New Frontiers: The Origins and Content of New Work, 1940—2018

Technological change transforms economies and labor markets, reshaping the types of jobs that are available, the wages they pay, and the skills they require. Much recent empirical work has documented the displacement of labor from existing job tasks by automation technology, including robots and software. However, this literature has primarily treated the set of human job tasks as finite and static, implying that as automation proceeds, labor is confined to an ever-narrowing scope of activities. Casual observation and historical evidence suggest this is at odds with reality: as employment in (previously) labor-intensive sectors such as agriculture, textiles, mining, and manufacturing has eroded, new varieties of labor-demanding activities have emerged, including in medicine, software, electronics, healthcare, finance, entertainment, recreation, and personal care. These new activities require specialized human expertise, making them meaningfully distinct from pre-existing jobs. While now recognized in theoretical models on this topic, a major empirical challenge remains: there is almost no direct, consistent measurement of either the emergence of new work tasks in our economies, nor of the technological and economic forces that are hypothesized to give rise to them.

In “New frontiers: The origins and content of new work, 1940-2018” my coauthors and I answer three core questions about the role of newly emerging job categories (`new work’) in counterbalancing the erosive effect of task-displacing automation on labor demand. First, what is the substantive content of new work, and how has this changed over 1940—2018? Second, where does new work come from? And third, what effect does new work emergence have on labor demand?

We investigate these questions using a newly constructed database spanning eight decades of new job titles linked both to US Census microdata and to patent-based measures of occupations’ exposure to innovations. We find that the majority of current employment is in new job specialties introduced since 1940, implying that new work emergence is quantitatively important for understanding the evolution of employment. Moreover, the locus of new work creation has shifted from middle-paid production and clerical occupations over 1940—1980 to high-paid professional and, secondarily, low-paid services since 1980. This indicates that new work has contributed to seismic shifts in labor market inequality, and to the polarizing labor market opportunities for workers with and without college degrees.

We show that new work emerges in response to technological innovation—however, not all technology generates new work. We distinguish between innovations which are labor-automating and those that are labor-augmenting by comparing the textual content of patents with occupational tasks performed by workers, and with occupational goods and services produced. Patents which are textually similar to worker task descriptions may replace labor by accomplishing those same tasks with machines or software; whereas patents which are textually similar to occupational goods and services may increase the capabilities, quality, variety, or utility of the outputs of occupations, potentially generating new demands for worker expertise and specialization. We empirically confirm that innovations that complement the outputs of occupations (`augmentation innovations’) do yield new work, whereas innovations that automate existing job tasks do not.

Innovation is not the only driver of new work emergence: changing demands for occupational outputs also play a role. Leveraging positive demand shocks from demographic change and adverse demand shocks from import competition with China, we show that demand shifts that raise occupational demand increase the rate of new work emergence while adverse occupational demand shifts slow it down. This matters for understanding the diverse nature of new work—as new work engendered by demand shifts does not necessarily have high technological content—and highlights that demand shifts may have an impact beyond the sheer amount of work, on the demand for specialized human expertise.

Having documented where new work comes from, we finally consider how new task emergence, alongside task displacement, affect employment and earnings. Augmentation and automation innovations have countervailing effects on labor demand: augmentation innovations boost occupational labor demand while automation innovations erode it. These effects are present in both four-decade epochs of our sample and are evident in both the manufacturing and non-manufacturing sectors. This provides empirical evidence for a so-called race between task displacement and new task creation highlighted in theoretical frameworks. However, we find that the demand-eroding effects of automation innovations have intensified in the last four decades while the demand-increasing effects of augmentation innovations have not: this tentatively suggests automation may be pulling ahead in this race.

Several questions for future research are made urgent by these findings. First, is `new work’ more labor-augmenting than simply `more work’? The finding that augmentation innovations increase occupational wagebills by boosting both employment and wages suggests that `new work’ may be more valuable than `more work’—plausibly because new work demands novel expertise and specialization that (initially) commands a scarcity premium. Identifying and quantifying such premia will require direct earnings observations of job tasks and earnings of individual workers engaged in new work, something that is largely infeasible in our Census public use microdata. If, as we suspect, new work provides additional opportunities for skill formation and earnings growth beyond `more work’, then policies that foster new work creation may be of particular interest. 

Second, while our evidence establishes that augmentation and automation move labor demand in countervailing directions, we take the shifting locus of innovation over time as given. How elastic are the locus and pace of augmentation and automation to incentives, and which incentives matter most for steering the direction of innovation?

Lastly, will ongoing advances in Artificial Intelligence shift the balance of innovation towards more rapid automation across an expanding set of occupational domains? And to the degree that AI is not exclusively automating, what novel job specialties will it catalyze and what skill sets will it complement?

The paper, coauthored with David Autor, Caroline Chin, and Bryan Seegmiller, is forthcoming in the Quarterly Journal of Economics.