What the Industrial Revolution Tells Us about the Future of Automation
cacm.acm.org
cacm.acm.org
Industrial Revolution during the 18th and 19th centuries led to massive social upheaval. Average height of Englishman decreased before started to increase (based on military records). Workers started to organize, skulls were cracked. People were shot in the streets. Armies were ordered to kill workers. There were bloody civil wars and revolutions.
Industrial revolution was not possible without massive social change that created, land reforms, working class, middle class and increased private consumption and accumulation of capital. It was at times bloody, but in the Western Europe and North America continuity of government was maintained. Not so in Soviet Union or China.
Anti-Union violence, Union violence (union busting) communist violence, anticommunist violence was pretty intense at time even when the society as whole worked.
Let's use US as an example of well managed industrialization and social upheaval with small number of casualties. We can still list numerous conflicts with 10-100 deaths: The Molly Maguire trials (20 workers executed), Battle of Virden, 1898 Colorado labor war,Italian Hall disaster, Ludlow massacre, Thibodaux Massacre, Pullman Strike, West Virginia Coal Wars, Rock Springs massacre, 1905 Chicago Teamsters' strike, Steel Strike of 1919, The Herrin Massacre, 1922, Battle of Blair Mountain, Hanapēpē massacre, Harlan County War, Memorial Day massacre of 1937.
Future of Automation will create wealth and prosperity beyond all our dreams, but it requires massive political struggle that settles issues like privacy, rent seeking, ownership of personal information.
Violence in the street may be positive sign that automation is finally changing our society.
Many of the emigrants of the 1800s ended up settling "new" land in the US. But there's vast territories around the world that are extremely inexpensive. Since the 1800s we have grown more dependent on the services provided by the state in urban areas.
Nowadays, you really can't.
Somehow I think the owning class will come up with ways to make this illegal or impossible.
That's what the poster above meant by "the primary income sources from open spaces now requires massive amounts of capital to be competitive", and owning these robots would be the massive amount of capital; you'd be able to sustain yourself because (and if) you'd have a lot of very expensive robots, not because you just had some land; and you'd also be able to sustain yourself by just investing the value of these robots.
Although those places are not very conductive to farming...
Alaska is largely government owned. It's empty, but that doesn't mean that you can take a chunk of land there.
So there WAS vast expanses available by that time, even without intentional genocide (which also happened).
2. There is still a huge amount of open space in America. In 1800, there were only 5 million citizens, compared to 323 million today.
3. Natives lived all over the continent. Europeans landed where the ocean ended.
A large percentage of the land battles of the War of 1812 were fought between the US and natives because the US wanted the land of the natives to become open space so the natives allied themselves with Britain.
Lewis and Clark would not have survived or even made it to the coast if they had not had help from most of the tribes they met along the way. Of course we displaced all of the natives they met eventually.
I just read https://en.wikipedia.org/wiki/Lies_My_Teacher_Told_Me so that may bias my opinion.
I highly suspect people are much more interested in the first, but not really the second. Interests who stand to benefit from the second are selling them as a package deal and we will end up only getting the second. In other words, the arrest machine will keep running, but businesses will be free to fleece customers, government and employees with impunity.
Surprising though, regulation is intended to save capitalism, not hurt it, at least in the long term. If capitalism runs down a path where it no longer benefits the vast majority of citizens in any given country, there will be hell to pay resulting in the end of capitalism. That would be a shame too, properly regulated capitalism is a really great system.
The problem is is that a robot this advanced could also be used for very sophisticated bad purposes, so you're going to have lots of regulation and big corporations will only be able to use these robots and you'll have to buy things they produce with whatever meager wages they let you have.
When the prices crash a large number of people with robots will be unable to pay back the loan. Their robots will be repossessed. They will now be out of a job and with no robot.
The poor would get left over food from the kindness of people with robots? They'll certainly never work again.
I have an American friend who lives in Argentina who works online. The cost of living is such that he can work two months and pay his living expenses for two years. Those horribly low prices in Argentina are totally killing him. Not!
So what you are saying is the disaster is there will be so much food that people won't be able to eat it all and the prices will crash? Imagine if everybody had unlimited amounts of everything they wanted! How would we support prices?
This kind of thinking reminds me of the great depression where they would destroy food to keep up prices even as people were starving.
BTW, The novella "Manna" by Marshall Brain is kind of a good take on the future with robots and everything, though I don't really like his "happy" future.
BTW, thanks for giving me a hard time on this. I am getting to the point where this is going to be a bit of a manifesto and you're helping me flesh things out.
You're friend in Argentina:
* has the ability to get a job from home, which means it's probably a tech/outsourcing job in a first world country. That doesn't take care of 95% of the population. That doesn't fix the issues.
* In the robot scenario where robots are smart enough to understand markets/futures and can decide a completely different line of work, and then do that work, those remote jobs are gone. Your friend is out of work. He doesn't get a robot.
* the company that hires him has to compete for resources in their country of origin. Their local pay requirements determine what they think a "good deal" on labor is. That country would have robots as well and their wages would crater. Again your friend's cush life would disappear, unless he could buy a robot.
>what you are saying is the disaster is there will be so much food that people won't be able to eat it all and the prices will crash? Imagine if everybody had unlimited amounts of everything they wanted! How would we support prices?
Now you are getting to the root of the issue.
We produce enough food in the US that nobody should go hungry, yet we have swaths of kids that don't know where their next meal is coming from.
We have enough abandoned houses that nobody should be homeless, but people still live on the streets.
