AI, Brexit, and the fungible lives fallacy

by Edward Cone

I look at Brexit and see some serious technology-induced pain coming.

We keep building complicated things (trade blocs, currency unions, computer-powered companies) without acknowledging the costs that will come with their benefits, even when those costs are well understood, and then we act surprised when the people who pay those costs get hurt and angry. And then we have to pay the costs anyway, with interest.

Artificial intelligence and robotics are expected to squeeze a lot of people out of their jobs in the years ahead. The optimistic prediction is that over time enough new career options will emerge to replace those made obsolete by the machines. Yet even in that rosy scenario we know that many, many people are still going to get left behind.

That’s because human lives don’t always sync well with secular economic, technological, or political trends. We get a limited window of productive years, and a big shift – no matter how beneficial on balance or to workers in another country or in years to come -- can seriously mess things up for individuals, families, and entire communities.

imageMaybe you see this happening close to you. It’s obvious near my hometown of Greensboro, NC, where the boarded-up main streets and rusted factories along both sides of the Virginia border show how globalization, automation, and regulation have hammered local economies for a generation and counting. (Photograph taken near Eden, NC by Lisa Scheer.)

Here’s the same reality through an academic filter: “Alongside the heralded consumer benefits of expanded trade are substantial adjustment costs and distributional consequences…Adjustment in local labor markets is remarkably slow, with wages and labor-force participation rates remaining depressed and unemployment rates remaining elevated for at least a full decade after the China trade shock commences.”

This kind of dislocation played some role in the Brexit vote. And while the UK never joined the Eurozone, the legitimacy of the EU was further undermined by its adoption of a common currency without a political structure capable of dealing with the easy-to-predict problems that have since immiserated tens of millions of people for nearly a decade.

So that’s my concern about the coming revolution in AI and robotics – less the impact of technology itself than our inability to plan for it honestly and holistically. A cost/benefit analysis may well tilt toward the benefits, but we know the costs on the ground will be substantial nonetheless. Yet we are not prepared to mitigate the inevitable damage, or even to discuss it seriously. Another sophisticated system is being constructed without adequate attention to its downside.

The pushback against the under-engineered mechanisms of the EU is messy, and may get a lot worse. But the reach of next-generation technology will be broader and the dislocations it causes may be much larger. Now would be an excellent time to recognize that human lives are not fungible goods.

Edward Cone is deputy director of Thought Leadership at Oxford Economics