- The digital revolution is driving three trends in working life: individualization, increasing dependence on technical systems, and the erosion of the boundary between personal and professional space.
- Questions remain about the number of jobs available in a digital economy. Automation and artificial intelligence appear to be making some roles obsolete while creating the need for others, still vague and ill-defined.
- It is suggested that the labor market is becoming polarized between the high value-added jobs of the digital economy and various forms of precarious work (“bad jobs,” “microjobbing,” etc.). These trends go against the interests of the middle classes and the so-called intermediate occupations.
1 The world of work is changing, and academia is taking notice. Certain authors maintain that the technological and organizational revolution we are witnessing today will have a positive effect on society, helping workers unleash undreamt-of productivity and presenting them with a far greater range of options. Others warn that our working lives will be ruled by anonymous digital platforms, significantly curtailing individual autonomy over the organization of tasks. The technological upheaval underway has been said to herald both untold freedom and a new form of collectivized society.
2 Here, I do not claim to settle any debates but to present the often-conflicting evidence that must be our starting point for drawing inferences about how digitalization will reshape the world of work. After (1) offering a definition of the digital economy, I will outline the primary shifts we are seeing, with an emphasis on those pertaining to (2) how work is organized and (3) the overall employment rate.
What do we mean by the digital revolution?
3 While the word “digital” is everywhere we look, it is rare to come across a concise definition. The digital economy is at once a set of technologies, a way of organizing production, and a culture. Technology allows any type of information, without exception, to be codified systematically into a series of 1s and 0s. The organization of production has become dependent on two technologies: computers (to process coded information at lightning speed) and the internet (to distribute that information near-instantaneously, at virtually zero cost). Together, they open up a vast panorama of opportunities for integrating all kinds of information, stowing it away on servers, processing it with algorithms, and sharing it through networks in ways never imagined in all of human history.
4 Digitalization also implies a cultural shift and a totalizing vision of “digital society.” Instant communication generates new behaviors, while social media challenges the institutions in which we have traditionally placed our trust. Digitalization radically alters consumption patterns in a pervasive market-driven economy, by lowering transaction costs and spawning an ever-growing number of platforms for reviewing products and services, bidding and auctioning, and calibrating prices for everything from travel and hotel bookings to training courses and investments.
5 Now, economic production looks very different. Boston Consulting Group identifies nine defining innovations of the digital revolution: big data and analytics, autonomous robots, simulation, horizontal and vertical system integration, the Industrial Internet of Things, cybersecurity, the cloud, additive manufacturing (using 3D printers to build up products layer by layer), and augmented reality (the possibility of operating within a virtual environment).  Finally, having penetrated the inner core of the productive system, digital technologies are transforming how we work.
6 The abundance of literature on this subject points to three main trends: 1) the individualization of work, 2) increasing worker dependence on technical systems, and 3) an erosion of the boundaries between private and professional space.
The individualization of work
7 We now have the technological capital to individualize and autonomize human work more than ever before, thanks to access to vast quantities of information. Rifkin  believes that we are witnessing the most momentous revolution since the birth of capitalism. He argues that the technological capital required to create wealth—which now happens through processing information, creating connections, and collaborating with others—is within the reach of anyone equipped with a basic laptop and a smartphone. That covers 87 percent of the French population. Thanks to online recruitment and trading platforms, the transaction costs of finding new clients have plummeted. It is now much easier for anyone to sell their services, whether that’s running an online store or offering professional advice. With more autonomy to organize one’s work, individuals are free to explore and develop their creativity, remote working being the basic starting point. But individualized production is not reserved for entrepreneurs and freelancers; it can also be seen among salaried employees, albeit in more subtle forms. At the furthest end of the spectrum, online labor platforms like Amazon Mechanical Turk  allow their many thousands of users to carry out one-off “microtasks” on an earn-as-you-go basis (a model sometimes referred to as “crowdworking”).
8 The relationship between supply and demand is becoming more fluid, as digital tools allow individuals to offer new kinds of services: 46 percent of new companies registered in France in 2016 were involved in B2C online sales, short-distance transport, or consulting.  Yet, digitalization has not fueled an explosion in freelance work: 11 percent of workers in France were self-employed in 2016, down from 20 percent in 1980. Like all developed economies, the United States has seen a slight upturn in the numbers opting to go it alone since 2010, but this is far from the radical shift we might expect, amid all the media hype and a somewhat loose approach to the facts. Salaried employment remains the dominant model of formal economic participation, and that is not about to change.
