Technology Will Not Save Us: The Decline of Manufacturing Jobs and the Rise of the Sharing Economy

The results of the recent U.S. presidential election delivered a jolt to the nation’s punditry, particularly its data-driven segment. The pollsters and analysts had all but assured the narrow victory of Democrat Hillary Clinton, but the winner was the unorthodox Republican Donald Trump. As soon as Trump’s election was certain, the statisticians whose models had not predicted this turn of events began to explain how it had happened. The most important demographic group, it appeared, was one that Trump had courted and Clinton had ignored: the white working class. Though Trump voters did not necessarily have below-average incomes, many of them had less education than the electorate at large. Jim Tankersley of the Washington Post echoed the thoughts of many journalists. “Whites without a college degree—men and women—made up a third of the 2016 electorate,” he wrote in the aftermath of the election. “Trump won them by 39 percentage points…[and] they were the foundation of his victories across the Rust Belt, including a blowout win in Ohio and stunning upsets in Pennsylvania and Wisconsin.” The dean of the statistician pundits, Nate Silver, summarized this conclusion by saying that “it appears as though educational levels are the critical factor in predicting shifts in the vote between 2012 and 2016.”

When There Are No Jobs

Clearly, Trump tapped into the economic and cultural resentments of this group. There are historical and cultural reasons why it was specifically white working-class voters who provided Trump’s margin of victory: African Americans have traditionally voted Democratic and Trump alienated Hispanic Americans with his anti-immigration platform. But the American working class, writ large, has every reason to want a drastic change in nearly every major trend in American economic life. In the United States, middle class wages have stagnated since the 1970s; those earning the median wage have seen their real pay increase by only 6% over the last thirty-five years. (Definitions and categories regarding class in the United States vary significantly from one source to another, so one person’s “middle class” might be another’s “working class.” In general, the members of the group considered here are neither living in poverty nor working in what might be called “professional” positions.) Even as the economy has grown in this period, wages have not. Jennifer Kaiser, an Indianapolis legal assistant interviewed by National Public Radio, explained her situation this way. “I get a raise, but every year it's lower and lower. The past few years, it hasn't even been a raise that's comparable to the cost of living.” The overall result is a slow but consistent downward pressure on that standard, accompanied by a persistent anxiety that a sudden financial jolt—such as an illness, accident, job loss, or the need to take in a parent or adult child—can completely shatter a person’s life.

The economic phenomenon that serves in equal parts as cause and symbol of this working-class decline is the decline in the American manufacturing sector. In 1947, manufacturing accounted for 25.4% of the country’s economy; in 2015, that figure was 12.0%. More important is that fact that employment is down significantly. As Vaclav Smil reports in his book Made in the USA, 17.64 million people worked in manufacturing in the United States in 1990, while only 11.8 million did so in 2011. The vanishing opportunities for work in manufacturing is a significant factor in the decline of the American working class. The mean manufacturing wage is $20.56/hour (excluding benefits) for nonsupervisory personnel. Very few jobs in today’s economy allow someone without a college degree to earn that much money. Today, an estimated one in six American men of prime working age are neither working nor seeking work. These so-called “missing men” are no longer participating in the labor force, even as job seekers. A White House report seeking to explain this phenomenon found that “men choosing not to work…explains relatively little of the trend, while reductions in demand, especially for less skilled workers, may be an important part of the story.” These men, in other words, are not lazy or uninterested in working; there are simply no jobs that match their limited skills. The American working class is not thriving, and the lack of available manufacturing work is a major cause of its malaise.

Historically, factory work played the same role in the twentieth century United States that cheap land had in the nineteenth: it allowed those without connections or wealth to work their way into the middle class. For three generations after World War II, Americans without a college degree could expect to own a car and a home, send their children to college, and retire comfortably. This broad-based prosperity has come to define the American Dream, and economic mobility is one of the central myths by which Americans understand their nation’s culture. A recent study detailing how infrequently American children grow up to earn more than their parents, for example, was greeted with a response befitting a national crisis. A lack of good jobs for those without higher education is, in fact, a tragedy of suffering and wasted human potential. But it is also a difficult reckoning with the society that we actually have, as opposed to the one we like to imagine.

Manufacturing and Technology

It is no coincidence, then, that one of Trump’s major promises to “make America great again” was to “bring back” jobs in this sector. But will the new president be able to keep his promise? The premise upon which Trump’s pledge is based is that American manufacturing firms have been sending jobs to other countries. To the extent that this is true, he can save the nation’s manufacturing jobs by keeping them from leaving, or even bringing them back: witness his recent claim to have convinced the Ford Motor Company to abandon its plans to move its Kentucky factory to Mexico and his deal with the Carrier corporation to retain one thousand jobs in Indiana. Moreover, Trump believes that American economic policies—particularly “free trade” as represented by the North American Free Trade Agreement and the Trans Pacific Partnership, among other such agreements—have made it easier for these jobs to leave the country. But Trump’s underlying presumption is mistaken. The primary driver of the decline of low-skilled American manufacturing jobs is not trade, but technology. In the words of New York Times economic columnist Paul Krugman, “the shift away from blue-collar work is mainly about technological change, not globalization, and no amount of tweets and tax breaks will bring those jobs back.”

