Recently an Oxford University study entitled "The Future of Employment: How Susceptible are Jobs to Computerisation?" ranked 702 occupations by their likelihood to be automated in the next 10-20 years. 12 occupations were deemed to have a 99% chance of automation, including telemarketers, data entry keyers, library technicians, tax preparers, and insurance underwriters.
Of the 20 occupations least likely to be automated, 14 of them were in the Health or Emergency Services fields - the report notes that those occupations least likely to be automated involve tasks that computers cannot (yet) easily replicate - Perception and Manipulation (Finger Dexterity, Manual Dexterity, and Cramped Work Spaces or Awkward Positions); Creative Intelligence (Originality, Fine Arts); and Social Intelligence (Social Perceptiveness, Negotiation, Persuasion, and Assisting and Caring for Others).
Of those that are likely to be automated, it's not necessarily robots, but technologies like machine learning and big data. In fact, many of the technologies that can do this displacement have been around for years, like OCR.
From an HR perspective, the likelihood of role automation is as follows:
- Payroll and Timekeeping Clerks: 97% likelihood of automation
- Human Resource assistants (other than payroll and timekeeping): 90% likelihood of automation
- Human Resources, Training, and Labor Relations Specialists, All Other: 31% likelihood of automation
- Training and Development Specialists: 1.4% likelihood of automation
- Industrial-Organizational Psychologists: 1.2% likelihood of automation
- Training and Development Managers: 0.63% likelihood of automation
- Human Resources Managers: 0.55% likelihood of automation (intriguingly, far less likely to be automated than General and Operations Managers at 16%)
Technology is driving this automation, and we've certainly seen automation as a continuing trend over decades. Without over-simplifying it, there are both utopian and dystopian views on where the automation of work will lead us as a society - on the one hand, there are those who predict widespread technological unemployment "due to our discovery of means of economising the use of labour outrunning the pace at which we can find new uses for labour" (John Maynard Keynes, 1933). The less pessimistic school of thought is that either we will restructure our definition of a "full time job" (as happened 200 years ago with the introduction of the 8-hour day and banning of child labour in Britain), and/or that new professions will continue to emerge - ones that are less dangerous and mundane, expand into new areas previously not possible without the technologies we have at our disposal today, and are good for our society.
Keyne's technological unemployment view hasn't proven to be particularly prescient in aggregate over the past 80 years, despite huge advances in automation and computerisation (although technological unemployment does happen in some industries, and this is no less painful for there being more jobs elsewhere in the economy). Keyne's forecasts are reminiscent of the Malthusian Catastrophe proposed by Thomas Malthus (1798), which predicted that because population grows exponentially, but food production only linearly, then the world was headed for inevitable famine unless we depopulated. Yet today, food production per capita has actually risen due to scientific advancements and changed methods of production.
Home automation is another technology which has long been assumed to have the inevitable result of increased leisure time:
"Without delving too deeply into the science-fiction world of robots it is interesting to assess the implication of the definite increase in domestic automation. One obvious fact of reducing household tedium is the increased leisure time available to the future housewife and the consequent demand for improved entertainment and recreational facilities" - New Scientist, 24 May 1973, p.478 (emphasis added)
In actual fact, we know that leisure time (or at least, the perception of leisure time) is declining despite huge advances in home automation. In Australia in 1974, 28% of Men and 23.9% of Women reported "always feeling rushed" (Cities Commission Time Use Survey, 1974). By 2007, 35% of Men and 42% of Women reported always or often feeling rushed (Australian Bureau of Statistics Gender Indicators, 2013). Although the question is slightly different, it's clear that the "obvious fact" of increased leisure time may have been obvious, but wasn't a fact at all. From Feb 1978 to Feb 2014, Women's share of full-time employment rose from 27.6% to 35.5%, and total employment 35.3% to 45.8% (source).
Those who believe that automation will lead inevitably to widespread technological unemployment, then, would be wise to learn from these predictions of the past. Certainly the technology we have at our disposal now has automated some things, and displaced some industries and occupations - manufacturing, bookkeeping, bookstores, travel agencies, and most other intermediaries / agencies. But history has also shown that technology enables new occupations to emerge - usually occupations that are less dangerous, less mundane, and are good for our society. Vertical Farming / Urban Agriculture, Manufacturing using new materials like Graphene, Medical Advances including nanobots, predictive crime prevention, and predictive maintenance for engineering are some such areas of future growth. It's true that there's a painful re-structuring of our definition of work and leisure ahead - indeed, we're already in the middle of it - but a challenging future doesn't need to be a dystopian one.
I believe a more fundamental change to the nature of work is not the rise and decline of particular occupations - but the increasing atomisation of work - something I'll be posting on shortly.
What do you think? Is your job one that is likely to be automated? Have you already experienced or seen technological unemployment?