Branko Milanovic
Several days ago Steven Hill presented at the Graduate Center CUNY in New York his new book “Raw Deal: How the ‘Uber Economy’ and Runaway Capitalism Are Screwing American Workers”. It discusses (according to Steven’s presentation – I have not read the book yet) the decline of trade unions, the future of jobs and robotics. It struck me that there are (in his presentation as well as in most of what we read), when it comes to the future of work, two narratives that often seem contradictory. There is a narrative of job-automatization and robotics whereby most of our jobs end up being taken by the robots. Then there is a narrative of people working more and more hours as work intrudes into their leisure time: instead of taking it easy throughout the day as the first narrative implies, we would use our “free” time to rent apartments we own or drive our cars as taxis. According to the first narrative, we are in danger of having too much leisure time; according to the second, of having none.
Let’s consider the two scenarios in turn, and separately.
Suppose first that most of “routine” jobs are replaced by robots. This seems quite possible (from today’s perspective). If capital replaces labor easily, we shall have the elasticity of substitution exceed 1 (a point argued by Piketty in his “Capital in the 21st Century”), and the return to capital will not fall in proportion to the increase in the capital/labor ratio. This means that more and more of net income will belong to capitalists, and thus to the owners of the robot-producing companies. Suppose Google cars become ubiquitous, and Google drives (so to speak) Mercedes, Ford, Volkswagen… out of business. The money will be made by the owners of Google.
In that future, the distribution will move even further against labor, income inequality will increase, unemployment will be higher and there would be a generalized problem of finding a job. Since the economy will have become richer (there is just more stuff), there may be some kind of income support paid to everybody. Society would look as follows: lots of people on the minimal guaranteed income with loads of free time, the employed with incomes somewhat higher than the social minimum working mostly in services, and relatively few fabulously rich owners of the means of productions. Basically, Mark Zuckerbergs, personal trainers and the unemployed guys living in Miami.
How would the alternative scenario look? There we have robotics as well but what robotics does is to break jobs into the tiniest possible segments, parcel off each of these segments to people who are temporarily hired to do that tiny parcel alone, and then combines hundreds of such tasks into one final product. Instead of having a worker W working for a company C full time and doing a task T, such that on the task T there is a bilateral exclusive commitment between the worker and the company, we would have task T broken into T1, T2,… Tn, and branched off to workers W1, W2… Wn. Now the workers will in turn each also work on other tasks and for other employers: so the worker W1 will in the same day work on tasks T, Z and Y, each for a different company.
There would be multiplicity (non-exclusivity) on both sides: workers will no longer be committed to a single employer, nor will the employer depend for task T on a single worker. Since tasks are so segmented, it makes it possible to hire less professional and thus cheaper labor, often using their “free” time. This is why professional taxi drivers are now being replaced by guys who spend 1/3 of their workday as sales agent, 1/3 as bar tenders, and 1/3 driving their own cars as cabs. Or why hotels are in competition with people who rent their own rooms. Or why I might use every minute of my leisure time to do jobs for which I have no training and that would replace people who have trained for them and done them for years.
Under that scenario, we should see a dramatic reduction in specialization (say, vocational education would end), blurring of the difference between leisure and work, and pressure on the labor share. Everybody would be a jack of all trades and master of none. There would be only limited unemployment (since practically everybody could do some extremely simple tasks into which all jobs are divided).
But perhaps it may be better to think of the two scenarios as just one scenario that would combine lots of labor substitution with heavy segmentation of tasks (and much more intense labor discipline made possible thanks to automation). In that case, jobs to which we have become accustomed would cease to exist: lots of today’s functions would be automated, and for many others, “amateurs”, not professionals, would do them.
And we should not fall for the “lump of labor” fallacy: the amount of jobs is not limited to the jobs that we know today. There will be entirely new jobs that we cannot even imagine. Steven gave one such example that exists already now: “invisible girlfriend”. People pay to receive, at some intervals, text messages which are ostensibly sent by their girlfriends. In the eyes of other people their esteem goes up: many girlfriends compete for their attention. In reality, some mustachioed guy in short sleeves is writing these messages and getting paid for them.
Or to give another example: women whose job is to be surrogate mothers to either gay couples or heterosexual couples who cannot have kids. Now, that job did not exist until recently, that is until (a) legal changes allowed for surrogate motherhood (and also for gay marriage) and (b) technological progress that made “artificial” insemination possible. When I give this example, people often ask me: but can male auto-workers from Detroit become surrogate mothers? No, but this is always the case in the transition period when the occupation is on its way out. But after a while there would be no more auto-workers at all, and yes, some women can become surrogate mothers and have that as their main income-earning job.
Technology will create new jobs, and if anything, I think we shall have more to worry about not having any free time than having too much. As commercialization of our lives progresses, we shall perceive (as we already do) every hour spent without directly or indirectly contributing to more money as wasted. Unemployment will become impossible. Being unemployed implies that you are specialized and that there is a (relative) shortage of such specific jobs. But not so in a new economy: everybody can carry Thai food from one place to another, everybody can exhibit himself or herself naked on the Internet, everybody can open doors, pack bags, or even write blogs. No one would be unemployed and no one would hold a job.
This blogpost was first published on Branko Milanovic’ Blog.
This contribution is part of our project on the future of work and the digital revolution.
Branko Milanovic is a Serbian-American economist. A development and inequality specialist, he is visiting presidential professor at the Graduate Center of City University of New York and an affiliated senior scholar at the Luxembourg Income Study. He was formerly lead economist in the World Bank's research department.