Technological Unemployment and the Future of Work
Few topics are more important than the future of work given our justifiable fears of rising technological unemployment. How are job markets likely to evolve in our 21st century digital economy? This question has been widely discussed for years, but at the end of the day we don’t really have good answers.
One such discussion took place this past September at the The World Summit on Technological Unemployment in New York. Its agenda featured top experts, including Larry Summers,Robert Reich and Joseph Stiglitz. I moderated a panel on Technology and the Future of Work, where MIT economist David Autor and New York Times reporter John Markoff talked about their most recent work. The Summit explored many good ideas, but there were no definitive answers.
Mr. Autor’s most recent paper on the subject, Why Are There Still So Many Jobs? The History and Future of Workplace Automation, frames the topic through a number of important questions, including:
Why hasn’t automation already wiped out a majority of jobs?;
Will mid-skill, mid-pay jobs continue to decline?; and
What’s the impact of AI, robotics and other advanced technologies?
Substitution versus complementarities. Nothing illustrates the impact of technology on jobs like the dramatic decline in the U.S. workforce employed in agriculture – from 41% of the population in 1900 to 2% in 2000. Big declines also occurred in a number of other occupations. Automobiles reduced the demand for blacksmiths and stable hands; machines replaced many manual jobs in construction and factories; and computers displaced a large number of record keeping and office positions. Given that technologies have been automating human work for the past couple of centuries, why are there still so many jobs left?
The answer isn’t very complicated, although frequently overlooked. As Mr. Autor succinctly puts it: “tasks that cannot be substituted by automation are generally complemented by it.” Automation does indeed substitute for labor. However, automation also complements labor, raising economic outputs in ways that often lead to higher demand for workers. “[J]ournalists and even expert commentators tend to overstate the extent of machine substitution for human labor and ignore the strong complementarities between automation and labor that increase productivity, raise earnings, and augment demand for labor.”
Most jobs involve a number of tasks or processes. Some of these tasks are more routine in nature, while others require judgement, social skills and other human capabilities. The more routine, rules-based the task, the more amenable it is to automation. But just because some of the tasks have been automated, does not imply that the whole job has disappeared. To the contrary, automating the more routine parts of a job will often increase the productivity and quality of workers, by complementing their skills with machines and computers, as well as enabling them to focus on those aspect of the job that most need their attention.
The advent of automated teller machines (ATMs) in the 1970s is a case in point. By 2010, there were approximately 400,000 ATMs in the US. But, not only were bank tellers not eliminated in that interval, but their numbers actually rose modestly from 500,000 in 1980 to 550,000 in 2010, driven by two major forces. First, as a result of the reduced costs of operating a bank branch, plus bank deregulation changes, the number of bank branches rose significantly. Second, by reducing the routine cash-handling responsibilities of their personnel, bank branches became more involved in providing relationship-based services to customers, including credit cards, loans, and investment options.
Polarization in the U.S. labor market. Over the past two decades, Mr. Autor has published a number of papers on the impact of technology on US jobs. In The Polarization of Job Opportunities in the US Labor Market, he examined the changing dynamics of the U.S. labor market by looking at 3 different segments:
High skill, high wage: These jobs generally require the kinds of expert problem solving and complex communications skills typically seen in managerial, professional and technical occupations. Most such jobs are beyond the scope of computer substitution for the foreseeable future, but are very amenable to be complemented by sophisticated computer tools. Over the past 30 years, high-skill jobs have significantly expanded, with the earnings of the college educated workers needed to fill such jobs rising steadily.
Mid skill, mid wage: Most of these jobs deal with the kinds of routine tasks, whether physical or cognitive in nature, that can be well described by a set of rules and have thus been prime candidates for technology substitution as well as for offshoring to lower-cost countries. Many blue collar jobs, such as manufacturing and other forms of production, fall into this category. So do white-collar, information-based activities like accounting, record keeping, dealing with simple customer service questions, and many kinds of administrative tasks. Mid-skill jobs have been steadily declining, especially since 2000.
Low skill, low wage: These jobs generally involve physical tasks that cannot be well described by a set of rules that machines can follow. They include protective services, food and cleaning services, personal care and health care aides. Such jobs have been expanding, while their wage growth, particularly since 2000, has been flat to negative.
As summarized by Mr. Autor, “the structure of job opportunities in the United States has sharply polarized over the past two decades, with expanding job opportunities in both high-skill, high-wage occupations and low-skill, low-wage occupations, coupled with contracting opportunities in middle-wage, middle-skill white-collar and blue-collar jobs.”
A recent WSJ article noted that the 2014 median earnings of a typical male full time worker was $50,383 based on Census Bureau statistics. Back in 1973, the median earnings of a similar typical full time male worker was $53,294 measured in 2014 dollars to adjust for inflation.
Will job and wage polarization continue into the future? “My own prediction is that employment polarization will not continue indefinitely,” wrote Mr. Autor in his recent paper. “While some of the tasks in many current middle-skill jobs are susceptible to automation, many middle-skill jobs will continue to demand a mixture of tasks from across the skill spectrum… many of the middle-skill jobs that persist in the future will combine routine technical tasks with the set of non-routine tasks in which workers hold comparative advantage: interpersonal interaction, flexibility, adaptability, and problem solving.”
Impact of artificial intelligence, robotics and other advanced technologies. Recent advances in AI, robotics and other such innovations are taking the potential for automation to a whole new level. Our increasingly smart machines are now being applied to activities requiring intelligence and cognitive capabilities that not long ago were viewed as the exclusive domain of humans.
What can we expect going forward? Will there come a time in the not too distant future when sentient,superintelligent machines will so far surpass human capabilities that there will be few jobs left for humans? Not according to Mr. Autor. Despite rapid advances in the frontiers of automation, “the challenges to substituting machines for workers in tasks requiring flexibility, judgment, and common sense remain immense.”
Central to his argument is the concept of tacit knowledge, first introduced in the 1950s by scientist and philosopher Michael Polanyi. Explicit knowledge is formal, codified, and can be readily explained to people and captured in a computer program. Tacit knowledge, on the other hand, is the kind of knowledge we are often not aware we have, and is therefore difficult to transfer to another person, let alone to a machine.
“We can know more than we can tell,” noted Mr. Polanyi around 50 years ago. This seeming paradox succinctly captures the fact that we tacitly know a lot about the way the world works, yet are not able to explicitly describe this knowledge. Generally, this kind of knowledge is best transmitted through personal interactions and practical experiences. Everyday examples include speaking a language, riding a bike, driving a car, and easily recognizing many different objects and animals.
In the end, “the challenges to computerizing numerous everyday tasks – from the sublime to the mundane – remain substantial,… there is a long history of leading thinkers overestimating the potential of new technologies to substitute for human labor and underestimating their potential to complement it.”
Irving Wladawsky-Berger worked at IBM for 37 years and has been a strategic advisor to Citigroup and to HBO. He is affiliated with MIT, NYU and Imperial College, and is a regular contributor to CIO Journal.