>Technological progress won’t create mass unemployment ... In Krugman’s celebrated example, imagine there are...
Stopped reading right there. Hypothetical scenarios are not a replacement for empirical evidence. I can just as easily imagine a scenario that runs contrary to the one given.
This kind of baseless speculation cheapens economics as a science and is why a lot of people don't take macroeconomics seriously.
Yes, that's the conventional wisdom from economists.
The real problem is simple: there are a lot of people now, and more in the future, who cannot generate enough value by working to be self-supporting, let alone have a good life. They're competing with machines which are better than they are.
The conventional wisdom used to be that more education would fix this. That failed. Unemployment and underemployment of college graduates is high and growing. Trade schools like ITT, which went out of business yesterday, were even less effective. The number of educated people needed is limited, and we have more than enough.
There was the assumption that new kinds of jobs would appear. They did, but not in large numbers. Information technology is only 1.8% of US employment, down from 2.2% ten years before. The only big growth area is health care and social assistance, which is a consequence of an aging population.
Where this takes us, under standard economic assumptions, is the favela model - giant slums where most people are just surviving, with a small core group that's doing well. This is what many third world countries look like. The US is headed there, but is more spread out - the decaying slums are far from the successful cities. (That's where Trump voters come from. They're angry, and they found a demagogue. That's common worldwide, and seldom ends well.)
Remember, capitalism is not optimizing for individual income. The fitness function is return on capital.
Alternatives? Welfare was tried from 1960 to the mid-1980s; it ended up with third generation welfare mothers in giant housing projects. Having a large non-working group of poor people leads to a dysfunctional society. Higher minimum wage? Worth a try; the places that are doing it now should be monitored to see who gets laid off. Basic income? Same problem as 60s welfare. WPA-style government as employer of last resort? Maybe; Saudi Arabia does that, and they have a massive number of useless workers.
Perhaps machine learning can also disrupt economists and politicians and come up with better theories, models and solutions how to best re-allocate labour, capital and education for common good :)
Societies in general are quite adaptive, but with robotics/ML one issue is that the shift will potentially happen within one generation - it's harder to retrain/adjust adults.
The real problem is simple: there are a lot of people now, and more in the future, who cannot generate enough value by working to be self-supporting, let alone have a good life. They're competing with machines which are better than they are.
Ok, where's my Robo-maid who's better than any human? How about my Robo-Taco-Truck, my robot laundry guy, and my robot driver? As other nations demonstrate, these are all things that humans could do.
People aren't competing with machines, or with each other. Valuable work is just not getting done for a variety of reasons: people prefer to live off welfare/disability fraud, people are afraid of summary execution by the police for doing certain valuable jobs, etc.
Technological unemployment may happen some day. But it hasn't happened yet.
>Technological unemployment may happen some day. But it hasn't happened yet.
That's the crux of the matter: Whether "some day" happens sooner rather than later. Also, what to do when "some day" comes.
You know it's coming, there's big money and serious effort into making things like Robo-Maid, Robo-Taco-Truck, Robo-laundry-guy, Robo-Driver, etc., a practical reality. We know that Robo-Burger-Maker is rather real today: http://www.eater.com/2016/7/1/12077990/robot-burgers-san-fra....
Your blanket statement regarding on why "valuable work is just not getting done" is overall a side-issue. Welfare may be gaming a system. Disability fraud is certainly criminal. Let's add to that: Living off a pension; Born a "trust-fund baby"; Fully retired on investments. There are plenty of other reasons that certain somebodies do not seek employment income, and perhaps do not have to.
People have good reason to fret about the availability, attainability and retain-ability of gainful employment. There's plenty of history and recent events to make such fretting reasonable (e.g. http://www.usnews.com/news/articles/2016-02-04/what-happened...).
If you read the statistics cited in your usnews article you'll see they aren't disagreeing with me. According to them the issue is that there are lots of people who are "neither employed nor actively looking for work".
That's what I'm saying too, in addition to the fact that many folks could find work if we didn't illegalize the work they would do (e.g. hair braiding, selling tacos). There are many reasons why people don't work today, but being replaced by machines is not one of them. So whatever solutions we implement today should reflect the problem of today: useful work isn't getting done.
