When robots do the work, is it time for a guaranteed income?

Last of an occasional series on the future of work.

An experiment is underway in northern California to see what happens if every month people get free money.

The project, sponsored by a Silicon Valley start-up accelerator and venture capital fund called Y Combinator, aims to find out if people sit around and play video games or whether they still seek to work for additional income. It will look at whether the handouts, $1,500 a month, make people happy.

These may seem like odd questions for a for-profit technology firm to explore, but they relate directly to one of Y Combinator's other interests: robots.

Consider one of Y Combinator's other big ideas: there's a business opportunity to make automation technology far less "brittle." Today's robots need adjustments, they break and can't fix themselves, high-speed manufacturing lines get jammed, and so on. A lot of high-wage people still work on the typical shop floor. Smarter robots would make those people unnecessary.

So Y Combinator and others in Silicon Valley are looking at what happens to workers then, and they think the guaranteed income could be part of the solution. They are clearly right to be pushing this idea forward, because if their vision for the economy comes to pass there might not be nearly enough paid work for everybody to do.

The details vary depending on who is talking about it, but the idea is the same, a guaranteed amount of money for people and families, more or less with no strings attached.

What theorists on both the right and left have in common, however, is that few of them seem to frame the issue as a solution to the looming problem of losing paid work to machines.

Where this idea is really catching on is among those working in technology, tech leaders as diverse as Tesla CEO Elon Musk and Microsoft co-founder Bill Gates. They are looking ahead, 20 or more years down the road, to when employers collectively will have millions more machines performing tasks they pay people to do now.

One well-known estimate earlier this year, by the McKinsey & Co. consulting firm, suggested that about half of what people are paid to do every year in the global economy could get automated, and by what McKinsey called "currently demonstrated technology." Only one in 20 of the 800 occupations it studied would disappear completely, but maybe six in 10 jobs could have at least 30 percent of their activities automated.

Without some change in how people get money, the spread of automation will be a worrying prospect for all of us who count on paychecks to buy the things we need in life, from secure housing to fresh food.

Doug Clinton, one of the founders and managing partners of Loup Ventures of Minneapolis, is another person who is keeping his eye on the thinking on universal basic income, calling it "the most logical" way for people to fund their living expenses as work increasingly gets replaced by machines. Clinton also has a good view of what's happening because he's been looking at making investments in artificial intelligence, robotics, virtual reality and other technologies.

The work that will be left to the people includes directly caring for other people and the creative fields, but even these might not be stable career paths. Clinton used the analogy of current YouTube stars, noting how a relative few of them, of the many millions of people who upload videos to share with the world, can make a consistently good living. The creative job of the future becomes more of an avocation than a vocation, something meaningful to do yet still not paying the rent.

Clinton may have rose-colored glasses on, but he makes the point that as a society we should be more than rich enough to provide a basic level of income for all, with the economic payoff coming from the productivity gains enabled by artificial intelligence and robotics. We really could have a far more leisurely life, with smart machines doing our heavy lifting.

"Our view is universal basic income is the ultimate outcome of capitalism, it's really a success of capitalism," Clinton said. "In a free market system the most competitive option should win, and the most competitive option in the future will be some sort of robot or AI-based system."

Clinton and his partners aren't sure how a universal income benefit would best be funded or how much would be needed. They imagine some sort of robot tax will be tried, perhaps on a portion of incremental profits generated by the adoption of artificial intelligence or robotics. They do know it's past time for policymakers to start thinking seriously about various ways this could be done.

"Now is the time to start experimenting, and trying to figure out what can work," Clinton said. "In 10 or 20 years, once potentially significant numbers of jobs are automated, at that point it'll be too late."