Vast swathes of the employment market - from the bottom to the top - are under threat from automation, and the result will be mass global unemployment. Silicon Valley entrepreneur and technology guru Martin Ford explains.
Hardly any time seems to lapse between scaremongering stories in the middlebrow media claiming that robots are going to take over the world and throw us all out of our jobs. In the usual course of events, these reports can be taken with a healthy pinch of sodium chloride, but when American commentator Martin Ford says this could actually be the case, it’s probably time to sit up and listen. Ford has spent the past quarter of a century in computer design, having founded his own software development firm in Silicon Valley, and is recognised as one of the first contemporary writers with an engineering background to be taking the issue of technology-led unemployment seriously.
Most of us will remember from our history lessons that fear of technology has been with us ever since we first mounted a wheel on an axle, or started to set movable type. In the early 19th century, a group of British textile workers named themselves after the saboteur Ned Ludd, and embarked on a spree of vandalism calculated to disrupt industrial production. This reaction to the onset of modern technology was taken so seriously by government of the day that it responded by making ‘machine breaking’ a capital crime. Yet none of this, according to Ford, can hold a candle to the situation we find ourselves in now. “I describe it in terms of the boy who cried wolf,” he says. “There has historically been a series of false alarms. Only this time there really is a wolf relating to this issue.”
The issue to which Ford is referring is that of intelligent algorithms making white-collar jobs obsolete. He says that professionals such as travel agents, data analysts and paralegals are currently in the firing line. In the near future doctors, journalists, taxi drivers and even computer programmers will be replaced by ‘robots’. He says that if we do not radically reassess our economic and political structures, we risk a catastrophic implosion of the capitalist economy.
The idea that such a revolution could polarise society, coupled with its attendant prosperity or catastrophe, is the central argument of Ford’s latest book, ‘The Rise of the Robots’. It’s an argument that certainly gained traction with the judges of the 2015 Financial Times and McKinsey Business Book of the Year Award, who bestowed on it first place. “And even if I am wrong,” says Ford, “and the issue isn’t as imminent as I have made out, it’s only a matter of time. It’s inevitable that machines will get better at doing more routine tasks. That’s when we will have reached the tipping point. I think that we have reached the point where we need to address an issue that has, admittedly, been raised prematurely before.”
Many of the observations in ‘The Rise of the Robots’ relate to the intersection of technology and society. On the one hand, as the mundane clerical duties of our lives become automated, it follows that we should find ourselves with more free time. However, it also follows that a society put out of balance by machines doing work for us will find it hard to generate sufficient personal income to give our free time high value. This is already seen in the much-debated trend towards ‘zero hours’ employment contracts, where individuals are retained only when they are needed and left to fend for themselves when they are not.
Ford agrees: “What this means is that basically, as a society, we are facing a choice.” He goes on to say that on one side of the coin there is “a techno-optimistic view, widely held in Silicon Valley by people who are very successful and who have made a lot of money. In my view, these people tend to be naively optimistic in assuming that everything will just work out and that the market will somehow adjust itself. But I don’t believe that. I think it is possible that there could be a utopian outcome where people have to work less, but I think it will take a political and economic adaptation to make that happen.”
If this doesn’t happen, Ford thinks we will be heading for a more dystopian case, characterised by “extreme inequality and where people are simply left out of society. And that is especially worrying in the US, where we don’t have the social safety net that seems to be present in Europe.”
At this point, Ford raises some of the further issues he has addressed in his book. Developing the theme of stagnation in society, he points out that in the first decade of the 21st century, there was zero net gain in US employment. “This is the first time this has happened since the Great Depression. In the UK, employment levels may be high, but this comes at the cost of substantially reduced real incomes and productivity.” One of the reasons for the flattening out in real terms of the job market is that “machines are no longer tools used to increase the productivity of workers - they are workers. We have seen job replacement in blue-collar sectors already, such as in fast-food restaurants, warehouses and supermarkets.”
Ford thinks that with the economies of the UK and the US becoming less efficient at creating new jobs, it’s not just the blue-collar sector that is under threat. “If you look at the bar graph of employment in the US decade by decade, it resembles a descending staircase. There is a clear structural trend towards fewer jobs being created. Technology has had an impact on that for a while, and it’s getting more dramatic.” Ford posits a model where 500 ‘solid middle-class jobs’ are lost, only to find “that you might get them back, eventually, in terms of numbers. But they won’t be of such high quality. The interesting thing about this is that if you then look at the new jobs that have been created, and you look at the emerging technology, you can see that these are going to be impacted on too.”