If we ever reach the future where I have a Robo-Maid and where Robo-Burger-Flipper is more than just a gimmick, then we can come up with a solution tailored to tomorrow's problems.
Similarly, if you are fat eat less. If you are underweight eat more. An obese person eating more because some day they might be underweight is really just making excuses for gluttony.
>According to them the issue is that there are lots of people who are "neither employed nor actively looking for work".
I wonder if we're talking past each other. Perhaps robots aren't the top reason for the apparent worldwide decline in labor participation (http://data.worldbank.org/indicator/SL.TLF.CACT.ZS ). But robots have brought economic dislocation, especially in the manufacturing sector. The worry has been out for a long time on whether displaced workers can find other gainful employment in reasonable time (http://www.bls.gov/opub/mlr/1982/09/art2full.pdf ). Note this is not about getting merely any job, but a job that can bring a satisfying life. As the automation has grown more capable, so has this concern (https://hbr.org/2014/12/what-happens-to-society-when-robots-... ).
Consider the full quote regarding the "neither employed nor actively looking for work"[1]: It critiques the official unemployment rate as shrouding a dysfunctional employment scene. Gallup says: "The official unemployment rate, which cruelly overlooks the suffering of the long-term and often permanently unemployed as well as the depressingly underemployed, amounts to a big lie."
The usnews article also cites other stats on wage gains, shrinking middle class, growing low-income, etc., all to illustrate its main point: "The Pew study considers a widening wealth gap to be driving the wage growth disparity, as more and more workers find themselves at the extremes of the income spectrum. And this gap is generated in part by the unavailability of what Gallup considers to be good jobs."
This pressure for "good jobs" has brought movements such as raising the minimum wage or implementing universal basic income. Invariably this leads to questions on whether the wages for low-skill jobs should ever exceed their apparent economic worth in the name of a "living wage". From there inevitably robots are brought into the picture (limited, gimmicky, or otherwise), if for no other reason than to suggest that low-skill workers may price themselves out of a job.
Putting it all together - robots on top of widening wealth gap on top of shrinking middle class... - well people worry, you know?
====
[1] The full quote is: Gallup also is among a host of organizations and individuals to criticize some of the government's statistical methodologies in recent years. The national unemployment rate of 5 percent, for example, doesn't keep track of the more than 94 million Americans who are neither employed nor actively looking for work. And while at least some of those not counted in the labor force are either in school or retired, there's also a sizable portion that simply hasn't actively looked for a job in the last four weeks. People who give up looking for employment altogether aren't actually counted as unemployed, even though they may have been without a job longer and in greater need than anyone who's regularly sending resumes out.
Robots may bring about dislocation in specific industries, but that's just a disingenuous way of saying "people refuse to change industries or accept jobs they don't like". That's fine, but that's not robots taking all the jobs - just the specific job that specific people want.
Gallup is also being disingenuous when it uses words like "permanently unemployed" to refer to people not seeking work.
Invariably this leads to questions on whether the wages for low-skill jobs should ever exceed their apparent economic worth in the name of a "living wage".
If only there were some sort of Earned Income Tax Credit that could provide less productive people subsidies to maintain a wealthy lifestyle [1] while also giving them incentive to work...
Unlike minimum wages and BI, EITC doesn't exacerbate the problem of people refusing to work and it costs a lot less too. It won't solve technological unemployment, but it will solve your "satisfying life" concern as well as my "people refuse to contribute to society while making demands on it" concern.
[1] I refuse to use the term "living wage", since billions of my fellow humans (luckily on the other side of an imaginary line, so we can ignore them) earn far less than this without dying.
Why not teach people skilled trades and crafts instead of sending them to get a STEM degree? That will keep people busy while UBI keeps them fed, housed, and healthy.
I'd like to offer some quotes from the book "Economics in one lesson", a book from the 40s that is still very relevant today. Specifically, from the chapter called "The curse of machinery":
Among the most viable of all economic delusions is the belief that machines on net balance create unemployment.
