The Economics of Our Robot Future

On Sunday morning of SkepchickCON I was one of the panelists on the faux award winning Cities of the Future panel along with Ryan Consell, Desiree Schell and Shawn Otto. We discussed a whole bunch of issues and problems that may come up in a future society such as global warming, resource constraints, and increased urbanization, but by far my favorite part was when we discussed robots. Right now we’re living in the infancy of robotics. Robots are already creating many of our manufactured goods, not to mention cleaning our carpets, driving our cars, playing jeopardy and even playing their own version of the World Cup. Even so, outside of certain industries, robots remain mostly in supportive roles. They are something humans use in order to make us better at our jobs rather than to replace our jobs. However, in the future as robots get smarter and more independent they may take over most of our work for us, freeing us humans up to do useful things all day like read, think, make art or more likely just sit around playing video games or watching tv.

However, as someone steeped in economics, when I think about this robot future it doesn’t seem quite as wonderful as it may appear. In fact, a future where robots are doing almost everything for us creates a lot of problems in determining how we divvy up our societal resources.

We can’t look into the far future, but we can look to the past to make predictions about what our future may look like. If we go back to the Industrial Revolution, with the advent of steam power many human jobs were replaced by much more efficient machinery. Although this may have caused job loss in the short term, leading to movements such as the Luddites who protested the replacement of their jobs by machines, the greater efficiency lead the economy to grow and entirely new industries popped up with new jobs that only humans could do. It’s possible that a future robotics revolution could take this same path. Even as most of the jobs we know about today get replaced by robots, entirely new industries that we can barely even imagine may open up. People living a century ago never could have imagined the field of computing and all the fields that fast and efficient computer technology has created. My own job as a data scientist barely existed even a decade ago. As technology advances some fields become obsolete while others are uncovered. Who knows what types of jobs we may be doing alongside robots a hundred or even a thousand years from now.

But wait a second, what about the future where we get to sit around playing Assassin’s Creed IV all day while our robots do our work for us? If robots taking over our current jobs just mean we’re going to move into new types of jobs, we’ll never have the ideal future where we don’t need to work any jobs at all. The thing is, just because in the past new fields have always opened up as others became obsolete doesn’t mean this will always be the pattern in the future, and in fact, we don’t necessarily want it to be. Isn’t a future where we can do whatever we are passionate about while our robots do everything we dislike something we should be striving for? Although it does sound great in many ways, there are many things that would actually cause this future to be a particularly bleak one.

In almost every society on Earth, we have a capitalist system in which individuals do work and are paid for that work with currency that we use to buy things we need to survive and be happy. Even in countries that have a communist or socialist government, society is still mostly run by a capitalist system in which people work for pay. In a future in which robots are doing most of our jobs for us, where would we get the money we need to buy the books we plan to sit and read all day let alone the food and shelter we need to survive? In a world without jobs, how could we buy all the things that the robots are making for us? How could we take advantage of our myriad of free time when we have no way to gain resources or money? Sure, the people who own the robots will be getting all the benefits gained from the robots (since robots don’t generally demand a salary), but who would they be selling their robot-made goods and services to when no one in our society has any money to buy from them? And without anyone buying their products, they wouldn’t be making enough to keep up with maintaining their robots and would likely go out of business. In other words, a robot future could actually lead to economic collapse.

The problem is that a society without jobs is a society that cannot sustain a capitalist economy. If we want to live in a future where robots are doing all our work for us so we can sit around and enjoy free time to follow our passions, we would need to come up with a new economic system that allows people to have resources without having jobs. For example, maybe rather than have all the robots owned by some giant monopoly, the government would divvy up the robots with everyone getting a certain number of robots. That way my robot can go to my job each day and do all my data scientist work for me while I continue to cash the paychecks. However, as one of the audience members in the Cities of the Future panel pointed out to me, this sounds an awful lot like communism and that never seemed to work out for anyone.

