Department of Strategy and International Business, University of Birmingham
Employment is in a continual process of creative reconstruction. Existing jobs are destroyed as innovations in technology and processes reshape the nature of work.
On Monday 25 March 2019, the Office for National Statistics (ONS) published a report which estimated the impacts of automation on the English labour market. It is important to highlight that this type of analysis can only ever be an estimation rather than a predication.
The headlines from this report are as follows:
- Around 1.5 million jobs in England are at high risk of some of their duties and tasks being automated. The estimate is that automation will alter around 7.4% of jobs in England
- Women, young people, and those in part-time employment are at higher risks of automation.
- The proportion of jobs at risk of automation decreased from 8.1% to 7.4% between 2011 and 2017.
- The ONS analysis is not about jobs but tasks performed across the whole labour market and an assessment of the probability that they will be automated.
- The three occupations that are at the least risk of automation are medical practitioners, higher education teaching professionals and senior professionals in educational establishments.
- The risk of automation is partly related to geography as different parts of England have different labour market structures.
There are many points to be made about this estimation:
First, the analysis is reliant on probability scales and words like ‘most likely’. This is an analysis of the future based on a series of assumptions, which are used to inform the development of a set of predications. This is one of many such assessments.
Second, automation is transforming labour markets, but no one can yet make an informed predication of the likely impacts. Medicine and medical practitioners are already experiencing the impacts of automation. There will come a time when medical diagnostics is a predominantly automated process. For higher education teaching professionals, it is important to differentiate between teaching and research. Georgia Tech, for example, has been using virtual teaching assistants (TAs) on an online artificial intelligence (AI) course since 2016. Jill Watson, the virtual assistant, was introduced during the spring of 2016 and students were not informed of her identity until the final day of the class.
Third, the ONS argues that the three occupations with the highest probability of automation are waiters and waitresses, shelf fillers and elementary sales occupations. This highlights one of the problems with this report. The difficulty is in removing people from service encounters. There is no question that automation plays an important role in the world of automated e-commerce warehousing. A good example is Best Buy’s most famous ‘employee’, Chloe who is based at this electronics retailer’s Chelsea location in Manhattan. Chloe works 24/7, and she is a robot. Nevertheless, some service encounters will remain centred on relationships between people and customers and will not involve the replacement of people with algorithms or robots.
Fourth, a revolution is underway that is transforming some occupations. Retailing and financial services are at the forefront of this revolution as e-commerce and e-banking shifts some retail experiences from the high street to automated warehouses and back offices that are accessed via online shopping platforms. Nevertheless, for the time being, many of the jobs considered to be most at risk from automation will continue with only minor alterations.
For all companies, the key decisions to make regarding automation revolve around three related issues:
- Value for money in relation to the investment required to automate a task.
- Impact on sales and customer experiences.
- Hard-to-fill vacancies and labour costs.
All companies should try to act responsibly, and a responsible solution to automation involves understanding the trade-offs between creating local jobs for local people and replacing employees with algorithms.
- ONS (2019), ‘Which occupations are at higher risk of being automated’