Let us turn to Adam Smith’s Wealth of Nations. The first chapter [...] is called “Of the division of labor,” and on the second page of this first chapter the author tells us that a workman unacquainted with the use of machinery employed in pin-making “could scarce make one pin a day, and certainly not twenty,” but with the use of this machinery he can make 4,800 pins a day. In the pin-making industry there was already, if machines merely throw men out of jobs, 99.98 percent unemployment.
Arkwright invented his cotton-spinning machinery in 1760. At the time it was estimated that there were in England 5,200 spinners using spinning wheels, and 2,700 weavers - in all 7,900 persons engaged in the production of cotton textiles. The introduction of Arkwright’s invention was opposed on the ground that it threatened the livelihood of the workers, and the opposition had to be put down by force. Yet in 1787 [...] the number of persons actually involved in the spinning and weaving of cotton had risen from 7,900 to 320,000, an increase of 4,400 percent.
[...] Technophobes will assert: “That may have been all very well in the past, but today conditions are fundamentally different; and now we simply cannot afford to develop any more labor-saving machines.”
If it were indeed true that the introduction of labor-saving machinery is a cause of constantly mounting unemployment and misery, the logical conclusion to be drawn would be revolutionary, not only in the technical field but for our whole concept of civilization. Not only should we have to regard all further technical progress as a calamity; we should have to regard all past technical progress with equal horror.
There is also an absolute sense in which machines may be said to have enormously increased the number of jobs. The population of the world today is four times as great as in the middle of the 18th century, before the Industrial Revolution had got well under way. Machines may be said to have given birth to this increased population; for without the machines, the world would not have been able to support it. Three out of every four of us, therefore, may be said to owe not only our jobs but our very lives to the machines.
Greater abundance devalues the relative productivity of labour to machinery. But in the contemporary case, there is not enough demand (ie. low/no economic/population growth in the rich world) to valorise the greater abundance, hence the relative devaluation of labour is to a much greater extent.
In the same way that computers can beat the best human chess and go players, machines are beating a large (and ever increasing) proportion of human labour. This is lost ground. We will end up doing labour for 'fun' (ie. non-economic ends) since we cannot compete on economic grounds.
But in the contemporary case, there is not enough demand
This is a strong assertion to be made, especially in general like that. There is not enough demand of what, specifically? I agree that current interventionist economic policies are restricting growth, but to conclude from this that "there's not enough demand" is incorrect in my opinion.
machines are beating a large (and ever increasing) proportion of human labour
Just like the cotton-spinning machinery did in the past. We could not compete with it either, and yet the result has been more productivity, hence more jobs, better wages and the ability to support a larger world population.
I think a big part of this misunderstanding comes from the idea that we are competing with the machines when we are in fact making use of them.
This is a pretty good theory but it lies on certain implicit assumptions that are becoming untrue. One is that there is abundance of energy and resources to exploit, so there is always more to be done (demand for labor) to transform it. But with automatization, we are coming to a state where limitation of economic activity is not labor, but resource consumption (mainly energy).
Also, what is the definition of job? But what the people worrying about automation are saying, there will not be more jobs that satisfy both of these conditions:
(1) Economic efficiency - the job is required to sustain the same level of production.
(2) Sustenance - the job must be capable to give person doing it enough money so he can live on it and participate in society.
For instance, we can always create jobs that break the condition (1), because we can always have more people being security guards or checking expense reports. Or we can always create jobs that break the condition (2), you know, you can always be a personal servant to somebody for really miserable pay.
I would actually argue that we are well past the point where (1) is true. But since our societies are so obsessed with jobs, we cannot really openly admit it.
I'm not sure if I understood your points correctly. There is always more to be done. There is not a fixed set of things humanity needs that will one day be fulfilled. We didn't know we wanted cars, computers and smartphones until they were invented. This is regardless of scarcity, possibly even because of it (scarcity of resources will drive the need to be more efficient handling them).
I also don't understand your two points since (1) automation increases efficiency and thus production levels and (2) as a result of that increased productivity, there is a raise in wages.
