A new study says that nearly half of all American jobs may soon be performed by robots. And the White House has just announced the formation of “the Advanced Manufacturing Partnership Steering Committee ’2.0,’” which it describes as “part of a continuing effort to maintain U.S.leadership in the emerging technologies that will create high-quality manufacturing jobs and enhance America’s global competitiveness.”
That seems like a good idea, but it raises a number of questions. There is only one labor representative on the committee, as compared to eleven corporate CEOs, and it would be good to know why. What’s more, labor isn’t acknowledged in the President’s statement that “industry, academia, and government must work in partnership to revitalize our manufacturing sector.”
That’s unfortunate, because the working people of America should have a strong voice in designing the future of our manufacturing sector. In fact, that role is more important than ever, as a manufactured product — a range of devices commonly described as “robots” — may change the face of work in America.
Defense contractors are heavily represented on the Committee. Its Chair is Wes Bush, CEO of Northrup Grumman. David Cote, head of Honeywell, also represents a defense-heavy corporation. Other corporations represented include Dow Chemical, Caterpillar, Siemens, Applied Materials, and Alcoa.
The lone labor voice belongs to Leo Gerard, head of the United Steelworkers.
The corporations on that list had the sympathetic ear of Mitt Romney, the Presidential candidate who considered them “people.” Romney, like many conservatives, believed that today’s runaway unemployment was structural – the permanent result of technology and other factors – rather than cyclical, and therefore treatable through time-tested macroeconomic methods and government policies.
In fact, as economist Brad DeLong notes, Romney replied to a question about how he would address cyclical unemployment by listing six approaches to treating the structural kind. (DeLong described it as a “bad plan” and said that only two of the ideas were any good. I’d argue that none of them were.)
Last year’s committee report, the product of version “1.0,” proposed more funding for R & D, innovation, and education. Many of its ideas were quite good, especially regarding clean energy, but none should be mistaken for solutions to today’s economic problems.
The report also included a wish list for corporations: a deceptively-packaged call for tax “reform,” that is, tax breaks for corporations who are already paying historically low real tax rates; “streamlined” regulations based on a misguided but fashionable vision of “cost/benefit analysis” which minimizes social benefits; and more of the same trade “liberalization” that has already dealt a harsh blow to the American workforce.
The Robots Are Coming
It’s probably true: Automation will almost certainly transform the American workforce. The robots are coming – but they’re not here yet.
Both parts of that statement are important. Certain vested interests would benefit if policymakers concluded that today’s high unemployment was due to the displacing effect of technology. That would mean there’s not much anyone can do except offer some job training and hope for the best. Monetary policies and government jobs wouldn’t help a bit.
But the robots haven’t arrived, at least as of this writing, even though liberal blogger Kevin Drum wrote a widely-circulated essay earlier this year suggesting that they had. “(W)e should be very alarmed,” wrote Drum. “… the evidence suggests that–slowly, haltingly–(robot job displacement) is happening already, and we’re simply not prepared for it.”
Unfortunately, as Jared Bernstein points out, there’s no evidence for that. “Most of the weak employment outcomes with which we’re currently stuck are cyclical,” notes Bernstein, adding: “…as (economist) Dean Baker likes to point out, productivity growth has slowed in recent years …”
Under those circumstances it’s hard to argue that the cause is constantly-improving technology. Drum’s probably right, however, when he warns that “the hollowing out (will probably start) with desk jobs in places like accounting or customer support.”
Bernstein concludes that the robot issue “bears watching,” and he’s right. A new report suggests that nearly half of the nation’s jobs could eventually be vulnerable to automation.
No, they’re really coming
A new study by Carl Benedikt Frey and Michael A. Osborne, of Oxford University’s Programme on the Impacts of Future Technology and Department of Engineering Science, concluded that 47 percent of current US jobs were vulnerable to automation.
