Pethokoukis, Economics, U.S. Economy

How to stop the robots from taking all our jobs: A Q&A with Noah Smith

Image Credit: shutterstock

Image Credit: shutterstock

What if smart machines do take all our jobs — or at least a lot of them and depress wages? What can policymakers do to make sure that as many Americans as possible (a) still work and have an opportunity for earned success, and (b) get more of their income from owning capital as opposed to selling their devalued labor? Those are two of several topics discussed in the latest Ricochet Money & Politics podcast. My guest was Noah Smith, an assistant professor of finance at The State University of New York at Stony Brook. Smith blogs at Noahpinion and writes occasionally writes for The Atlantic. Here are some edited highlights of our chat:

I recently interviewed Tyler Cowen, and we talked about his new book Average is Over, a book where he addresses how automation is affecting and will continue to affect the US labor market. Now sort of the killer stat in his book is that he thinks about 15% of workers, the tech savvy ones, will do great in the future, but there will be stagnant wages for everybody else. And of course some people say we’re already seeing this effect in job polarization where there’s a lot of demand for high-skill jobs and low-skill jobs, with the middle sort of a bit maybe being off-shored either overseas or to “Robotland.”  I want to read something you wrote for The Atlantic which I think relates to that:

For most of modern history, two thirds of the income of most rich nations has gone to pay salaries and wages of people who work, while one third has gone to pay dividends, capital gains, interest, and rent to people who own capital. This two thirds-one third division was so stable that people began to think that it would last forever, but in the past ten years something has changed. Labor’s share of income has steadily declined, falling by several points since 2000 and now sits around 60% or lower.

So why is this happening, and what do you see as sort of the longer term impact?

Well so everybody’s talking about this rise of the robots, skill-based inequality. But I actually think that this is not as big a story as people are claiming, at least in the past decade because one of my mentors from Michigan, Mike Elsby, has written a recent paper that basically shows that a lot of it was just because of the end of the Cold War. Half the world’s labor supply was locked up behind the Iron Curtain, with not much capital, or with a bunch of wasted capital, and so once, you know, China and India and other countries basically opened up their labor markets, we had this giant dump of labor onto the global economy and of course that is going to drive down labor’s share and is going to drive up capital’s share of income, at least temporarily. I’m not saying that this is all of the story, but I think this is a lot more of the story than people realize. There’s this giant labor dump from China and India and other countries. But going forward I think it’s something we’ll have to worry about more.

Do you more or less agree with Tyler’s thesis going forward?

Well it sort of rings true to me, and I think Tyler is very good at saying things that ring true. It’s just really really hard to predict because we’ve been here before when you had automation of a lot of crafts 100 years ago. You had a lot of people predicting, you had- or 200 years ago- you had the Luddites, you had a lot of people predicting mass unemployment and it didn’t happen. But just because it didn’t happen that time doesn’t mean it won’t happen this time. So I think people are right to be worried, right to be concerned, and we need to ask what do we do if this really does happen.

Let me quote something else that you wrote:

The big question is what do we do if and when our old mechanisms for coping with inequality break down? If the endowment of human capital with which people are born gets less and less valuable, we’ll get closer and closer to that Econ 101 example of a world in which capital owners will get everything. A society with cheap robot labor would be an incredibly prosperous one, but we will need to find some way for the vast majority of human beings to share in that prosperity or we risk the kinds of dystopian outcomes that now exist only in science fiction.

And in this Atlantic article, you go through some of these things we can do to deal with that situation. Now one of them you mentioned is that it should be easier for regular people to own their own capital, their own private army of robots. What do you mean by that?

Well, what I actually mean is that when George W. Bush was talking about the Ownership Society — even though he might not have gotten very far in actually creating that — I like the phrase.  If you do read science fiction — and I do — some of these people in the 1980s were predicting these far futures where everyone kind of has these armies of robots. Vernor Vinge wrote some books where that happens. So basically you can imagine everyone sort of owning their own small firm making stuff for other people. And basically each person is a firm owner and a capitalist. And  if the capital share went to you know 80% or 90% of income, then to have that income be equitably distributed you’d either have to run things through the government and have the government redistribute or you’d have to figure out some way to have capital be owned much more widely than it is now. Because now capital income goes very disproportionately to rich people and so you know poor people don’t own very much capital. And so if you were really going to redistribute the robots you you’d need some way to fix that or you’d have to have the government redistribute incomes.

I would prefer the first rather than the second. But you’re saying that  more of us would need to be entrepreneurs or at least small business owners in some fashion and we should make that easier. How can we make that easier?

Well one thing is education. I learned a lot of a lot of skill-based kind of things, but I never really learned how to start my own business in school. I think that if that if capital income keeps going up, there’s going to be a lot more people who need you know entrepreneurial skills and so education both college and at the, the K-12 level might change to emphasize that to train us to own robots, own and command our armies of robots, to do things like basically own and run a business instead of simply doing some skilled work for someone else.

And even if you don’t completely buy the story of the futuristic robotic society, a lot of these solutions to that scenario strike me as just kind of good ideas anyway.

I agree. People like owning their own businesses. It’s something that a lot of people used to do back in the day before the Industrial Revolution and the rise of giant corporations.  I think it gives people a feeling of freedom and empowerment. I don’t know how much we can scale it, but I think we might look for ways to try to scale it out.

And to the extent we can scale it, you suggest at least making it easier for individuals who don’t want to be entrepreneurs just to own stock in companies.  But as you know it’s harder and harder to do that as you have a lot of companies which are going private. How can we get companies to go public so people can enjoy sort of the fruits of owning capital?

I worry that Sarbanes-Oxley has driven companies toward going private and becoming, you know, master limited partnerships or whatever. In this case I’m little better than your average newsreader in terms of what the government needs to do in order to encourage companies to go public more. But I think that if you have really large companies, then you really want to try to encourage those to go public so that ownership can be widely distributed and, but I don’t know how to do that.

If people own capital and feel like they have a stake in companies, they feel more empowered. The whole reason to keep people in jobs in the first place, to keep people working would be that people feel valuable from working.

But a better proposal than that one is actually wage subsidies, government wage matching, also called a negative income tax. We would be putting our thumb on the scales between humans and robots to keep humans in work that in a perfectly free market they wouldn’t be doing. When a company offers you wage, the government matching would have already done behind the scenes. Someone comes and offers to pay me $20 an hour, the government is paying $12 of that. I would be making $8 an hour, but I would feel like a person who making $20 an hour. Unlike the Earned Income Tax Credit where you get a check from the government based on how much income you earned, I think people would feel a lot better in term of the framing of it if the government matched their wages instead.

That is why people really dislike handouts but really love the minimum wage even though it doesn’t make any sense economically. The minimum wage really distorts the economy more than handouts. With a minimum wage, they feel like that is money they deserve even if the government legislates it. People’s preferences are people’s preferences. Actually you could get a really efficient wage subsidy if you had local governments subsidizing local wages with land value taxes and then you could tune it to the local cost of living. But now we are getting into the realm of policies so smart they will never actually be done.

Follow James Pethokoukis on Twitter at @JimPethokoukis, and AEIdeas at @AEIdeas.

One thought on “How to stop the robots from taking all our jobs: A Q&A with Noah Smith

  1. Jobs and wages will simply transition to other fields. Artists and musicians will likely become better paid and more numerous. Jobs like fine dining servers will require top people skills and become more highly valued and paid. All kinds of other fields that require human interaction will boom. (massage therapists, psychologists, yoga instructors, etc.)

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>