Increasing dependence on technical systems
9 These newly enfranchised workers, whether employees or self-employed, are nonetheless tied to the technical systems that enable their “freedom.” Although each individual may work entirely alone, the work itself is always managed through some kind of digital platform. It is the platform that facilitates connections, packages individual tasks, and determines the type of work available: to achieve any real autonomy, users must master the algorithms that coordinate every aspect of their role.
10 A tension therefore arises between the flexibility that allows them to take advantage of these platforms in the first place and a dependence on the tempo and workflow their algorithms impose. The neologism “Uberization” has captured this tension and brought it into the mainstream. California-based Uber works with thousands of private-hire drivers, allowing each of them to “be their own boss.” But total dependence on a platform that sets the price and pace of work makes this a precarious, even alienating, existence.
An erosion of the boundaries between private and professional space
11 Digital work adheres to a production logic characterized by constant, global flows. Both employees and freelancers can log in to the networks they need and earn a living at any time and from any location. Digital work blurs the once-clear lines between professional and nonprofessional activities. Until recently, domestic, voluntary, and community-based activities all belonged to the private sphere, whereas professional life was split into freelance and salaried work.
12 Digitalization is muddying these distinctions. First, let’s look at businesses: today, 20 percent of employees on permanent contracts work variable or nontraditional hours, 30 percent expect to work on Sundays, 10 percent are free to choose their own hours, and 17 percent work remotely.
13 As the concept of the working day becomes more fluid, traditional hierarchies are wavering. Until recently, ascending the corporate ladder was dependent on competence, thus concentrating knowledge in those empowered to make decisions. This distinction between advanced and entry-level knowledge has become devalued in a world where information is freely and instantly available, and organizations hoping to foster innovation do so precisely by adopting an open-access model. Today’s employees feel less invested in a corporate project than in a space that nurtures (or neglects) their own personal development.  In the world of human resources, this radical transformation of workplace hierarchies is the source of much debate, particularly when it comes to Generation Y, said to be put off by old-school pecking orders and destined to become the managers of tomorrow.
14 In nonprofessional work, the implications of the digital revolution are even more profound. Online platforms make it easier to contribute to collaborative projects and causes; this is how, for example, in less than ten years we ended up with the largest encyclopedia ever created (Wikipedia), after several million internet users donated their expertise free of charge across an unfathomable variety of topics. Free, open-source resources allow thousands of internet users to join and benefit from all kinds of initiatives in ways that overlap with what was once regarded as “professional” activity. This applies as much to free software (Linux, Mozilla, Apache) as to amateur music or film productions that have completely redefined their respective industries, forced to adapt to the irrepressible new marketplace for free peer-to-peer services. Amateur and commercial services are now in direct competition.
15 The countercurrent to this is a marketization of work that was previously unpaid. Internet users can now supplement their income by using so-called capital platforms. These platforms open up opportunities to monetize personal assets that can be rented out for short-term use, such as cars (Blablacar) or apartments (Airbnb). Picking up a hitchhiker or hosting a friend in a spare room, once voluntary acts of kindness, now hold the promise of financial reward. The boundary between domestic time and professional time is flexing as a result, in an interesting echo of preindustrial practices. For millennia, the distinction between private and professional space was simply nonexistent. It was only the rationalization of production in factories and offices in the nineteenth century that instituted the work/life partition.
16 What is genuinely new is not so much the lack of distinction between the personal and professional spheres, but the radical destabilization of working rhythms. Workers are more in control of their hours, but also constrained by the technical and organizational system in which they are embedded. A classic example is the constant deluge of emails that often require workers to respond outside of their preferred hours (meetings across time zones, emergency management accelerated by the sheer speed of data processing, and so on). There is an obvious friction between growing demands for remote working and the right to disconnect.
How the individualization of work creates tension between freedom and technical dependence
17 In short, we are witnessing a dual polarization: yes, many can look forward to greater individual autonomy, skill sharing, creativity, and self-direction; more scope for innovation at the individual and collective levels; and a cornucopia of opportunities to monetize their expertise. But only if they consent to be bound by the parameters of technical platforms that make it possible to interact, collaborate, and supervise each other. This creates a heightened dependence on platforms, technologies, and digital networks—and on those who control their algorithms and objectives. Whether we take an optimistic or pessimistic view of the future depends on the gravitational pull we attribute to each of these two poles.