However, it is not just the bottom rungs of the employment ladder that will be affected by machines capable of running any kind of predictable task. The entire spectrum of the job market will come under pressure, “and if technology has an impact on the higher end jobs, then our entire thinking will become up-ended.”
Ford notes that when lower-end jobs are lost, the solution has always been education. “There was always this idea that you could send people back to school where they can learn to make their way up the skills ladder and perform tasks that require more intellectual input. But the paradox today is that in many cases, it is the higher-end jobs that require information manipulation that are under pressure. This is where you have someone sitting behind a computer, doing repetitive tasks, generating the same reports or the same analysis. In many cases, this is easier to replicate than more manual jobs that require high levels of dexterity, hand-eye coordination and interpersonal interaction.” The sort of jobs that university graduates are likely to want, involving high levels of computer time “are exactly the sort of jobs that can be easily automated.”
Big firms, small workforces
Ford switches his attention to the new digital monoliths - Google, Amazon, Facebook and Twitter - which are characterised by disproportionately small workforces compared to their turnover. He then goes on to compare General Motors with Google. “General Motors at its peak employed 850,000 workers globally. Google has less than 5 per cent of that total. And yet, after adjusting for inflation, Google generates something like 20 per cent more turnover than General Motors ever did.” Not only is Google incredibly efficient as a business, “but also has more positive impact on society in other ways, such as the development of the driverless car. This is what the economy of the future is going to look like increasingly.” This example calls into question the whole notion of ‘creative destruction’, where old industries are disrupted in order to make way for new ones, and where the assumption is that the new order will hire those who have been put out to grass onto the job market. “Yes, they will. But not many of them.”
The Death of Moore’s Law
Ford agrees with industry commentators who think that Moore’s Law is slowing down. “Clearly it has a fundamental limit when it comes to the number of transistors that you can pack onto a chip. But there are technologies being explored, such as vertical stacking on three-dimensional chips, which may extend the lifespan of Moore’s Law. In the future, there is the potential for disruptive technologies such as quantum computing. But there is definitely a scenario where we are seeing a slowing down or stagnation of Moore’s Law. But my argument goes beyond this. I also think that Moore’s Law is something of a crutch: if engineers know that hardware is going to double in its capacity every two years, then maybe they don’t have to think so creatively about software development and other areas where there could be potential performance improvements. One such obvious way would be to take cheap commodity processors and find new ways of linking them together. For example, we are already seeing designers using the cheap chips from videogames to build neural networks for deep learning. We’re beginning to see more creativity, rather than just relying on computers to double in speed every couple of years.”
Ford is in no doubt that relying on computers, rather than looking for more creative solutions, will lead to “more and more inequality”, which was precisely the objection the Luddites levelled against the incorporation of emerging technology without reference to the employment market. So does he have a proposal?
“What makes more and more sense to me is to have a situation whereby everyone has a guaranteed income whether or not they have access to a job that pays the living wage.
“You hear more and more people talking about this now. But beyond that - and this is a problem I highlight very clearly in my book - we should keep in mind a key concern about the ultimate impact of widespread automation: workers are also consumers. We need people out there in the market of sufficient income to be able to purchase the products and services produced by the economy. We are seeing evidence of this already. If you look across Europe, there is talk of deflation for the simple reason that the bulk of the population doesn’t have enough income to go out there and become vibrant consumers.”
As our conversation develops, we touch on the influence of competition from the BRIC economies (Brazil, Russia, India and China), and in particular India’s strategy specifically geared towards the electronic capture of American and European jobs. Rather than discuss skills shortages, we discuss the fact that in Europe approximately one third of graduates are overqualified for the jobs they find themselves in. Meanwhile in China, 43 per cent of the workforce is overeducated, leading to the inevitable conclusion that education reform cannot solve the problem of technology-led unemployment and underemployment. Ford thinks that we need a “radical overhaul of our entire system.”
How does the author feel about a future that he seems to find dystopian? “I would describe myself as a short-term pessimist, but a long-term optimist.” He clarifies this by adding that in the long run there is the potential to adapt to technology-led unemployment.
“If you look at the techno-optimistic view, then there is a future where nobody needs to have a job that they hate, while nobody should have to do a job that is dangerous. But how we get to that point depends on us figuring out a way of adapting, so that everyone in society gets a share of the benefits that the technology brings. But in the short run, politically and economically this is a tough problem, and I am very doubtful that we will make the necessary adaptations without some sort of crisis.” *