In capitalism, a private person/entity produces and someone else has to PAY to use what's produced. The issue isn't that there won't be enough to go around. The issue is that our current system will allow lots of people to starve and do without even when there is excess.
You used the term libertarian. That might be why you're getting all the push back. Maybe I'm misinformed on the topic. Libertarians believe an unregulated market and human morality will make everything better. Those two things combined gave us the triangle trade, raw goods (America) --> manufactured goods (Europe) --> slaves (Africe). We also got child labor, poisonous food being sold, no education, life threatening work conditions, etc.
>This kind of thinking reminds me of the great depression where they would destroy food to keep up prices even as people were starving
Being against this strategy is the antithesis of capitalism. If I give away food, prices drop. If I reduce the supply then prices will go up.
During the great depression, prices dropped so low that all farmers were going to go bankrupt. Farmers themselves were dumping crops and rioting to keep crops off the market to try and increase prices. Farmers were marching on courts because they were demanding that foreclosures stop.
If you let farmers flood the market with goods (lowering prices), while allowing them to refuse to pay their debts, then what kind of economy do you have? It's not capitalism.
My concern is that the coming wave of automation will recreate the great depression x10. The population at large doesn't see it coming, and maybe I'm a chicken little. 30% of the US population think poor people are just lazy, and that if they just worked harder they'd be better off.
When automation comes full force, then our economic and political system will change. The question is what will it change to, and what will the transition look like. There's a happy path and a bloody path.
This is what I keep harping on with folks. I think automation will hit and hit hard and our current economic/political structures are not ready for it or won't be able to adapt fast enough (hence your depressionx10 or just a massive inequality gap).
We already have 'more productive' economies but we have stagnant wage growth, increasing rental/housing costs due to people leaving for cities since there are little/no economic prospects in rural areas now, and, as you said, logistical problems in fulfilling basic necessities. Not to mention mental models that say that these 'unemployed' folks are bad and should just retrain/go to bootcamp/move elsewhere.
Consumption-wise, we can buy whatever we want. Shitty food is cheap, electronics, etc, all cheap. But some of the base foundations in Maslow's hierarchy don't come very cheap anymore and it's harder and harder to get to the higher reaches.
* My friend is not a rockstar programmer. He's about average, and in San Francisco he was living in Bayview and getting his house and his car regularly broken into and barely scraping by. Now he lives like he makes 5x the money he did in SF and his taxes are vastly lower since he's making less money, but living better. This is one of the largest mysteries of modern capitalism. Why are the prices for things so vastly different in different countries?
* If the robots can run the futures market and do general purpose programming then we probably won't even need money because we will have perfected central planning. Instead the robots will be playing a game called "The Humans" which is like when we play "The Sims" with the goal of trying to make the Humans happy. This will go on at a superhuman level of intellect and we won't even be able to analyze what the heck is going on, much like we can't really know why Alpha Go makes the moves it makes. We just won't turn it off because we can't live without it. This is the good or bad future depending on the ultimate fitness function, which in the good future is make everybody happy and not where the fitness future is based on deep ecology (DE) [1] which would prefer a 95% population reduction.
* My friend might not be a web developer, but he could probably get a job working on designing some guy's Burning Man float who owned a robot for enough to live well since it would require an absolutely trivial amount of effort to live since there would be so much supply.
* As far as Libertarianism goes, your formulation seems to be that that means no rules. No not really. I don't think ANY libertarian that is respected by the libertarian community would think slavery or poisoning people should be legal so you really have got a strawman there. Besides if we had no rules at all the wealthy Deep Ecology people could get together and decide they wanted to exterminate everybody they don't need to maintain things with killer robots who unlike in the movies have perfect aim and tactical strategy.
* The foreclosures during the great depression were due to collapse of bank reserves due to fractional reserve banking. You had short panics in the 19th century before the fed and fractional reserve banking, but not depression like scenarios of rapid credit contraction as the money multiplier work in reverse. This is unfortunately a very very deep rabbit hole side issue that requires a 30 page essay to get through. In China, which is probably more capitalist than the U.S, they have a different system. They have tons of companies that have bad debts and the government prints money out of thin air to selectively bail them out without austerity forced on the taxpayer, which is why they haven't had a credit crunch that has predicted every year for the last 30 years.
* The happy path of the future AI automation is normal people get access to these robots and use them to produce unlimited prosperity. The bloody path is only the elite get access and they decide to wipe out everyone else with killer robots because they find them to be irritating useless eaters.
Empires fell during this time in western europe. And we had a civil war in the US.
> Let's use US as an example of well managed industrialization and social upheaval with small number of casualties.
You forgot the mass extermination of natives in the US due primarily to industrialization. And the civil war which was bloodier than all the other wars we participated in combined.
The nations first to industrialize benefited immensely at the expense of others. But it was extremely bloody in the US as well.
Industrialization is what prompted and allwed ships to bring millions of europeans to the US. It's what allowed railroads to ship millions to the midwest and the west. And of course the desire for oil/resources and farmland/etc. Industrialization is what allowed our armies to wipe out and clear the natives west of the mississippi.
>And the civil war which was bloodier than all the other wars we participated in combined.
Only because the causalities from both sides count as American casualties, and because we developed better medical treatment and sanitation technology.
Far more civil war soldiers died of disease than combat, which was mitigated in later wars through advanced technology. The number of US combat deaths was higher in WWII.
This is an oft-repeated genocide denial. Yes, disease killed many natives but that's not what committed genocide against the natives and stole their land. This is the same genocidal denial peddled by some who want to excuse nazis by claiming the jews died of disease in the holocaust.
