Automation and the Future of Work

AUTOMATION AND THE FUTURE OF WORK

AUTOMATION AND THE

FUTURE OF WORK

Aaron Benanav

Th is paperback edition first published by Verso 2022

First published by Verso 2020

© Aaron Benanav 2020, 2022

An earlier version of this text appeared as Automation and the Future of Work,

New Left Review, nos. 119, Sept.–Oct. 2019 and 120, Nov.–Dec. 2019.

All rights reserved

The moral rights of the author have been asserted

1 3 5 7 9 10 8 6 4 2

Verso

UK: 6 Meard Street, London W1F 0EG

US: 20 Jay Street, Suite 1010, Brooklyn, NY 11201

versobooks.com

Verso is the imprint of New Left Books

ISBN-13: 978-1-83976-132-4

ISBN-13: 978-1-83976-130-0 (UK EBK)

ISBN-13: 978-1-83976-131-7 (US EBK)

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress

Typeset in Garamond by Hewer Text UK Ltd, Edinburgh

Printed and bound by CPI Group (UK) Ltd, Croydon CR0 4YY

Contents

List of Figures and Tables

Preface

   1. The Automation Discourse

   2. Labor’s Global Deindustrialization

   3. In the Shadow of Stagnation

   4. A Low Demand for Labor

   5. Silver Bullets?

   6. Necessity and Freedom

Postscript: Agents of Change

Notes

Index

Figures

Figure 1.1. Labor Share of Income, G7 Economies

Figure 1.2. Productivity–Wages Gap, OECD Countries

Figure 2.1. French Manufacturing Sector

Figure 2.2. Global Waves of Deindustrialization

Figure 2.3. Deindustrialization in China, India and Mexico

Figure 2.4. World Manufacturing and Agricultural Production

Figure 3.1. French Manufacturing and Total Output Growth

Figure 3.2. World Manufacturing and Total Production

Figure 4.1. Unemployment Rates in the US, Germany, and Japan

Figure 4.2. OECD Index of Employment Protection

Figure 4.3. Service Sectors in the US, France and Italy

Figure 4.4. Service Sectors in Thailand, Mexico and South Africa

Figure 5.1. Gross Government Debt-to-GDP

Tables

Table 2.1. Manufacturing Growth Rates, 1950–2017

Table 3.1. Manufacturing and GDP Growth Rates, 1950–2017

Table 3.2. Capital Stock and Labor Productivity Growth Rates, 1950–2017

Preface

THE INTERNET, SMARTPHONES, AND social media have already transformed so much about the way we interact with each other and come to know the world. What would happen if these digital technologies moved off the screen and increasingly integrated themselves into the physical world around us? Advanced industrial robotics, self-driving cars and trucks, and intelligent cancer-screening machines appear to presage a world of ease, but they also make us uneasy. After all, what would human beings do in a largely automated future? Would we be able to adapt our institutions to realize the dream of human freedom that a new age of intelligent machines might make possible? Or would that dream turn out to be a nightmare of mass technological unemployment?

In two New Left Review articles published in 2019, I identified a new automation discourse propounded by liberal, right-wing, and left analysts alike. Asking just these sorts of questions, automation theorists arrive at a provocative conclusion: mass technological unemployment is coming and can be managed only by the provision of universal basic income, since large sections of the population will lose access to the wages they need to survive.

In this book, I argue that the resurgence of the automation discourse today is a response to a real trend unfolding across the world: there simply isn’t enough work. This chronic labor underdemand is manifest in economic trends such as jobless recoveries, stagnant wages, and rampant job insecurity. It is visible as well in the political phenomena that rising inequality catalyzes: populism, plutocracy, and the rise of a new, sea-steading digital elite—more focused on escaping in rockets to Mars than on improving the livelihoods of the digital peasantry who will be left behind on a burning planet Earth.

Pointing with one hand to the homeless and jobless masses of Oakland, California, and with the other to the robots staffing the Tesla production plant just a few miles away in Fremont, it is easy to believe that the automation theorists must be right. However, the explanation they offer—that runaway technological change is destroying jobs—is simply false. There is a real and persistent under-demand for labor in the United States and European Union, and even more so in countries such as South Africa, India, and Brazil, yet its cause is almost the opposite of the one identified by the automation theorists.

