Mapping Migration in Georgia
Visualizing Regional Migration Trends in the Peach State
Over the last few months, I’ve written extensively about migration trends nationally, and also about the specific migration stories of my home state, Kentucky. As I’ve done this, I’ve been excited to get feedback from readers asking about other states: what stories can we tell about Illinois, or Iowa, or Oregon, or North Carolina, or, as I’ll be covering this week, Georgia? I don’t know as much about these states as I do about Kentucky, but I’m hoping to learn more as I research their migration patterns. For this week, I’m going to write a few posts analyzing migration in Georgia.
A major caveat: I’ve argued again and again that migration is extremely local and specific, and resists sweeping generalization. I stand by that argument. But that also means that explaining migration requires a great deal of local knowledge about businesses, climate, traffic patterns, culture, and innumerable other potential variables. I don’t have that local knowledge for many of the states I’m going to cover. I have tried to speak to locals and read up on the states themselves, but there will undoubtedly remain gaps in my understanding. To that end, I encourage locals to tell me where I’m wrong, what I’m missing, and how I can explain these migration trends better.
Dividing Georgia: How to Identify the Regions
My first task for studying Georgia was to identify meaningful sub-regions. This was a genuine challenge for two reasons: first, I had to learn about Georgia, and second, once I did, I realized there are really two different Georgias: Atlanta, and not Atlanta. The Atlanta metro area has 5.5 million residents in a state with just 10 million people total. Short-range moves within the metro area could be larger than the entire migration flows of some small states. I tried to find a way to subdivide Atlanta for incorporation into statewide trends, but the result was an unmanageable fractal pattern of migration sub-regions within Atlanta, and major statistical biases based on sample size issues. Thus, later this week, I will have a single post dedicated entirely to migration within the Greater Atlanta area. For today, however, I’m just calling the whole area by one name.
To define regions, I consulted three original sources: Google Maps, to get a sense of transportation and development networks; geological survey maps, to get a sense of underlying geographic regions; and Census-designated metropolitan area maps. Once I had rough outlines defined, I began researching specific cities and counties to check my assumptions, trying to make sure I hadn’t made any obviously incorrect categorizations.
I then allocated migration flows to their respective regions, and checked my results again: did any “marginal” counties that could reasonably included in either of two regions create major changes in results? Were any of my findings excessively sensitive to how I categorize one or two debatable counties? Were there any counties with extraordinarily robust bilateral migration flows for which I could not account? This process lasted for a while as I moved one county or another into different regions, re-assigning migration flows data each time. If you think this sounds boring, you have no idea how dull the reality actually is, although I did get to learn lots of trivia about random Georgia counties. Given that the state has the 2nd most counties of any state (Virginia is first, Kentucky third) and among the smallest average county size, I spent a lot of time researching why county lines fell where they did. The end result is a map of Georgia with 10 major regions.
Georgia’s Migration Regions: Central Cities and Rural Hinterlands
Georgia’s 10 “migration regions” are relatively uniform in terms of population and size, with the exception of Atlanta (very populous) and South Georgia (very large). Aside from Atlanta, populations range from 360,000 for Greater Columbus to over 700,000 for South Georgia.
I’ll briefly head-off a few objections to my regional definitions, to try and explain some of my most debatable choices.
Atlanta Exurbs: My definition of Atlanta is much smaller than the Census-designated metro area. Several counties along Greater Atlanta’s borders with Greater Columbus, the Blue Ridge Central Georgia, or North Georgia are counted in the Atlanta metro area as “exurban counties,” but, for my purposes, are counted in totally different regions. My designations are based on three considerations: commute times, road and development networks, and migration flows. In most cases, these counties had relatively stronger migration ties to the regions I designated than to Atlanta. They also tended to have over an hour-long commute time into Atlanta, or had no major road establishing a direct link. They also tended to have much lower population density. For all those reasons, I elected to count them as outside of Atlanta.
Savannah and the Lowlands: Cities like August and Columbus got their own regions, but Savannah didn’t, despite being larger than either. Furthermore, Savannah does actually have fairly distinct migration patterns that differentiate it from the rest of the Lowlands (it tends to lose migrants to other states, while the rest of the region gains). However, treating Savannah as its own region left no reasonable way to categorize the rest of the coastal lowlands: they have very distinct migration patterns relative to South Georgia or the Upper Coastal Plains, and also probably too small of population to be a statistically meaningful group in their own right. Furthermore, geological, climatalogical, and biome maps show that the lowlands from Savannah to Florida share many common features that may impact migration: climate, ocean access, scenery, demographics, etc. Thus, for statistical reasons, and because of geographic similarity, I kep Savannah lumped in with the rest of the lowlands.