> Only because the causalities from both sides count as American casualties, and because we developed better medical treatment and sanitation technology.
What's your point? Dead is dead. Should they not be counted?
> Far more civil war soldiers died of disease than combat
This is the case in most wars. Not just the civil war. Most of the deaths in ww2 were due to disease/starvation/etc rather than combat. Should we not count those?
The current consensus is that disease was the primary cause of depopulation. Perhaps you should read some of the current work in the field.
That doesn't change the fact that European colonists conquered and subjugated the remaining Native population.
>This is the same genocidal denial peddled by some who want to excuse nazis by claiming the jews died of disease in the holocaust.
Nonsense. You're going against the scholarly consensus here--just like holocaust deniers.
>This is the case in most wars. Not just the civil war. Most of the deaths in ww2 were due to disease/starvation/etc rather than combat. Should we not count those?
You held up the Civil War as an exemplar of terrible wars, saying that it was "bloodier than all the other wars we participated in combined". The implication is that it was somehow worse than other wars.
The point is that this is a useless comparison to make and taken out of context you can draw conclusions that don't fit the facts. Previous wars were less bloody because they involved fewer combatants. Later wars were less bloody because we developed new sanitation techniques. And ww2 was much worse in terms of civilian casualties. You can't really draw conclusions about the time period or the social upheaval caused by the industrial revolution based on civil war casualty numbers.
As I said, that's just genocidal denier's distracting tactic. Also, the consensus is that most victims of the holocaust died of disease/starvation. That doesn't change the fact that the genocide took place. You keep harping on it as if the disease caused the genocide. It didn't. It was european colonization. If europeans didn't come to america and steal native lands, americas would be full of natives - disease or not.
> That doesn't change the fact that European colonists conquered and subjugated the remaining Native population.
And the genocide. Keep denying it all you want, it doesn't change reality.
> Nonsense. You're going against the scholarly consensus here--just like holocaust deniers.
The only one denying genocide is you.
> You held up the Civil War as an exemplar of terrible wars
No. I said it was the worse was america experienced. I didn't say it was an "exemplar of terrible wars".
> saying that it was "bloodier than all the other wars we participated in combined".
Because it's true.
> The implication is that it was somehow worse than other wars.
Yes. It was the worst war america fought in. It was the bloodiest. It caused the most damage. It resulted in the most casualties in absolute terms and as a percentage of the US population. If you can find a war that was worse for america, please let me know.
> Previous wars were less bloody because they involved fewer combatants.
No kidding.
> You can't really draw conclusions about the time period or the social upheaval caused by the industrial revolution based on civil war casualty numbers.
Sure you can. Especially in terms of the civil war and its place in american political and military history.
Listen, you are a genocide denier. And now you are arguing against the civil war facts. I'm finished here. You are free to believe what you want.
European colonization brought disease, but disease spread west faster than settlers, or even explorers.
Disease is the cause of the Native American population decline. Without disease the settlers wouldn't have been able to push the remaining Native peoples off of their lands. In the vast majority of cases disease was a byproduct--the settlers didn't intend for hundreds of thousands of people thousands of miles from them to die.
There were many cases where Europeans intentionally killed Native Americans, in some cases intentionally spreading disease.
You seem to have a very difficult time with nuance. Yes European colonizers caused millions of deaths, but the vast majority were caused indirectly without the knowledge of the colonizers. Yes the Civil War caused the highest death toll in US history, but only because both sides were considered American (despite the fact that they weren't American during the war).
>Listen, you are a genocide denier.
Please try to be a little more civil. This is hacker news not reddit.
>And now you are arguing against the civil war facts. I'm finished here.
I never argued against any facts you presented. My argument is that these facts are useless without context.
> I'm finished here.
I doubt it. I'm fairly certain you're going to reply to this.
The cotton gin is an example of an automation removing the need for labor (deseeding cotton) and subsequently creating a greater need for labor elsewhere (planting and harvesting). Of course, in this example, the increased labor demand ultimately resulted in 720K deaths.
The Linotype was a win for printing workers. Typesetting had been bottlenecking the process, and with fast typesetting, book and newspaper prices came down, and the volume of printed material increased enormously. That's the good case.
The glass-bottle blowing machine is the not-so-good case. Bottle-blowing used to take a team of five people trained to work together and manipulate molten glass, blowpipe, and mold with tight coordination. This was a skilled trade and paid well, but bottles were expensive. Automatic bottle-making machines could make bottles rapidly with minimal labor. Bottle sales went up, and the bottle business grew, but the skilled trade was dead. Most of the workers were shoveling sand in one end or taking bottles out the other. This didn't take much skill. Operating the machines was not too hard. You probably needed a maintenance guy who really understood them, but only one for many machines. So most of the skilled jobs went away, replaced by low-paid entry level jobs.
The worst case was the stone planer. Back when brick was the key building material, stone lintels were used over doors and windows to provide structural support for the bricks above. Stone was chipped by hand by big guys with big hammers and big chisels to make flat, rectangular stone beams. The stone planer was a powered machine for squaring off a block of stone. Like a wood planer, but heavier-duty. No more need for big guys with chisels. Huge increase in productivity. But the market didn't grow, because stone lintels were a minor building component and cheaper stone lintels didn't mean more demand for them. So most of the workers lost their jobs.
Those three outcomes still apply to many cases of computerization.
This may be the best argument for basic income. Why don't we just admit that we just haven't a clue how things are going to pan out? Things could be fine, or things could get very nasty. Looked at this way basic income is not analogous to social welfare/security – when we admit that we have no clue how the future is going to unfold then basic income looks more like social insurance.
