Automation could lead to prosperity and a shorter working week, or to mass under-employment and poverty. Charles Seaford sets out a path to the former.
Academics at the Martin School at Oxford University estimate that about one third of UK jobs could be automated in the next few decades, with other estimates in line with this. There will be new jobs as well, of course, just as there were when enclosures threw people off the land and when computers replaced armies of filing clerks. But will there be enough? And if there is less work to do overall, can we arrange things so that everyone gets a good balance of work and leisure?
Economists tell us that entrepreneurs will make use of the released resources to launch new, innovative products and in this way soak up the excess labour. The problem is, wages may need to be pushed down for enough of these products to be viable. This is the existing trend, with automation already reducing demand for so called ‘mid-level’ skills and mid-level wages. In any case, job-creation is not guaranteed: a World Economic Forum Global Risk Report argues that the process of innovation led job-creation may be slowing, while in 2015 new enterprises in the US created 40% fewer jobs than they did in 1995. As a senior government economist has put it, there is simply no reason to expect sufficient demand for the products that a fully employed workforce could produce.
What is to be done? A shift to a four-day week would help and there are some fairly familiar policy measures that could nudge us in this direction. Labour market regulation and public sector employment practices could make it easier for employees to choose to work less (as in the Netherlands, where half the labour force works part time). A universal basic income, perhaps financed by a tax on robots and other software, could reduce the need many have to work long hours. Stronger trade unions could negotiate shorter hours in return for productivity increases.
However this kind of policy may not be enough. We may also need to influence choices: between leisure and money, and between different ways of spending money.
People are more likely to choose leisure over money in more equal societies. There is a positive correlation between inequality and average number of hours worked in a year in the 20 most prosperous OECD countries, and there are reasons for thinking this is causal: inequality means more poor people, who work long hours to make ends meet; at the same time how much people value income, and thus the choices they make, is affected by inequality levels. So traditional redistributive measures may be part of the answer.
But economics is not everything. There is a nagging worry that people will not find enough to do in their extra time without spending more, so sending them back to work. At least some people may be deterred from working and consuming less for fear of boredom and lack of meaning. Will they miss the companionship and sense of purpose provided at the office?
Hence the role of institutions such as the Church, the freemasons, guilds, political parties, volunteer run charities and all kinds of club, which have always been a source of both companionship and purpose outside work. Perhaps the critical ingredient they provide is discipline. At work, you can do well or badly, and to do well you have to bow to the discipline of the organisation or profession or craft, at the same time challenging it when appropriate. You can also do well or badly at school or university, as a member of the Church, or as a freemason, member of a political party, or member of a football team – and again to do well you have to accept a discipline. It is this that distinguishes active participation from passive consumption. Educational institutions are particularly important, for they develop the skills needed to master the disciplines of a meaningful and active life outside employment.
What about how people spend money? How is that relevant? Even if we move to a four-day week, we will still need jobs that cannot be automated, that can command decent wages and that can be sources of fulfilment. If all that people ever buy is beautifully designed, but automatically produced objects, music downloads, dry-cleaning and take away pizza, then things are not looking good. Jobs will polarise, with highly paid professionals designing the beautiful objects and performing and delivering the music downloads, while badly paid servants do the dry-cleaning and deliver the pizza. If, however, people take up various evening classes, buy hand-made furniture and go on holiday in family run hotels, there will be more independent, mid-level jobs.
Government can neither breathe life into social institutions nor force people into evening classes: this is not a plea for Soviet style regimentation. However, it can work with those institutions and with business as part of a project to influence choices. That requires a degree of partnership that we rarely see, but one that may be necessary if automation is to meet its positive potential.