What are the long term implications of AI for Labour, Productivity and Economic Equality?

What are the long term implications of AI for Labour, Productivity and Economic Equality?

Last week, Citi Research in conjunction with the Oxford Martin School released their latest report in a long term series looking at some of the global challenges of the 21^st century.

In an earlier report, it was estimated that around 47 per cent of jobs in the US were at risk from automation. In this report, the authors expanded the analysis to look at other countries.

The report cites 3 reasons why the impact of technology change on the economy would be more significant in future than the impact we have seen historically:

1. The pace of change has accelerated. This is highlighted by the rate in which internet access has been rolled out globally and more recently, the growth in penetration of smart phones.

2. The scope of technologic change is increasing.

3. The benefits of technological change are not being widely shared.

Connectivity is expected to move from people to things. The rapid and accelerating rate of data capture will increasingly require the use of machines to process given the sheer volume. Automation of processes will accelerate, but so will the scope of the applications for automation.

The development of artificial intelligence means that machines will be increasingly capable of replacing humans in a wide range of roles due to improvements in learning algorithms combined with the ability of machines to process vast amounts of data.

The report goes on to suggest that the impact of increased automation in replacing existing jobs will be greater in developing countries than in developed countries. It estimates that 77 per cent of jobs in China and 69 per cent of jobs in India are at high risk of being replaced by automation as opposed to an estimated average of 57 per cent for the OECD according to the World Bank.

This is due to developing countries being more exposed to manufacturing industries, with their main source of competitive advantage being low wages. Increased labour productivity through automation reduces this competitive advantage. Deflation in the capital cost of implementing automation will also reduce this labour cost competitive advantage over time. The affordability of automation is further enhanced in the near term by falling capital costs for companies as a result of a sustained period of low interest rates. The falling cost of automation means jobs that rely on low wages to remain competitive are seen as the most automatable in future.

This has significant implications for the sustainability of high economic growth in developing markets as automation makes their labour cost advantage redundant and technology such as 3D printing encourages companies to keep manufacturing closer to home to reduce freight costs. As a result, developing countries will need to implement policies to accelerate their transition toward valued added services driven economies.

The falling cost of automation will mean that the rate of commoditisation of labour, and the deflationary impact it has on unskilled wages is likely to increase. At the same time, as technology advances, the range of jobs that automation can replace will expand. This will be enhanced by the increasing ability of machines to learn and problem solve as artificial intelligence becomes increasingly sophisticated.

Therefore, as artificial intelligence develops, it will also increasingly replace labour in service industries.

Another implication is likely to be a widening gap between the rich and the poor. Additionally, with an increased proportion of jobs being replaced by machines, there is the potential for the proportion of the population that is considered wealthy to narrow. The productivity improvements from increased automation will accrue to the shrinking proportion of the workforce that continues to have skills that cannot be replicated by machines.

In this environment, it will be critical to maintain a skill base that remains ahead of the capabilities of automation. This means more frequent retraining is likely to be required, to build, adapt and develop new skills to stay ahead of the curve. The days of education being completed by the age of 20 and staying in the same field/job for 40 years are over.

This puts a spotlight on education and the need for an economy to invest the ability to continuously develop the skill base of its workforce.

For people in developing economies, as demand for labour falls in the manufacturing industry there is likely to be greater focus on investment in education to build a more value added services based economy. However, as previously noted, low skilled services industries are already becoming automated. As a result, an individual will need to build developed market skills in terms of creative capabilities to be in demand. This is likely to increase the demand for and value of education services in more advanced economies. The value of an educator’s brand will become increasingly important as a differentiator.

Labour needs to focus on areas in which automation is more problematic. This is in creative intelligence, social intelligence as well as perception and manipulation. Creativity is likely to be an area that will be difficult for machines to replicate for some time given the ability of the human mind to think laterally and to use a deep pool of knowledge in using judgement in decision making. Similarly, it will be difficult for machines to replace humans in occupations that require empathy such as palliative healthcare and customer service.

It also means that regulations need to enable and encourage change in an economy. Labour flexibility and mobility will become increasingly important, as the demand for labour and skills changes over time.

One of the biggest challenges will be in managing the growing social divide. We are already seeing the political implications of the widening gap between the rich and the poor through the polarisation of politics. The challenges of wealth redistribution are likely to increase significantly over time.

This has a lot of implications for investing as well. Clearly the education industry is likely to remain a winner. But it also implies that to maintain your standard of living without relying on Government driven wealth redistribution, individuals need to be able to benefit from the increase in productivity without suffering from or to offset the potential for redundancy of their labour skill set. Hence investing in creative, value adding companies that can not only utilise or provide automation technology, but also retain that value for shareholders (rather than having it competed away) can provide some protection in relative living standards.

Stuart Jackson is a Senior Analyst with Montgomery Investment Management. To invest with Montgomery domestically and globally, find out more.