Kathryn Anne Edwards

It's 53 separate state and territory programs.

They don't coordinate benefits.

They don't coordinate benefit size, benefit eligibility.

They've got 53 aging computer systems that don't talk to each other.

And they don't cover workers that are not in W-2 employment.

The benefits themselves are incredibly meager.

They don't go to the majority of the unemployed or replace even a majority of their wages.

And it's very much a program that was designed around temporary seasonal layoffs and manufacturing in the 1930s.

It's well-funded to a certain degree relative to— because it's a social insurance program, so it's funded by payroll taxes, so it never runs out of money the same way that, you know, if it were funded by the largesse of the federal government, it could very easily have been ended 15 years ago.

But the federal government walked away from careful management of this program in the 70s, and we have suffered every recession because of it.

So a new dynamic unemployment system, I have—

like hours worth of ideas about it.

I'll let you pick what you want to hear, but we could, I've always thought that the scariest thing about AI is losing your job.

Well, I can't change AI, but I can change how you feel about job loss if there was a system of unemployment that you felt confident in.

Of course.

I'm so glad you asked.

So, yeah, problem one, it needs to be federal.

You need to have a federal system that incorporates not just W-2 employees, but the self-employed, your contractors.

Number two is that it needs to be triaged to look more like what we know unemployment to look like, which is a lot of people who are unemployed for a pretty short amount of time

Some people that are unemployed for a longer amount of time but still get a job, and then the very long-term unemployed that need to get retrained or move into a different occupation.