Updates on a Few City Populations
Grading the Forecast Is Fun!
It’s been a while since I blogged! I’ve been doing lots of writing elsewhere, so the blog has taken a backseat!
But I wanted to take a second to check in on a few cities I’ve written about recent and see how recent population estimates impacted them. Basically, I want to assess how my takes have held up over time. I’ll score on the following basis:
- Did any numeric claims or expectations continue to remain essentially true?
- Did any new data fall within any forecast ranges provided?
- Does the essential core claim being made in the original post still hold with new data?
Is Pikeville Dying?
Let’s start with one of my more widely-read takes: that Pikeville, KYis a uniquely strong performer in Appalachia. Specifically, I want to update this chart, and see how it looks:
So what’s happened since?
Well, new data, plus a few small methodology changes that I think are better for this, we get the following estimates:
First of all, municipal Pikeville’s population fell sharply. However, as I explained in the original post, municipal populations for small municipalities like Pikeville can be very misleading due to annexations: and Pikeville has made many annexations in its history.
The colored lines are more accurate, however, and they reflect time-consistent geographic units corresponding to Census tracts. What stands out here is that Inner Pikeville, the core of the town, had growth, but that growth now appears to be stalling out. The Pikeville suburbs also saw robust growth, but that growth has turned flat in the most recent data.
So is Pikeville’s overperformance a sham? Not so fast! Let’s compare Pikeville to Pike County, for the current estimates vs. my old estimates.
What we can see is that my current estimates suggest Pikeville’s growth occurred somewhat earlier in the 2000s than previously suggested, and that growth has now basically stopped, but it remains much larger than in 1990. But Pikeville’s growth has stalled even as the rest of Pike County has seen an accelerating pace of decline.
Here’s growth rates in Greater Pikeville minus the growth rate in the rest of Pike County.
Pikeville’s growth-overperformance is noisy, but pretty reliably positive. In other words, Pikeville’s exceptional performance is continuing, it’s just that the baseline against which it is exceptional is getting worse and worse.
So let’s score this.
Were the basic numbers I provided right with new data? No, they weren’t. The new data shows a substantial downward turn in Pikeville’s municipal population which I did not expect, and a flatter trend in the Pikeville area than I’d previously expected. So that’s 0/1.
Did any new data fall in previous forecast ranges? I didn’t show a graph on this one, but the answer is “no.” My forecast range was too high, though this was partly because of a formula error I just found while reviewing this. But even adjusting for that, my range was just a bit too narrow. That’s what I get for using just 1 standard error as a range I suppose! So that’s 0/2
But do these objective errors actually change the core conclusion that Pikeville is reliably and significantly overperforming the rest of its region, and that this advantage is continuing or increasing? Turns out, these errors don’t change the conclusion! It still holds! So that’s 1/3. But the one I got right is the most important one.
So I score my Pikeville-Exceptionalism take as “True in general, but some of the details were wrong.”
Is Cincinnati Making a Comeback?
And Is It Beating Pittsburgh?
I wrote recently that Cincinnati has an extremely impressive record compared to other Rust Belt cities, and that it is a far better candidate than Pittsburgh for “Rust Belt Revivalism.”
But was I right? At the time, we didn’t have complete 2017 numbers. Let’s see how they shake out. In particular, I want to see if my estimates of the municipal areas were correct. So here’s the chart I’m updating:
And here’s what it looks like with actual 2017 data in:
Ah, whoops, alas, the changes are too small to see on this scale! Let’s zoom in.
Do you see what I see? Because what I see is that vintage 2017 showed very similar trends to what I predicted and in fact, if anything, Pittsburgh’s population is a bit lower than I expected, and Cincinnati’s a bit higher. In other words, Cincinnati is indeed making a demographic comeback, Pittsburgh isn’t, and Cincinnati will surpass Pittsburgh before the next Census, possibly even in 2018.
How do I score my take? Well, I basically got the numbers right within a reasonable range of statistical error, I basically got the forecast right, and the core conclusion is if anything even more robust than I suggested. Ergo, 3/3. This take can be ruled “Completely correct, and Lyman is awesome.”
Rolling total thus far, 4/6, but, more broadly, 2 basically correct takes, zero basically wrong takes.
Now I have a third city that I assessed and deserves looking at.
Is Hartford Struggling?
I wrote a while back about the demographic and financial troubles of Hartford, CT.
Basically, my take was that Hartford’s financial troubles aren’t only due to local policy idiosyncracies, but are also due to durable demographic weakness which is unlikely to change rapidly.
So how do the 2017 estimates stack up? Well, I won’t show a graph, but basically, Census raised their estimate of Hartford’s population in every year back to 2010, but they didn’t change the trajectory of population, which is fairly steeply downwards. Their revision only amounted to adding about 0.3% to the population, vs. a 0.4% decline in just 2016–2017. Notably, virtually all of Connecticut got an upward revision in population. As I’ve discussed elsewhere, this was due to changes in how the Census Bureau estimated the foreign-born population in the 2017 vintage.
Overall, the revisions were small and didn’t change any of the relationships I discussed (1/1), the direction of change in 2017 was as predicted (down) and in a similar magnitude as recent years (2/2), and so none of my core conclusions about long-run bearishness on Hartford are changed (3/3). In other words, Hartford still faces serious demographic difficulties.
That makes me 7/9 on my takes. I’ll take it!
Pikeville is still performing remarkably well given the headwinds it faces. Cincinnati is making a serious population comeback, even as much-heralded Rust-Belt-turnaround Pittsburgh languishes. Hartford’s decline is continuing. All my city-specific takes last year were basically correct. I also had a post about Lexington, KY: Lexington’s growth is continuing apace, and furthermore, it got a small upward revision in its back-years. I also wrote about Huntington recently, following up on a bigger piece about the Huntington-Ashland-Ironton Tri-State Metro Area. In both pieces, I’ve argued that these areas may need to make peace with decline, as decline is likely to continue. Lo and behold, Huntington’s population was revised down appreciably in 2017, and decline is continuing apace. Ashland’s population was also revised downwards, and is showing a very rough 2017. The same story goes for Ironton. In other words, pretty much every single one of my local-area takes, including some outside-the-conventional-wisdom takes on Pikeville and Cincinnati, have turned out as expected. It’s been a good year or two of forecasting!
I’m 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 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.