Data scientist, according to a 2012 Harvard Business Review article, is the sexiest job of the 21st century. Given that its authors are Thomas H. Davenport and D.J. Patil, the declaration is hardly surprising. Nonetheless, the IT industry is besotted with the term, and is expecting huge shortages of skilled workers.
A less well-known paper, “The future of employment: how susceptible are jobs to computerization?,” was published less than a year later by two Oxford academics (data scientists, perhaps). The authors, Carl Benedikt Frey and Michael A. Osborne, examine 702 detailed occupations in the U.S. labor market and estimate that about 47 percent of total U.S. employment is at risk from computerization “over some unspecified number of years, perhaps a decade or two.” Nearly 100 occupations, covering a wide range of skills from manual to mental, showed a 95 percent or higher probability of computerization.
The probability of computerization, or automation more generally, is not the only factor that determines whether jobs will be lost or not. The availability and cost of skilled labor, government regulations and social factors all play a role. One fact is indisputable, however: Automated solutions, whether hardware or software (or a combination of both), are getting cheaper, faster and more versatile at an ever-increasing rate.
In a market economy, where cost of production, including labor, is a major component of competition, ever cheaper and better automation will replace labor wherever possible. For example, Uber’s research into autonomous vehicles is clearly aimed at removing the cost of driver labor from their bottom line. Extrapolate vehicle autonomy to buses, trucks, etc. and you see the potential to decimate the near 4 million driver jobs in the U.S. labor market of 135 million. Add another 4 million retail cashiers, 4 million fast food counter workers, 8 million financial, information and record clerks, and the potential impacts become clear.
Economists, for the most part, are dismissive of the so-called Luddites. Past experience, they say, shows that in all prior technological revolutions, displaced workers re-skilled and took up jobs that had not existed previously. These economists seem immune to the reality that when today’s technologies create new jobs, they do so in far smaller numbers than the ones displaced.
Even technologists who should know better, with the honorable exception of Martin Ford and a few others, see only the benign outcomes. Technology, they say, has consistently improved living standards, health and longevity everywhere it has been introduced, an argument explored at length by Erik Brynjolfsson and Andrew McAfee in “The Second Machine Age.” True though this may be, there is one condition: The general public must be able to afford the good life on offer.
Competition, of course, decreases the cost of goods by reducing production costs through improved technology and the reduced labor costs it allows. So far, this impact on labor has usually been mitigated by the movement of displaced workers to other sectors of the economy and their continued earnings. The shift to services jobs in the Western world is just the latest example of this shift.
Automated solutions, whether hardware or software (or a combination of both), are getting cheaper, faster and more versatile at an ever-increasing rate.
However, today’s technological advances affect every sector of the economy. Where, for example, do the 20 million workers mentioned above move when their jobs are automated? They take lower paid jobs if available or they join the ranks of the unemployed. At a certain point, the rate of decrease in purchasing power within the economy will drop below the rate of decrease in the cost of goods. The economic and social consequences of this crossover have been repeatedly demonstrated in the Rust Belt cities of the U.S. Extrapolated to the entire Western economy, the outcome could be catastrophic.
Emerging economies fare little better. The surge of manufacturing jobs are quickly automated as living standards rise and workers demand higher wages, as has been seen already in China. Increasingly, the transportation and logistics costs of shipping products from offshored jobs become less attractive than highly automated production onshore. Some Third World countries may not even make it to the manufacturing plateau.
Mark Andreessen’s well-known essay “Why Software Is Eating the World” misses the point. Software (and hardware) is actually eating the social and economic foundation of the world. Automation is threatening to break the cycle of money flow between consumers and producers in the market economy by impoverishing the consumers who are identical to the workers whose cost is being eliminated. However, this voracious automation genie is well and truly out of the bottle and cannot be returned by conventional thinking.
With the cost of production of goods continuing to decrease as ever more jobs are automated, breaking the link between work and spending power seems to be the obvious, and perhaps only, solution. Automation continues to increase overall economic wealth. The issue that is coming to the fore is that it is inequitably distributed, with far greater rewards to capital rather than labor.
An idea that is starting to be whispered once again is that of a basic income, or guaranteed income, an idea espoused by Dr. Martin Luther King, Jr. as far back as 1967. Of course, some will call it communism; for many, it flies in the face of the “Protestant work ethic” that underlies much of Western culture. Certainly, it implies a redistribution of wealth earned by capital to those whose labor has been disenfranchised by automation and the competition for lower-cost production.
But, as basic-income advocate Scott Santens, puts it, “It’s not about stealing from the rich and giving to the poor, it’s about a minimum amount of access to resources.” In fact, it becomes a necessity if we are to preserve a market economy in “The Zero Marginal Cost Society” described by Jeremy Rifkin, although he advocates a somewhat different approach, with capitalism displaced by a system he calls the Collaborative Commons. Martin Ford in “Rise of the Robots” also discusses how a new economic paradigm might look, favoring a guaranteed income, as well.
Whatever the balance of the outcome, it’s pretty clear that jobs will never be the same again for the majority of people. A benign conclusion would be that people would work at jobs in which they were interested rather than only for the money. Others might devote themselves to continuous education, spiritual growth or hedonism; all would be equally valued. A new golden age of leisure and contentment. Unfortunately, the path to this utopia may well be strewn with thorns and barricaded with barbed wire, as the rich and established order clings to the past model of the market economy.
Indeed, data scientist may well continue to be the sexiest job of the 21st century. It will certainly be among the last…