This puts me in mind of a thought I had this morning. Why are narratives compelling? Because they explain events. In the past before formal methods allowed us to develop accurate models humans used a narratives even where they are inappropriate. Now, because models have been so successful the pendulum has swung the other way and we tend to over-rely on model based explanations when really we have no clue.
What am I saying? Just suppose that the "elephant curve" has been brought about by automation or the 4th industrial wave as the article calls it. We have many many people predicting outcomes based on economic and social models that may fail to apply in spectacular fashion. Maybe what would be better would be to institute basic income to ride out digital automation while models are tested and refined. But also let us collect stories from now and from the recent past to use as cautionary tales whose main purpose is emphasize the point "we don't really have a clue and anybody who tells you they do is talking out of their ass".
Dafuq? What do they think marxism was a reaction to?
Have they never heard of the theory of alienation? https://en.wikipedia.org/wiki/Marx%27s_theory_of_alienation
To avoid the HN snark police (oops), here is a relevant passage:
> They are definitely right about the long period of painful adjustment! The aftermath of the Industrial Revolution involved two major Communist revolutions, whose death toll approaches 100 million. The stabilizing influence of the modern social welfare state emerged only after World War II, nearly 200 years on from the 18th-century beginnings of the Industrial Revolution.
The thesis seems to be that we should consult historians, not just economists, on this subject.
But it's details in general I agree that Economists are the least interesting people to ask.
Hans Rosling did a great job explaining all that progress.
Yes peasant revolts happened before this. There was an old cruel aristocracy, and then a new, industrial aristocracy which was inexperienced, fumbling and cruel. Industrialization wasn't the first harsh transition in human history, but who argued it was?
Like, what's a few dead employees between friends?
> Many economists say there is no need to worry. They point to how past major transformations in work tasks and labor markets – specifically the Industrial Revolution during the 18th and 19th centuries – did not lead to major social upheaval or widespread suffering.
interesting that it also correlated with emergence of national identities in Europe (https://en.wikipedia.org/wiki/Rise_of_nationalism_in_Europe). There is some parallel here to the nationalistic anti-globalization movements starting to rise today.
>They point to how past major transformations in work tasks and labor markets – specifically the Industrial Revolution during the 18th and 19th centuries – did not lead to major social upheaval or widespread suffering.
Which is then followed by this in the 4th paragraph:
>The aftermath of the Industrial Revolution involved two major Communist revolutions, whose death toll approaches 100 million.
18th and 19th century didn't replace humans it replaced either animals or mechanisms (water-mills) or laborious human physical processes.
The replacement we are looking at today is about replacing not just the solutions of the 18th and 19th century but about replacing human laborious intellectual processes.
In the 18th and 19th century you could still point to the human mind and say well at least they can't do that.
But this is now an argument that is becoming less and less true.
What a lot of people who try to claim automation isn't doing anything to jobs forget that most jobs doesn't actually require the whole human being they only require parts of what a human can do. A job is a function in a process even some of those who seem to require the whole human (like experts)
The real way to look at this is too look at what types of jobs requires general human intelligence and which ones require specialized human knowledge or intelligence and what we find is that the cleaning lady is more likely to keep her job for a while while the radiologis function in the process of detection can be compressed into much fewer (and almost none in the long run) people.
And compression is IMO the right way to think about the job market and why it's going to be less and less an area that humans are going to have any dealing with.
In other words automation is about replacing higher and higher levels of abstractions. I don't know of any level beyond human intellect and that is why you can't make comparison with 18th or 19th century progress.
The last people to listen to IMO are the economist as they treat technology as an externality when it's really at the core the job market discussion.
Sure we might end up in some post scarcity society but getting there needs serious attention as it affects how we tax, where we tax, who we tax and to what extent.
Currently politicians listen to economist, they need to listen to technologist way more.
Most everyone on this site has the education and skills to stay ahead of the AI revolution for some time. But many other people already have fallen behind. If the only things you know how to do can be done by machines at less cost than a living wage, then what hope do you have to compete?
Empirical data shows the opposite. Machines have been getting more capable for at least 100 years, and job creation has soared. Even if you just look at a snapshot of data from today, the societies with the most capable machines are also the societies with the best standards of living and lowest unemployment rates.
1) Population growth
2) Globalization
3) Technological acceleration
4) Cost of switching to automation
5) Monetary policies to keep economy going
6) The way employment are measured
Unemployment rates aren't what you think they are. In the US 95% of the jobs created since 2010 are temp jobs not actual structurally solid jobs. Furthermore the wages are stagnating while the cost of living is soaring.
So you are looking at this from the wrong angel. Empirical data all points to the fact that humans in normal jobs makes less and less money because their role as a function of the given industry are diminishing. Only a small group of people actually experience growth (and accelerating too).
The canonical example I recall is junior legal discovery lawyers who used to pore over paper documents. Such routine tasks are automated and a smaller number of more senior lawyers who do the work of integration and interpretation.
https://mobile.nytimes.com/2017/03/19/technology/lawyers-art...
https://www.economist.com/news/special-report/21700758-will-...
Humans are social animals; we are each interested in ourselves, and interested in each other. It was true before we developed intelligence and it remains true today. We want to be healthy, socially connected, and entertained; we can never get enough of those. And they can't be automated--a robot can't be healthy for you, or feel emotions for you.
There is no reason to doubt that an economy that is 30% health care, 30% social communications, 30% entertainment, and 10% everything else, can work. Or even an economy that is 99% entertainment, and 1% everything else. Comparing today to even just 200 years ago, we're already well on our way.
