That computers are ubiquitous in contemporary life is self-evident. The share of information processing equipment and software in private, nonresidential investment rose from approximately 8 percent to more than 30 percent between 1950 and 2012, with the largest leap occurring between 1990 and 2000.1 It is hard to think of a prior historical episode where a single category of capital investment came so rapidly to dominate all others, now accounting for close to one in three business investment dollars.
Given their ubiquity, it is tempting to infer that there is no task to which computers are not suited. But that leap of logic is unfounded. Human tasks that have proved most amenable to computerization are those that follow explicit, codifiable procedures—such as multiplication—where computers now vastly exceed human labor in speed, quality, accuracy and cost efficiency. Tasks that have proved most vexing to automate are those that demand flexibility, judgment and common sense—skills that we understand only tacitly—for example, developing a hypothesis or organizing a closet. In these tasks, computers are often less sophisticated than preschool-age children. The interplay between machine and human comparative advantage. allows computers to substitute for workers in performing routine, codifiable tasks while amplifying the comparative advantage of workers in supplying problem-solving skills, adaptability and creativity. Understanding this interplay is central to interpreting and forecasting the changing structure of employment in the U.S. and other industrialized countries. This understanding is also is at the heart of the increasingly prominent debate about whether the rapid pace of automation threatens to render the demand for human labor obsolete over the next several decades.
This paper offers a conceptual and empirical overview of the evolving relationship between computer capability and human skill demands. Autor begins by sketching the historical thinking about machine displacement of human labor, and then considers the contemporary incarnation of this displacement—labor market polarization, meaning the simultaneous growth of high-education, high-wage and low-education, low-wages jobs—a manifestation of Polanyi’s paradox. He discusses both the explanatory power of the polarization phenomenon and some key puzzles that confront it. Autor finally reflects on how recent advances in artificial intelligence and robotics should shape our thinking about the likely trajectory of occupational change and employment growth.
A key observation of the paper is that journalists and expert commentators overstate the extent of machine substitution for human labor and ignore the strong complementarities that increase productivity, raise earnings and augment demand for skilled labor. The challenges to substituting machines for workers in tasks requiring flexibility, judgment and common sense remain immense. Contemporary computer science seeks to overcome Polanyi’s paradox by building machines that learn from human examples, thus inferring the rules that we tacitly apply but do not explicitly understand.