How Technology Is Destroying Jobs
technologyreview.com
technologyreview.com
The problem is that technology is being centralized by a few owners. A computer in every pocket, 3d printers and CNC machines are very near to us.
Companies grow bigger and bigger, not because they are more efficient, on the contrary, but because they could black mail society to sustain it. Too big too fail became the motto of the new era, banks and industry companies so inefficient with huge losses that nobody could touch it without breaking society itself.
Those companies had done well, who needs to make cars when you can take $50 billion by the government to speculate in the stock market? A stock market that is fueled by printing dollars by a central bank(inflating and raising taxes of millions people).
Don't make me started with software patents in order for a few to own all ideas. Society is collapsing to sustain unproductive parasites.
It will have to get worse before it gets better.
This essay on four possible futures along the axes of scarcity and equality is highly relevant.
http://jacobinmag.com/2011/12/four-futures/
Does a post-scarcity economy, the sort implied by democratizing production via 3D printers and cheap energy, necessarily imply a more equal society or can the rentier class find a way to extract profits via patents and intellectual property?
Conversely, if we end up constrained by natural resources — energy, feedstocks, labor, etc. — is it even possible to move to a more equal society?
I think in the end, the inventor of a technology is somewhat responsible for figuring out what the people he/she will displace can do.
For example, if I'm about to put a whole industry out of work ('disrupt'ing it), I think it would be reasonably responsible for me to offer training or some kind of conversion for the workers of that industry (so they can use /license the tool I have made).
I'd say, not necessarily.
Think about the impending disappearance of professional driving. Automation may not be total, but it will probably reach a point where a single human for 50+ trucks is enough (just have the truck communicate with its command centre). So basically, the professional driving profession will likely shrink by a factor of 10 to 100.
Now, what's a ex-driver to do? I doubt Google could, nor should, have an answer.
Anyway, the reduction of necessary labour, which technology permits, should be good news. If it's not, that's probably because our societies are ill-equipped to deal with such progress.
If you're going to make an atomic bomb, you should consider the devastation it might cost. While this ISN'T an atomic bomb, if you're going to so heavily disrupt a field such that many people will find themselves out of work, at least giving those people some thought seems to be the responsible choice.
On your last point, societies are always ill-equipped to deal with rapid progress. In fact most of it is dismissed until it can't be ignored. However, the problem of introducing change in a way that will not hurt/severely degrade the life quality of large parts of society is worth tackling -- I don't think people should just throw care to the wind and disrupt areas in which displaced employees are unlikely to be able to bounce back.
Not that you shouldn't release the technology, just that some thought should be given to the effects of your innovation.
I feel in todays world many people are slow to realize that they need to get more technical. It's not like there isn't "jobs" or "work" that needs to be done, it's just smarter, more technical work, with no clear-cut answers due to the ever increasing amounts of data we can collect and process. Think about it like this: "Programming skills in 2010s are the Excel skills of the 1990s".
You are aware you've just posited free will? I think that's such a significant claim that it needs some sort of support. Because things like family education level and income seem highly determining for lifetime outcomes - in a way that would seem to support the idea that free will doesn't exist.
A: "Are you going to school tomorrow?" B: "Well, statisticians tell me crime rates are incredibly high; so I probably won't make it to school. Why go, it might as well not exist?"
The point of A is lost — get an education. It's not all teet-sucking techno-commerce; those statistics say nothing about how peoples came to build those systems in the first place of which those [EDIT:] quantificational terms like "income level" describe.
Moreover, a revolution does not require that anyone die off, but that a system, or status quo, ends.
The more data we get, the better we've been able to predict what people are going to do. Uncertainty reduces if you assume that people's choices are determined. But free will - what does that get you; what does that even look like as a prediction of the distribution of behaviour? It's an empty premise, like saying phlogestine.
The idea that everyone, many of whom have serious problems with motivation and the practical concerns of doing so, should or can just choose to educate their way into the workforce pays no attention to the difficulties faced by in doing so. It's a denial of the reality in which many people live and how they interact with that.
I do some charity work in my free time, many of people there can't even work out how to attach their CV to their job applications for god's sake. Many of them have serious mental problems. Many of them feel entitled to benefits. Many of them just have no motivation for any sort of learning - never really having seen benefits from it. They're not going to sit down and do umpteen years worth of self-education. Free will, choice, these are not serious answers to the sorts of problems that we face as a society.
Any study you're going to look at that measures "lifetime outcome" (if you can measure such thing, is Nikola Tesla a bad outcome because he didn't have much money at the time he died?) will have to look at a span of 50 years at least. So a study done right now would have to start for people born around 1963 at best - a time very, very different from the world nowadays.
no doubt, there's now plenty of opportunity for tech-workers, who will profit from the process (at least as long as their jobs don't get automated away, too). but the wages are already stagnant also in this sector - when the entry of new workers will undercut current engineers' negotiating position (and it will, as engineering becomes the only safe-heaven on the labour market), they will gradually face similar fate as the rest of the workers. the real winners of automation are the owners of capital.