You're right of course, but you're not addressing an actual argument. No one is saying that.
What people are talking about are AI's replacing doctor's and other health care professionals, replacing Hollywood and the rest of the entertainment industry, etc.
We as consumers could still be consuming all of these things, but that says nothing about who is producing it.
I'll give you an example: people still watch bicycle racing. We all know a motorcycle is more advanced technology that could easily beat any human on a bike. So what did we do? We invented rules out of thin air that exclude motorcycles, just so that we can keep watching humans agonizingly grind their way up a mountain on a bike. Because we find that entertaining.
There's motorcycle racing too of course but again--who's on the back of the motorcycle?
A long-held concern is that we'll get rid of all the old jobs, but no new jobs will be created. My point is that even in an economy in which all physical needs are provided for free, there will still be plenty of reasons to create new jobs.
All they need to do is develop ex medical solution that keeps you healthy and last I looked no one made money feeling or be able to detect your emotions.
Many economists say there is no need to worry. They point to how past major transformations in work tasks and labor markets – specifically the Industrial Revolution during the 18th and 19th centuries – did not lead to major social upheaval or widespread suffering. These economists say that when technology destroys jobs, people find other jobs. ...
As one economist argued: "Since the dawn of the industrial age, a recurrent fear has been that technological change will spawn mass unemployment. Neoclassical economists predicted that this would not happen, because people would find other jobs, albeit possibly after a long period of painful adjustment. By and large, that prediction has proven to be correct." ...
These economists are definitely right about the long period of painful adjustment! The aftermath of the Industrial Revolution involved two major Communist revolutions, whose death toll approaches 100 million. The stabilizing influence of the modern social welfare state emerged only after World War II, nearly 200 years on from the 18th-century beginnings of the Industrial Revolution. ...
Today, as globalization and automation dramatically boost corporate productivity, many workers have seen their wages stagnate. The increasing power of automation and artificial intelligence technology means more pain may follow. Are these economists minimizing the historical record when projecting the future, essentially telling us not to worry because in a century or two things will get better?
The author's answer is, WE SHOULD WORRY, because the adjustment already is and in the near term will continue to be painful. It's hard to disagree with him.
This article brings to mind John Maynard Keynes's most famous quote: "Economists set themselves too easy, too useless a task, if in tempestuous seasons they can only tell us, that when the storm is long past, the ocean is flat again."[1]
For a long time, I complained that the US couldn't afford such a thing, but we've been running a huge deficit for decades, and no one seems to care, so why should we let a budget hold us back from things like a living wage and free higher education and a single-payer insurance system and any other entitlement you want to tack on?
It's funny, because I lean liberal and I don't. I think there will be plenty of jobs for a long time. I don't want to put words in your mouth, but is it possible what you're saying reflects some sort of bias you have about the inherent value of people who are less _____ than you? Can you say a little bit about what these supposedly useless/unmarketable people are like, what they have spent their time doing and why their skills have no value?
My hypothesis is different than yours: The pendulum in the employer class has swung towards authoritarian control over the last 50 years. Employers have tried to organize both their business and society so that employees either have fewer choices, or believe they have fewer choices.
Employers have organized things such that people are afraid to leave their jobs, through norms like employer-provided health insurance. They have pushed important work like customer service into an "hourly" category, which is considered unskilled, where wages can remain stagnant.
These norms benefit all employers, and can't easily be "shorted" by a single organization paying their employees their true value. To extract the true value from those workers you'd need to have a continually expanding management apparatus to provide space for those employees to grow into, and you run into scaling issues. Therefore the mispriced labor can only be "shorted" by a federated system of many businesses making the play together.
But the pendulum is swinging back. People are waking up to the idea that if they can contract themselves out to multiple clients, they can have leverage to extract their actual market rate from employers.
And software will eventually provide the market for the federated play I describe.
The people who are living this lifestyle ("gig"-only, or multiple part time jobs) seem to dislike it. Their income is unpredictable and highly variable, while their expenses are consistent and difficult to meet.
Please note that I'm not talking about software development consultants - I'm talking about Uber drivers and service industry hourly workers.
I'm puzzled by the question. Aren't we already seeing it? When it takes a 4-year degree to do an office clerk's job, doesn't this already indicate that people who can't get a degree are being phased out of the job market? If the employers with menial positions have enough selectivity in the job pool to weed out people without a degree -- for jobs that do NOT require them -- doesn't that, in itself, show that we're running out of jobs for people? I'm being simplistic for the sake of brevity, but it seems to me that there's a whole class of people (non-college-grads) that are being weeded out of the job market as we speak.
I think the old systems of employment have mostly run their course and are on the decline, but new systems based on loosely federated networks of independent contractors will replace them. Right now the tools are weak, but will get better. So we're in a kind of gap right now, where the old system has optimized on a local maximum that is leading to unemployment, but the new system hasn't quite formed to take over those employees. Increased unemployment will increase the pressure for the tools to stand up.
How many people has Google Apps for Business already put out of work? I've "gigged" setting up this sort of thing for small companies with Linux, and continue to do it for my medium-sized church. However, this market has changed DRAMATICALLY within the past 5 years. My church payed the money to switch. Even the Fortune 150 I work for has now outsourced its email and messaging, from in-house Notes, to hosted Office365. That represents HUNDREDS of internal admins that are simply no longer needed in one of the largest manufacturers in the country. Even someone with experience and specialized knowledge in basic networking and services (email/file & print/DNS/DHCP) will have a hard time finding "gigs" going forward with Google Apps and Office365 in the market. It's just too easy! A kid who grew up with a smartphone is all you need to set it up.
