Globotics. Telemigrants. White-collar robots. To describe the future of work, Richard Baldwin is developing a new lexicon.

The professor of international economics at the Graduate Institute in Geneva warns that we are unprepared for the ways in which new technology is changing the nature of globalization. Baldwin’s new book, The Globotics Upheaval: Globalization, Robotics, and the Future of Work, is a natural follow-up to his 2016 book, The Great Convergence. Three years ago, he explained how a third wave of globalization—a collapse in the cost of the movement of people thanks to technology—would be the most disruptive, because it hits workers in the service sector.

Baldwin’s new book, published earlier this year, breaks down what this disruption will entail. He echoes the warnings about 800 million jobs being displaced by automation in the next decade or so. While artificial intelligence will cause much of the upheaval, there is another factor: telemigration. The global gig economy, also known as the human cloud, where job tasks are broken up and done by freelance workers all around the world, is already a $82 billion industry, according to advisory firm Staffing Industry Analysts.

Image: McKinsey

“Automation and globalization are century-old stories. Globotics is different for two big reasons. It is coming inhumanely fast, and it will seem unbelievably unfair,” Baldwin writes. “Globotics is injecting pressure into our socio-political-economic system (via job displacement) faster than our system can absorb it (via job replacement).”

Governments are failing to protect workers, companies are advancing the disruption, and workers are unwillingly complicit in it all. But Baldwin is no Luddite. Digital workforces and telemigration should rightly be prominent in the labor market of the future, he says. Quartz spoke to Baldwin in London about how the real problem is how ill-equipped countries like the US and the UK are for these changes. The conversation has been edited and condensed for clarity.

Quartz: Firstly, how do you describe globotics?

Baldwin: Hopefully it’s a memorable word. It smashes together globalization and robotics. I was tired of people talking about robots every single day and what they can do but completely missing the idea that the same technology was transforming globalization.

When I say globots, I mean both the white-collar robots, which is software that can do stuff that only humans could do before, and telemigrants, who are basically telecommuting from abroad. Don’t forget that telemigrants are real humans. There’s nothing they can’t do except be in the room.

What are some of the digital technologies that enable this?

Machine learning is the key. Let’s first talk about automation and then I’ll talk about globalization.

In automation, the key is machine learning, which is database, pattern recognition. Machine learning has given computers cognitive capacities that they did not have before 2016 and we’re all familiar with them—speaking, reading, writing, reading handwriting, generating PowerPoint presentations, etcetera. Those were in some sense gateway skills, which are exactly why humans were involved in many service office jobs. Somebody had to open up the email or answer the phone and figure out what people wanted. Now, robots can do that. These are what I call white-collar robots. There is not one killer application yet. The one that’s closest is maybe RPA, robotic process automation. And then there’s a higher level one like [IBM’s] Watson and [IPSoft’s] Amelia.

These computer skills came over into globalization through progress in machine translation. And there’s a whole set of telecommunication advances which made remote people less remote. That isn’t really digital technology, it’s just that people are rearranging the way they work to slot in remote workers. But these things have basically made it possible to break up a service sector by having people far away do the same tasks. In the US, there was some estimate that 40% of people telecommute at least once a month. Their companies, once they’ve set the whole thing up for domestic telecommuting, are going to realize they could get at least some of those tasks done by people overseas for less.

So this wave of globalization doesn’t have an immigration debate attached to it? It’s about moving people’s capabilities without moving people.

These telemigrants are coming for service jobs and a wall won’t stop them! It’s good news if you want to keep out migrants because you can have the work without the workers.

“You can have the work without the workers.”

The other side of this is that anti-migrant governments are not thinking about the right immigrants, or maybe they don’t care about white-collar immigrants. In any case, they are substitutes. In countries like Japan, which really doesn’t want immigrants, they can use telemigrants to do things like teach English or childcare. It’s an explicit policy of using telemigrants instead of migrants.

In the book, you describe the freelance economy as kind of a precursor to telemigration.

(Laughs) Yes, there’s an element of a Greek tragedy to it. By trying to be more flexible with work, you thought you were getting control of your life. You could come home and take care of the kids while handling a few emails. But you actually arranged it so you ended up out of work because a telemigrant took your job for much less. It’s a Greek tragedy where the hero is essentially enabling their own downfall. Workers are paving the road to global competition.

By being more flexible, we’ve actually shot ourselves in the foot?

Globalization always means more opportunities for the nation’s most competitive citizens, but more competition for the least competitive citizens. The trouble is that in the services sector there are a lot of uncompetitive people. Many of these jobs have been non-tradable.