- cloud labor, mechanical turk style. Hackers need to turn the tasks done in businesses into small bite size chunks that can be done by masses of unskilled workers. This is purely a software task that hackers relish.
- training workers cheaply and effectively, khan academy/udacity/coursera style.The business model for this should not be charging students (an evil practice carried out by colleges) but instead charging businesses for access to those workers.
- decentralizing production. Making what is produced by large factories produceable in the home. 3d printers, home robots, and so on. These are also mostly software problems of controlling mechanical parts.
That there is joblessness is mostly because hackers are not thinking creatively and working hard enough. Help will not come from other segments of society, they are incapable of doing so because they lack the key skill of which we hackers are famous for - inventiveness.
Looking at the average earnings on these sorts of platforms it would seem like a challenge to eek out even a meagre existence in a first world country.
This is also a much bigger opportunity for startups than traditional eyeballs model. A startup that can pull it off could easily be worth a -trillion- dollars. Apple, Microsoft, Google grew large just by taking $100-200 a year from a billion users. Imagine a mechanical turk that grew to a billion users - it could be taking $500 a month from users while paying them $5000 a month. Truly immense potential.
Higher skilled sellers who can demand higher prices will want to avoid these platform for that reason. They don't want to have the platform owner taking a cut when they are capable of finding customers on their own. This leads to a sort of lemon market where the buyers who are willing to pay the highest rates for the best workers will tend to avoid these places altogether.
So yes, they are commodities, but if they are a lot of them, and a lot of work to go around, wages rise. We saw this with the industrial revolution, especially assembly lines. Mid 20th century you could live pretty well as an unskilled assembly line worker, but then the work dried up.
Most modern work, knowledge work, hasn't been turned into assembly line yet. Many would argue that's because it can't be, but I disagree with that, much of enterprise software is about turning it into that (workflows,processes). It just requires more of the same. There are actually very few cloud labor startups (mturk, crowdflower, mobilworks) with at most a 100 programmers, one reason being that it doesn't have the network effects moat that VC's look for. Robotics is another casualty of that. How do you defend your cloud labor,robotics company from competition? That's why it will have to be built from the ground up by bootstrappers.
It also makes the competition for knowledge work go international which might increase pay in developing countries but lower it in first world countries.
Higher pay rates exist for two reasons, unfair monopoly positions (e.g. bill gates and the OS monopoly of the 90s, actors/athletes), and because there is complex work that hasn't been deskilled yet. Not much can be done about the former except regulations, the latter is where programmers can change the situation. Demand is everywhere - whether it's fulfilled by cloud labor, or by specialized labor - is the question.
Either way, I think we can agree that immense opportunities exist for startups to profit from this space.
If you equalize wages you also need to equalize the cost of living which is something that doesn't seem to be happening because a lot of people have a lot of money locked up in land etc that they are not going to want to see devalued.
If deskilled work does not cover rent payments then it is not much good to anyone.
Yes, you're totally right - land will get devalued. The older generations in richer countries will lose the most relatively. This is not a bad thing fundamentally (it reverses the rich get richer trend), and is actually what's already happening, but people don't feel it. Every story that is heard about poverty reduction in China is also a story of westerners getting relatively poorer - though in absolute terms everyone is getting better quality of life.
I'm not sure that the older/richer will lose out. They have a lot of political clout that they will bring to bear to keep value in their homes and retirement assets. A lot of this will be hoarded rather than redistributed so demand will continue to outstrip supply in many areas.
Deskilling benefits everyone, is leads to more equitable wealth distribution. The people inventing the assembly lines (hackers) will be paid more than the assembly line worker (cloud laborers), but that's a fair reward for the effort. They won't be much richer (just like mechanical engineers who invent factory equipment) because fierce competition will prevent it (there aren't network/platform effects that lead to gates/zuckerburg).
The difficulty is transitioning these labour resources from tasks now mechanized, to new endeavours that society is interested in solving.
Short to medium term, one angle is figuring out how to actually get people paid for the new stuff - there is a lot that creates value that is difficult to capture.
There are a lot of factors that contribute to fewer jobs. When it comes to technology, yes, technology destroys jobs. But never on a net bases. 100 years ago 40% of our economy worked in agriculture. Today it's less than 2%. Yet people aren't longing for the days of farming. Technology alone is not the culprit.
Productivity is the ratio of units of output per unit of input, where you can abstract input as being either capital or labor. Indeed, productivity continues to grow as a consequence of technological advances. When mankind started using oxes to plough fields, landworkers' productivity rose significantly. As we develop more tools to aid us in daily tasks, less labor - and sometimes even less capital - is required to perform those tasks.
