As a journalist I’m often told that a trained monkey could do my job. While there’s probably some truth to that claim, I’m not all that concerned about a Planet of the Apes style takeover of my occupation. Trained monkeys don’t worry me; trained robots do.
Or at least they did. For years I’ve heard my vocation is on the list of jobs where robots are already replacing humans (along with shepherds, telemarketers, and bartenders). But is that really true? A working paper by James Bessen of Boston University School of Law argues that occupations that use computers grow faster, not slower and that computer automation is not a source of significant overall job losses.
Many occupations have been eliminated because demand for the occupational services declined (e.g., boardinghouse keepers) or because demand declined because of technological obsolescence (e.g., telegraph operators). But Bessen found that since 1950 there is only one occupation whose “decline and disappearance” can be largely attributed to automation: elevator operators.
Most automation of jobs is only partial, not complete. Bessen explains why that is a key difference:
This distinction between partial and complete automation might seem irrelevant when many or most of the tasks of an occupation have been automated. However, the economic difference between being mostly automated and being completely automated can be critical. Complete automation implies a net loss of jobs; partial automation does not. During the 19th century, 98% of the labor required to weave a yard of cloth was automated, however, the number of weaving jobs actually increased. Automation drove the price of cloth down, increasing the highly elastic demand, resulting in net job growth despite the labor saving technology. Similar demand responses are seen with computer automation. Consider, for example, the effect of the automated teller machine (ATM) on bank tellers. The ATM is sometimes taken as a paradigmatic case of technology substituting for workers; the ATM took over cash handling tasks. Yet the number of fulltime equivalent bank tellers has grown since ATMs were widely deployed during the late 1990s and early 2000s (see Figure 1). Indeed, since 2000, the number of fulltime equivalent bank tellers has increased 2.0% per annum, substantially faster than the entire labor force. Why didn’t employment fall? Because the ATM allowed banks to operate branch offices at lower cost; this prompted them to open many more branches (their demand was elastic), offsetting the erstwhile loss in teller jobs.
Even partial automation can lead to jobs losses, of course. But as a whole, automation tends to merely shift the need for human labor from routine, low-skill tasks to more creative, high-skilled functions. Automation lead to fewer elevator operators but more elevator designers, engineers, and repairmen.
This shift ought to be lauded by Christians. While we should rightly be concerned about the employment prospects of low-skilled workers, we should not become nostalgic for the mind-numbing, back-breaking work that automation has made obsolete. Too often we treat “jobs” as if they were an inherent good (at least if they pay a “living wage.”) But not all jobs are created equal. Some jobs that may benefit our neighbors’ bank account may also be crushing their soul.
The rapid adoption of computerized automation has the potential to increase job satisfaction for entire occupations that have previously been dangerous, dirty, and demoralizing. In looking at the future of work, we therefore must look not only at the wages that a job will pay but also the price such work requires of our neighbors. We can let the robots take over the parts that a machine can do so that we may use our God-given human abilities for more ennobling tasks.