Technology Doesn’t Always Beat Labor

“History discloses no tragedy more horrible than the gradual extinction of the English handloom weavers,” wrote Marx in 1867. Everywhere he looked, humans were being displaced “by the rapid and persistent progress of machinery.” A century and a half later, we are once again preoccupied with the potential of new technologies to put people out of work. Last year MIT’s Erik Brynjolfsson and Andrew McAfee highlighted the perils (along with the promise) of our accelerating digital future in their book The Second Machine Age. This year, XPrize founder Peter Diamandis and journalist Steven Kotler wholeheartedly embrace technology’s power in their (less compelling) book Bold. This manual for radically innovative entrepreneurs offers advice on how to launch “exponential organizations”—that is, ones whose impact is disproportionately larger than their head count. After all, who needs human labor when you’ve created artificial intelligence or a globespanning network that can harness the wisdom of the crowd? Books like these tap into the fear many people have about technology’s role in society: that lots of wealth will be created but will be distributed very unequally. However, other commentaries paint a more optimistic picture and see opportunities to progress in the 21st century without leaving workers behind. Further Reading The Second Machine Age Erik Brynjolfsson and Andrew McAfee W.W. Norton, 2014 Bold Peter H. Diamandis and Steven Kotler Simon & Schuster, 2015 Zero to One Peter Thiel with Blake Masters Crown Business, 2014 Learning by Doing James Bessen Yale University Press, forthcoming Creating a Learning Society Joseph E. Stiglitz and Bruce C. Greenwald Columbia University Press, 2014 Perhaps the most talked about is Zero to One, by PayPal cofounder Peter Thiel. While extolling the benefits of monopolies, Thiel drives home another message: Technology is a complement to labor. “Software engineers tend to work on projects that replace human efforts, because that’s what they’re trained to do,” he notes, but that’s hardly the only way to move forward. The software of Palantir, a company Thiel cofounded, for example, bolsters the work of intelligence analysts by sifting through very large data sets and highlighting suspicious activity. Unfortunately, that complementary arrangement doesn’t happen quickly or automatically, according to the economist and former software executive James Bessen. In his soon-to-be-released book, Learning by Doing, Bessen returns to Marx’s 19th-century weavers to prove that as humans work with new technologies over the long term, they improve them and boost their own fortunes in the process. So, yes, when the power loom was invented, in 1785, it shifted weaving from farms to factories, instantly increasing productivity yet leaving workers’ wages flat for decades, as Marx noted. But he failed to predict what happened next: From 1860 to 1890, weavers’ pay more than doubled. That’s because the value of any technology is unlocked incrementally, Bessen argues, perhaps over a generation, through on-the-job learning. Weavers working with early power looms produced two and a half times as much cloth per hour as their predecessors who used handlooms; 80 years later, they produced 50 times as much. It’s therefore the adopters and adapters of a technology—not its inventors—who create much of its value. Still, wages don’t rise until the skills needed to operate a technology are standardized and able to be easily taught to workers. Once power looms matured and factories became more uniform, weavers could credibly threaten to take their skills elsewhere—and command more money as a result. Bessen offers several recommendations to speed up skills training in modern times: increased investment in community colleges, vocational education, and retraining programs for displaced workers, along with company-sponsored training and development programs that help workers learn new skills and gain experience with new technologies. So what’s the role of government in such a scenario? After all, if the return on skilled labor is really so high, wouldn’t companies have ample incentive to invest in training themselves? The answer is not necessarily, according to Joseph Stiglitz and Bruce Greenwald in Creating a Learning Society. In this sweeping work of macroeconomic theory, the authors contend that firms in a competitive market invest in only those innovations that allow them to capture the most benefits. They skimp on green tech because they do not bear the cost of pollution, and they overspend on labor-saving tech because they don’t bear the cost of unemployment. The fiercer the competition, the worse these market failures. And so policy makers do have an important role to play: not only in funding research but also in creating compelling incentives for socially useful innovation and discouraging the spread of harmful technologies. Together, these books offer a partial path forward: Reject the obsession with job-eliminating technology in favor of a focus on complementarity. Help workers acquire new skills, and craft an industrial policy that focuses as much on adoption as on creation. But if the pace of technological innovation is speeding up, today’s newly standardized skills will soon be obsolete. Maybe such an agenda could have helped the weavers, but no one truly knows if it will work in an era of exponential change.

Technology Doesn’t Always Beat Labor

Partner Center