During the relatively prosperous past decades, organizations have grown heavy, and were able to do so because of a continuously rising product demand. People started over-working and over-consuming. Then there was an external shock, media started talking about a financial crisis, people started being more considerate of their consumption in the face of job insecurity, and businesses that grew too heavy needed to get back into shape. This is a process that has significant feedback effects. The process of business getting back into shape means the cutting non-contributing jobs (think of excess layers of management), which by itself causes a growth in productivity. In this sense, employment growth and productivity growth can be negatively correlated, and it is not necessarily "technological advances" that drive productivity growth.
Telling us that technology is destroying our jobs is the same as telling us that China is stealing our jobs, though they may affect different industries (I don't see Chinese laborers replacing our butchers.) The thing is that the working population is very flexible, and despite offshoring certain jobs the average American still works over 50 hours a week. Population growth hasn't justified job growth in the US for the past 40 years, so perhaps a stagnant job growth is a good thing in that it draws things back to "normality".
There will be new jobs, but fewer and fewer people qualify. Network effect and superstar economy mean that the distribution of workforce compensation will change dramatically. Majority of workers will be worth less. More of them will drop below the point where it's not worth to hire them.
The economy is doing great, productivity is doing great, but there are increasing numbers of people with obsolete skills who can't find work and are struggling to survive this improving economy, through no fault of their own. A small number of people are taking all the benefits of the new economy - entrenching inequality.
The problem is with people, not with the economy.
Interestingly, muscle jobs are more-or-less immune as they have already been automated as much as is reasonable.
13. I.e., since 2000.
Artificially low interest rates == artificially high amount to spend on machine replacing labor == labor earnings go down as percentage of the economy.
When is the last time the minimum wage was raised? The last time the work week shortened? Those laws are in place because of technology eliminating human labor. Those legal changes were the solution to a problem technology created. But they have not been updated while technology is galloping away. We blithely crush low-skill, entry level jobs because they are easy to crush.
Technologists have a responsibility to mitigate the impact of technology. But the answer isn't to restrain technology. The answer is to deliver benefits to society as a whole, and that comes from restructuring our economy to benefit workers.
The truth is that much of the manufacturing here in the US is already largely automated, and many of the few jobs left in manufacturing are a few higher skill positions that probably pay an hourly wage near or above $20 already. Any job that's going to be eliminated is going to be eliminated in the next few years whether you raise the minimum wage or not. Instead this argument against a higher minimum wage means that every worker who job cannot be outsourced because it is some service industry job for example is being held down financially on account of the country losing jobs that might be automated eventually anyway.
On top of that a $20 minimum wage would also probably create much more demand by many of those who today are earning less than $20 per day.
Unless we were to subsidize companies when they hire more workers, raising minimum wage would increase un- or under-employment.
Isn't it more likely to be a political problem ? I am thinking capitalism and neo-liberal tenets are tainting technologies and those philosophy hold the responsibility for the current situation. The job/resource/supply/demand pool, despite what pro free market supports believe, isn't limitless.
The hope is that basic income would make part time work sustainable, and give more leverage to workers. Possibly making minimum wage unnecessary.
One solution would be a negative income tax, where you take some threshold X and basically cut low income people a cheque for (X - Income) / Y where Y is something like 2.
That way there is never a financial incentive not to work as every $ you earn will make you at least $0.50 better off than you were before (where Y = 2).
Under a complicated benefits system you often have incentives not to take low paying or casual work in case it disqualifies you for government benefit. In many cases the rules around this can be complicated, so even the effort required to figure this out can be a disincentive.
Not in France. Here, you're supposed to look for a job (and prove it). If you don't, no welfare for you.
I believe few countries have a welfare system that is conditioned on nothing else than near-zero revenue.
Yep.
The main problem here is that most people don't understand incentives and what motivates them. They think everyone's just out to work as little as possible. Selling a negative income tax or basic income to the public, or any system where you provide anyone with "three hots and a cot," as my father put it, is impossible, regardless of how much we need it. According to them, we'd be irreparably damaging the social fabric.
— "Towards a Liberatory Technology", Lewis Herber [Murray Bookchin]
«Unfortunately, the singularity may not be what you’re hoping for. By default the singularity (intelligence explosion) will go very badly for humans, because what humans want is a very, very specific set of things in the vast space of possible motivations, and it’s very hard to translate what we want into sufficiently precise math, so by default superhuman AIs will end up optimizing the world around us for something other than what we want, and using up all our resources to do so.»
— http://www.reddit.com/r/Futurology/comments/y9lm0/i_am_luke_...
> In other words, in the race against the machine, some are likely to win while many others lose.