We, as a society, are already leveraging the efforts of a relative tiny few to alleviate the work of thousands, even hundreds of thousands, just in IT services, and this grows exponentially by the year. Twenty years ago, the trade mags were raging on "cloud" everything. It took 10 years to figure out that it just meant "someone else's computer." Now it means a web portal to configure everything you need and take your money.
I'm also good at hardware (microcontroller-based machines) and business domain programming. Sure, I figure I'll have work until I'm dead, but I see what's happening with Arduinos and Raspberry Pis, and the "internet of things." I know it's crap, but when Joe Blow can buy the junk from Amazon, throw it in his home and office, and it does essentially what it's supposed to do, what's the difference to him? That trend is also going to continue to scale out so that more and more things that would have taken custom hardware and software will be recognized as a market need, and product-ized.
At each step, more and more people just aren't needed to make what everyone wants or needs, because of market dynamics like the parent alludes to: all home appliances being made by less than a handful of companies. All entertainment being copyrighted by 4 companies. All IT being controlled by about 6. All cell services in the hands of (essentially) 2. All broadband in 4. On and on it goes. All endeavors trend towards monopoly in a capitalistic system, and I don't see any way for the US government to reign that in at this point, with regulatory capture being operative phrase of the day.
I'm glad someone's hopeful about how my grandchildren are going to find their place in society. I'm holding my breath.
You suggested that the commenter might be biased against those people and feel above them. Those who oppose a living wage for everyone are the ones who tend to have this attitude because they believe that poverty is the individual's fault.
I'd argue outsourcing and austerity are to blame.
Take a look at your infrastructure once in a while. It's crumbling and it's not robots who are rebuilding it, it's nobody.
And that company that builds that iPhone of yours? Uses a LOT of very fleshy robots.
And Tesla, which just built a single factory designed to produce 500,000 vehicles a year with ONLY robots? People are watching. If this succeeds (and I suspect it will), there's going to be a LOT more attempts at such a thing.
Tesla uses far more workers than NUMMI employed to build far fewer cars. In 1985, its first full year of production, NUMMI had 2,470 employees and produced 64,764 vehicles — about 26 vehicles per worker per year. By 1997, it had 4,844 workers and produced 357,809 vehicles — about 74 vehicles per worker per year.
Tesla, on the other hand, had between 6,000 and 10,000 workers in 2016 and manufactured 83,922 vehicles. That puts its vehicle-per-worker number between 8 and 14, about one-seventh the efficiency of NUMMI at its peak.
http://www.autonews.com/article/20170611/OEM01/170619951/tes...
I like Tesla, but Elon often talks about the future like it exists today. One day they'll be able to completely automate that factory, but first they need to become as efficient as the other auto manufacturers that already exist today.
Automation of production is something that all of auto manufacturers have pushed... and Tesla is far behind on that metric. I would not assume they get to full automation before Ford, GM, Toyota, etc.
Aren't they opposites?
> For a long time, I complained that the US couldn't afford such a thing, but we've been running a huge deficit for decades, and no one seems to care,
One mistake that people make is believing that no deficit is a good thing. Running a deficit is like borrowing to fund college.
Another is that money is worth something. If as a generation all the wealth was owned by rich retired unproductive people, but the only way that food is generated is from poor working people, the elderly are screwed. Savings are not value, they are a hope that the future generations as a whole will be kind enough to work for old people based only on how much old people did when they were productive.
If future generations decide they will not pay any attention to old people's money, what happens then? What do I care that in 1960 you worked 60 hour weeks saving $500,000 in assets. Why shouldn't my generation decide "lets reduce the value of the dollar to 10% of it's current value, and lets put a massive tax on assets to redistribute to working people (say swapping income tax for an asset tax). That way I only owe rich people 1 year's salary in debt, not 10, my wage would simply have a zero added to the end (as would prices). The tax relief would mean a real wage increase too.
> Aren't they opposites?
Not OP, but not necessarily. Basically, Libertarians strongly support civil liberties in many forms. That can mean supporting drug legalization (liberal) and also gun rights (conservative). It can also mean supporting an individual's right to discriminate in private business transactions (men's or all-white golf clubs, hetero-only bakeries) which would likely be viewed as a really conservative social stance.
Many Libertarians also strongly support States' Rights, which tend to lean conservative.
>Aren't they opposites?
It's sad that everything about American politics casts every issue in black and white, either/or, polar-opposites mentality, and must be labelled. I was just trying to "get on paper" for context with my post.
> Running a deficit is like borrowing to fund college.
That may have been the most unfortunate example you could have used, since so many people are not seeing the value of their education in the marketplace pay off that investment in a reasonable period. Regardless, I understand what you're saying, but if you're trying to tell me that the deficit that the US federal government consistently runs is a planned, strategic effort, I'm not buying it. ;-)
Having said that, government debt is a little different from personal or business debt for a couple of major reasons:
1. Government's don't die. If they do, debt is the least of your worries.
2. Governments can control inflation and print money.
Having said that, I'm not agreeing our amount of debt is good or even ok, it's just there are different things to consider and it isn't really analogous to personal or business debt.
You just rediscovered something that Marxists had concluded long ago: class wealth is a product of political control, not the other way around. If a class of people takes control of government, they can rewrite laws and the economic system to favor them, instead of the previous ruling class. That's the whole basis of the so-called "dictatorship of the proletariat".