Undivided South Georgia: A strong case could be made for dividing South Georgia into two regions, one in the west, one in the east concentrated along I-75. The eastern counties tend to be more densely populated and have higher total migration volume, while the western counties are more sparsely populated. The eastern counties adjacent to I-75 saw net interstate migration of over 3,000 people per year, while the western counties of South Georgia lost over 1,000 people per year to other states. However, both regions saw similar in-state migration of about 1,400 people per year. Furthermore, both areas have very similar industrial structure, demographics, climate and geography, history, educational and medical infrastructure, etc. An argument could be made that the I-75 corridor is different, but the new information generated from such an additional segmentation of the data seemed small to me compared to the amount of additional work it required.
Total Domestic Migration: Columbus Dominates, Atlanta Vanishes
Most regions in Georgia have positive total domestic migration. Of the regions with negative total migration, only the Valley and Ridge Country has out-migration greater than 0.1% of the population. It is important to note that Greater Atlanta is actually losing migrants, driven largely by major in-state out-migration. Once international migrants are included, Greater Atlanta has almost exactly balanced migration.
Columbus’ extremely high total migration is largely due to Fort Benning: the 2005 BRAC process expanded the garrison at Fort Benning, which already plays host to over 120,000 civilian or military employees and contractors and their families. While Columbus does get in-state migration, its real strength is in interstate migration. Augusta and the Coastal Lowlands likewise have migration records that are dragged far above what would otherwise be the case due to the presence of large bases (Fort Gordon and Fort Stewart, respectively).
In-State Migration: Atlanta Loses, the Blue Ridge Wins
In terms of in-state migration, Atlanta is the clear “loser.” I don’t usually focus on “winner” or “loser” language for migration, but Atlanta’s case is genuinely concerning. I looked at migration by various demographic factors, and found that Atlanta’s migration trend is essentially a story of out-of-state whites displacing blacks to other parts of the state. While racially-biased migration isn’t necessarily a problem, it may be when it leads to one race being systematically expelled from regions experiencing economic growth to regions with much higher poverty, unemployment, and other negative characteristics. The Yankification of Atlanta, in other words, is also a racial process.
In-state migrants are displaced to many regions: Columbus, North Georgia, South Georgia, the Lowlands, and the Blue Ridge all receive migrants, with the Blue Ridge receiving the highest in-migration rate, 0.65% per year.
Interstate Migration: Military Bases, Good Cities, and Feeder Schools
Interstate migration in Georgia can be explained by looking at military bases, education, and Atlanta’s ongoing appeal as a pleasant, affordable city. Greater Atlanta has just a 0.12% net interstate migration rate, but that comes to about 1/2 of all interstate migrants into the state. Meanwhile, Columbus has higher net migration than Atlanta in raw numbers or on a percentage basis, but just 1/7 the gross in-migration.
Columbus, Augusta, and the Lowlands all have major military bases. If net migration associated with the military base counties were set at zero, Georgia’s net migration would fall by 50%. This is an extremely important statistic. Pundits may claim that migration to Georgia is about weather or taxes or jobs or culture: but the reality is that half of Georgia’s migration record can be explained by military deployment decisions.
North Georgia loses migrants, especially young, college-educated migrants. This is because the University of Georgia at Athens produces an enormous amount of undergraduates qualified for graduate programs or specialized work, but Georgia doesn’t have enough higher-level programs to absorb them all.
Total Migration Volume: Military, Military, Military
I’ve commented on the military several times, but must do so again. The regions in Georgia with the highest total migration volume are all regions with major bases, and migration in counties containing those bases drives the regional trend. In Greater Columbus, over 15% of the population will migrate into or out of the region in any given year. For comparison, migration churn rates in the military-heavy parts of Kentucky are about half as high.
Interestingly, Atlanta doesn’t have that much overall migration volume. In any given year, most Atlantans (Atlanteans?) are more likely to stay in place than other Georgians.
The classic story about southern-state migration is one of weather or taxes. Conservatives claim low taxes attract migrants, liberals say its just the weather. But a cursory study of Georgia’s actual migration trends would indicate it might be neither. If other southern states have as much military presence as Georgia, that could be a huge part of the “Sunbelt” migration trend.
In my next post, I’ll explore what’s going on within Atlanta. Then, over the next several weeks, I’ll try to continue regional spotlights for Kentucky and Georgia, and also begin doing state summaries like this one for other states.
Start the series from the beginning!
If you like this post and want to see more research like it, I’d love for you to share it on Twitter or Facebook. But what’s just as valuable for me is if you click the recommend button at the bottom of the page. Thanks!
Follow me on Twitter. Follow my Medium Collection at In a State of Migration. I’m a grad student in International Trade and Investment Policy at the George Washington University’s Elliott School. I like to write and tweet about migration, airplanes, trade, space, and other new and interesting research.