Over the past several decades, manufacturing has been completely made over by advances in robotics and other automation technologies. Technology and increased efficiency has been a significant part of manufacturing practice since Frederick Winslow Taylor performed his “time and motion studies” on industrial workers and Henry Ford pioneered the assembly line for automobiles. But advances in silicon chips and robotic technology have allowed for unprecedented increases in output in the last twenty years or so. These advances are usually measured in terms of “efficiency” or “productivity,” which is, in economic terms, the amount of output one can get from a given input. If a firm can produce more goods with less capital or a smaller number of machines, it is more efficient. Unfortunately, this conception of efficiency treats workers as just another input: one way for a firm to increase efficiency is to use fewer workers to produce the same amount of goods.

The more productive workers become, then, the more likely they are to be put out of work. One study attributes 87.8% of manufacturing job losses between 2000 and 2010 to increases in productivity and only 13.4% to trade. (The numbers total over 100% because domestic demand was responsible for creating a small number of jobs in the period.) Firms today are simply able to produce more goods, with fewer humans, than they could in the past. As a result, they will hire a significantly smaller number of people. (A human welder earns about $25/hour, while the total cost for a robot welder is closer to $8/hour.) In becoming more efficient, workers have made themselves obsolete.

Footage from a Kia factory in Slovakia. Notice how few humans appear in this video. Source: YouTube

Former Secretary of Labor Robert Reich tells a story about visiting a new factory that “had been lured to a Midwestern state by a governor who had spent millions of taxpayer dollars subsidizing the project.” But when Reich toured the factory, he did not find the hive of activity that he expected. Instead, there were “about a dozen people sitting at computer terminals, typing instructions to the computerized machine tools and robots.” There is little disagreement on this point, even between those of markedly different ideological persuasions. On the right, Mark J. Perry of the American Enterprise Institute writes that “it’s the steady march of technological progress that has allowed the US to manufacture output at record levels in recent years with the same number of factory workers as in 1941.” On the left, Krugman wrote, via social media, that “the story of U.S. manufacturing is basically one of high productivity growth allowing demand to be met with ever fewer workers.” Writing in the New York Times Magazine, Binyamin Appelbaum summarized the issue this way. “From an economic perspective,” he wrote, “there can be no revival of American manufacturing, because there has been no collapse.”

Can Technology Be the Answer?



There are many who believe that this decline in good-paying jobs for Americans without college degrees is merely what Joseph Schumpeter called “creative destruction.” Under this characterization, capitalism is a dynamic process by which innovation constantly swept aside older ideas and practices that no longer served a useful purpose. While this might be a difficult process for those who had emotional attachments to the old ways of doing things, it would provide more benefit to more people in the long run. Interpreting the decline in manufacturing jobs this way suggests a painful adjustment for some that will nonetheless be the harbinger of something new and better. Some believe that it is already clear what that new, better thing is going to be. In this rosy future, they argue, not only will manufacturing jobs be obsolete, but so will jobs themselves. Instead, workers will string together a series of deals with different entities to perform one task or another at a negotiated rate.

This new way of understanding the world of work is referred to as the “sharing” or the “gig” economy. These umbrella terms group together several disparate practices, but they generally embody two specific concepts. The first is that the traditional notion of a “job” no longer applies. In the “sharing” economy, one rents out a privately owned good (say, a room in one’s house) for short-term usage. The “gig” economy is based on people performing certain discrete tasks (such as giving a person a ride to the airport in one’s own car) for a relatively small amount of money. (The terms “gig economy” and “sharing economy” technically denote different types of transactions. The former is a form of contract labor and the latter a short-term rental. But their economic logic and human consequences are identical, as is their reliance on new forms of technology. Here, the terms will be used interchangeably from this point forward.) The sharing economy requires no specific work hours or locations, nor does it provide health insurance, vacation and sick days, or steady income. In this new model, the person who was once thought of as a “worker” is now better understood as an “entrepreneur” and the one-time “employer” is now a “customer.”

Proponents cite this as a major virtue of the model. In April, for example, the ride-sharing company Uber announced the results of a lawsuit in which it was sued by its drivers. Despite the fact that Uber was ordered to pay them $84 million, the firm used that occasion to trumpet how much its “drivers value their independence—the freedom to push a button rather than punch a clock, to use Uber and [its competitor] Lyft simultaneously, to drive most of the week or for just a few hours.” The firm quoted one driver as saying “I wouldn’t even want to be an Uber employee,” as opposed to an independent contractor. “I would quit if they tried to make me an employee, because I value my freedom as an independent contractor too much.”

