Why People Think There Is a Sunbelt
It All Comes Down to the Denominator
In yesterday’s post, as well as the historical series leading up to it, I made very clear that I don’t buy the Sunbelt/Frostbelt distinction as being particularly meaningful. Physical geography does not determine economic geography. And I think I laid out a pretty good case that popular narratives about the regional geography of migration are wrong.
But that raises a question. If the Sunbelt/Frostbelt dichotomy is so obviously wrong, then why does the narrative persist?
There are lots of reasons this could be. Maybe people propagate it as a convenient political artifact allowing them to make otherwise unfounded claims about electoral politics or preferred state policies. Or maybe it accurately describes the experiences of a few major cities where many media elites reside. Or maybe people just don’t look at the data very closely. Or maybe, if I may say so myself, my work was just really innovative and nobody has ever strung together that kind of time series before. Maybe.
When 1% Is Not 1%
Comparing State Migration to National Population
Or maybe there’s a difference between 1% migration in California, and 1% migration in Vermont. My focus was on net migration rates, where the rate was based on net flows divided by prior-period state-level population. But while local policymakers may care about the rate, absolute numbers matter for certain economies of scale, for some Federal programs and designations, and the size of a given migration flow compared to national population is important for writers at nationally-read publications.
In general, the broad trends should be similar in both cases. There won’t be anywhere that loses migrants based on the state flow rate, but gains when divided by national population. But the relative size of, and thus the emphasis we assign to, a given state’s migration balance may be dramatically altered. Whereas dividing by state size lets us compare different states apples-to-apples, dividing each state’s net migration balance by the whole national population let’s us see how big that flow is in the national scheme of things. Even if a state has 10% net inflows, if it’s an itty-bitty state, it may be perfectly reasonable to ignore it when writing about national trends.
So how do things change when we divide by national population? First, a look at the states.
The above chart tracks the net decennial migration balance as a percent of the prior-census population of the entire United States. I included every state that had any period that cracked 0.5% absolute value. Only 8 states met that criteria.
Hold the phone — what? Yes, you heard me, only eight states since 1900 have seen sufficiently extreme net migration to have a decadal migration balance equal to 0.5% or more of national population. The most extreme single-state net migration flow ever observed was into California during the 1920s. About 1.8% of the nation’s population moved into California during the 1920s, on net.
Okay, so what can we see in these 8 states? Well, first of all, except for Kentucky and Oklahoma, they’re among the most populous states. Oklahoma’s high inflows related to its “Land Rushes” and it being the last “frontier” opened. Kentucky’s high outflows are not as simple to explain, but, in both cases, extreme flows subside before WWII.
We can see that Florida commanded high migration in all periods, but highest from the 1950s to 1990: that sounds like the right window for the Sunbelt! And California commanded its highest migrations from the 1920s to the 1960s, which also seems like it could fit. Texas has seen rising migration recently, but New York has too.
Overall, this presentation seems to fit a little better with the Sunbelt/Frostbelt story… but maybe we can get an even closer fit.
Behold the Sunbelt
In This Case, Size Matters
Well, lookee there! The South dominates migration, and that domination has persisted from the 1970s to the present, led off by striking rises from the 1930s to the 1960s. And look: the Rust Belt and the Mid-Atlantic states are the two lowest over that period! Those other areas like the Western Sunbelt, New England, the Plains, the Northwest, they all basically break even around zero, so probably aren’t worth mentioning.
By presenting regional net migration as a share of national population, we find a significant shift in national population from the Rust Belt and the Mid-Atlantic to the South. That’s the Sunbelt/Frostbelt story! Found it!
I hesitated to post this because it’s sort of giving fodder to a view with which I vehemently disagree, but the point here is that if you want to show that there’s a Sunbelt and a Frostbelt, you do it by aggregating states fairly carefully, and then dividing flows by national population instead of local population.
So we found the Sunbelt and the Frostbelt. But does this data actually tell us anything useful?
Politics vs. Policy
The Two Uses of Migration
Long-time readers know my disdain for the Sunbelt/Frostbelt narrative. They also know my explicit aim in running this blog: I want to offer useful information to help policymakers make better decisions, especially about state and local policy.
For that kind of policy information purpose, dividing migration balances by national population is actively misleading and inappropriate.
When policymakers want to know the role migration plays in their area, it doesn’t help to say, “Well, X% of national population moved in,” because that X% would have a hugely different impact in Vermont or California. California could absorb inflows of 1% of the national population. Vermont would explode. So as far as advising policymakers goes, the “correct” variable is the local net migration rate (or, for various purposes, gross rates, inflow/outflow rates, replacement rates, but always something tied to other local scaling variables). That allows apples-to-apples comparisons across local areas, and gives policymakers a sense of the real scale of the challenges they face.
But if your goal is to talk about shifting electoral votes and voter demography, national population shares could matter more. Even at 5% net migration, it could a long time for North Dakota or Delaware to get more electoral votes. What we care about for shifting electoral votes and House representation is the share of national population. And for that kind of conversation, we absolutely need to look at these shares. For example, I find that domestic migration is likely to boost Texas’ share of national population by 0.4% in the next Census, and Florida’s by 0.33%. Meanwhile, New York’s will decrease by 0.4%, California’s by 0.2%, and Illinois’ by 0.25%. Naturally, immigration and natural population increase offset these declines, so don’t take them as gospel, but these “national shares” give us a sense of the scale of migration.
For comparison, one House seat is worth somewhere between 0.05% and 0.5% of national population depending on the state and how long it’s been since the last apportionment. In other words, these net migration flows are sufficient that, if not offset by other facts, California, Illinois, and New York could lose multiple House seats to Texas and Florida purely due to domestic migration.
So what’s the right way to view this data? Well, if you’re informing policymakers, use locally-tied migration rates. In 90% of cases, local-tied rates like the net migration rate or the replacement rate are the correct variable to look at. You should use national-population-based-rates if and only if you’re actually talking about the net redistribution of population, especially for electoral college purposes. And even then, exercise caution. There is huge variation within the “Sunbelt” and the “Frostbelt,” and identifying them as monolithic groups is almost certainly misleading.
See my previous post, announcing my new Podcast and presenting my recent foray into historical research.
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. Or, just as valuable for me, you can click the recommend button at the bottom of the page. Thanks!
Follow me on Twitter to keep up with what I’m writing and reading. Follow my Medium Collection at In a State of Migration if you want updates when I write new posts. And if you’re writing about migration too, feel free to submit a post to the collection!
I’m a graduate of the George Washington University’s Elliott School with an MA in International Trade and Investment Policy, and an economist at USDA’s Foreign Agricultural Service. I like to learn about migration, the cotton industry, airplanes, trade policy, space, Africa, and faith. I’m married to a kickass Kentucky woman named Ruth.
My posts are not endorsed by and do not in any way represent the opinions of the United States government or any branch, department, agency, or division of it. My writing represents exclusively my own opinions. I did not receive any financial support or remuneration from any party for this research. More’s the pity.