18 However, in each case, the stage is set for further individualization, undermining the concept of work as a collective endeavor. Not only can we contribute to meetings or projects “remotely”; we can also train others or diagnose illnesses without meeting face to face, by relying on technology to manage contacts, evaluations, pending tasks and decisions, and any necessary readjustments. Yet, one of the fundamental aspects of work is how it builds communities: a coexistence founded on a common purpose. Today’s “virtual communities” demand only a loose affiliation of their members, the corollary to the autonomy promised by a capitalist society that is becoming more and more “liquid,” in Bauman’s sense of the word.  As a society, we should keep a close eye on this phenomenon.
Changes to employment
19 Work is organized and carried out within a structure of defined professional roles. The question of how the digital economy will change the nature of these roles draws us into a tangle of uncertainty and debate, often complicated further by false claims about the “end” of salaried employment, if not work itself.  In fact, we can be confident that salaried work will continue to prevail in the developed world for at least a few more decades, and will continue to grow and flourish in developing economies. There is no reason to suppose that the end of work, or of salaried employment as its dominant form, is in any way imminent. The impact of digitalization on employment levels, that is, the number of salaried employees or self-employed workers (or the number of hours of paid work), is a different matter entirely. Here, economists are divided.  In a 2014 survey of 2,000 experts from all over the world, half agreed that artificial intelligence and automation would lead to a net loss of jobs, whereas the other half believed that any jobs destroyed would be replaced elsewhere. 
Toward a new occupational landscape
20 There is a lot of room for debate; we are talking about two novel technological phenomena, automation and artificial intelligence, coming together to transform employment in ways that cannot be deduced from past experience. Research by Daron Acemoglu and Pascual Restrepo at MIT concludes that between 1993 and 2010, each new robot replaced on average 5.6 employees, and that one new robot for every 1,000 employees reduced employment by 0.34 percent and salaries by 0.5 percent.  Will new generations of robots have a deeper impact? A much-discussed study by Carl Frey and Michael Osborne at the University of Oxford examined more than 700 different professions, concluding that, over the next two decades, 47 percent of tasks in the United States and 37 percent in the United Kingdom would be affected by automation.  In terms of net job destruction, Arntz, Gregory, and Zierahn predict losses of 9 percent for the United States, 10 percent for the United Kingdom, 9 percent for France, and 7 percent for Japan.  In the French context, analysis by France stratégie downplays these figures by pointing to the banking sector, demonstrating that previous phases of automation led to a restructuring of professional roles. 
21 It is difficult to say what artificial intelligence will mean for highly technical roles. A team from Oxford and Yale surveyed 352 scientific experts in this field with a view to identifying key trends. The results suggest that machines will be capable of outperforming humans in tasks such as translation by 2024 (compared to amateurs), writing high-school essays (by 2026), driving a truck (by 2027), working in retail (by 2031), and performing surgery (at some point after 2050).  In the past, only low-skilled jobs were vulnerable to automation. We have now reached an inflection point where computers are capable of performing calculations and interpreting data with such sophistication that certain high value-added skill sets could be made obsolete. Automation will have a powerful impact on managerial and specialist roles, and we do not yet know what new ones will emerge. 
The uncertain future of jobs
22 In the face of the digital revolution, skilled professionals can no longer afford to specialize too tightly. Knowledge is becoming more inclusive than exclusive, the role of the expert being to make creative connections between data points available to all. For some authors, such as David Deming, this emerging environment will usher in a new generation of interface and network-management roles.  In parallel, expertise and power will lie with those skilled in rapid data management by devising ever-more elaborate processing algorithms.
23 The prestige attached to knowledge (traditionally enjoyed by teachers, researchers, politicians, and doctors) will be displaced, as the content of that knowledge becomes less valuable than the production of algorithms capable of automating the interpretation of information. Changes in the perceived value of work, particularly intellectual work, will trigger a reshuffle in the labor market, perhaps even a “digital exodus” to rival the “rural exodus” of the past. 
