For all the economic turbulence – let’s use that neutral word – that Brexit is likely to create, the impact of leaving the European Union is nothing compared to what is about to happen to our economies. The twin forces of automation and artificial intelligence (A.I.) will dwarf whatever economic challenges we currently face. To give a historical perspective, we are like Henry VIII’s courtiers, worrying about our break with the Church of Rome while ignoring Guttenberg’s printing press.
Few people seem to have recognised this. A handful, including Paul Mason and Matthew Lawrence, have noted that Brexit is an unhelpful distraction. But more surprising is that barely anyone has argued that being outside the EU might actually be a distinct advantage in the face of the looming A.I. revolution.
Obviously there will be problems. The headwinds of a Brexit-induced skilled worker shortage plus inflation would hamper the A.I. sector, perhaps more than most. There’s also a chance some of our snappy start-ups will head to Berlin, which is fast becoming the EU’s tech hub. But taking back control – if that means being flexible, dynamic and bold – could be exactly what the sector needs.
How precisely to do this will depend on your politics. Theresa May would presumably look for an active government-led industrial strategy. Our departure from the EU opens up some opportunities on this front: the government could aggressively use tax and investment in a way not currently possible within the EU, a move mirroring the early days of Silicon Valley which was a centre of US government investment in military technology. EU state aid rules prohibit member states from distorting free competition between states with tax reliefs or resources. Infrastructure, financial and other incentives are particularly effective in the tech sector because companies can relocate fairly quickly and easily as they tend to have few fixed assets.
We would be building on a position of relative strength. While there are problems (despite our excellent higher education, we under-invest in robotics compared to the rest of Europe) the UK is already better endowed than other European countries with four key inputs for the sector: a strong higher education infrastructure, an investment community with deep pockets, an existing cluster of technology companies (a source of experienced talent) and good ties with the critical US market. A tilt towards the tech and AI sector would also encourage more broadly a closer relationship between industry and education: developing new research departments and incentivising closer relationships between academic departments and private industry. (Furthermore, current emerging technologies and associated new business models are game changers across a broad swathe of industries beyond just A.I.)
So far, so obvious perhaps. But there is also a predictably depressing way this could play out. Shortfalls in the rest of the economy might lead the UK government into a desperate clingy embrace with this emerging sector. And while that might yield some short term benefit, without equally careful planning about how to respond to the social disruption of A.I., any economic benefit could be swiftly wiped out by social costs.
There is now a small library’s worth of predictions about what AI will do to our economies. The Economist, MIT Review and Harvard Business Review have all recently published articles about how the economy is on the brink of transformation. Over the last year or so various forms of A.I. technology, teamed up with robotics, are making inroads into brick-laying, fruit-picking, burger-flipping, banking, trading and driving. Machine learning (a branch of A.I. in which machines use examples to learn how to mimic human decision making) can, for example, already outperform the best doctors at diagnosing illness from CT scans, by running through millions of correct and thousands of incorrect examples real life doctors have produced over the years. Potentially no industry will be untouched.
The effects will be dramatic. President Obama’s team suggested driverless cars would dispense with 3 million jobs pretty soon. According to the Bank of England, as many as 15 million British jobs might disappear within a generation. The best account I have heard is from a former adviser to the French Government. He reckons ten per cent of jobs in France will be wiped out in ten years. Half of jobs will be unaffected. The remaining 40 per cent will be humans and machines working together. Although the timing might be out on some of these predictions, the direction of travel is clear.
No-one know exactly where this leads. But it will have an effect on the shape of our economies. David Autor, an MIT economist, reckons we could be heading toward a ‘bar-bell shaped economy’. There will be a few lucrative tech jobs at the top of the market, but many of the middling jobs – trucking, manufacturing, logistics, factory work – might wither away. They will be replaced by jobs that can’t be automated, many highly skilled (and highly paid) – but also very many in the low paid service sector. There will be new jobs – who imagined app developer would be a profession – but will they be the same sort of jobs? Will they be in the same places, or clustered together in already well-off cities? Will this mean far more inequality than now? What does this mean for the idea that you can get ahead with hard graft?
How government manages the transition is as important as the race for jobs and growth. And again flexibility and agility will be vital. The EU certainly has far more clout than a single country (the Commission recently fined Google €2.4bn for a breach of anti-trust law – although Google is appealing). But the mighty behemoth can also be over-regulating and ponderous. It has been criticised often for over-regulating digital platforms. Post-Brexit UK might be better positioned for swifter, and more innovative action on the regulatory and social consequences of this new industrial revolution. Any industrial strategy will need to analyse new forms of inequality – the tech sector has a tendency to cluster – and make sure it doesn’t simply exacerbate existing inequalities, even if that means giving some regions an advantage over others. UK society will need to accommodate for that ten per cent of jobs which will be vapourised: empty promises about re-training for the knowledge economy won’t be enough. It will need sufficiently bold opportunities for in-work retraining for the vast numbers of people whose workplaces will be transformed. (One suggestion I’ve heard is a universal education income: everyone should be paid through central taxation for one day training every week. This seems more plausible than the increasingly discussed universal basic income.) It will even need to experiment with robot taxation or other ways of making up for any revenue shortfalls as the gig economy, electric cars, and the rest continues to grow. Perhaps new forms of collective ownership might be tried too – for example, a local authority run ride sharing app for regulated taxis. And it will need to do all this with the agility, speed, and willingness to experiment that matches the dynamism of the A.I. sector itself – rather than a clunking, if powerful, bureaucracy.
The changes to our economy are going to be significant: and it can be a source of great economic opportunity when we go it alone. But the changes to society will be equally dramatic. To only think about jobs, jobs and more jobs would be self-defeating and short-sighted. It is also, unfortunately, the most likely response. But it doesn’t have to be.
Jamie Bartlett is Director of The Centre for the Analysis of Social Media
