Robot Shift: Congress Ponders Day When Machines Replace Workers

Daily Labor Report® is the objective resource the nation’s foremost labor and employment professionals read and rely on, providing reliable, analytical coverage of top labor and employment...

By Chris Opfer

June 3 — The robots are coming, and at least some lawmakers in Congress expect them to make a lot of jobs obsolete when they get here.

“There are assembly lines where you already look out on a field of robots that can work 24-7, they don’t need health insurance, they don’t take vacations and they don’t get sick,” Sen. Dan Coats (R-Ind.) told Bloomberg BNA. “I think it’s coming faster than anybody realizes,” he said of the shift to machine workers.

Coats and other lawmakers are starting to think about “technological unemployment,” or the replacement of workers with self-checkout aisles, driverless vehicles, medical robots and other automations. They have different ideas about how to prepare for the coming workforce overhaul, but most agree that it's going to take new thinking about how people are educated and trained.

They might want to get started soon.

“Technological progress has been putting people out of work since at least John Henry,” Andrew McAfee, a research scientist at the Massachusetts Institute of Technology, told Bloomberg BNA. “What that progress also does is open up new opportunities.”

Cars of the Future Mean Fewer Drivers

There's a black Uber car winding its way through the streets of Pittsburgh these days. The only difference between this Ford Fusion and the hundreds of thousands of ride-share vehicles operating elsewhere around the globe is that no one's behind the wheel.

Uber has already helped change the way many people work, pioneering the new gig economy by giving drivers the ability to work when they want and where they want. Now the company appears to be at the forefront of another employment revolution: the replacement of those drivers with machines that can do the job more safely and for less cost.

The Pittsburgh test car—developed by engineers at Uber's Advanced Technologies Center—marks the company's first foray into driverless vehicles, technology that is expected to eventually make actual human vehicle operators unnecessary. Although the highways won't be turned over to robots overnight, many observers say the technology could eventually eliminate a wide variety of other transportation and shipping jobs, including delivery drivers, truckers and even airline pilots.

“Long haul trucking seems to be the kind of thing that will be automatable,” Tom Mitchell, who chairs Carnegie Mellon University's machine learning department, told Bloomberg BNA. Maybe long haul flying will follow suit, he said.

Still, not everyone is ready to hand the keys over to a computer. “We feel very strongly that nothing is going to replace the skill and experience of a human driver,” International Brotherhood of Teamsters spokesman Galen Munroe told Bloomberg BNA.

Munroe and other sources spoke with BNA in a series of interviews conducted between May 26 and June 3.

New Technology Allows Machines to Learn

Driverless cars aren't the only technological advances expected to shake up the labor market in the years to come. Innovations such as automated kiosks, robot warehouse workers and mechanical surgery assistants are already doing jobs that used to be done by human employees.

Leading the automation movement is technology called machine learning, which gives robots and computers the power to develop new capabilities based on observation. The filters that capture spam e-mail and move it out of your in-box are arguably one example. Algorithms that help machines design software, diagnose medical conditions and analyze the stock market are others.

The question that machine learning developers set out to answer is “how do we build computer programs that automatically get better with experience,” Mitchell said. The technology took a big step forward earlier this year when a Google robot beat one of the world's top Go players in the ancient Asian strategy game that had long been considered impervious to machine mastery.

MIT's McAfee said that development surprised even the robot's creators.

“It's really important because no one on the planet can explain how to play Go at a very high level, even the people that are playing it at a very high level,” McAfee said. “The team that created the robot basically said ‘here are lots and lots of examples of Go games and here are a good outcomes and bad outcomes,' and the machine used that information to master the game.”

The same artificial intelligence technology creates the potential for smart robots that not only adapt to new working environments, but also pick up new skills along the way.

Red Tape Versus Safety Net

The robots that will replace workers on highways and factory lines and in bank teller booths draw on new forms of automation, but they raise an old problem: advances in technology have been moving workers out of jobs since at least the days of the industrial revolution.

Mid-skill workers—in positions that involve routine physical and mental tasks—are those whose jobs are most threatened by robots. Observers fear shifts toward filling these jobs with machines will further the hollowing out of the middle class that has happened as a result of stagnant wages in the face of rising productivity.

They also say artificial intelligence will seep its way into both entry-level, low-wage jobs and those at the higher end of the earnings spectrum that often require advanced education and training. “What we're seeing is that the effect is going to be felt more broadly,” McAfee said.

Many of the policy solutions being considered to address the situation aren't exactly on the cutting edge. They fall over familiar political fault lines about just how much government should be involved in business, a divide exposed when the Joint Economic Committee invited McAfee and other experts to discuss automation at a May 25 hearing.

