Widgets Magazine


Automation, Robots and the Disappearing Worker

Days ago, I came upon a letter sent to The Economist, written by a jobless man whose desperation is hard to miss:

“I am young and unemployed and face a lifetime on the dole. Why? This morning I collected my jobseekers allowance from my bank, where I have it paid directly into my account. I did not see a cashier, but withdrew money from a cash point. Then I went to the supermarket… I scanned the items at a self-serve till, no need for a check-out assistant. I went home, switched on my Chinese computer and applied for jobs online. I do not send letters through the post; e-mail is more convenient. I then shopped online, I rarely use local shops. Who can I blame for the lack of jobs?”

A century ago, the same question had been on the minds of a group of workers in Yorkshire, Lancashire and Nottingham. In an act of rebellion, they took it out on spinning jennies and power looms, smashing the machines that had left low-skilled, low-wage laborers without work. Posterity calls them “Luddites,” a word that is today synonymous with being an old fuddy-duddy, decidedly anti-technology and anti-progress.

In Stanford and in Silicon Valley especially, being labeled a Luddite is almost tantamount to being sent into exile. After all, who can be anti-technology in the heart of techie paradise? Yet slapping the luddite label on the unemployed does not make the problem go away. Being so close to Silicon Valley, we see technological progress creating a seemingly endless stream of lucrative jobs; what we don’t see, however, is how it is also eliminating other types of jobs and leaving the typical worker worse off than before.

There is no doubt, for one, that workers are increasingly being squeezed out by robots and automation. And we’re not talking only about jobs at the lowest end of the pay scale– what were once considered “middle-class” jobs are also being hollowed out as your property agent gives way to mobile applications like Airbnb and Trulia and your baggage check-in staff at the airport are replaced by self check-in kiosks. With the advent of MOOCs, community college lecturers might in time find themselves confronting superstar lecturers who put their own necessity into question.

Today, a whole class of workers is being rendered irrelevant as technology like the Internet, big data and artificial intelligence are automating many routine tasks. It’s not as simple as robots replacing workers– digital processes are creating new processes that enable us to do more with fewer people and making human jobs obsolete at a faster pace the skills and organizations can catch up.

Under such circumstances, the market does what it does best: It rewards whoever adds the greatest economic value captured by the price mechanism. And they are, inescapably, owners of capital and machines that bring about greater productivity and profits. With capital-based technological change comes a notable shift in income away from labor: in the United States, the share of compensation in gross domestic income is at a 60-year low, and the share of middle-class income has fallen from 62 percent in the 1980s to 45 percent today. Wither the American dream– there is probably no worse time to be a worker with no special skills.

Who is to be blamed? The popular rejoinder proffered by governments all over has been uniformly disingenuous: market forces. The inexorable forces of market competition, so the story goes, has led to innovations that increase productivity, and international trade has put downward pressures on wages.

For many governments, especially those insisting on welfare minimalism, the sole corrective has been to promote labor productivity: The onus is always on the workers to play catch-up with the robots. But increasingly this is not going to work, because better education will not do much to increase incomes or reduce inequality as long as productivity increases of the machines outpace that of the worker– which it most certainly will.

“Market forces” is a convenient scapegoat because, being sufficiently nebulous, it doesn’t hold anyone responsible and creates the illusion that the plight of the middle-class is ‘inevitable’ in the face of unstoppable technological advancements and globalization. But if it is true that automation is efficient, it is simply untrue that it got there because of the market.

Much of the most important innovations were a result of public sector investment. Silicon Valley, for example, did not come about through private capital– before there was Silicon Valley there was Microwave Valley, which was essentially a federal project specializing in electronic intelligence for CIA and the military. Stanford had research labs working for the CIA, and several engineering doctoral theses were actually classified.

Before Google and Facebook became poster children for Silicon Valley, the largest employer in the valley had been Lockheed Martin. In short, what is now the world’s hotbed of innovation once started out as Uncle Sam’s experiment.

If Silicon Valley and all the technological disruptions that have made less-skilled workers obsolete is a result of government-driven market distortion, then the hollowing out of the middle class is a failure of government, not “technology” or “market forces”.

Luddites past and present are not anti-technology in the abstract– rather, the real struggle is against the restructuring of social relations at their expense. Historically, technology both creates and destroys jobs; increasingly, though, the costs of technological transitions are going to fall on the workers and the less skilled.

It is no coincidence that the United States is seeing a more unequal distribution of wealth despite tremendous increases in economic productivity. For governments and technological optimists (which we have no lack of in Stanford) alike, there is a need to re-visit the assumption that technological progress is a good in and of itself that can be allowed to eclipse notions of fairness and social betterment.