The part you're missing is that a higher wage is given to one person rather than many people, that you need capital to produce (hey, we need more chips, so now everyone with a hammer makes one).
Can you actually make a car? How many people are involved in making a Honda Civic vs Ford Model T?
See the quote above. Automation has historically increased wages AND the number of available jobs.
No, I can't make a car, and have no interest whatsoever in making one. I could, however, buy a Honda Civic today in the US for more or less the same price (adjusted for inflation) as a Model T, but thanks in no small part to automation, it's a much more efficient, safe and comfortable car. And still the auto industry directly employs 1.5 million Americans now vs. 450k in the late 20s.
I don't see your point? Some demand may be limited by personal income, while some is limited by corporate income and so on. This will cause producers to focus on things that are more in demand, while bringing down the price of lower demand products.
But this is just demand/supply dynamics. I don't see how automation is relevant here since that is valid with or without it.
"(2) as a result of that increased productivity, there is a raise in wages."
That stopped years ago. Ask any auto worker.
There's no economic reason that increased productivity should lead to an increase in wages. The benefits of increased productivity accrue to capital, not labor. What produced the increase in wages with productivity was labor activism, and fear of government intervention.
It's hard to imagine now, but there was a time when American companies were seriously afraid of what workers could do, both through unions and politics. Communism was considered a serious competitor to capitalism from about 1930 to 1980. That's what kept wages rising.
Once Ronald Reagan successfully made the case for unbridled capitalism, that ended. Now the US has "just in time" workers, few unions in the private sector, and weak worker protections. Competition between employees keeps wages down.
Of course there are economic reasons for wage increases with productivity. They are the same reasons that explain any other increase in prices.
First, employers don't want to lose their more productive employees, and they have to compete with other employers offering jobs at higher pays. Second, once productivity rises, employers increase their earnings and can therefore hire more. This creates more demand for employees, and the price of work rises.
It's in fact the view that companies should be "afraid" that create regulations and restrictions that are usually what's responsible for reduction of wages and quality of life in general for workers.
Finally, I find it really unbelievable that someone can claim in a serious conversation that the US have anything remotely similar to "unbridled capitalism".
I've often wondered about returning to more rural, subsistence-farm living. I'm very likely naive about a number of factors involved, but with the massive agricultural advances of the last 100 years it seems like setting up a subsistence farm for a moderate sized family/community should be do-able.
The Amish do it fairly well, and they even "handicap" themselves by not employing many of the modern farm automation capabilities.
Frankly, I think machinery isn't necessarily the best explanatory factor. Have we compared it against levels of private indebtedness, concentration of asset ownership, labor-union density, and aggregate demand as factors in the behavior of the economy broadly and labor markets in specific?
The article is blithe about how we overcome economic dislocation. During the first industrial revolution, farmers were made obsolete and had to crowd into cities looking for wages. Children worked in factories and poorhouses. I don't know if this was worse than farm life, but the art of the period depicted it as such.
Some people never recover from economic dislocation. They either abstain from having children because they sense how marginal they are, or they have families anyways and raise children with few opportunities.
I am using genetic propagation as a measure of human success, which of course is arguable, but which means a lot to many people. More importantly if a human line dies out, there is no voice for it anymore. Such losses are silent and largely ignored by the historical record. So this economist is gravely undercounting the human suffering that technological change can bring.
Saying that farmers were made obsolete kind of glosses over the real reason that they were forced into cities: enclosure. Land that they were previously free to use for subsistence farming was taken and combined into larger landholdings. Had they been properly compensated for the loss of the land, they would have been less desperate to rush to cities.
34 comments
[ 3.8 ms ] story [ 34.2 ms ] threadStopped reading right there. Hypothetical scenarios are not a replacement for empirical evidence. I can just as easily imagine a scenario that runs contrary to the one given.
This kind of baseless speculation cheapens economics as a science and is why a lot of people don't take macroeconomics seriously.
First happens because sociological data collection is hard, slow and expensive. Second usually due to various kinds of politics and lobbying.
The real problem is simple: there are a lot of people now, and more in the future, who cannot generate enough value by working to be self-supporting, let alone have a good life. They're competing with machines which are better than they are.
