Work, Out of Reach

Work, Out of Reach

Daniel Susskind, Gresham Professor of Business

24 February 2026
 

The idea of technological unemployment

In 1931, the great British economist John Maynard Keynes published a collection of essays, entitled Essays in Persuasion. One of the most important was ‘Economic Possibilities for Our Grandchildren’. Here, Keynes reflected on what the world would look like a century from the time he was writing (roughly, now). This essay has proven to be influential for many reasons. But perhaps most importantly, it is here that Keynes coins the term ‘technological unemployment’, defined as ‘unemployment due to our discovery of means of economising the use of labour outrunning the pace at which we can find new uses for labour’.

Frustratingly, Keynes says little more about the idea of technological unemployment in his essay. Instead, he gestures to the ‘revolutionary technical changes’ that were taking place at the time to persuade his readers that this was an argument worth taking seriously. It may be that this was effective: he was writing at the close of the Roaring Twenties, a time when new technologies – from antibiotics to the aeroplane – were tearing their way through society, much as we might feel about AI today. But the result is that we are left none the wiser about the most important question: why might we not find new uses for human labour in the future, as Keynes worried?

Frictional and structural problems

In this fourth lecture, I begin where Keynes ended. I start by setting out two types of technological unemployment, two different ways that people might find themselves without work because of the remarkable technological changes that are now unfolding. To do this, I put to use the ideas and arguments that I set out in the first half of this lecture series – why automation anxiety has proven to be repeatedly mistaken in the past, how new technologies actually impact the labour market, what to make of the remarkable advances in AI that have unfolded in the last few years.

The first type of technological unemployment is ‘structural’ – here, there are not enough jobs for people to do, full-stop. This is the type that tends to capture the headlines in the popular press: think how often you read the headline, WILL ROBOTS TAKE YOUR JOB?, or some variation on that theme. But there is also a second type of technological unemployment, a ‘frictional’ one – here, there are enough jobs for people to do but for various significant reasons, people might not be able to do those jobs. As I will explore, it is this latter type of technological unemployment that is likely to matter most for now and in the medium-run. In this lecture, I want to focus on this phenomenon and set out its different causes.

The three mismatches

The first cause of frictional technological unemployment is the ‘skills mismatch’. This is the most familiar explanation for the phenomenon – here, people lack the right skills to do the jobs that have to be done. The difficulty now, as I explore, is that there are good reasons to think these skills mismatches are getting larger and harder to solve.

The second cause of frictional technological unemployment is the ‘place mismatch’. Here, people do not live in the particular geographical location where jobs are being created. At the start of the Internet era, there was a moment when people thought that these sorts of worries about place would no longer matter – people wrote about ‘The Death of Distance’ and how ‘The World is Flat’. But today, in looking for work, for various reasons I explore, the place that you live matters more than ever.

The final cause of frictional technological unemployment is the ‘identity mismatch’. Of the three causes, this is the most underexplored – yet perhaps the most significant. Here, people have a particular identity, a specific conception of themselves, and they are willing to stay out of work in order to protect that identity. This mismatch can take various forms, and I look around the world to show how it is unfolding.

The idea of frictional technological unemployment may not resemble some of the more alarming imagery associated with the future of work. Yet it would be a mistake to dismiss it on those grounds. For those who find themselves unable or unwilling to do the available work – whether because they lack the right skills, do not live in the right place, or have an identity that sits at odds with that work – the distinction between there not being enough jobs, and not being able to do the existing jobs, is not particularly meaningful. From their point of view they are out of work, and that is what matters.

© Professor Daniel Susskind 2026

References and Further Reading

Autor, David. ‘Why Are There Still So Many Jobs? The History and Future of Workplace Automation’, Journal of Economic Perspectives. 29: 3. 2015.

Goldin, Claudia and Lawrence Katz. The Race Between Technology and Education. 2008.

Keynes, John Maynard. Essays in Persuasion. 1931.

Moretti, Enrico. The New Geography of Jobs. 2012.

Susskind, Daniel. A World Without Work. 2020.

Susskind, Daniel. ‘Technological unemployment’, in The Oxford Handbook of Economic Governance ed. Justin Bullock et al. (2022).