The second important aspect of the gig economy is that it is made possible by modern information technology. It would be impossible to make even a decent part-time income selling twenty-dollar car rides to strangers, or renting a room in one’s house for one or two nights, if a person had to line up each of those transactions one-by-one. Sharing economy arrangements are based on technology that links, in real-time, those that need something with those who have it. Prices go up and down instantaneously based on supply and demand of both gigs and workers. Time writes that Uber has become the fastest-growing startup in history. “Yet to hear the company tell it, Uber has done this without hiring a single driver; its role is simply providing software that allows willing parties to connect.” By the same logic, Airbnb is not a hotel, but a service that brings together those with empty rooms and tourists looking for a place to stay.

Advanced technology, market-driven innovation and futurism are all the rage in Silicon Valley today. So it is no surprise that business and technology media love the sharing economy. Forbes calls it “an economic revolution that is quietly turning millions of people into part-time entrepreneurs, and disrupting old notions about consumption and ownership.” The breathless tone is typical of the genre. Wired featured a piece of “partner content” from the CEO of gig-economy firm Fiverr calling the gig economy “the force that could save the American worker.” The Harvard Business Review published an article by an instructor in the M.B.A. program at Babson College who tells her students to “prepare themselves to be independent workers, not full-time employees” because “full time jobs are disappearing” and “both companies and workers prefer and choose the gig economy’s more flexible and independent work arrangements.”

What these advocates fail to incorporate into their analyses, however, is the fact that the disappearance of traditional jobs and the rise of the gig economy are, for the most part, terrible developments for the worker. If a person has no chance to get a job, then he she had better figure out how to work the gig economy. But this is not at all the same as arguing that that worker is better off in this brave new world. Advancing such an argument is both logically unjustified and ethically callous. Thus the advocates of the gig economy seldom go quite that far. Instead, they tiptoe up to that line with their content, and let their tone do rest of the work. The celebration of “innovation,” “flexibility,” “disruption,” “adaptability” and the like is thus the most offensive part of the entire intellectual program in support of the sharing economy.

As independent contractors, gig-economy workers are responsible for their own expenses. That includes, for example, taxes, insurance (including now-legally-mandated health insurance, which was often provided by an employer in the Paleolithic job economy) and equipment upkeep. Since the sharing economy is so new, there are no reliable metrics that track employment and earning numbers in the sector. The lack of reliable information allows for the generation of a lot of myths about the lofty rates of pay associated with gig economy work, tall tales sometimes fueled by sharing economy firms themselves. All of the lofty numbers being bandied about regarding sharing economy workers should therefore be taken with a very large grain of salt. A Washington Post reporter calculated that, for a driver who took in $62,000 a year working with Uber, “what’s left over after paying Uber’s commission, [his/her] gas, car maintenance, health and car insurance expenses, and [his/her] federal income and self-employment taxes" is about $27,600. Emily Guendelsberger, then a reporter for the now-defunct Philadelphia City Paper, began driving for Uber to get to the bottom of their wage claims. (Guendelsberger’s original article is an excellent exposé on the difference between the hype and the reality of gig-economy earnings. Unfortunately, in another negative outcome of the disruption wrought by information technology, her paper has since gone out of business and the article is no longer available online. The citation used above is to a Business Insider article reporting on Guendelsberger’s original article.) After keeping a meticulous log for 100 rides, and calculating her expenses, she concluded that if she worked “10 hours a day, six days a week with one week off, I'd net almost $30,000 a year before taxes.” Uber seems to generate the largest amount of controversy, coverage and data, but its problems do seem representative of the sharing economy as a whole. Sarah Kessler wrote up a damning, Barbara Ehrenreich-style essay about her own short time trying to live as a gig economy worker. Kessler worked primarily through TaskRabbit, and her experience suggests that instability, anxiety and pay that would be illegally low for an actual employee are par for the course.

The American manufacturing sector once supplied the most essential ingredient in the recipe for a broad-based middle class: good-paying jobs available to anyone with a high school education. Since technology has decimated that employment path, it would seem particularly fitting if Silicon Valley would create a new, 21st century backbone of a more modern sort of middle class. But wishing does not make it so, and the sharing economy is not the answer. It is insulting to the American worker to argue that he/she should rejoice at the creative destruction that replaces a stable, secure job with a precarious series of poorly-paying gigs. This new economy can only exist because people with no better options are forced to participate in it. The extent to which it thrives, then, is the extent to which the middle class continues to deteriorate. There are better options for facing our future.