Roanoke is losing population two ways. What that means for the Evans Spring debate.

Want to read more about Virginia’s changing demographics? We’ve collected all our coverage in one place.

Spoiler alert if you haven’t seen the classic 1942 film “Casablanca”!

After Humphrey Bogart’s character shoots and kills a German major, the French captain who witnessed the murder diverts police by telling them: “Round up the usual suspects.” That wasn’t the first use of that phrase — it apparently originated in the world of New York police and criminals in the 1930s — but it was the usage that popularized the phrase.

It’s also a phrase that comes to mind when I rummage through the latest population estimates from the Weldon Cooper Center for Public Service at the University of Virginia. One way to understand what’s going on beneath the surface of population gains and losses is to look at the two ways localities are either gaining or losing: Do they have more births than deaths or the other way around? Do they have more people moving in than moving out or the other way around?

There are 19 localities that lost population both ways — “double losers,” I call them, not to be pejorative, just accurate. Of those 19 localities, 17 are what I’d call the usual suspects — mostly rural communities in Southwest and Southside that have been losing population for a long time. I’ll have more to say about those double losers in a future column — teaser: some of them actually have some positive trends going down deep in the numbers. Today, I’ll turn my attention to the two outliers: Radford and Roanoke.

Radford is the easiest to explain. Virtually all of its population decline is due to out-migration. The city saw 727 more people move out than move in, while deaths outnumbered births by 23. While the data doesn’t say who those people moving out were or what their motivations are, it’s hard not to conclude that this is simply a function of Radford being a university town. Of course, people are going to be moving out of town. Even Charlottesville had net out-migration and nobody says Charlottesville is a market failure of a city. I’m inclined not to be alarmed by Radford’s figures.

That leaves Roanoke, which is more curious. Roanoke, which dropped to a population of 94,911 in the 2000 census, has been very proud of gaining (more accurately, regaining) population in the two decades that followed. In the 2010 census, the city was counted at 97,032; in the 2020 census, Roanoke moved back over the 100,000 mark for the first time in 40 years with just 11 to spare: 100,011.

However, the annual estimates since then have shown the city’s population declining and this latest round of data was no exception: Roanoke’s population is now put at 99,045, a drop of 966 people, or 1%, since the last census.

Why is this? Why has Roanoke’s growth suddenly stopped and gone in reverse?

Let’s dig a little deeper and see what we can find.

As noted above, Roanoke is one of those double losers — it’s got both more deaths than births (683) and more people moving out than moving in (283).

The good news for Roanoke is that it’s a pretty explainable configuration.

Most localities in the state have more deaths than births, a consequence of both an aging population and declining birth rates. As cities go, Roanoke is on the older side of things, with a median age of 38.1, as compared to 34.4 for Richmond and 28.6 for Lynchburg. Yes, Lynchburg has the demographic advantage of being a college town, but the point is, Roanoke’s population skews older — which naturally means more deaths and fewer babies.

While it would be better to have more births than deaths, Roanoke shouldn’t be surprised by having more hearses than baby carriages; the localities in the state that have more births than deaths are almost entirely in the urban crescent. In demographic terms, no locality can stop people from dying and, realistically, localities aren’t going to reverse declining birth rates, either. That’s not just a national phenomenon, it’s one that we see across much of the world. That means the only thing a locality can do to make up that deaths-over-births deficit is to attract more young adults, who then proceed to have babies. Hold that thought.

The more worrisome figures are the out-migration numbers, because those do tend to reflect whether a community is a market success or a market failure — people really do vote with their feet.

Now here’s the curious thing: Through last year, Roanoke had net in-migration — 67 more people moving in than moving out from 2000 to 2022. Something has happened in the past year to turn that 67-person surplus into a 283-person deficit. What would suddenly cause such an outflow?

This data doesn’t tell us, so we’re left to surmise some things we know from other data. 

First, maybe we shouldn’t hang too much on a single year’s dataset. When we look year by year at the data from the previous decade, we see some years with net in-migration and some years with net out-migration. However, for the previous decade, Roanoke wound up on the plus side, with net in-migration of 342. In the first three years of the current decade, the city’s new net out-migration has almost wiped that out, so perhaps we ought to pay some attention.