In reality, rates of labor-productivity growth are slowing down, not speeding up. That should have increased the demand for labor, except that the productivity slowdown was overshadowed by another, more eventful trend: in a development originally analyzed by Marxist economic historian Robert Brenner under the title of the long downturn—and belatedly recognized by mainstream economists as secular stagnation or Japanification—economies have been growing at a progressively slower pace. The cause? Decades of industrial overcapacity killed the manufacturing growth engine, and no alternative to it has been found, least of all in the slow-growing, low-productivity activities that make up the bulk of the service sector.

As economic growth decelerates, rates of job creation slow, and it is this, not technology-induced job destruction, that has depressed the global demand for labor. Put on the reality-vision glasses of John Carpenter’s They Live, which allowed the protagonist of that film to see the truth in advertising, and it is easy to see a world not of shiny new automated factories and ping-pong-playing consumer robots, but of crumbling infrastructures, deindustrialized cities, harried nurses, and underpaid salespeople, as well as a massive stock of financialized capital with dwindling places to invest itself.

In an effort to revive stagnant economies, governments spent almost a half century imposing punishing austerity on their populations, underfunding schools, hospitals, public transportation networks, and welfare programs. At the same time, in a world of ultralow interest rates, governments, businesses, and households took on record quantities of debt. They did not do so to invest in our digital future, as former Federal Reserve chairman Alan Greenspan argued they would in the midst of the late-1990s tech bubble. Instead, firms mortgaged their assets to pay off shareholders, while poorer households borrowed in an effort to make ends meet.

These trends have left the world economy in an incredibly poor position as it faces one of its greatest challenges: the COVID-19 recession. Dilapidated healthcare systems have been overrun with patients, and schools have closed that were for many children vital sources of basic nutrition (and for parents, of much-needed childcare). Heavily indebted companies watched their stock values plummet, at least initially, at rates not seen since the Great Depression. Unemployment rates rose significantly across the world, and stratospherically in the United States, leaving large parts of the population unable to pay for food, medical care, or housing. In spite of massive monetary and fiscal stimuli, weak economies are unlikely to bounce back quickly from the shock. It is easy to see how over the long term, the COVID-19 recession will accelerate what are by now long-unfolding trends of rising economic insecurity and inequality.

It is precisely for this reason that it is so important to reflect on today’s automation discourse. Automation theorists offer a utopian reply to our dystopian world. Remove the They Live reality-vision glasses and return for a moment to the world of fantasy inhabited by these authors. In it, we all work less (like the victims of the present recession) yet have access to everything we need to make a life; we spend more time with our families (but not because we are in imposed isolation); the elderly jog through parks wearing new exoskeleton jumpsuits (rather than dying in hospital beds); and the air has been cleared of smog because we are transitioning rapidly to a world of renewable energy (rather than because factories have been shuttered and people are no longer driving cars). With the exception of the exoskeleton jumpsuits, all of this is possible now if we fight for it. We can already achieve the post-scarcity world that the automation theorists invoke, even if the automation of production proves impossible.

My interest in this topic arose from two distinct sources, one in the deeper past and the other more recent. Like many of the automation theorists, I grew up in the 1980s and ’90s reading science fiction novels and watching the spacefaring communists of Star Trek: The Next Generation tour the galaxy. My father, who inspired these interests, was himself a researcher in the field of automation. Like many of his peers, he left a career in academia to try his luck in the startup culture of the 1990s. Some people made a lot of money in those years, but many more did not: most internet startups went bust, leaving their overworked engineers with little to show for their efforts. Interning with him at a different company every summer of high school—writing HTML and Javascript—I decided that there was little promise of happiness to be found in the digital economy, so I devoted myself instead to studying the history of economic growth and unemployment, the twin engines of prosperity and insecurity in the contemporary economy.