Republicans and businesses want to reduce regulatory red tape, which they say restricts innovation and entrepreneurship. They point to moves like the Affordable Care Act, environmental regulations and the Labor Department's new rule expanding overtime eligibility as examples of those obstacles.

“I think the biggest challenge is for government to recognize that trying to micromanage this and put in policies that overregulate it is something that we need to avoid,” Coats, who chairs the committee, said.

Democrats and labor groups, on the other hand, tend to favor stronger bargaining rights, higher pay floors, wage insurance and other worker-friendly measures as an antidote to mechanical outsourcing.

“How can we harness this engine of prosperity while making sure that the benefits are widely shared?” Rep. Carolyn Maloney (D-N.Y.), the committee's ranking Democrat, said, citing her primary concern during the hearing.

Some are even starting to talk about a universal income program, in which government pays its citizens a certain modest amount simply for being citizens. The idea has been compared to the natural gas royalty fees distributed to Alaska residents and is currently being considered in Finland.

Replacing or Enhancing?

McDonald's CEO Steve Easterbrook told investors May 26 that mandatory minimum wage increases won't eliminate the fast food giant's need for human workers in many service positions. He did say, however, that more computerized kiosks may wind up taking customers' orders.

“It may change the nature of the jobs in the restaurants because frankly technology is something that our customers are embracing, whether it’s through their phones or a kiosk,” Easterbrook said at the company's annual shareholders meeting.

There's plenty of concern to go around about robots putting people put of work altogether, but many observers say it's just as likely that machines will simply make humans' jobs easier. New technology will also mean new positions to take advantage of that technology, according to Mitchell.

“Technology is creating new types of jobs at the same time it's eliminating others,” Mitchell said. “I think the main thing that it's doing is augmenting workers’ performance.”

Personal assistant and scheduling technology, for instance, can make it easier for workers to do their jobs and free up more time for substantive work.

Self-driving vehicles could revamp food delivery service by allowing human “drivers” to instead cook up a meal in the back seat of a self-driven car on its way to a customer, Mitchell said. The same technology could also usher in other on-demand, door-to-door services.

‘Just in Time' Training

Whatever the future holds, one policy area that appears ripe for bipartisan agreement is training—or, in many cases, retraining—workers for new career paths.

That includes shifting workers to jobs that still require a human touch, like elder care, teaching, event planning and some sales positions. It may also mean experimenting with technical and other educational opportunities that train workers to pivot from one task to another, completely different one.

Predicting the kinds of jobs that the market will demand isn't so easy. That's why Mitchell and McAfee said the focus should be on “just in time” continuing education models that can make training available to workers to develop new skills quickly and on demand.

“People aren’t going to have one job and stick with it throughout a five-decade career,” Mitchell said. “Probably, people will have multiple types of jobs and they're certainly going to need to be able to retrain and learn new jobs throughout their careers.”

Mitchell said that means teaching students to be flexible, beginning at earlier stages of education. It may also mean making training programs more nimble, including by loosening some accreditation standards.

Although lawmakers often differ in their opinions about where the training program money should come from, they've generally agreed to pour additional funds into workforce development and apprenticeship opportunities.

“The key now more than ever is to look for ways to help workers adapt to disruptions in their industries, including changing education and investing more in adult retraining efforts, with initiatives like President Obama's proposal of free community college,” Rep. Don Beyer (D-Va.) told Bloomberg BNA via e-mail.

Into the Great Unknown

The rise of the machines also raises some existential questions about the meaning of work and what people would do with themselves if they didn't have to toil away at job 40 hours or more each week.

Some posit that the move to machine work could spawn some sort of renaissance period, in which people devote their time to creative endeavors, passion projects and charity activity. Others say work is such an integral part of the human experience that many will want their jobs even if they don't need them.

“Work is not just a matter of doing something that you dislike for which you are compensated,” Adam Keiper, a fellow at the Ethics and Public Policy Center told the committee. “It can also be a source of pride and dignity, self-definition, meaning and purpose in human life.”

Of course, some have also raised concerns about the mechanical student eventually surpassing the human master, or robots getting too smart for their own good. Maybe the machines will get to a point where they start asking for paid time off, complain that Jimmy in accounting is making more money or decide they're leaving the 9-to-5 world because they're better suited for a life in the arts.

To contact the reporter on this story: Chris Opfer in Washington at copfer@bna.com

To contact the editor responsible for this story: Susan J. McGolrick at smcgolrick@bna.com