Some of you may have seen the piece I did at New Geography showing the shift in migration trends for selected cities: http://www.newgeography.com/content/002514-back-city

 

I thought I'd share similar graphs for the 3C's.  Unlike say Philadelphia their central counties (which is what I define as "core") do not truly map to the urban core, but it's still a useful measure.

 

Index of migration from suburb to core:

 

 

Index of migration from core to suburb:

 

 

Total number of net migrants from core to suburb:

 

 

I trust you'll think this is good news.

 

It sure "appears" to be good news on all "3-C" accounts!  Thanks for presenting these provocative and positive graphs, arenn.

 

Got any data that goes farther back? Projecting forward, it looks like the net movement is still from core to suburbs, but the pace is slowing and the net movement will be toward the core starting in 2012 or 2013.

Cool, thanks.

 

Is this based on census estimates for these years, though? Those estimates turned out to be wildly inaccurate.

I have data going back to 95-96. It is based on IRS tax return statistics, so those are real people, not estimates.

 

It will be interesting to see what happens when the economy recovers. Obviously the suburban housing boom has taken a big hit. But I do find it interesting that out-migration started falling before the housing bubble popped.

 

It is based on IRS tax return statistics, so those are real people, not estimates.

 

 

Actually, its based on real exemptions, not quite real people.  Unfortunately, it also leaves out many low income households since they may not file returns.  But still, a nice tool.  I wish they would publish zipcode-to-zipcode, and not just county-to county.

 

 

^Yeah, the zip to zip data would be bad@ss.  Thanks arenn for posting.  It's good news in only the most parochial sense, since the probable driver is negative home equity (i.e., household financial distress).  Interestingly, a recent paper by some fed researchers found that negative equity was likely reducing intra-regional mobility even more than inter-regional mobility.  And, it wouldn't be good news if the decline in net core-to-suburb out-migration were tied to a corresponding increase in net out-migration from the whole region (e.g., more people in Cuyahoga County deciding to move, not to Medina, but to Texas), but I'm pretty sure the net migration from the Northeast and Rust Belt has also slowed, due in part to negative equity, but also to the decimation of sun belt economies and slower household formation.

Yes, I agree out-migration has been depressed due to the housing market. Some of the return migration might also be due to fringe suburban foreclosures forcing people to come back closer to the core to rent. But there's nothing wrong with pocketing good news where you can find it. I don't like to try to "explain away" bad data, so in a similar vein I don't want to pooh-pooh good data.  And the out-migration declines started well before the bubble popped. In fact, they were declining even into the peak of the bubble, which I find interesting.

 

Yes, I agree out-migration has been depressed due to the housing market. Some of the return migration might also be due to fringe suburban foreclosures forcing people to come back closer to the core to rent. But there's nothing wrong with pocketing good news where you can find it. I don't like to try to "explain away" bad data, so in a similar vein I don't want to pooh-pooh good data.  And the out-migration declines started well before the bubble popped. In fact, they were declining even into the peak of the bubble, which I find interesting.

 

This may be why there was so much hype in a "return to cities" before the Census.  The trends were already suggesting a decline in people moving out to the suburbs, but like many things, the Census revealed that the trend was not as strong as hyped.  Still, there *is* a trend, and it is a trend that we haven't seen in several decades.  I think some people were just too fast on their predictions.  It's great to see all 3 of Ohio's major cities involved in this trend, even if recent economic problems have helped it along.   

^^I don't mean to explain away the trend, just to put it in a context of household financial misery, which likely means its not due to any fundamental change in neighborhood-type preference.  It's "good" for the cities, but not necessarily for the families driving the trend.  I definitely agree that the housing stock of core counties is poised to benefit from increased rental demand.  And you're right that the early onset of the shift is a bit of a puzzle, but could be the conflation of intra-regional and inter-regional migration patterns.  For example, increased migration out of the region altogether in 2004 and 2005 might also have cut into net intra-regional migration.  Do you have the gross out-migration data by year for the core counties?  Or aggregated outmigration data for each core county? 

 

Seriously though, thanks for posting- this stuff is very interesting, even if the interpretation is contested.

i think that issue is best answered by comparing the sinking out-migration trend before the financial meltdown vs after it. i would assume the former would be for good pro urbanist reasons & the latter more due to staying put due to money/job problems. regardless, until maybe we think about it a little more i just take this data as positive news for the core cities.

Strap hanger, I'm not sure what you mean for those statistics, but I have over 150 variables of data from the IRS migration files, including tons of derived statistics such as the core to suburban stuff. I'm pretty confident I have what you're interested in. Of course, I'm in the business of selling that :)

^Sorry about that, sloppy typing and word formation...

 

For example, increased migration out of the region altogether in 2004 and 2005 might also have cut into net intra-regional migration.  Do you have the gross out-migration data by year for the core counties?  Or aggregated outmigration data for each core county?   gross or net out-migration by year for the 3C core counties, regardless of whether or not the other county was in the same region or in another region?

 

I totally agree that it's interesting to break out migration into intra- and inter-regional trends, but I wonder if looking at one without the other misses the full picture.

Brain starting to hurt like grad school.

 

Powerful numbers, though.