Frey and Osborne take note of some surprisingly rapid developments, such as Google’s development of a self-driving car, then assess jobs based on the level of “creative intelligence,” “social intelligence,” dexterity, and other skills in which humans currently have the advantage.
They conclude that, in the first wave of automation, “most workers in transportation and logistics occupations, together with the bulk of ofﬁce and administrative support workers, and labour in production occupations, are likely to be substituted by computer capital.”
The authors add that, “more surprisingly,” service, sales, and construction jobs will be increasingly vulnerable to automation. Forbes magazine is even automating some of the tasks formerly done by human journalists. (Time to dust off that novel.)
They conclude that “generalist occupations requiring knowledge of human heuristics, and specialist occupations involving the development of novel ideas and artifacts, are the least susceptible to computerisation.” That heterodox category includes CEOs, mathematicians, and poets.
Which profession do the authors conclude is most vulnerable to automation? Telemarketers. That will satisfy some people’s vindictive streak, but it will also create more unemployment.
The manufacturing sector is already automating its workforce, and it would show considerable foresight on its part if it were to address this issue holistically and in depth. A case in point: The New York Times recently ran a story about textile manufacturers who are moving their plants back to the United States to save on shipping costs, but are heavily automating them to save on labor costs.
The article points out the consumer appeal of “made in the USA” products, along with the bragging rights politicians can earn for bringing these industries back home. But, as the Times points out, “it means jobs, but on nowhere near the scale there was before, because machines have replaced humans at almost every point in the production process.”
So far, “Made by Humans” does not have the same kind of sales appeal that “Made in America” does – although it may be a campaign worth considering. But US sales of manufacturing goods will only be robust as long as workers have jobs and are earning enough money to buy them.
Corporate executives nowadays routinely prefer the quick income jolt, for something that will look good on the next quarterly statement. That’s precisely the kind of shortsighted thinking that led the President’s last manufacturing committee to call for tax cuts and regulatory breaks while ignoring cyclical unemployment. And that’s precisely the kind of thinking they should avoid.
American Giant, one of the brands named in the article, sells well-made clothes that appeal to people with decent working-class jobs – the kind of people who work in textile mills. The fewer the jobs, the lower the sales for American Giant.
What about the robots?
That’s the question of the day: What should we do about automation? If we ignore its long-term implications, the US labor force will take the first hit. But most corporations will eventually get their turn in the barrel.
It should be noted that predicting the future is a risky business. History is filled with inaccurate predictions about the ‘world of plenty’ that technological advances will bring. Scott Winship of the Brookings Institution (who is convinced that robots don’t cause unemployment) notes that, despite the dire predictions, median income rose and unemployment fell after computers became ubiquitous.
He has a point. We don’t know what will happen. We only know what might happen. But it would be foolish not to plan for that. A rational planning process would look something like this:
- Include all segments of society in the conversation, not just corporate interests. Labor has a key role to play.
- Avoid becoming so engaged in hypothetical future events that we ignore the crisis right in front of us. Today’s unemployment emergency involves cyclical, not structural, unemployment.
- Act, not as Luddites or technophobes, but as wise stewards of our own future - a future which must include a healthy middle class, as well as corporate executives and engineers.
- Target government resources toward technology that is job-producing, as well as wealth-producing. Clean-energy technology is an excellent example of that. The President’s first manufacturing committee had excellent ideas for “green tech,” which produces well-paying and hard-to-automate jobs retrofitting homes and commercial buildings across the country.
- Raise the minimum wage, to ensure that every working American can earn a living wage during this period of displacement.
- Learn from the errors of globalization, This time we should remember to distinguish the inevitable from the merely possible, and to consider all the alternatives before embarking on a course of action.
Automation is not a tidal wave, sweeping everything in its path. It’s a process which we can shape and direct toward the best possible outcome for all the affected parties. That’s a category which includes all of us. Everyone deserves a place at the table.
Richard “RJ” Eskow
Friday, 27 September 2013