marginal costs paying for past inputs, capex, and labor as well as present inputs and depreciation. Except you're not really paying the past labor directly, but instead via the firm that owns the trade-secrets and IP created by the past labor. So there's labor in there eventually, but alienated through enough layers of indirection that the same employer gets paid a bunch of times over for one design's work of actual work.
Meanwhile, the guy who did the actual work is probably hunting for a new job because he doesn't get paid for use of his old work.
Hmm... we need IP reform.
I've never seen any value in this argument. If I make Bugattis, I gain nothing by paying my employees enough to buy a Bugatti. In a two-person one-good economy, yes, I can't hire you to milk the cow for $1 and then charge you $2 for the milk, but this isn't going to lead to you starving among spoiling milk. The market will clear, the prices of labor and milk will equalize, because $2 is more than the market will bear.
With robots the price of the milk will go very low because it's so easy to make, meaning I will be willing to sell it for very little in return. You can get a near worthless job like collecting pretty rocks and buy milk with that.
This is making a huge amount of dangerous assumptions. With robots the price of milk might skyrocket, for example, if land becomes more desirable for other purposes that robots unlock driving up feedstock costs. We are already seeing farmland costs increase significantly. As long as land/homes are a good investment expect costs to increase.
That's really the threat; that all 'useful' labor will be conducted and collected upon by a very few, and how do we as a society handle that? I haven't seen an economic model situated in capitalism to explain how that can work.
And, the "two person economy" is a reductive argument which misses the point: Ford had to sell cars to people in the income bracket of his workforce. He did not drive the price of labour down below the cost of his cars because he knew that he needed people in that income bracket to be able to buy Model-T cars (he extended favourable terms which probably was a trap, but thats sort of beside the point: he wanted them to be car owners and buy cars) _ I say Ford because he was anything BUT a socialist economist on this matter. He did NOT like organised labour, and fought hard to avoid it. At points, he deliberately paid his workers above scale, to force his opposition out of business (and then subsequently probably drove their rate of delivery up with his massive commitment to Time and Motion Tailorism models of production lines)