For decades, technological development has stirred both excitement and fear in the public. In The Work of the Future: Building Better Jobs in an Age of Intelligent Machines, David Autor, David A. Mindell, and Elisabeth B. Reynolds delve into this topic by examining the effects of technological advancements on the U.S. economic landscape and labor market. The book begins with a foreword by Robert M. Solow and an introductory chapter by the authors. The remaining chapters are split into two parts, with part I exploring the evolution of the U.S. workforce and the current technologies predicted to reshape it, and part II presenting policy recommendations for achieving shared labor market prosperity in conditions of technological change.
The idea of blaming the economy’s problems on technology is not new. In the book’s foreword, Solow discusses the topic of high unemployment and the public’s tendency to seek simplistic explanations for it. Whether or not these explanations lay the blame for joblessness on an unskilled workforce or a technological takeover, they typically focus on a single cause. As noted by Solow, however, the problem that has to be tackled is “not to pick a cause but to reckon how much weight to attach to each of a list of causes…more training or even better training [for workers] does not necessarily lead to higher employment.” In other words, Solow’s position is that high unemployment can be attributed to many causes that are not mutually exclusive.
In the introductory chapter, Autor, Mindell, and Reynolds lay out the premise of the book, stating that their research on advancements in artificial intelligence (AI) and robotics does not support the idea that automation displaces workers: “Our research did not confirm the dystopian vision of robots ushering workers off factory floors or AI rendering superfluous human expertise and judgment…. We found a labor market in which the fruits are so unequally distributed, so skewed toward the top, that the majority of workers have tasted only a tiny morsel of a vast harvest.” What the authors refer to is the phenomenon of real wages trailing real productivity since the 1970s, a development that has caused low- and middle-skilled labor to fall behind.
Setting aside the COVID-19 pandemic, the authors argue that the current economic situation is complex and not solely due to technological advancement. However, they also caution that today’s labor market is not ready for the widespread adoption of AI and robotics, because such adoption may exacerbate the gap between wages and productivity. With this in mind, the authors delve into the question of whether technology-driven increases in productivity can improve living standards for the broader population or whether they benefit only a small group of people.
In part I of the book, Autor, Mindell, and Reynolds offer insights from history, noting that, contrary to popular belief, technological change has created new jobs and not led to job extinction. Although this observation may seem counterintuitive given that technology and automation are often seen as threats to human labor, the authors explain that automation creates three forces that generate new job opportunities. First, as some work tasks are automated, workers become more productive in tasks that are not automated. Second, automation boosts productivity and income, with higher income then driving up consumption and demand for labor in sectors supporting that consumption. Third, automation fosters the creation of new types of work and new industries. In support of this account, the authors compare the number of available jobs in 1940 with those available today, revealing a staggering contrast, with today’s jobs far exceeding those available in the past. The authors also note that job creation has not been limited to jobs with technological focus, such as those in information technology or engineering, but has instead spanned a wide range of occupations, from mental health counselors to fitness coaches.
With respect to the current status of technology, it should be noted that most AI systems deployed at the time of the book’s publication were specialized and able to solve only limited problems. However, while machine learning software and AI tools are still in their infancy, they are changing rapidly and finding many mainstream applications. Although the commercialization of new technology takes time, sometimes even decades, I think the themes covered in this part of the book are posed to evolve quickly, with new technologies assuming a more prominent role in shaping the economic landscape.
In part II of the book, the authors entertain different pathways to achieving shared prosperity while coping with technological change. This discussion focuses on training and education, suggesting that efficient and quality training in all sectors of the economy is essential for worker success. In laying out specific options for reshaping training and its accessibility, the authors emphasize the importance of providing adequate educational funding, strengthening regional commitment to training, and generating new ideas for training content and delivery.
The authors also harp on poor job quality as a contributor to the gap between wages and productivity. As productivity in the United States has trended up in recent decades, job quality and wages have risen at a slower rate. To illustrate this development, the authors share Organisation for Economic Co-operation and Development data showing that low-skilled Canadian workers are paid 26 percent more than low-skilled American workers, although it is doubtful that Canadian workers are 26 percent more productive than their American counterparts. According to the authors, promising options for revitalizing job quality include redesigning unemployment insurance, gradually increasing the minimum wage, and giving workers voice in business and economic decisions.
Some of the reforms proposed in the book also involve institutional change, with the authors explaining that government support is what ultimately drives economic emergence. Pointing to past defense efforts and the work of agencies such as the National Aeronautics and Space Administration and the National Science Foundation, the authors argue that, historically, government research and development have been critical in sparking job and technological innovation in all mission areas. This discussion calls for an increase in government research spending, which would fuel productivity across sectors. One recommendation that I found particularly interesting is a proposal for a tax reform that would encourage employer investment in labor.
The authors close the book by asserting that technological progress and robots will not replace humans. However, they also note that worker success in achieving new heights in innovation will require job security and institutional changes that protect the dignity of the worker. The overall message of the book is optimistic: “The remarkable history of American innovation was powered not by fear or fatalism, but by a profound sense of possibility.” This statement resonates with me because it reminds me of the original drive behind progress. We should continue to look at innovation through a lens of wonder and opportunity.
As a labor economist, I find The Work of the Future to be reassuring. Amid alarming headlines and information overflow, the book convincingly alleviates my ambivalence toward the subjects of robotics and AI. The authors also successfully articulate the problem of the wage and productivity gap as a threat to low- and middle-skilled labor (independent of technological innovation). Structuring the book around this theme creates a solid foundation for its arguments and recommendations. The book will benefit not only economists but also all workers who wish to gain deeper insight into the future of work. In addition, the book may be useful to policymakers, government officials, and innovators who seek to understand the relationships between their objectives.
United States Department of Labor
