Thread by @bencasselman

In the early 2000s, the news was full of stories of a coming jobs apocalypse: Globalization had come for manufacturing. Now it was coming for white-collar jobs too.

15 years later, the numbers are in: The apocalypse never happened.
nytimes.com/2019/09/27/bus…

In a new paper, @ModeledBehavior revisits some of the most influential offshoring work, Alan Blinder’s 2007 paper that categorized jobs by how easily they could be sent offshore. Blinder estimated that a quarter or more of U.S. jobs would be “offshorable” within a decade.

But a decade later, @ModeledBehavior finds no relationship between how “offshorable” a job was and subsequent job growth.
Some easily offshored jobs shrank. But others grew.

To be clear, offshoring was not a myth. Millions of office jobs were sent to India, the Philippines and other places where labor was cheap. And it’s still happening today.

But those losses were offset by growth elsewhere in the economy. And at the same time, companies discovered offshoring was harder than it looked. There were language barriers, cultural barriers, logistical barriers. There was also political backlash.

But Blinder (& others) wasn’t wrong about technology making these jobs easier to do from a distance. It turns out there *is* a relationship between how easy a job is to offshore and the share of people in that job that work from home (in the U.S.).

The work-from-home share is still small. But it hints at a larger pattern: Instead of sending jobs offshore, (some) companies are moving them to other, cheaper parts of the U.S. Satellite offices, remote workers, domestic outsourcing, independent contractors...

There’s an tantalizing possibility here: that the technologies we worried would destroy jobs in the U.S. could instead be a force for geographic convergence, pushing back against the widening economic divide between big cities and everywhere else.

To be clear, that hasn’t happened yet. There’s some evidence the geographic divide isn’t growing quite as quickly as it was, but little evidence it’s shrinking. So we will see.

One last thought: The offshoring debate may carry lessons for another much-discussed threat, that of automation. It’s remarkable how similar today’s stories on automation are to the offshoring stories of 15 years ago.

Of course, just because one threat didn’t prove as bad as feared doesn’t mean the next one won’t be worse. And there are obvious differences between automation and outsourcing (and also concerns about how the two interact).

Still, as @SusanLund_DC told me:
“The lesson is, change is evolutionary, not revolutionary... And because it’s a slow change, it gives people and companies a chance to adjust so we never saw the mass exodus of jobs to India.”

Anyway, much more in my story: nytimes.com/2019/09/27/bus…
And in @ModeledBehavior’s report: upwork.com/press/economic…