What are the Robots Doing? Rebalancing our Inequality Intellectual Portfolio

March 4, 2015

A blog post responding to a blog post responding to a blog post. Who says the blogosphere is dead?


Recently I wrote about Larry Summers demolishing an argument about robots and our weak recovery on a panel. Jim Tankersley called up Summers to further discuss the topic, and put his interview online as a response meant to correct and expand on my post. But I don’t think we disagree here, and if anything Summers’ interview shows how much the consensus has changed.

Before we continue, I should clarify what we are talking about. When people talk about “the robots,” they are really telling one of three stories:

1. Technology has played an important role in the economic malaise of the past 35 years, broadly defined as a mix of stagnating median wages, increased inequality, and weakening labor-force participation.

2. The Great Recession has led to such a weak and lackluster recovery in large part because of technology. In one version of this story, technology is simply taking all the jobs that would normally be found in a recovery. As the AP put it, “Five years after the start of the Great Recession, the toll is terrifyingly clear: Millions of middle-class jobs have been lost… They’re being obliterated by technology.” (President Obama himself often mentioned this story throughout the dark period when unemployment was much higher.)

Another, more popular, version is that workers simply don’t have the skills required for a high-technology labor force. A representative quote from the Atlanta Fed President Dennis Lockhart in 2010: “the skills people have don’t match the jobs available. Coming out of this recession there may be a more or less permanent change in the composition of jobs.”

3. We are moving to a post-work economy, one where robots substitute for human labor in massive numbers and fundamentally change society. Here’s an example. We may or may not be seeing the first hints of such a change now, depending on the story.

The story I said Summers (as well as David Autor) demolished is the second. There’s no evidence that we are having a technology renaissance right now, or that technology has contributed in a major way to the weak recovery, or that a skills gap or other educational factor is holding back employment, or that highly skilled workers are having a great time in the labor market. The arguments against this story from the original post are pretty damning, and Summers either reiterates them or doesn’t walk them back in the Washington Post column. (Let’s leave the third story to science fiction speculation for now, noting that the second story getting demolished means it isn’t happening now, and that it’s hard to imagine robot innovation when labor is so cheap and abundant.)

However, Summers does argue for the first story as well, the one in which technology has played a role in the malaise of the past 30 years. As he tells the Post, “In the 1960s, about 1 in 20 men between the age of 25 and 54 was not working. Today, the number is more like 1 in 6 or 1 in 7. So we have seen some troubling long-term trends, and they appear to be continuing trends.” Summers also notes, “to say that technology is important is not to say that technology is the only important factor, or even that it is the dominant factor.” He mentioned this as the conference as well; Brad DeLong and Marshall Steinbaum noted it in their posts.

Intellectual Portfolio Rebalancing

When we think of the economic malaise of the past 30 years, we should probably think of it as a combination of technology, globalization, sociology, and public policy. Tankersley wants to emphasize technology as a piece of this story, and I agree it should be there.

But here’s what I find interesting. Whenever we have a portfolio of ideas, some ideas get more weight than others. And what strikes me about this conversation is how much technology and skills have been deemphasized relative to other stories since the Great Recession, especially those of public policy.

This is a pretty quick and important change. Almost ten years ago, Greg Mankiw could write, “Policy choices […] have not been the main causes of increasing inequality. At least that is the consensus, as I understand it, of the professional labor economists who study the issue.” Brad Delong also said in 2006 that he “can’t see the mechanism by which changes in government policies bring about such huge swings in pre-tax income distribution.” Skill-biased technical change (SBTC) and technology were assumed to cover the entire inequality story.

That consensus is weaker now than it was then. Certainly the argument for SBTC, while always shaky, has taken a hit. You can see it with Summers himself in the Washington Post, where he notes that “changing patterns of education is unlikely to have much to do with a rising share of the top 1 percent, which is probably the most important inequality phenomenon.”

Meanwhile, more and more inequality research is focused on institutional factors, ranging from marginal tax rates to the minimum wage to the inefficiency and growth of the financial sector to deunionization.  And as the Mankiw quote hints, 10 years ago you’d be less likely to hear, as Summers says at the Washington Post, that a “combination of softer labor markets and the growing importance of economic rents” are an essential part of inequality spoken with the same confidence as you see here. I read that as a major change of the consensus.

This is a major rebalancing of our intellectual portfolio of inequality stories, a change that I think is opening up a much more rich and accurate description of what has happened. I hope the research and conversation continues this way.