If a car is made in Britain that’s because it is globally competitive to do so. The manufacturer could always have made it somewhere else. Whereas people in the service sector in London, there’s never been any international competition.

There is some international job competition in London.

So there is trade in services and particularly business services. These are competitive. But there’s a huge number of people working in London offices and restaurants and bars, etcetera. Technically speaking, in order to be competitive, your productivity has to more than compensate for your wage compared to the foreigners. Somebody working in a pub in London is earning maybe £10 an hour, but they are doing exactly what somebody in Nairobi would be doing. They are no more productive, but the price is higher and the price is higher because it’s a non-traded service. The number of people who are uncompetitive in the service sectors in rich countries is way higher than it is in the manufacturing sector.

You see this in the legal industry already. Software robots can read hundreds of documents, hundreds of emails. These things will look for patterns, just like young lawyers used to do.

One of the companies that you mention a lot is Blue Prism, the British firm that offers Robotic Process Automation (RPA). You quote some of their marketing material, which says:

“Imagine a different kind of workforce. A workforce that you can teach countless skills. The more it learns, the more efficient it becomes. It works without ever taking a vacation. It can be small one day or large when your business hits a spike. And it frees up your best people to really be your best people. Meet the Software Robots—the Digital Workforce.”

That sounds vaguely sinister.

It is. I’m quite sure they’re going to have to change that when the upheaval starts. They are selling to companies the promise they’re going to undermine all these average-skill office workers.

What causes the upheaval?

People will find their jobs being disrupted. They already are, but they don’t know who to blame. And it’s possible that they never really do figure it out because it’s very disaggregated. I think I’ve started to detect that it’s digital technology that people are reacting against. The big issue will be if the white-collar workers connect with the blue-collar workers who’ve been hurt by robots in globalization for the last 20 years: then we could have a really holy mess.

Several studies show technology will disrupt jobs, but they also argue that almost as many jobs will be created.

“I’m a pessimist in the short run, an optimist in the long run.”

There’s a mismatch of job displacement and job creation. Job displacement is being driven at the speed of digital technology, which is explosive at this moment. And displacement is the business model for the AI geniuses and all those companies. All of them are hoping to get rich by displacing workers, not by creating jobs. Creating jobs is much slower. So, at least in the next few years, the displacement will outstrip the creation. But it’s not the direction of travel which is wrong. It’s just a mismatch of speed. I’m a pessimist in the short run, an optimist in the long run.

In the book, you suggest this backlash could lead to a new political movement. You write, “perhaps the movement will come to be called “shelterism”—not antiprogress, just a little shelter from the storm.” To me, it describes a sort of intersection between right-wing populism and left-wing socialism.

There’s a presidential candidate in the United States called Andrew Yang. He says people are trying to replace us with software robots and foreigners working abroad. He understands the globotics upheaval. He’s never going to get elected. But I think that the left-leaning Democratic candidates will find it almost irresistible to redirect the anger, which exists in America, away from China and away from immigrants and towards technology. Big tech will become the scapegoat.

You say you are a long-term optimist, but that’s probably only if governments can work to avoid massive job displacement, right?

I’m worried about the United States in particular. There’s qualitatively nothing different about this transformation. People are having to change jobs because of technology and that’s about a three-century-old story. The key difference is it may be happening faster than it did before.

Some governments are better at it than others. The UK is not the best, but much better compared to the United States. In the US, if you lose your job and you have a health problem, you’re likely to go bankrupt and lose your house or your kids have to stop going to college. So it’s stuff like that in the United States that is just so stark. In the [European] continent they have these active labor market policies, which are already engaged and are helping people adjust.

In past disruptions, some of the workers displaced from manufacturing ended up stepping down into worse jobs into the service sector.

I think that’s how it’s going to happen now.

What would people step down into today?

Uber drivers. Paralegals become Uber drivers! There will be a downgrading in the service sector, but it’s worthwhile pointing out that service workers are intrinsically more flexible than factory workers were. So while there will be a downgrading of skills, they will stay in the service sector and there won’t be mass unemployment.

What kinds of jobs will be created?

AI is giving sophisticated thinking skills and certain types of pattern recognition to average-skilled people. This will lead to semi-professional jobs between doctors and nurses, between lawyers and paralegals, between architects and draftsman.

A level that doesn’t currently exist.

That level doesn’t exist because it will make people a lot smarter and you don’t need a university degree to use this AI. It’s just dead simple.