The Covid-19 pandemic has accelerated the adoption of cutting-edge technologies. From contactless cashiers to welding drones to “chow bots” — machines that serve up salads on demand — automation is fundamentally transforming, rather than merely touching, every aspect of daily life. This prospect may well please consumers. Forsaking human folly for algorithmic (and mechanistic) perfection means better, cheaper, and faster service.
But what should workers — who once provided these services — expect? Can they also benefit from technological progress? If so, how?
The labor market impact of technology is often viewed through the lens of job creation or job destruction. Economists — with near ubiquity — treat technology as being either labor displacing or labor reinstating. If technology displaces workers, jobs are lost. If technology creates (or reinstates) work, jobs are created. Under this dichotomy, the key question is whether technology creates more jobs than it destroys. The World Economic Forum estimates that by 2025, technology will create at least 12 million more jobs than it destroys, a sign that in the long run, automation will be a net positive for society.
Technology’s job boosting ability is often touted by tech advocates. Take Waymo, the Google backed startup developing driverless taxis. In recent years, the company’s sensor-laden, white minivans have become a common sight in some American suburbs. However, mobility sans driver raises concerns about job losses. What will otherwise would-be cab (or more likely Uber and Lyft) drivers now do? Waymo’s response? Take up new jobs created by self-driving technology, gigs like self-driving fleet technicians, rider support operators, and software engineers. “We can be helpful as a company that creates jobs,” noted one Waymo exec.
Job creation isn’t everything, however. Equally important is what workers can earn for working those jobs. Do wages rise or fall owing to technological progress?
Bots Can Boost Wages
Wages — conventional economic theory posits — are dictated by supply and demand. When jobs require specialized skills, wages rise because fewer people can meet demand for these skills. Wages also rise when workers are — regardless of requisite skill — scarce because there are fewer people available to supply their labor. This explains why pilots earn more than plumbers, chemists more than cashiers. Pilots require more specialized skills than plumbers and chemists are (in part because of pricy education requirements) less plentiful than cashiers.
Work by the late Alan Krueger hinted at automation’s wage-boosting abilities. Krueger found that computer-savvy workers — laborers who worked alongside automation — commanded wage premiums of 10 to 15% more than their computer-illiterate counterparts. Economic historian James Bessen has suggested that over the past two centuries, wages have risen 10-fold owing to technological progress. Bessen credits wage growth for many ordinary workers to new technology. It’s an encouraging story, but unfortunately, it’s also an incomplete one.
Bots may well boost wages, but they can also depress them. Daron Acemoglu and Pascual Restrepo recently found that laborers displaced from jobs owing to automation are often forced to compete with other workers for whatever jobs are left. For example, clerical workers who have been replaced by automation may subsequently seek employment in sectors that have not been automated; say retail work. Their entry into the retail sector causes wages in this sector to drop as clerical and retail workers undercut one another for employment.
But even these findings don’t fully capture the wage impact of automation. The transportation sector — which my colleagues and I have been studying closely — provides a vivid example of yet another way in which technology can depress wages. During the birth of commercial flying, pilots commanded a minimum salary of $2,000 annually ($30,000 today). However, aviators willing to fly at night could earn between at least $2,400 and $2,800 annually. The reason? Night flying was considered more dangerous. Back then, taking flight after dusk required specialized skills and temperament, attributes that were in short supply. Firms responded by paying hefty salaries to aviators who had these attributes.
However, as technology improved — air traffic-control systems became more mature, aircraft engines more reliable, and cockpit displays more accurate — the risk associated with night flying diminished. Lower risk lessened the necessity for specialized skills and temperament required to manage that risk. The result? A gradual elimination of the skills-based wage premium. Today, pilots who fly at night earn no more than those who fly during the day. Nor do they command a wage premium for flying over hazardous terrain (like mountains). That’s something early aviators could also count on for added earnings because it was considered more dangerous (and hence required more skill).
The wage-hindering effects of technology have been observed in other industries. Cab drivers in London could once command a hefty wage premium. The reason? Becoming one wasn’t easy. Would-be cabbies had to demonstrate encyclopedic mastery of London’s streets, and few could do so. The result was an earnings boost for cabbies: Scarcity after all, breeds value. But along came Uber. The ride-hailing giant equips its drivers with a powerful smartphone app that provides turn-by-turn instructions on where to go and how to get there. Landmarks, street names and routes are all laid out in meticulous detail.
That should benefit aspiring cabbies. And it did. Uber has — since its 2012 entry into London — created jobs for more than 40,000 drivers. It has afforded these drivers an opportunity to “make money and support my family,” as one driver told the BBC. But by purging the need for specialized knowledge, by making ferrying passengers around London easier, the Uber app also purged the need for encyclopedic mastery that had historically commanded a wage premium. The result? Accusations (and plenty of litigation to boot) that the ride-hailing giant underpays its drivers.
Automate with Caution
Technology can boost earnings particularly when using that technology demands specialized skills and knowledge. But bots can also depress wages by making some jobs easier to perform. If a job is simple, anyone can do it. And if anyone can do it, why pay some workers a premium? When the market demands fewer skills, workers with anything extra become less valuable.
This prospect may please firms. Paying workers less is a surefire way to boost margins. But this strategy is also risky. Technology does not purge the need for human labor but rather changes the type of labor required. Autonomous does not mean humanless. Technology can and will fail. And when it does, firms will confront the prospect of making nice with the very workers who — during automation’s better days — were shortchanged. In 2018, “Flippy,” a hamburger-flipping robot was forced to the sidelines after one day after being unable to keep up with customers orders. The restaurant’s response? Asking human cooks to step in.
Automation can increase productivity, improve efficiency, and reduce errors. Robots can, and should, occupy professions that are too risky for human workers to perform, offer little in the way of purpose, and deprive human workers of the joys of free living. Machines have — as Bertrand Russell aptly noted — “given us the possibility of ease and security for all.” Ignoring this reality, reasoned Russell, makes us foolish, “but there is no reason to go on being foolish forever.”
Yet the long-term benefits of forsaking humans for robots are hardly guaranteed. Firms stand to lose cash should the productivity benefits of technology adoption dwarf costs. These costs (and there’s always a cost) are typically discounted by firms keen to show solvency. But adopting bots can push firms further into the red. Technological singularity — the idea that machines know all, and can fix call — remains, despite what we’re told, a long, long way away.
Firms should consider this reality when adopting technology. Execs should ask themselves three questions when scrutinizing the value of bots. First, what can’t technology do? Technological valor may be dizzying but it too — much like humans — has limits. What are they? Second, how do those limits impact the operation? Investing in tech can boost productivity but only up to a point. What does that point look like and is it acceptable to shareholders? And third, how does the cost of overseeing technology affect is value proposition? Technology should be observed and kept in check. This is particularly true in safety critical industries like transportation, energy, and healthcare. What is the cost of doing so and how does it impact a bot’s cost advantage?
Asking these questions may reveal surprising answers about when (and under what conditions) forsaking human muscle for algorithmic prowess makes sense. There is, as Nicholas Carr notes, no economic law that says that everyone, or even most people, automatically benefit from technological progress.
This blog was originally published on Harvard Business Review.