He’s right in that a communist society really would not work in today’s world where our jobs are done by people. If everyone is assigned a job and everyone gets the same amount of resources no matter how well you do at that job, then there is no longer any incentive to do your job particularly well. If you are wondering how bad things can get under a communist government, just ask the 36 million Chinese who starved to death in a 3-year period under the rule of the Chinese Communist Party. However, these problems are due to issues with people. Robots are not humans and do not have the same considerations. Robots could work and work all day and night as long as they are being maintained and given energy to consume. They will do their jobs at a top-notch level with no paycheck and no complaints and no need for health insurance or even good working conditions. Nowadays we complain when we feel the companies we work for treat their employees like robots, but robots never complain about being treated like robots (unless you program them to complain, of course). A future society that chooses an economic system closer to socialism than capitalism may not do so bad in a robot world. There are obviously other things to consider, such as who decides who gets what and the consolidation of power. Future generations will perhaps come up with some sort of hybrid system that is able to democratically divvy up resources gained from our robots.

If we do end up with this idealized robot future, I don’t know what type of economic system we would have but I know for certain that we wouldn’t be able to sustain the capitalist system we have today. Our future society will need to come up with a new way to decide who gets what so we could take full advantage of all the time we have to do new and exciting things while robots do all our boring work.

Featured image of a robot at the Museum of Science & Industry in Chicago by Jamie Bernstein

Jamie Bernstein

Jamie Bernstein is a data, stats, policy and economics nerd who sometimes pretends she is a photographer. She is @uajamie on Twitter and Instagram. If you like my work here at Skepchick & Mad Art Lab, consider sending me a little sumthin' in my TipJar: @uajamie

Related Articles


  1. Well, the Great Leap Forward wasn’t really the fault of “Communism” per se. It was rather a problem that’s inherent to top-down, bureaucratic management styles that fetishize meeting specific numeric targets above viewing those targets in a local or larger rational context–basically exactly the way most corporations are run internally today. Rather than laziness being at fault, you had local cadres who were rewarded for making production targets (and could be severely punished if they did not), so they practiced ‘creative accounting’ to inflate their crop yields; they had villagers melt down useful tools and clear all the local (wood) fuel resources to make useless pig iron (because the quotas made no distinction between that producing good-quality steel); etc. This is a universal problem whenever maximization of metrics take the place of holistic problem analysis.

    As for the main premise of the article, I’m in total agreement. I’ve long been saying that robots, 3D printing, etc. will not bring us into a post-scarcity society. From the perspective of any one of the last thousand generations of our ancestors, we should already live in a post-scarcity society. We can build stronger buildings faster than ever before; our clothing takes less physical effort to produce; our crop yields have never been higher. Post-scarcity is never defined by production levels, but only by how that production is distributed. And since we’ve decided to distribute it solely to those who already command the vast majority of our resources–well, that Star Trek future is still light-years away.

  2. Eh, robot futurists I think generally take a shortsighted view of what robots will automate when they mention that robots will automate our whole lives. Automation is a force multiplier for humans which allows us to do new things that weren’t able to be done before.

    Computing is massive automation, and sure, did switch operators and typists get run out of business because of it? Yes. In their place though are graphic artists, computer programmers, thousands of freelance writers and editors, the internet, electronic gaming, scientific analysis tools, statistical analysis tools.

    Automating a few jobs to obsoletion made many many times more jobs. Robots I imagine will create more opportunity for a capitalist society, not less.

  3. //The thing is, just because in the past new fields have always opened up as others became obsolete doesn’t mean this will always be the pattern in the future//

    I’d say there’s an *overwhelming* chance this will be the pattern.

    //we don’t necessarily want it to be. Isn’t a future where we can do whatever we are passionate about while our robots do everything we dislike something we should be striving for?//

    This is a narrow view of passion that’s only thinking of today’s life. As new jobs open up, new passions will as well.

