What Happens When Robots Eliminate All Our Jobs?

Paul Krugman says we’re running out of jobs—because they’re all being automated. But can we sit back and let bots do all the work without going into another Great Depression? And is a new kind of welfare the solution?

In his weekly column for the New York Times, Paul Krugman talks about a problem that is likely (or at least should be) on the minds of many technologists: Companies are making more money today on the back of fewer laborers, and it’s not just low-skilled workers who are being displaced:

...some of the victims of disruption will be workers who are currently considered highly skilled, and who invested a lot of time and money in acquiring those skills. For example, the report suggests that we’re going to be seeing a lot of "automation of knowledge work," with software doing things that used to require college graduates.

More alarmingly, unlike in the past, displaced workers aren’t finding new jobs. In fact, given America’s prolonged unemployment problem, it now seems likely that those jobs will never come back. They’ve been replaced: Sometimes by cheaper workforces overseas, but increasingly by technology.

So what do we do about it? Krugman’s solution is to create a new kind of welfare that guarantees a basic level of income for everyone, paid for by taxing corporate profits that are increasingly being driven by automated workforces. That might make sense in the short term, but it’s incredibly myopic.

A more interesting suggestion for temporarily mitigating the problem is to decrease the length of the workweek from 40 to 30 hours, which was successfully implemented by Kellogg’s, the cereal-maker, in the middle of last century. In 2000, France reduced its workweek from 39 to 35 hours with few ill effects, and President Nicolas Sarkozy’s reinstatement of the 39-hour week did nothing to help the French economy when it tanked during the 2009 fiscal crisis.

Reducing the length of the workweek has two effects: it brings more people into the workforce by creating new jobs to replace the lost hours, but it also incentivizes workers to choose not participating in the economy over consumption, which is exactly the opposite of what most people choose today. In fact, the Orion article argues that working long hours is actually less efficient, and driven by the need to consume excess production:

By 1991 the amount of goods and services produced for each hour of labor was double what it had been in 1948. By 2006 that figure had risen another 30 percent. In other words, if as a society we made a collective decision to get by on the amount we produced and consumed seventeen years ago, we could cut back from the standard forty-hour week to 5.3 hours per day—or 2.7 hours if we were willing to return to the 1948 level.

If more people start to choose leisure, it could help wean the country off of the notion that everyone needs to be maximally employed for the economy to function. Getting used to that idea is the remedy that most economists worried about the issue suggest as a long-term fix:

(Columbia University Economist Jeffrey) Sachs and (Harvard Labor Economist Lawrence) Katz are somewhat more hopeful, but their optimism is based on the politically problematic proposition that the United States can adopt wage and income policies similar to those in Scandinavian countries.

But those solutions still don’t address the root of the problem: we simply don’t need as many people to work anymore. Why? Mostly, because jobs and profit are not inherently connected. Humans have traditionally been the most efficient way to produce goods, but with increased automation, that’s starting to change, and that’s what Krugman is really writing about.

What happens when humans no longer need to work for the economy to function? That depends on whether or not you believe scarcity is inevitable. Most economists who do propose solutions similar to Katz and Sachs’ aggressive wage policies, and argue for a best-case scenario where we are all partially employed. But there are a rising number of people who believe scarcity might not be permanent.

Izabella Kaminska of FT’s Alphaville argues that, eventually, technology will advance to the point that we will be able to produce goods virtually for free. Labor will be removed from the system altogether, and everyone will be provided with a baseline of goods and services to keep them alive and allow them to pursue what makes them happy. In his book Makers, Cory Doctorow envisions a society where small teams of hackers invent new economic systems as they go, using new accessible production tools like 3D printers and easy-to-assemble microprocessors.

In these scenarios (Makers' rough parts notwithstanding), the future is literally up to us to invent, and for technologists, that idea ought to be the most intriguing of them all.

[Photo by Flickr User Alden Jewell]

Add New Comment


  • Kevin Kent

    I'm not overly concerned because the economy is always evolving--we'll eventually economically evolve around these future issues. But today what we can do is try to limit mass illegal immigration where we bring in millions of low skilled workers each year whose services will be largely obsolete in 20 to 30 years. If we implemented a large scale guest worker program to meet present demand and effectively shut our southern border down, then the outlook for the U.S. would be a lot less dire. 

  • Jtybenson

    Please. Learn just a little about basic economics before you try to write an article about it. And Paul Krugman is the LAST person you want to learn from. People MUST produce in order to consume. Socialists eventually run out of other peoples' money. Jobs evolve- what happened to the buggy whip manufacturers? The same screams were heard when cars were invented. Aren't there millions MORE jobs in the USA than 100 years ago, before we had much or our technology and innovation? Isn't it better that we don't have to do the crummy jobs robots do now?

  • Gabe Stein

    I suggest you actually read the article before commenting, especially the part where I say Krugman's idea "might make sense in the short term, but it’s incredibly myopic," and then spend the rest of the article talking about all the shortcomings of all of our present-day ideas for coping with this potential problem.

  • libertywins@odummerlooses.com

    I hope someday the robots will go to work growing food and shipping it to the market and be our servants... robots could be our mechanical slaves and I would not feel guilt or shame for a piece of non organic material to serve this world and real humans.

  • Tricia Billingsley

    A more realistic solution to mass human obsolescence is provide in this fascinating little book:

    Robot Nation: Surviving the Greatest Socio-economic Upheaval of All Time by Stan Nielsen

    To ensure human welfare, he seems to suggest the following 

    1) While robots are stupid, tax their use in order to support those permanently displaced by them.

    2) When robots are smarter, program them so that they "voluntarily" act to ensure human well being

    3) When robots are smarter than we are, use robotic technology to enhance human performance to make us competitive (basically merge humans with robots)

    Nielsen shares the view that the overwhelming majority of people will eventually be living lives of leisure supported by robot-generated wealth.

  • libertywins@odummerlooses.com

    I never quite understand the need to merger of robots and humans. Maybe you want to keep a body to house your organic brain in when your body dies... I think mankind could enjoy the robots running the world and making sure they wait on us hand and foot ... the robots don't have a soul . ...