Contact Chi Ling Chan at chiling@stanford.edu

About Chi Ling Chan

Chi Ling, Chan ('15) is a junior majoring in Political Science and Symbolic Systems. On campus, she presently runs The Stanford Roundtable where she facilitates conversations on science, technology, society and more broadly, the human condition. In her free time, she writes. Chi Ling can be contacted at chiling@stanford.edu.
  • CommentBot

    This is a well-argued piece. I doubt anything will ever change, though. Beneath everything are more fundamental human traits. For fun, here are some good reads on this subject from various 20th-century thinkers: Kurt Vonnegut’s Player Piano, Bertrand Russell’s “In Praise of Idleness” (available online), Einstein’s “Why Socialism?” (also online).

  • AnonymousCoward

    I would argue that minimum wage hikes also accelerate this progress, the more a business needs to pay for low-skilled labor… the greater the likelihood that they will invest in automation to replace those workers.

  • Terry

    Latest robot can pick strawberry fields forever | The Japan Times http://www.japantimes.co.jp/news/2013/09/26/business/latest-robot-can-pick-strawberry-fields-forever/#.Uky_ThFeKJZ.twitter

    It is going to happen whether ppl want it or not–so they better figure out asap how to learn new job skills–many classes taught online and junior colleges cheaply!

  • Z

    Who is to be blamed? It is called progress ie a better mouse trap—you cant stop it and why would you?

  • Ed

    Daddy what is a truck driver? ATA: Self-Driving Trucks Are “Close To Inevitable” http://bit.ly/1g5lQtX Wise up Chi, quit drinkin the koolaid you liberal denier wacko!

  • Ed

    If Silicon Valley and all the technological disruptions that have made less-skilled workers obsolete is a result of government-driven market distortion, then the hollowing out of the middle class is a failure of government, not “technology” or “market forces”.<—–WRONG! Chan do you get out much from behind your puter?

  • Guest

    I might soon enter training for a military specialty that, I’ve been told, is on track to be entirely replaced by a computer in a dozen years or so. It’s a complicated problem. Hundreds (thousands?) of people will lose what was once a steady job. But each of those people consumed roughly 1.5 million dollars to be trained, not to mention at least 2 million dollars in salary and benefits over the course of a career. I wouldn’t be surprised if the taxpayer expense ran close to 5 million all told. That’s a lot of money. And there will likely be fewer casualties as a result of the new technology, at least for the “good guys.” But that’s all theory- when push comes to shove, can you program the computer to perform as well as a human? The last thing the government needs is to spend billions of dollars building subpar systems.

    “Progress” will likely win out. I (selfishly) hope it takes a while.

  • Gowtham

    I think what’s needed is that individuals should have access to capital and the ability to be entrepreneurial/innovative on the own without the need of huge capital. The govt. has made laws against raising investment from your friends. You can be an investor in your friend’s startup or any startup of your choice only if you have at least 1 million dollars of cash: http://www.sec.gov/news/press/2013/2013-124-item1.htm. Not possible when most of the country is in debt. Only recently did the govt. pass the law that allows crowdfunding, but that still limits the total amount an individual can invest so greatly that you wont benefit as an individual. Also the stock market is rigged against the individual: We pay a commission per trade, and cannot do high frequency trading. We don’t have as much information as the wall street firms, nor do we have all their talent and advice. These laws are VERY convenient for the rich people, and it is the rich who fund the govt campaign. The poor are left to use only labor as a means of growing their capital. Sad but true. America is not truely a free country. It is a country where the people with money rule. All the laws are made to favor the rich, and work against the poor. No way will I call that fair. The rich can double their wealth by simply giving you education or housing loan. Do you have that option?

  • Gowtham

    And yea, I don’t even have to mention how patent offices work. Companies who have money and can afford lawyers abuse the patent system by filing thousands of patents on the most ridiculous aspects that they call innovation. The real innovation by individuals really have little chances of being protected. No lawyers. Just read a few patents for yourself and you will see what I mean. A small company cannot afford court fees and will of course be killed even if they were right,

  • Chi Ling

    A capitalist system will always work in favor of those with capital and property – that much is nothing new. True not just for America but for all capitalist societies; what is exceptional about US has been the rate by which inequality has increased over the years. The problem at hand is whether the US would entertain a more inclusive capitalist system through redistribution – but we have seen that calls for redistribution have been constantly drowned out by narratives that more welfare for the poor is antithetical to American work ethic. Until people start realizing that the rich didn’t get there by their own effort alone things are unlikely to change.

    And on money hijacking politics, Lessig made a pretty powerful speech on that sometime back: http://www.ted.com/talks/lawrence_lessig_we_the_people_and_the_republic_we_must_reclaim.html

  • Chi Ling

    That’s what artificial intelligence is trying to do, although it is all very experimental right now (as with all new technology). What would be interesting to see is whether AI develops in a way that permits division of labor, so that humans and computers can do different things and trade these strengths to realize more gains. In this scenario AI is not substitute for humans, but rather a compliment.

    As for your military specialty – if you already know that it is on its way to becoming irrelevant because of a computer program, then the smart thing to find what it is that the computer can’t do (at least in the next 10 years) and be damn good at it. Better yet, learn to program it, if it’s something up your alley.