The conventional wisdom used to be that more education would fix this. That failed. Unemployment and underemployment of college graduates is high and growing. Trade schools like ITT, which went out of business yesterday, were even less effective. The number of educated people needed is limited, and we have more than enough.
There was the assumption that new kinds of jobs would appear. They did, but not in large numbers. Information technology is only 1.8% of US employment, down from 2.2% ten years before. The only big growth area is health care and social assistance, which is a consequence of an aging population.
Where this takes us, under standard economic assumptions, is the favela model - giant slums where most people are just surviving, with a small core group that's doing well. This is what many third world countries look like. The US is headed there, but is more spread out - the decaying slums are far from the successful cities. (That's where Trump voters come from. They're angry, and they found a demagogue. That's common worldwide, and seldom ends well.)
Remember, capitalism is not optimizing for individual income. The fitness function is return on capital.
Alternatives? Welfare was tried from 1960 to the mid-1980s; it ended up with third generation welfare mothers in giant housing projects. Having a large non-working group of poor people leads to a dysfunctional society. Higher minimum wage? Worth a try; the places that are doing it now should be monitored to see who gets laid off. Basic income? Same problem as 60s welfare. WPA-style government as employer of last resort? Maybe; Saudi Arabia does that, and they have a massive number of useless workers.
Somebody had better think of something.
..and someone did think of something:
http://userpage.fu-berlin.de/~roehrigw/kennedy/english/
"Information technology is only 1.8% of US employment, down from 2.2% ten years before."
Isn't this caused by globalization and outsourcing?
Well said. :)
Societies in general are quite adaptive, but with robotics/ML one issue is that the shift will potentially happen within one generation - it's harder to retrain/adjust adults.
Ok, where's my Robo-maid who's better than any human? How about my Robo-Taco-Truck, my robot laundry guy, and my robot driver? As other nations demonstrate, these are all things that humans could do.
https://www.chrisstucchio.com/blog/2016/robots_didnt_take_ou...
People aren't competing with machines, or with each other. Valuable work is just not getting done for a variety of reasons: people prefer to live off welfare/disability fraud, people are afraid of summary execution by the police for doing certain valuable jobs, etc.
Technological unemployment may happen some day. But it hasn't happened yet.
That's the crux of the matter: Whether "some day" happens sooner rather than later. Also, what to do when "some day" comes.
You know it's coming, there's big money and serious effort into making things like Robo-Maid, Robo-Taco-Truck, Robo-laundry-guy, Robo-Driver, etc., a practical reality. We know that Robo-Burger-Maker is rather real today: http://www.eater.com/2016/7/1/12077990/robot-burgers-san-fra....
Your blanket statement regarding on why "valuable work is just not getting done" is overall a side-issue. Welfare may be gaming a system. Disability fraud is certainly criminal. Let's add to that: Living off a pension; Born a "trust-fund baby"; Fully retired on investments. There are plenty of other reasons that certain somebodies do not seek employment income, and perhaps do not have to.
People have good reason to fret about the availability, attainability and retain-ability of gainful employment. There's plenty of history and recent events to make such fretting reasonable (e.g. http://www.usnews.com/news/articles/2016-02-04/what-happened...).
That's what I'm saying too, in addition to the fact that many folks could find work if we didn't illegalize the work they would do (e.g. hair braiding, selling tacos). There are many reasons why people don't work today, but being replaced by machines is not one of them. So whatever solutions we implement today should reflect the problem of today: useful work isn't getting done.
If we ever reach the future where I have a Robo-Maid and where Robo-Burger-Flipper is more than just a gimmick, then we can come up with a solution tailored to tomorrow's problems.
Similarly, if you are fat eat less. If you are underweight eat more. An obese person eating more because some day they might be underweight is really just making excuses for gluttony.
I wonder if we're talking past each other. Perhaps robots aren't the top reason for the apparent worldwide decline in labor participation (http://data.worldbank.org/indicator/SL.TLF.CACT.ZS ). But robots have brought economic dislocation, especially in the manufacturing sector. The worry has been out for a long time on whether displaced workers can find other gainful employment in reasonable time (http://www.bls.gov/opub/mlr/1982/09/art2full.pdf ). Note this is not about getting merely any job, but a job that can bring a satisfying life. As the automation has grown more capable, so has this concern (https://hbr.org/2014/12/what-happens-to-society-when-robots-... ).