24 The changing nature of high value-added jobs will force workers to adapt with an agility to match the exponential productivity gains achievable through automation. If we subscribe to the idea of “creative destruction,” society is locked in a race against time to come up with new roles in a context where growth is weak and capital productivity soaring. This calls for massive retraining policies—but retraining in what? For the moment, the question remains completely up in the air.
The rise of portfolio careers
25 If opportunities for salaried employment are receding, “nonprofessional” work may offer a promising alternative. A study by JPMorgan Chase estimates that active use of labor platforms (Uber, crowdworking sites, etc.) generates a replacement income equivalent to 12 percent of a regular salary, whereas capital platforms (e.g., Airbnb, Blablacar) boost overall income by up to 15 percent.  In other words, platform-mediated piecework helps compensate for lost earnings, but work associated with renting out property contributes to income growth—good news for those who own capital.
26 As the content of work continues to change, there appears to be a shift toward portfolio careers, where supplementary income sources are offset by the growing precarity of opportunities in the labor market. Countries with low unemployment also tend to have a higher rate of part-time working: comparing France and Germany, it transpires that France posted a higher unemployment rate (10.5 percent versus 4.8 percent) but a considerably lower rate of part-time employment (18 percent versus 26 percent) than Germany, where, in the period between 1999 and 2011, “all job creation can be explained by the increase in part-time roles.” 
A squeeze on the middle classes
27 Until relatively recently, poverty was associated with being out of work. But today it is just as likely to strike among the gainfully employed. A new cohort has emerged: the “working poor.” An international study by Brady, Fullerton, and Cross reveals that among individuals or families living below the poverty line, 30 percent are in work in the United Kingdom, and as many as 50 percent in Germany and France and 65 percent in the United States. 
28 One consequence of this is a lively debate over what it now means to be middle class. For the last two hundred years, economic growth has been predicated on the expansion of the middle classes. In developed countries, income distribution is smoothed out by the existence of a very significant majority of middle earners. So, in France, 95 percent of the population have a monthly income of between €900 and €3,100, with a median of €1,772 in 2014. 
29 Yet, many commentators argue that the transformations of work will leave median earners and middle-class households worse off. In the United States, very high earners have seen their incomes rise rapidly, whereas middle- and lower-income households have seen theirs stagnate.  In the “winner-take-all society,”  the majority of wage growth is monopolized by a handful of exceptionally high-skilled jobs, many of them in digital industries. If this trend persists, the labor market will be reshaped in ways that exacerbate unequal distribution of the gains of economic growth, just as we saw in the 1990s and 2000s with inflated remuneration packages for finance roles.
Economic outlook and societal issues
30 Today’s techno-social transformations of work have led to more than just a new economy: society as a whole has been altered by the digital technical system and the disruptions to working life it implies. In summary, this new economy is defined by 1) a productive system organized around two poles: dominant centralized platforms and the decentralized use of productive assets right down to the individual level; 2) a dramatic shift toward the individualization of work in the context of a regimented technological system; 3) the obsolescence of certain jobs and the emergence of new ones, the overall balance being uncertain; and 4) a transformation in how all kinds of work are remunerated, with more individualized contracts and supplementary income streams.
31 These evolutions will have a profound effect on the middle classes, who in previous decades relied on skilled jobs to get ahead. There is a real probability that current social equilibria will be unsettled, as a whole swathe of the population employed in white-collar and management roles find their value in the labor market undercut.
32 More generally, the market will continue to permeate more and more aspects of our lives, and work will become increasingly commodified. The demarcation between state and market, so clear-cut in the 1980s, has been replaced by a new opposition: between an economy where the technical system works for us (technological decentralization; an upsurge in local initiatives thanks to the ease of collaboration; open-source resources enabling entrepreneurship at a low cost of entry), and one where we work for an all-encompassing technical system, with human beings just one factor of production among many (preponderance of global platforms; centralization of big data management; massive data processing driven by artificial intelligence).
33 The implications are therefore not just economic and social, but also societal. The digital economy is transforming social relations. Corporate control of personal data, the ascension of all-powerful multinational companies, robotics, and the effects of vast social networks on human relationships all paint a picture of a new kind of civilization, deeply rooted in the transformations in the world of work we are beginning to observe today.
Michael Rüßmann et al., “Industry 4.0: The Future of Productivity and Growth in Manufacturing Industries,” Boston Consulting Group, 2015, https://www.bcg.com/publications/2015/engineered_products_project_business_industry_4_future_productivity_ growth_manufacturing_industries.
