Keith Romer

Yeah, the old housing projects had really separated low-income people from the rest of their community.

In the new Hope Six projects, there was, by design, a mix of different kinds of housing.

Some public housing, some housing for people who made too much to qualify for public housing, and some full-on market rate housing.

So the question for Raj and his team was, did Hope Six actually help to reduce poverty and improve upward mobility?

And to answer those questions, they got their hands on these massive data sets.

So different kids lived in Hope Six developments for different amounts of time.

Some were there for only like a year, some for their entire childhoods.

And according to the team's calculations, the average kid who spent their whole childhood in a Hope Six development, they would see a roughly 50% boost to their income.

I saw some evidence suggesting, I think, including in your paper, that the Hope Six projects were much more selective in who was allowed into them.

What would you say to that concern, that these kids are just systematically different?

And to a lot of housing advocates we talked to for this story, this second explanation seemed really plausible.

In a lot of places, there was this whole screening process to get into the new HOPE VI developments.

Some people who had lived in the old housing projects simply weren't allowed to move into the new buildings.

To get around this problem, the team went back to an idea from earlier studies, an idea about how neighborhoods shaped the way kids' lives turn out.

So yeah, it looked like Hope Six caused better outcomes, but it could still be the case that families who stayed a long time in Hope Six were just different from the families who stayed a short while.

Whoo, plenty of money, tip of the hat.

That's a huge award, Keith.

OK, so yes, they were now pretty confident that what they were seeing in the data was real, that there was something about growing up in the new Hope Six public housing that made a big difference.

But to be able to use their findings to shape policy going forward, they really needed to sort of hone in on what was actually causing this improvement in outcomes.

Basically, it wasn't clear what part of Hope Six exactly was causing the big difference in outcomes.