Consider the full quote regarding the "neither employed nor actively looking for work"[1]: It critiques the official unemployment rate as shrouding a dysfunctional employment scene. Gallup says: "The official unemployment rate, which cruelly overlooks the suffering of the long-term and often permanently unemployed as well as the depressingly underemployed, amounts to a big lie."
The usnews article also cites other stats on wage gains, shrinking middle class, growing low-income, etc., all to illustrate its main point: "The Pew study considers a widening wealth gap to be driving the wage growth disparity, as more and more workers find themselves at the extremes of the income spectrum. And this gap is generated in part by the unavailability of what Gallup considers to be good jobs."
This pressure for "good jobs" has brought movements such as raising the minimum wage or implementing universal basic income. Invariably this leads to questions on whether the wages for low-skill jobs should ever exceed their apparent economic worth in the name of a "living wage". From there inevitably robots are brought into the picture (limited, gimmicky, or otherwise), if for no other reason than to suggest that low-skill workers may price themselves out of a job.
Putting it all together - robots on top of widening wealth gap on top of shrinking middle class... - well people worry, you know?
==== [1] The full quote is: Gallup also is among a host of organizations and individuals to criticize some of the government's statistical methodologies in recent years. The national unemployment rate of 5 percent, for example, doesn't keep track of the more than 94 million Americans who are neither employed nor actively looking for work. And while at least some of those not counted in the labor force are either in school or retired, there's also a sizable portion that simply hasn't actively looked for a job in the last four weeks. People who give up looking for employment altogether aren't actually counted as unemployed, even though they may have been without a job longer and in greater need than anyone who's regularly sending resumes out.
Gallup is also being disingenuous when it uses words like "permanently unemployed" to refer to people not seeking work.
Invariably this leads to questions on whether the wages for low-skill jobs should ever exceed their apparent economic worth in the name of a "living wage".
If only there were some sort of Earned Income Tax Credit that could provide less productive people subsidies to maintain a wealthy lifestyle [1] while also giving them incentive to work...
Unlike minimum wages and BI, EITC doesn't exacerbate the problem of people refusing to work and it costs a lot less too. It won't solve technological unemployment, but it will solve your "satisfying life" concern as well as my "people refuse to contribute to society while making demands on it" concern.
[1] I refuse to use the term "living wage", since billions of my fellow humans (luckily on the other side of an imaginary line, so we can ignore them) earn far less than this without dying.
Among the most viable of all economic delusions is the belief that machines on net balance create unemployment.
Let us turn to Adam Smith’s Wealth of Nations. The first chapter [...] is called “Of the division of labor,” and on the second page of this first chapter the author tells us that a workman unacquainted with the use of machinery employed in pin-making “could scarce make one pin a day, and certainly not twenty,” but with the use of this machinery he can make 4,800 pins a day. In the pin-making industry there was already, if machines merely throw men out of jobs, 99.98 percent unemployment.
Arkwright invented his cotton-spinning machinery in 1760. At the time it was estimated that there were in England 5,200 spinners using spinning wheels, and 2,700 weavers - in all 7,900 persons engaged in the production of cotton textiles. The introduction of Arkwright’s invention was opposed on the ground that it threatened the livelihood of the workers, and the opposition had to be put down by force. Yet in 1787 [...] the number of persons actually involved in the spinning and weaving of cotton had risen from 7,900 to 320,000, an increase of 4,400 percent.
[...] Technophobes will assert: “That may have been all very well in the past, but today conditions are fundamentally different; and now we simply cannot afford to develop any more labor-saving machines.”
If it were indeed true that the introduction of labor-saving machinery is a cause of constantly mounting unemployment and misery, the logical conclusion to be drawn would be revolutionary, not only in the technical field but for our whole concept of civilization. Not only should we have to regard all further technical progress as a calamity; we should have to regard all past technical progress with equal horror.