Jeremy Rifkin, The Third Industrial Revolution: How Lateral Power Is Transforming Energy, the Economy, and the World (London: Palgrave Macmillan, 2011).
Amazon Mechanical Turk (AMT) is an online labor platform launched in 2005, which connects autonomous workers with clients seeking help with straightforward tasks (evaluation, writing reviews, word-for-word translation, etc.) paid on an hourly basis. The average rate is $2 per hour. In 2017, AMT had 500,000 registered users in the United States, 20 percent of whom were using it as their main source of income (source: www.mturk.com).
Félix Bonnetête and Nicolas Bignon, “La création d’entreprises en 2016,” INSEE première 1631 (2017).
Denis Pennel, Travailler pour soi: Quel avenir pour le travail à l’heure de la révolution individualiste? (Paris: Seuil, 2013).
Zygmunt Bauman, Liquid Modernity (Cambridge: Polity Press, 2000).
Jeremy Rifkin, The End of Work: The Decline of the Global Labor Force and the Dawn of the Post-Market Era (New York: Putnam Publishing Group, 1995).
For more on this ongoing debate, see Daniel Cohen, The Infinite Desire for Growth, trans. Jane Marie Todd (Princeton, NJ, and Oxford: Princeton University Press, 2018); and Bernard Stiegler, L’emploi est mort, vive le travail! Entretien avec Ariel Kyrou (Paris: Les Mille et une nuits, 2015).
Aaron Smith and Janna Anderson, “AI, Robotics, and the Future of Jobs,” Pew Research Center, 2014.
Daron Acemoglu and Pascual Restrepo, “Robots and Jobs: Evidence from US Labor Markets,” Boston University, March 17, 2017, https://economics.mit.edu/files/19696.
Carl Benedikt Frey and Michael Osborne, “The Future of Employment: How Susceptible Are Jobs to Computerisation?” The Oxford Martin Programme on Technology and Employment, September 17, 2013.
Melanie Arntz, Terry Gregory, and Ulrich Zierahn, “The Risk of Automation for Jobs in OECD Countries: A Comparative Analysis,” OECD Social, Employment and Migration Working Papers 189 (2016).
Nicolas Le Ru, “L’effet de l’automatisation sur l’emploi: ce qu’on sait et ce qu’on ignore,” France Stratégie, Note d’analyse 49 (2016).
Grace Katja et al., When Will AI Exceed Human Performance? Evidence from AI Experts, May 2018. https://arxiv.org/abs/1705.08807
Roland Berger Strategy Consultants, Les classes moyennes face à la transformation digitale. Comment anticiper? Comment accompagner? 2014. Richard Florida sparked a fierce controversy with The Rise of the Creative Class (New York: Perseus Book Group, 2002), which popularized the notion that these new occupations would all revolve around artistic creation and intellectual innovation.
David Deming, “The Growing Importance of Social Skills in the Labor Market,” NBER Working Papers 21473 (2015).
On this topic, see chapters 12 and beyond of Pierre-Yves Gomez, Intelligence du travail (Paris: Desclée de Brouwer, 2016).
JPMorgan Chase Institute, The Online Platform Economy in 2018: Drivers, Workers, Sellers, and Lessors, 2018. https://www.jpmorganchase.com/content/dam/jpmc/jpmorgan-chase-and-co/institute/pdf/institute-ope-2018.pdf
Nicolas Costes, Laurence Rambert, and Emmanuel Saillard, “Temps partiel et partage du travail: une comparaison France/Allemagne,” Trésor-Eco 141 (2015). Translator’s note: Unless otherwise stated, all translations of cited foreign-language material in this article are our own.
David Brady, Andrew Fullerton, and Jennifer Moren Cross, “More Than Just Nickels and Dimes: A Cross-National Analysis of Working Poverty in Affluent Democracies,” Social Problems 57, no. 4 (2010): 559–85.
Insee, “Les niveaux de vie en 2014,” Insee première 1614 (2014).
Thomas Piketty, Capital in the Twenty-First Century, trans. Arthur Goldhammer (Cambridge, MA: Belknap Press/Harvard University Press, 2014).
Robert Frank and Philip Cook, The Winner-Take-All Society: Why the Few at the Top Get So Much More Than the Rest of Us (London: Penguin Books, 1995).