Reducing th3 dollar's value will me1n that your imports will be ten times more expensive.. And for the tax, it will mean that all the people who can invest abroad instead of in the US will do it before the tax is applied..
That's not an economy that looks like it has permanently automated away net 4% jobs.
That's an economy that looks like it automated away some jobs, created more new jobs, but companies can't find workers to fill them.
>Many economists say there is no need to worry ... specifically the Industrial Revolution during the 18th and 19th centuries – did not lead to major social upheaval or widespread suffering.
but economists don't foresee job losses in their own ranks
The optimist in me says that there's been a period of 70 years or so without major wars between different state powers, and maybe MAD will hold up -- but in the theme of prologues repeating themselves...
However, the rest of the things on the list are pretty much guaranteed.
If there's a threat on the horizon, it's probably the one we discuss the least right now.
1: https://en.wikipedia.org/wiki/The_Eighteenth_Brumaire_of_Lou...
> This time is different
If economists said that as a rule, automation improvements lead to better jobs for horses you would call them crazy. But when they say it as a rule for humans it suddenly makes total sense. Horses were replaced by cars. Humans can be replaced by AI.
Is it?
What happened in the past was that people stopped being handy with a scythe and started getting handy fixing tractors. That's "replacing human labor entirely" in the exact same way that replacing a welder with someone who can program a welding robot is.
How is this different? The supposed AI which is better than every human at every task is sci-fi. If we're talking about real AIs and real robots we're just not there, and strides in that direction continue to be spotty, as they have been for the last 30 years.
What happened is that people stopped being farmers. The number of farmers per capita dropped by something like 95%, compared to a century ago.
People stopped being farmers and were able to pursue other industries. But there were still farmers, and a whole support economy of new jobs to support the new method of farming. AI is going to be the farmer driving the tractor, the person who repairs the tractor, the system that manufactures new tractor parts, the person that drives it, a large amount of the support designing new tractors (even if there were 100 competing brands of tractors there are a limited number of people required for research and design and maintenance of the automation (not enough to replace 2.1 million farmers, let alone the millions of factory workers, mechanics, and delivery people), and the AI can even do most of that themselves too soon).
Today, if I have no modern job, where can I procure food? The land is all spoken for. I can't just start a farm in someone's backyard.
We should also listen to ecologists.
I have a legitimate question about the usefulness of this statistic.
Everyday living has dramatically improved since the 1970s. There's been massive improvements in medical care, for starters. Communication and finding information has never been better thanks to computers, the Internet, search engines, wikipedia, etc. Entertainment has never had more options, and the options have never been this good. There's smartphones, of course: They made the internet accessible everywhere, and provide a platform for some incredibly useful apps (e.g., Uber/Lyft). There's been dramatic innovation since 1970, including: microwaves, GPS, home video, text messaging, MRIs, HIV treatment, smoke detectors, and on and on and on.
These technological improvements are not reflected in income statistics, but they're extremely important to every-day living.
Are there statistics that try and encapsulate technological improvement along with the normal measures of economic improvement? It's totally possible that wages are stagnant, but you get triple the utility out of your money than you would've in the 1970s.
Likewise, for those at the bottom of the job market, employers offer far more modest benefits to further your development (like tuition reimbursement), thereby closing yet another door to upskilling and advancement.
Hmm, how is "living standards" defined exactly?
It seems like many of the technological advancements of the last 40 years have become cheap and affordable by almost everyone. Even smart phones are relatively inexpensive and widely used. 64% of individuals making less than $30k a year own a smart phone, for example [1].
Life seems to be getting better for all income levels. Though, this isn't easy to track statistically. The citation I was originally responding to (an article on stagnating income vs inflation [2]) uses CPI as the measure of inflation. But the CPI only tracks the prices of basic commodities like food and fuel. It misses a huge part of our lives.
1. http://www.pewinternet.org/fact-sheet/mobile/ 2. https://fred.stlouisfed.org/graph/?g=eUHS
Nonetheless, the "economist consensus" is that there "is need to worry" and the article linked to is a (well written!) exception to the current groupthink.
This is in spite of the inconvenient fact that labor productivity is down.
As for the "widespread suffering caused by automation" - I don't see ANY evidence that it was caused by unemployment caused by "too much automation". The industrial revolution induced more suffering primarily because of new weaponry that presaged industrial-scale slaughter.
As before, the real job stealer isn't a robot, it's a man in a suit executing a trade deal with countries that still practise literal slavery, blaming your lack of a job paying > 5¢ an hour on a robot.
What is it that's new and so threatening automation-wise that is going to have drastic effect on ppl's lives comparable to the industrial revolution?
As for retail, i got curious, so: "Stores are closing because of the rise of e-commerce and shifts in how people spend their money," said Deb Gabor, CEO of Sol Marketing a brand/retail strategy consultancy. "Shoppers are devoting bigger shares of their wallets to entertainment, restaurants and technology, and spending less on clothing and accessories." [1]
Retail has revenue per employee of ~$450K. Restaurants have about $150K revenue per employee. So money is shifting to the more labor intensive industries, and maybe this is true even while including tech and entertainment.
But this is just a small scale, relatively to what could be coming. And people's spending patterns may not be so supportive of humans.
[1]http://www.houstonchronicle.com/business/columnists/tomlinso...
Ok, fair enough. So then what are those large and fast-paced industries that are under the immediate and severe threat due to automation?
I mean for something to be on par with industrial revolution there has to be a pretty profound impact all over the place, the kind of impact that the existing ways of doing things can't adapt to at all (kind of how retail has been adapting to e-commerce with mixed results over the course of the past 15+ years).
