Keith Romer

At some point, Matt started explaining this puzzle to Carol Naughton, a woman he knew who ran a big housing nonprofit.

Carol's from Atlanta, and she told Matt that she had noticed that not all hopesic sites in Atlanta were the same, that some of them did much better than others.

So just to underline a little bit what Matt is saying here, not all hope six developments were created equal as far as this whole income mixing idea went.

If a hope six development was near richer neighborhoods, the kids in public housing there tended to do really well.

If surrounding neighborhoods were poor though, the kids in hope six development saw no gains.

This strongly suggests that Hope 6 caused better outcomes, not because it improved the housing or got rid of lead paint or whatever.

It suggests that these new neighborhoods fostered more social integration between kids from different backgrounds.

When that crucial ingredient was missing, the Hope 6 kids did not see better outcomes.

Now, this is all pretty astounding.

Low-income kids do way better when they interact with or are friends with higher-income kids.

This is a crucial finding in their paper.

And it left us wondering, like, why?

So we asked Raj, why would these social interactions matter?

Theory number two is about information.

Information that you might get by like, you know, hanging out with your higher income friends or their families.

in part because of some legitimate criticisms of how HOPE VI was implemented.

The biggest criticism was that it knocked down a ton of public housing that it didn't replace.

Almost 100,000 units were demolished, but only 55,000 public housing units were built back in their place.

Tens of thousands of families were functionally kicked out of public housing by HOPE VI.

So yeah, HOPE VI had some big issues, but it also integrated low-income kids with high-income kids, and that had a really positive impact on their future outcomes.