Short-Term Pain, Long-Term Gain

Professor Yuk-fai Fong is under no illusions about the challenges that automation presents to workers around the world. Robots, automation and artificial intelligence (AI) are threatening jobs – indeed, in 2019 Oxford Economics predicted up to 20 million manufacturing jobs will be lost to robots by 2030. Last year, economist Daron Acemoglu of the Massachusetts Institute of Technology showed that from 1990 to 2007 in the US, each additional robot in manufacturing replaced an average of 3.3 workers and the use of robots lowered wages by 0.4 per cent over the same period.

“I agree that AI machine learning and increased automation will create short-term pain and some people’s jobs will be replaced,” Professor Fong said. “But as a society, I don’t think we should slow down that process. It’s about embracing it and finding ways to help those who are affected. Because even if we are in a grimmer scenario in which automation displaces more jobs than it creates, it may not be worse in the long run.”

This is because the alternative would be for new technologies to mainly enhance our productivity, which would not necessarily improve our lives. “Think of our phones. They make us work longer hours so we’re never away from work. And because our time is more productive, the market operates in a way that our time gets utilised.”

Displacement, on the other hand, would force workers and governments to adjust to new realities, such as fewer jobs for people and lower wages. For example, a large reduction in jobs could prompt governments to reduce the working week so there could be enough jobs to go around. That would have the additional benefit of improving work-life balance, he said.

And while this would mean wages would drop, so would prices.

Market equilibrium

“As an economist, we want to look not only at the direct implications, but what we call the equilibrium. We believe the market will settle in a certain way and when it does, there are different adjustments that will be made,” he said.

“We are still going to consume a mountain of commodities – keep that in mind. Machines and AI may compete for our jobs, but they do not compete for our products, they are not consumers. All these products produced at a massive scale would come down in price to a level that the market clears.”

Inequality would be a concern because the small number of people who own new technologies would have first-mover advantage and likely become or maintain their positions as monopolists. But they could never consume everything that their technologies produce – products would still need to be affordable to the general populace.

And while prices go down, free time would increase. People would have more time to seek out services provided by humans. “We don’t long to interact with machines, we long to interact with people,” he said.

This optimistic picture depends, however, on individuals and governments being flexible and prepared. Governments should take responsibility for helping people transition to a more automated society, he said. Individuals should retrain if they see their field is becoming automated.

Testing adaptability

In any case, Professor Fong does not envision a jobless future: “Having a job by itself gives us a lot of fulfilment and satisfaction that we are contributing to society.” Because of that, he would rather governments provide retraining and time-limited unemployment benefits than a universal basic income.

“I don’t want to paint an excessively rosy picture. What I would say is that in the long run, we will be in a better world and it will not just be a small number of people who are going to benefit. Most people will benefit. But I am mindful there could be short-term pain,” he said.

More immediately, he has also taken note of the workplace changes that have arisen during the COVID-19 pandemic, which is testing people’s capacity to adapt. Online meetings and classes have shown it is possible to achieve things remotely, but he cautions that the circumstances are unusual. “People already had prior relationships which has meant their online meetings can be productive. In three or four years, when there is a turnover of staff and new people are hired, how can you establish a productive relationship with a stranger?”

Nonetheless, he foresees a future of ‘blended working’ where part of the week is spent working remotely and part is at the office. “The pandemic has been a long experiment. I feel it is impossible that there will be no permanent impact on work arrangements. It is not like SARS, which was limited in scale and only lasted a few months. This time it’s different. It just makes sense for us to retain some of the online component of work that has been improved upon, to save our time.”