Where Is the Rust Belt?
I’ve written before about how we define the Appalachian Region. It’s a tricky subject.
But recently, I’ve been asked to think about how we define the Rust Belt. The term “Rust Belt” is used to refer to all sorts of places. In response to my post identifying Cincinnati with the Rust Belt, several commenters asked for clarification: which cities do I think are, or are not, in the Rust Belt?
Well, that’s a tricky question. The Rust Belt isn’t precisely the midwest. Nor is it precisely anywhere with industrial decline.
As I see it, the Rust Belt has 3 major characteristics:
- Large historic and current role of manufacturing, especially heavy manufacturing, like steel. It should be “Rusting,” i.e. experiencing industrial decline. But manufacturing should still be a big part of its economy. Alternatively, some parts of the Rust Belt are not manufacturing-intensive, they are mine intensive. So counties with coal mines active as of 1983 (the oldest data I could find that was already machine-readable, complete, and county-coded) will also show up as “Rusty.”
- It is a contiguous “belt” or region, not just random splotches. So if we get places that both show the statistical signs of being “Rust Belt,” but there’s a vast non-Rusty place between them, only one is likely to actually be the Rust Belt. We can fill in some gaps, but only if they are fairly small.
- It has to have that special midwestern something. For my purposes, that “something” is the Lutheran share of county-level population in 1950, around the time when Lutherans were at their peak level of national population.
I’ll break these into three maps: a map of all my manufacturing components together, a map of the “mine belt,” and a map of Lutheranism in 1950. Then, I’ll combine them together for a general index of Rustyness. Finally, I’ll fill in the blanks, simplify, and present my map of the Rust Belt.
Mapping the Rust
Let’s start with manufacturing. I have 3 key measures of manufacturing. The first county-level value of iron, steel, and sailing vessel production from the 1880 Economic Census. After 1880, data on this topic was no longer produced. 1880 is quite early for much Rust Belt development as it is pre-car. But it should at least give us a first pass of places that should definitely be included. Any county with ironworks gets 1 point.
Then we look at the manufacturing employment share of population in 1940. Any county with at least 1% higher employment in manufacturing that the national average gets a point. I then do the same for BEA’s 2016 county-level estimates.
So a county can get 3 points if it had ironworks in 1880, above-average manufacturing in 1940, and above-average manufacturing in 2016. I then give it another point if its manufacturing population share declined between 1940 and 2016 faster than the nation on the whole, and an additional point if it managed to complete a hat-trick of 1880, 1940, and 2016 manufacturing measures, basically as a way to single out places that meet all of my criteria.
Here’s the resulting map:
As you can see, there are at least 3 distinct manufacturing zones. In the south, we see the upper-south textile areas. In the northeast, we can see a zone around Massachusetts, which is also textile heavy, with a side of shipbuilding. Then we get a swathe from Pennsylvania, hrough Ohio, to Wisconsin, with a slash down the Ohio valley as well.
That is the Rust Belt. Remember my point about geographic continuity? I’m tossing out New England and the upper South. They may be Rusty, but they are not part of the belt. They are their own thing.
So next let’s look at another key question: coal mines! Heavy industry is key to the Rust Belt mythos, and heavy industry is heavily reliant on coal for energy. So anywhere with coal mines we should expect to be tied in to the Rust Belt Ecosystem.
This map includes some extra weighting for 1880 steelworks and for the overall manufacturing index, but is hugely driven by coal mines. You can see two major coal belts that tie into the Rust Belt: Appalachia, and the Illinois/Indiana/Western Kentucky cluster. The string through the Great Plains and the Mountain West mines are not of interest to me.
Finally, we can look at my indicator of cultural Rustyness: Lutheranism!
Here I dropped the binary to look at the actual data. Blue-green indicates more-than-national-average Lutheran share, brownish indicates less. As you can see, there are several distinct clusters of Lutherans. Down in Texas there is a cluster of Lutherans. There are a few smatterings in the Carolinas and the Great Valley. Then a very big group in Pennsylvania. Then not so many in western PA, but a very big group in northern Ohio, into northern Indiana. There’s another group in northeast Michigan. And then a vast clump of Scandinavians and Germans dominate the upper midwest.
The Carolina and Texas Lutherans don’t interest me. The others do!
So, when we combine all of our indexes together, we get:
There’s a more-or-less consistent band of shaded counties from Pennsylvania to North Dakota. There are variations in the intensity of course. And there is another band in the upper south, and smatterings in New England and out west… but one belt really stands out.
So let’s shade it in a bit better to make a smoothly-defined region.
The dark green areas indicate places that I think are indisputably Rust Belt. The paler areas are places where I think there’s a case to be made for it, but the data also suggests there are various substantial missing factors to be definitively included in the Rust Belt.
There were some surprises for me. For example, the big swathe of low-manufacturing-intensity counties cutting through Illinois up into Indiana, versus the two peninsulas of Rust-Beltery terminating in St. Louis and Owensboro, both surprised me. I also also surprised by how weak many of the indicators were in the upper Ohio River valley around the OH/KY/WV tristate area.
Anyways. I’m sure others will disagree with one or another of my Rust Belt identification choices. But I think this is as rigorous, defensible, and useful a map of the Rust Belt as exists anywhere.
I’m an an Advisor at Demographic Intelligence, the nation’s leading producer of rigorous national- and regional birth and marriage forecasts. I’m also a Research Fellow at the Institute for Family Studies, a Senior Contributor at The Federalist, and an I write periodically for Vox’s Big Idea column. I’m a native of Wilmore, Kentucky, a graduate of Transylvania University, and also the George Washington University’s Elliott School. My real job is as an economist at USDA’s Foreign Agricultural Service, where I analyze and forecast cotton market conditions. I’m married to a kickass Kentucky woman named Ruth. I am not paid one penny by anybody for this blog post.
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!
Check out my Podcast about the history of American migration.
DISCLAIMER: 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.
