New York’s mayor has a plan to address the displacement of workers through automation: Bill de Blasio wants the federal government to enact a “robot tax.” De Blasio’s proposed law stipulates that when a company introduces labor-saving automation, such as a forklift on a loading dock, it would have to pay the federal government five years’ worth of payroll taxes for each worker that the innovation displaces. The employer would have to find new jobs for the workers at their same pay or give them severance. Further, the law would invent a new federal agency to determine which jobs were eliminated by automation and how much robot tax employers owe. De Blasio has even come up with a name for the agency: the Federal Automation and Worker Protection Agency (FAWPA). Separately, he would have Washington eliminate all tax incentives for innovation; though he uses the word “automation” instead of innovation, it amounts to the same thing.

The money raised by the robot tax will presumably help cover the taxes that the displaced workers no longer pay and help ease their transition into either new jobs or permanent retirement. But the assumption that the hands of workers replaced by automation will fall idle—though popular in some circles, and especially with de Blasio’s fellow Democratic presidential candidate, Andrew Yang—flies in the face of everything we know about innovation and its broad effects. Centuries of statistical and anecdotal data show that innovation, even if it initially displaces some workers, will create many more new employment opportunities, advancing overall productivity and increasing output. Since the industrial revolution began more than 250 years ago, business and industry have absorbed wave after wave of labor-saving innovation, yet economies have nonetheless continued to employ on average some 95 percent of those who want to work. If innovation, whether spinning machines or the robotics of the 1960s, had really thrown people out of work permanently, as people today fear that Artificial Intelligence will, then the portion of the population at work would have fallen consistently. That hasn’t happened.

It’s nearly impossible to imagine what new areas of employment will emerge in response to any wave of innovation—and always easier to look at who is losing out right now. But as the overall statistical record shows, the new jobs do emerge. For instance, in the last decades of the twentieth century, word processing, email, and the Internet displaced thousands of typists, messengers, and administrative assistants. But a class of unemployable people never emerged because the same technology also facilitated, among other things, the emergence (or transformation) of Federal Express and other delivery services, including Amazon. Jets and trucks had long been around, but not the ability to ensure that packages get to the appropriate conveyance and arrive at their destination quickly—much less the ability to track each item along the way. Similarly, the introduction of automatic-teller machines threatened to throw thousands of bank clerks out of work, but the machines instead brought such profits to banks that they could employ more tellers—who, with the assistance of new technologies, could do more interesting, complex, and valuable jobs, at higher pay than before. Effects like these play out again and again.

De Blasio, Yang, and others point to studies that cite millions of unemployed as the result of robotics. One widely quoted survey published in the MIT Technology Review forecast that AI would eliminate 47 percent of the jobs in the U.S. and 35 percent in the U.K. by 2035. Yet, a study by the Organization for Economic Cooperation and Development took exception to those figures. It noted that the first study accounted for all the jobs that AI could do, without regard for economic realities. When the OECD considered only what work robots could do profitably, the jobs at risk fell to 10 percent in the U.S. and 12 percent in the U.K. More significantly, neither study considered—or had a basis for considering—the new positions that economies surely will produce, as they always have.

Apart from de Blasio’s lack of historical perspective and imagination, his law would also drive businesses overseas. An American company that saw opportunities to apply robotics to its operation would, if faced with de Blasio’s robot tax, simply build abroad. All the high-paying technical and maintenance jobs attached to the robots would depart as well. The company might close its onshore, non-automated operations, and since no robots were immediately involved, those layoffs would cost nothing in robot tax. Some American firms that have already gone overseas to take advantage of low wages have used the efficiencies of robotics to bring these operations back home. In the face of a robot tax, more firms will stay abroad.

Any proposal that creates yet another agency in Washington deserves skepticism. FAWPA’s power to determine a firm’s tax liability would invite abuse. Business managers would think less about production, profits, and efficiency and more about bureaucratic rules. Instead of focusing on their competitive edge, firms would spend time trying to disguise the employment effects of a decision to install newer, more effective equipment. De Blasio’s proposal would also encourage managers to seek waivers—which Washington administrative agencies invariably grant to favored companies or industries—by couching its innovation decisions in terms of keeping some jobs here instead of sending them overseas. It may all be for a good cause; it may even be true. But still, managers presumably have better uses for their time than spinning business decisions to align with political fashion.

DeBlasio’s ideas are wrong in so many ways and on so many fronts that this article can only outline them. In the unlikely event his proposals come to fruition, they would keep people mired in what any honest observer would describe as dreary, repetitive, and sometimes backbreaking labor. But it is even more likely that the proposals would fail to save such jobs, stifle innovation, and hinder the efficiency and competitiveness of American businesses. The loss of efficiency would, of course, make American products cost more, hurting living standards accordingly. One could say more about de Blasio’s ill-begotten proposal—but isn’t this enough?

Photo by Leon Neal/Getty Images


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next