As for what big industries are under immediate threat ? I'm not sure about how rapid, but in almost every industry you look you see a lot of real term deep innovation(from startups) that, if successful would greatly reduce jobs. Think not only automation but electric cars and mechanics and plant based meat and farmers.
That, combined with the fact that we are maybe starting to be able to mechanize the key skills of humans:dexterity and thinking, seem quite a huge change.
And maybe we can adapt - but we don't hear almost any ideas at all for many more new jobs, even with creativity of the net so maybe massive unemployment is a possibility?
It's both, really. For some retailers with brick-and-mortar shops online presence is a way to expand their market, for others - internet is slow death.
But again nothing drastic in terms of job reduction has happened overnight even though the whole "internet thing" changed the way things are done drastically af.
>> And maybe we can adapt - but we don't hear almost any ideas at all for many more new jobs...
It's hard to predict these things. But again I'm not buying the "zomg there will be riots" spiel. If anything there will be a whole set of new (or subset of existing) industries along the lines of how adoption of the internet-based tech spawned them.
Let's think about the Internet(in 1994) but in the abstract:
It was a new medium. New mediums create new advertising and content industries around them. So did the Internet. They also create many jobs to manage/improve the technical stuff - That's also true.
It's also a great distribution channel for software and goods. we've known this pretty early. and new distribution channels reduce a business cost thus increasing demand. They also create changes along the whole supply chain, so that's more work.
But the surprising fact about the net is it's ability to enable businesses targeting very tiny niches. But if you talked to a many direct marketing guys back in the day, before the net and explained to them about the technical possibilities of the internet(and moore's law and powerful computing) and let them think about it, one in a 10-100, would have told you: this could be great for small targeted businesses.
Or maybe that guy is working in prediction, or science fiction, etc.
But today we have blogs,and media business who would kill for an eyeball, so we should be able to hear from that smart guy .
So why aren't we hearing stories about all those possible new jobs ?
you tell me :) I'm not 100% sure why the automation narrative has primarily been of the "impending doom" variety.
I think just like with the internet - even if automation does take off big time there will be some "victims" but also new possibilities from the obvious ones - like manufacturing, supporting/updating, installing the "robots" to some things we (speaking for myself lol) can't envision until this actually happens.
This is starting to remind me of peak oil, where it's become an article of faith among some people that there's a disaster right around the corner, and no amount of failed predictions or facts pointing to the contrary are going to convince them otherwise.
https://qz.com/666311/why-americas-impressive-5-unemployment...
I think the problem isn't that automation kills jobs, it's that nobody cares about people who worked in these jobs for >10 years and don't know anything other.
I'm not saying we should allow everyone to become a pilot or medical doctor after working 30 years as a trucker, but I think we should at least try to work in this direction.
revive me in 2099 if i was correct.
1. The workers then were displaced in many (most?) cases by newly invented physical machines. Those machines had to be built in brand new factories, employing many workers. Building those machines greatly boosted the need for many resources, such as metals, lubricants, fuels, and wood, so indirectly created many jobs in those industries. The machines needed frequent and regular maintenance, spawning even more jobs and industries.
What's going on now seems different. Our automation nowadays involves much less new physical machines. As an example, consider self driving big rig trucks, which some think will be the first self driving vehicles to displace a large number of workers.
What is the difference between a self driving big rig and a regular big rig? The self driving one is essentially a regular one with some additions:
• some sensors,
• some motors or actuators to operate controls,
• one or more computers that run the driving software, and handle the interface with the human who is in charge,
• maybe some kind of wireless communication system so the driving software can receive orders from the trucking company if that's not handled by a human who accompanies the truck.
None of these things are new. What's new is putting them in a truck.
Let's focus on the computers. Let's be optimistic and assume that every single new big rig that comes off the assembly line is equipped from the factory for self driving, so every single one needs a computer. Let's further assume that they use a high end GPU to run the AI software.
Putting a high end GPU in every single new big rig in the US would raise total annual demand for high end GPUs by about 10%. That's not "build a new factory" territory. That's "authorize a little overtime at the GPU factory" territory.
It's similar for other things. Each big rig now has a few electric motors so the AI can steer and shift gears and brake...but as with GPUs that's just a small bump in the total demand.
Worse, many of these things, like the GPUs, the sensors, probably the motors, and probably the communications system are built in highly automated factories, so even if the factories have to kick up production with overtime or even add new shifts, it will be machines getting the new work.
2. During the Industrial Revolution we had large, rapidly expanding frontiers.
If you were in the developed part of your country and lost your job to a machine there was a good chance you could go to the frontier and find a similar job that was safe from the machines because the frontier lacked the infrastructure to support the machine.
Our frontiers now, to the extent that we even have any, are small and not expanding much if at all, and we've gotten a lot better at getting our machines there.
Unhelpful and counterproductive rhetoric.
This grand vision was unsustainable and led to millions of deaths.
Universal basic income is no less unsustainable (even government retirement and pension funds, which require much less money, are struggling everywhere). I really can hope that this false promise will not lead to similar catastrophes.
I am Russian, I went to a Soviet school, and I am quite aware what was and what wasn't a selling point for Communism. It was -- down to being mentioned in every socialist realism book, in nearly every major TV report, even as a running joke. There was a line in a popular Soviet movie: "no more worries, no more rat race, robots are toiling, man is happy" -- and this line was taken at a face value, as a defining feature of the future Communist utopia we were all building.
There's no reason that UBI can't be sustainable. And last time I checked, retirees are still receiving their social security checks.