In the aftermath of the 2008 crisis, I became involved in the social movements of my time, an experience I attempted to digest through conversation and collaboration with fellow members of the Endnotes collective. The unsigned, coauthored texts we wrote have greatly influenced the analysis to be found in these pages. It was through an encounter with two critics—Nick Srnicek and Alex Williams, whose Inventing the Future (2015) is a key example of the left wing of the automation discourse—that I discovered the intellectual ecosystem populated by the automation theorists, which led me back to my childhood love of science fiction and at the same time transformed my outlook on the future.

As I read book after book by the automation theorists, supplementing that still-growing reading list with forays into the utopian and science fiction literatures of the past, the conviction grew within me that, collectively, these authors had done more than anyone I have yet encountered to think through the logical organization of a postcapitalist society and to imagine the pathways by which we might get there. I disagreed with their analysis of the present but saw responding to their vision of the future as a way to develop my own, which by comparison with theirs was still of the dullest-possible grey. In the pages that follow, I explore possibilities for achieving a post-scarcity future without the full automation of production: by sharing the work that remains to be done in a way that restores dignity, autonomy, and purpose to working life without making work the center of our shared, social existence.

In the course of an exposition and critique of the automation discourse, I lay out a brief history of what has happened to the world economy and its workforce over the past fifty years, focusing on the origins and development of the present-day, chronically low demand for labor. I discuss the policy alternatives that aim to resolve this market failure—neoliberal structural adjustment, Keynesian demand management, and universal basic income—and sketch out a post-scarcity world against which they should be measured.

Writing this book has only further convinced me that turning the tide toward a more humane future will depend on the refusal of masses of working people to accept a persistent decline in the demand for their labor and the rising economic inequality it entails. Struggles against these outcomes were unfolding with increasing intensity across the globe before the COVID-19 recession, and have recently resurged. We need to immerse ourselves in the movements born of these struggles, helping to drive them forward. If they fail, maybe the best we will get is a slightly higher social wage in the form of universal basic income—a proposal governments are now testing out as a possible response to the present recession. We should not be fighting for this modest social goal, but rather to inaugurate a post-scarcity planet.

I could not have written this book without the support and friendship of many people, including: Perry Anderson, Arielle Angel, Elyse Arkind, Marc Arkind, Mia Beach, Dan Benanav, Ethan Benanav, Mandy Benanav, Jasper Bernes, Mårten Björk, Jan Breman, J. Dakota Brown, Jonny Bunning, Paul Cheney, Christopher Chitty (RIP), Joshua Clover, Chiara Cordelli, Oliver Cussen, Daniel Denvir, Andreas Eckert, Hugh Farrell, Adom Getachew, Maya Gonzalez, Daragh Grant, Lee Harris, Gary Herrigel, Joel Isaac, Felix Kurz, Rachel Kushner, Natalie Leonard, Jonathan Levy, Marcel van der Linden, Rob Lucas, Neil Maclean, Henry Mulheim, Jeanne Neton, Mary Ellen O’Brien, Chris O’Kane, Moishe Postone (RIP), Thea Riafrancos, Pavlos Roufos, Bill Sewell, Jason Smith, Maureen Smyth, Juliana Spahr, Zöe Sutherland, Ben Tarnoff, Sarah Watlington, Suzi Weissman, Björn Westergard, Gabriel Winant, and Daniel Zamora, as well as participants in the History and Theory of Capitalism Workshop and the Society of Fellows Workshop, both at the University of Chicago. I am especially grateful to Chloe Watlington, Robert Brenner, John Clegg, and Charlotte Robertson, who supported me in my research and writing every step of the way. Lastly, thank you to my editors at the New Left Review, Susan Watkins, Tom Hazeldine, Emma Fajgenbaum, and Lola Seaton, and at Verso, Tom Hazeldine (again), Duncan Ranslem, and Sam Smith. Thanks especially to Tom, who pushed this project along an accelerated timeline despite a world turned upside down.

This book is dedicated to Chloe Watlington.

Aaron Benanav

Chicago, June 2020

CHAPTER 1

The Automation Discourse

RAPID ADVANCES IN ARTIFICIAL intelligence, machine learning, and robotics seem set to