I think we have yet to see the worst of it for now. But make no mistake, huge sections of jobs are going to evaporate if not by 2019, then in the decade thereafter.

Not just elevator operators and dirigible pilots. A massive chunk of economic activity is going to look silly in a decade. Pretty much everything regarded as a summer job for high school teenagers is game, and that leaves a considerable number of normal, employable adults out in the cold too.

Magazines are pretty much already dead, and everything that went with them shrank. Music and movies have imploded hard enough to show signs of serious stress. Shops, department stores and big box retail are off balance, and will fall over before 2030.

It's going to be pretty rough. Think about your first job, and the summer jobs your high school classmates had. Will your kids hold the same jobs? I doubt it.

One can hope that the next generation of children won't have to hold a 'job' at all.

Not going to happen - unless all jobs disappear at about the same time.

The simple conclusion is that unless you have something to provide in exchange, no one will give you anything in return.

>He (Amartya Sen) presents data that there was an adequate food supply in Bengal at the time, but particular groups of people including rural landless labourers and urban service providers like haircutters did not have the means to buy food as its price rose rapidly due to factors

As long as someone owns the factory or the company, they expect to be paid. Thats not going to go away overnight.

And since robots are not going to be perfect, there will always be jobs that human beings will be employed to do. The employers will in turn expect to get their money's worth for those jobs.

Instead of jobs going away, theres going to be a concentration of wealth in the hands of a few.

I think summer camp jobs will probably still be readily available for teenagers in a decade. Maybe not working at a movie theater or something like that (cashier-intensive jobs), but a lot of jobs with interpersonal interaction will probably stave this off for a while yet.

> Think about your first job, and the summer jobs your high school classmates had.

Walking a dog - ain't going to be automated any time soon. Filling shelves in a supermarket - I am not aware of any automation on the way for this, though it may be. Whitewater rafting guide - no chance that is getting automated.

my first couple of jobs would all be easy to automate but it would cost more then $3/hour

What if employment law advances such to add liabilities to human employment? Then you might get payed $3/hour, but you costs include insurance, losses due to sick-days etc.

I guess, but you must also consider the upside as well. A human cashier is not just a cashier, but an ad-hoc customer service rep, security guard, market intel gatherer, etc.

I guess this is the "human touch" argument, whether this actually applies varies; but for "market intel" machines are way better at gathering information.

machines are best at monitoring what other machines are up to. you need to do both.

What is your predicted level of unemployment in the US by 2019, 2024, and 2029?

I think this question is more interesting if you talk about underemployment or people who are worse off than their parents at a given age.

Both of those numbers are already creeping up, economically coerced crap jobs aside.

I feel like that's a huge walk-back from a position of "huge sections of jobs will just evaporate." I mean, I realize that you're not the person who said "huge sections of jobs will just evaporate," but I'd like someone who does believe that, not something much milder, to put some numbers to it.

I believe huge numbers of certain kinds of jobs will evaporate, particularly whitecollar jobs -- just that the powers that be will implement modern sharecropping to nominally keep "jobs", even while the standard of living massively erodes and the middle class vanishes.

I believe we're already seeing that, with the job markets losing professional jobs but gaining service industry jobs.

I think it's a distraction to talk about employment numbers without talking about the quality of the jobs in question -- sharecroppers and CEOs both have a job, but their lives are radically different.

We're probably aged out of this discussion, but in case you're still reading: okay, fine. Provide some kind of quantitative prediction of what you're talking about. If unemployment numbers are the wrong measure, use Gini coefficient or something.

>Isn't that already the case?

Yeah, but it wasn't robots that got us here, it was rather more mundane - union busting, offshoring and outsourcing.

Robots are a rather convenient scapegoat though.

The financial crisis was simply symptomatic of that (middle class households were forced to swap rising incomes for more and more debt).

It seems like while the developed mature economies (Europe, NAmerica, NEast Asia) will fare okay if they can "retrain" their workforces, lesser developed economies will suffer because: developed economies will not need to export labor to cheap labor locales and two the demographics of developing nations mean they won't have jobs for those people. So either they put a lid on their pop growth or they will suffer from lack of jobs for all those people since developed nations will not be vacuuming cheap labor up and instead will automate those jobs in their economies.

We still have all the laid off manufacturing workers from the last 20-30 years that we couldn't retrain

here's hoping that this leads to less labour arbitrage and exploitation, and more developing nations investing in their own locally-beneficial industries.

By definition of global trade, the locally beneficial industries won't survive if the firms in the west take off - those firms will undercut even developing world prices and sell to those markets.

The question is what do those markets offer in return?

not necessarily, with logistical costs to transport goods. The alternative is that those firms start up local factories, but then that's helping the local economy. With digital goods, many Western products aren't translated or localised. Even with translation, there are a lot of cultural artefacts built into software products that may translate to a poor user experience in other cultures.