Friday, April 25th, 2014
The IRS recently released its place to place migration data for 2011. This data is reported at the county and state level, but I process it to enable analysis at the metro area level, which is something you rarely see. This is one of the many data sets available in my Telestrian system.
I’ve been playing around with it a bit and wanted to show a few trends. The first is slowing net out-migration to the suburbs, which is something I’ve highlighted before. Here’s a chart of net migration of people from the core county* to suburban counties in large Midwestern metro areas from 1995 to 2011 (the 1996 data is people who moved between 1995 and 1996):
I normalized this across metros by plotting the data as indexes. You’ll see that the trend was up, particularly during the housing madness, then turned south when the bust happened. All of these metros except Detroit have lower levels out net suburban out-migration today. Some of them are down to around only 25% of their 1995 value, and it’s not inconceivable they could see reversing migration if the trend continues. Keep in mind this is county level data, not city.
I decided to look also specifically at in-migration, people moving from a suburban county to a core county. That would help us see if the change is a result of declining out-migration or if more people are actually moving back closer to the center.
We do see that there’s been an increase in migration to core counties, though nothing of the magnitude of the net change. But there has been some modest increase in the numbers moving in. However, look closely: the second highest increase is Detroit – Wayne County, MI. So we can’t reflexively treat this as a pure sign of core health.
A Closer Look at the Twin Cities
Minneapolis-St. Paul had the biggest collapse in net suburban migration and the biggest increase in in-migration. I wanted to show the absolute levels where to give you a feel for what’s going on. This is an interesting case. Since these are twin cities and there’s no standard government definition of core county, I treat both Hennepin and Ramsey Counties as core counties. (Movements between them don’t count as suburban migration in or out). This makes the MSP case a bit different. But with that caveat, here’s the out-migration:
We see the rise and then fall, with the net result that people moving outwards has returned to the levels of the mid-90s, but hasn’t dropped much below that. Here’s in inbound:
Here we have close to a 50% increase in inbound migration from the suburban counties. The major rise seems to have paralleled the housing collapse, which could indicate people forced to move closer into the core for rentals because of foreclosure. The Twin Cities actually got hit hard in the exurbs during the bust, so it’s not surprising to see a change there. As housing markets normalize, it will be interesting to watch where this pattern goes.
Cleveland’s Great Divide
I’ve heard some people say that Cleveland is really the demarcation point between Chicago’s sphere of influence and New York’s, and that the West Side of Cleveland orients towards Chicago and the East Side towards New York. I can’t do an East vs. West Side comparison, but I did want to see how Cleveland’s migration patterns have changed with these two megacities over time. First let’s take a look at gross migration, or the total number of people moving in both directions between Cleveland and these cities. Gross migration is very important because two places could both have net migration of say 0, but one city pair might actually have nobody moving in either direction and the other pair might have 50,000 people moving in and 50,000 people moving out. The net value can obscure some of the most important human capital ties between cities:
As we see, Cleveland used to have greater church with Chicago, but now it is more equally balanced between the cities. Although there’s a slight uptick with New York, it appears mostly a decline vs. Chicago. This makes sense because as I’ve documented, it was Chicago that really boomed in the 90s, but suffered a lost decade in the 2000s.
Here’s the net migration:
This is interesting. Cleveland has experienced a net loss of people to Chicago every single year, whereas it actually attracted people on a net basis from New York. Is this because of lots of people moving to Chicago or fewer people moving from Chicago? I decided to take a look specifically at the in-migration:
Over this period of time, the two cities were actually very comparable in terms of how many people moved to Cleveland – almost equal in fact. So there must be something going on with out-migration. But rather than go that direction, I want to show another chart of in-migration, this time looking at income instead of people.
Here we see that although there’s a fairly balanced inflow of people from both megacities, Chicago is sending much more income to Cleveland. In a couple years there were big spikes, maybe indicating some high income movers. Things have evened out in the last few years, but historically Cleveland has been importing more income from Chitown than NYC.
This is just a quick look, but it shows you the type of relationship information between cities that can be picked apart a bit using this data. It was in part research based on this that led Cleveland State University to establish its Center for Population Dynamics with Richey Piiparinen to help Cleveland better understand its human capital flows in more detail. With the criticality of talent and human capital to urban success, understanding this with regards to your city is a must.
* Core for St. Louis is city+St. Louis County. Core for MSP is Hennepin+Ramsey Counties