  4. I think you’re missing a big part of the picture here: The entertainment sector. It’s extremely unlikely that we’ll be able to create robots who can come up with entertainment as compelling as humans can come up with (and if they can, then we might also be in the scenario that they’re sentient, and thus we shouldn’t be enslaving them). Unlike many aspects of the economy where there’s a natural cap to how many jobs can exist (eg. you can’t keep profiting off of farming when everyone already has more than enough to eat), entertainment can keep expanding through adding more and more variety. In fact, this has already happened – just look at the size of the entertainment sector now, compared to 100 years ago. So the first thing I’d predict is that the entertainment sectors of modern economies will keep growing, representing a larger and larger share of the labour performed by humans.

    However, there is going to be a cap. Generating entertaining material is skilled labour, and not everyone has the right skillset for it. Some of those people will have the skills for other labour that only humans can do (advancing science, philosophy, etc.), but many won’t. It’s likely that at some point, all (or most) unskilled jobs are going to be performed by robots of one sort or another. As such, I see us shifting to a middle-ground between capitalism and communism: We’ll see a base level of support for everyone, with people who can perform skilled labour that robots can’t gaining money and being able to buy perks above and beyond this basic floor. How high this floor is depends on what we can provide and what societies want to provide, though. The way things are going in the US, for instance, the floor will be as low as one can possibly stay alive on, but places like Iceland would make it as high as they could afford to.

    Which probably means, it’s extremely important to fight and keep fighting for a strong social safety net, as it’s going to be more and more relevant as time goes on. We have to keep reminding people that many are poor because they can’t find jobs (or better jobs), not because they don’t want to work, as that’s a situation which is going to become the case for more and more people. If you don’t fight now for better welfare for those who can’t get jobs, you could be the next one to lose your job to automation and have to rely on a low base-level of subsistence.

  5. This paper may be worth reading: The_Future_of_Employment.pdf Its authors determined how easily computerized they are with a help of a canonical breakdown of them into skills necessary for them. The more difficult-to-computerize ones fall into these categories: Perception and Manipulation, Creative Intelligence, and Social Intelligence. Its authors hand-scored 70 kinds of jobs for how easy they might be to computerize, and used them as a training set for a skills-to-vulnerability classifier. They found a good fit, and they then extended this fit to the rest of the 702 kinds of jobs that they looked at. They found that Perception and Manipulation are somewhat vulnerable, and that Creative Intelligence and Social Intelligence have low vulnerability. Looking at Infophile’s post, entertainment clearly requires Creative Intelligence and Social Intelligence, so by the authors’ rating, it’s nearly invulnerable to computerization. Management, business, finance, law, education, science, and engineering jobs are also nearly invulnerable, because of needing those two intelligences, or at least the more upper-level ones of them. The authors found that the more education a job requires, the less vulnerable it is.

    As to what to do about it, one solution would be what may be called the Solaria solution, In Isaac Asimov’s novel The Naked Sun, the inhabitants of that planet each live on estates taken care of by a big army of robots. I’ve seen a virtual version of that solution proposed: widespread stock ownership. Another approach would be taxation and downward redistribution of wealth, but the upper classes then moan and groan about how exploited they are. However, such a solution might be necessary to ensure widespread ownership of robots or stock.

  6. An excellent sf book that looks at this question is “Beggars in Spain” by Nancy Kress. Her premise is that society will bifurcate into two classes — a subsidized class that consumes entertainment and does little else, and “donkeys” who actually seek out and enjoy work for its own sake. Naturally, both classes tend to look down at one another. It’s kind of like a reverse Eloi/Morlock situation, actually, where the educated are the workers and the uneducated are the leisure class.

    But yes, I think it’s overwhelmingly likely that robots will transform our lives as much in the next 30 years as computers have in the last 30. And at least in the short term, this probably means that in the next generation of kids, the ones who don’t get a good education are really screwed.

  7. Robots do our jobs –> We lose jobs –> we don’t make money and don’t buy robots –> robot companies go bankrupt –> humans (re)fill the niche.

    Of course if it were just that simple. We have a greedy class of owners who would love to eliminate that “all too necessary human requirement”

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Back to top button