Second, let’s think of the life cycle of a city. Often young adults start out in an apartment in a city, then graduate, so to speak, to a home. We know from migration reports from the Internal Revenue Service that the most common destination of those moving out of Roanoke is Roanoke County. It’s entirely possible that what we’re seeing here is the natural progression of things: A young couple moves into the city, has a kid or two, and then moves out to a house in Roanoke County or beyond. If that’s the case, then the question we don’t know the answer to is whether that family moved out of the city because they didn’t like the city or because they couldn’t find suitable housing in the city, but did find it elsewhere. Hold that thought, too.

All we do know is this: Roanoke’s net out-migration from 2000 to 2023 is unusual when compared with other cities in the western part of the state. During that time, Martinsville had net in-migration of 226. Galax had net in-migration of 233. Salem had net in-migration of 246. Lynchburg had net in-migration of 293. Staunton had net in-migration of 434. Waynesboro had net in-migration of 580. Danville saw net in-migration of 815, a surge that puts it in the same neighborhood as Chesapeake, a city nearly six times its size. (I’ll look more closely at this in a future column.) In losing people through a net out-migration of 283, Roanoke is more akin to Hopewell, which had a net out-migration of 268.

Ultimately, I can’t tell you why Roanoke had this net out-migration; the data doesn’t speak to it. It seems odd to me because I see Roanoke as a quite appealing city. However, I can tell you what these numbers lead to. The Weldon Cooper Center projects that Roanoke will see its school enrollment fall by 4% over the next four years. That’s not the biggest decline in the state — some localities are in the double-digit range, with Buchanan County at -16% — but it is one of the bigger ones in the immediate area. Roanoke County’s enrollment is projected to fall by 2%, Botetourt County’s is projected to remain flat. Only Salem’s drop comes in larger, at -8%.

Now come the policy questions that Roanokers must deal with. First, are they OK with the city losing population again? If so, then there’s nothing they need to do — just keep in mind that a smaller population doesn’t necessarily mean fewer expenses, because some costs are fixed. It costs the same to keep Patrick Henry and William Fleming high schools open no matter how many students are in them, for instance. If Roanokers aren’t OK with losing population, then they need to decide what to do about that.

Demographically speaking, the city needs more young adults — that’s how you ultimately reverse declining school enrollment. The city also needs to figure out how to keep more people from moving out of the city than moving in. Is that a more robust economy? Is that more housing options, and maybe more affordable housing options? Is that something else? We could make a long list of possibilities.

I mention housing because that’s a problem we see cropping up in lots of communities, and it’s one that looming larger in Roanoke for one particular reason: the question of whether the 151-acre Evans Spring property should be developed. That’s the largest undeveloped tract of land in the city, and it’s in a prime developable location, beside Interstate 581 and across from Valley View Mall. It’s also next to a predominantly Black neighborhood with a long memory of being treated wrongly by the city, dating to urban renewal in the ’60s. We’ve seen a movement spring up to “Save Evans Spring” in the name of preserving woodland and green space. All this is complicated by the fact that the city doesn’t own the land in question; it’s divided into multiple tracts owned by multiple owners, all of whom already have the right to develop the property to some degree — they just haven’t.

It’s understandable why some believe Evans Spring should be developed: If you think the city needs new residents to reverse these population declines, then they need a place to live. If not here, where? The challenge is it’s not particularly popular to be in favor of cutting down lots of trees, especially when the city has declared this to be the Year of the Tree.

It’s also understandable why some want to leave Evans Spring untouched: You can’t really go out and create new green space. The challenge is that because the owners already have the right to engage in some development, if you really want to “Save Evans Spring,” then the city needs to buy the property and turn it into a nature preserve or park.

The fight over Evans Spring may seem political, or environmental, but it’s really demographics that are helping drive this — a declining population, and declining enrollment, and the desire to reverse both of those. If Evans Spring is spared, where should the city develop new housing? Or is that not something the city should worry about?

This is a good test of Roanoke’s priorities.