Adair Turner, director of the McKinsey management consultancy, spoke to deliver a speech on “Capitalism and Robots” in Bengaluru as part of a series of lectures at the Azim Premji University. Let us see what he has to say.
You appear worried about the implications of the automation and robotics revolution.
I’m reasonably convinced that information and communications technology is increasingly going to enable us to automate just about everything. There’s a very good McKinsey Global Institute report on which jobs are inherently automatable. And what comes out of it is, eventually we may be able to automate everything, but the things which are most automatable upfront are repetitive, predictable, physical activity, and that means manufacturing. In the past we worked out how to automate anything that involves hard materials like metal, the auto industry is highly automated. But increasingly we are working out how to automate soft materials – textile, apparel. It may be good for productivity growth, but the impact of it will be very unequal. Some people will do very well out of it, some will not.
In emerging countries, the people I’m most worried about are the countries which have very rapidly growing populations. That in particular means Africa. India is in an intermediate position – unlike China and other parts of Asia, it has not yet achieved population stability. Quite a big growth in working age population is coming – from 740 million today to 1.3 billion in 2050. The challenge is nothing like for Nigeria or Ethiopia, but it’s still a major challenge.
What do you say to people who say global unemployment figures have been relatively stable over the decades despite worries every now and then about technology taking away jobs?
We can always find jobs for people to do, but they may be low productivity, low income jobs. We can always automate an Indian factory and employ some poor people as guards and drivers. That’s the story of what is happening in British productivity. We have some functions that are being rapidly automated in manufacturing and bits of retail, but people are finding jobs in some extraordinarily low tech activities. Not sure if you know about Deliveroo drivers. It’s a guy on a bicycle, like a London rickshaw driver, he’s not paid very much.
The issue is not full employment, the issue is real wage rise. We’ll always find things to do, but will it be high paid jobs? The historical record tells us that in the long run its fine, but the long run can be a century. Economic historians at Oxford looked at real wages in the early 19th century, in the early industrial revolution, when we automated weaving, spinning. People found new jobs, but real wages fell very significantly, and we came very close to a violent social revolution in the 1840s. We had enormous numbers of troops and policemen with swords, beating down striking workers. We crushed working class revolt with military power. It was only in the latter half of the 19th century that the technology then turned around and began to deliver real wage growth to the totality of the population. But in the first fifty years of the industrial revolution, it impoverished the British working class.
And today the changes are happening at such a rapid rate that people may find it impossible to reskill at the required rates…
I’m hugely in favor of high skills and education, but we have to be wary of the idea that skills are the answer to everything. If we are in an environment where we don’t need people in factories, however highly skilled people are, they are not going to be in factories, they will need jobs elsewhere, and however many people may know how to compute a code, all the revenue from, say, writing computer games every year is going to accrue to a very small number who have written the best computer games and the most popular ones. That’s true for apps as well. Apps and games have a real winner-takes-all. Being the fifth best computer game coder is of almost no value at all.