Automation and artificial intelligence does not remove human labour from the economy, it shifts and creates jobs elsewhere.
Automation leads to various directions of change for workers, including:
- Who they are employed by;
- Where they are employed;
- How they are employed;
- Which sector they are employed in; and, most importantly,
- Which skills they need and use;
Although still far away, widespread adoption of advanced automation technology will reduce costs and increase productivity. For example, automation could reduce road haulage costs by 30%. Cheaper, more productive transport encourages more transport-intensive trade. Productivity gains are recirculated, jobs are reallocated, economic output increases, and new sources of wealth are created. In transport, this would sustain or create jobs:
- Alongside automation technology, performing different tasks that may require less or more skills; and/or
- At the local level, at the end-points of highly automated long-haul transportation, in last-mile operations and urban areas.
Intensification and control
New technology increases the speed and efficiency of transport. Human tasks alongside automated transit, and at the end-points, usually accelerates in tandem with the technology. Workers also tend to lose control over their work.
‘Blind passenger acceptance’, whereby Uber’s algorithms decide but hide the destination of journeys until drivers accept the request, takes control away from the worker. A tired Uber driver may prefer a short final journey towards home, but is coerced into travelling further and working longer.
Deskilling, informal work and falling wages
60% of all occupations have at least 30% of tasks (and skills) which could be automated, but the jobs will remain.Automation could both up-skill and de-skill transport workers, polarising the workforce. Middle-skilled employment will decline, whilst high- and low-skilled employment rises.
Workers employed alongside automation technology could have to operate more complex and hi-tech machinery. Train drivers, pilots and truckers may become supervisors, handling paperwork when cross borders or arriving in new countries, or operating complex computer systems that manage several vehicles and instructions. The higher skills and training required could therefore demand a more formal, permanent and well-paid workforce, and disincentivise the use of subcontracted, temporary labour.
However, automation can de-skill workers or rely on them only as a failsafe. As Uber has shown in private passenger transport, automation of some skills such as navigation has contributed to the misclassification of drivers, informal employment and falling wages.
Similarly, as transportation and logistics jobs shift away from long-haul, to warehousing, last-mile delivery and private transport, there is greater scope for the same worker misclassification and precarious employment. Bogus ‘independent contractors’ and informal work are growing in Amazon warehouses, package delivery by companies like Yodel and Hermes, and ‘gig’ delivery and taxi services such as Deliveroo and Uber.
Historically, standardisation of skills and training is needed for wages to grow after an employment shift. As automation is ongoing with new innovations looming, technology and training is not standardised. Skills learned at one job are not necessarily valuable to other employers, so overall wages do not rise according to demand.
Technological change is estimated to have caused half the decline in the labour share of national income over the past 40 years. Without government or union intervention, bosses and the owners of technology would increase their profits whilst workers endure menial tasks, informal work and falling pay.
If left unchecked, the digital employment shift could disproportionately affect workers with lower wages. Low wage jobs tasks (paid under £30,000) have five times the technical potential to be automated than high wage jobs tasks (paid over £100,000). Low wage jobs are far more likely to be de-skilled, informalized and displaced.
Automation could enlarge the rural-urban divide and accelerate urbanisation. Large cities exhibit increased job and skill specialisation as they house far greater managerial and technical professions, which are far less prone to automation. Long-haul freight and transportation outside of large cities is most prone to automation, whilst urban warehousing and last-mile delivery is relatively resilient.
46.8% all jobs performed by women are automatable compared to 40.9% of men’s jobs. If automation polarises jobs towards low- and high-skilled work, women disproportionately fall into low-skilled employment.