There is also an absolute sense in which machines may be said to have enormously increased the number of jobs. The population of the world today is four times as great as in the middle of the 18th century, before the Industrial Revolution had got well under way. Machines may be said to have given birth to this increased population; for without the machines, the world would not have been able to support it. Three out of every four of us, therefore, may be said to owe not only our jobs but our very lives to the machines.
In the same way that computers can beat the best human chess and go players, machines are beating a large (and ever increasing) proportion of human labour. This is lost ground. We will end up doing labour for 'fun' (ie. non-economic ends) since we cannot compete on economic grounds.
This is a strong assertion to be made, especially in general like that. There is not enough demand of what, specifically? I agree that current interventionist economic policies are restricting growth, but to conclude from this that "there's not enough demand" is incorrect in my opinion.
machines are beating a large (and ever increasing) proportion of human labour
Just like the cotton-spinning machinery did in the past. We could not compete with it either, and yet the result has been more productivity, hence more jobs, better wages and the ability to support a larger world population.
I think a big part of this misunderstanding comes from the idea that we are competing with the machines when we are in fact making use of them.
Also, what is the definition of job? But what the people worrying about automation are saying, there will not be more jobs that satisfy both of these conditions:
(1) Economic efficiency - the job is required to sustain the same level of production.
(2) Sustenance - the job must be capable to give person doing it enough money so he can live on it and participate in society.
For instance, we can always create jobs that break the condition (1), because we can always have more people being security guards or checking expense reports. Or we can always create jobs that break the condition (2), you know, you can always be a personal servant to somebody for really miserable pay.
I would actually argue that we are well past the point where (1) is true. But since our societies are so obsessed with jobs, we cannot really openly admit it.
I also don't understand your two points since (1) automation increases efficiency and thus production levels and (2) as a result of that increased productivity, there is a raise in wages.
Can you actually make a car? How many people are involved in making a Honda Civic vs Ford Model T?
No, I can't make a car, and have no interest whatsoever in making one. I could, however, buy a Honda Civic today in the US for more or less the same price (adjusted for inflation) as a Model T, but thanks in no small part to automation, it's a much more efficient, safe and comfortable car. And still the auto industry directly employs 1.5 million Americans now vs. 450k in the late 20s.
But this is just demand/supply dynamics. I don't see how automation is relevant here since that is valid with or without it.
That stopped years ago. Ask any auto worker.
There's no economic reason that increased productivity should lead to an increase in wages. The benefits of increased productivity accrue to capital, not labor. What produced the increase in wages with productivity was labor activism, and fear of government intervention.
It's hard to imagine now, but there was a time when American companies were seriously afraid of what workers could do, both through unions and politics. Communism was considered a serious competitor to capitalism from about 1930 to 1980. That's what kept wages rising.
Once Ronald Reagan successfully made the case for unbridled capitalism, that ended. Now the US has "just in time" workers, few unions in the private sector, and weak worker protections. Competition between employees keeps wages down.
First, employers don't want to lose their more productive employees, and they have to compete with other employers offering jobs at higher pays. Second, once productivity rises, employers increase their earnings and can therefore hire more. This creates more demand for employees, and the price of work rises.
It's in fact the view that companies should be "afraid" that create regulations and restrictions that are usually what's responsible for reduction of wages and quality of life in general for workers.
See also: http://www.heritage.org/research/reports/2013/07/productivit...
Finally, I find it really unbelievable that someone can claim in a serious conversation that the US have anything remotely similar to "unbridled capitalism".
The Amish do it fairly well, and they even "handicap" themselves by not employing many of the modern farm automation capabilities.
Some people never recover from economic dislocation. They either abstain from having children because they sense how marginal they are, or they have families anyways and raise children with few opportunities.
I am using genetic propagation as a measure of human success, which of course is arguable, but which means a lot to many people. More importantly if a human line dies out, there is no voice for it anymore. Such losses are silent and largely ignored by the historical record. So this economist is gravely undercounting the human suffering that technological change can bring.
Why should the market need the additional capacity? This is where i stopped reading.