Skip to Main Content
Speaking of the Economy
Lots of cars commuting between communities
Speaking of the Economy
March 15, 2023

The Daily Commute: Connections Between Communities

Audience: General Public

Santiago Pinto discusses his research on commuting patterns in the Fifth District and how these patterns may help us better understand the economic connections between rural and urban communities. Pinto is a senior economist and policy advisor at the Federal Reserve Bank of Richmond.



Tim Sablik: Hello, I'm Tim Sablik, a senior economics writer at the Richmond Fed. My guest today is Santiago Pinto, a senior economist and policy advisor at the Richmond Fed. Santiago, welcome back to the show.

Santiago Pinto: Hi, Tim. It's a pleasure to be back here and talk to you about our ongoing research.

Sablik: Today, we're going to be discussing a recent Economic Brief that you wrote with Sierra Latham on commuting patterns and economic connectivity in the Fifth District. We'll include a link to that Economic Brief in the show notes.

To start, how important are physical connections between communities and job markets?

Pinto: One important factor that has long been recognized to improve economic opportunities for residents in smaller urban and rural areas is their ability to interact with larger regional markets. It is well known that densely populated areas tend to offer broader ranges of local labor opportunities. They can also provide functions such as health services, education services and transportation more efficiently than places where the population is more sparsely distributed.

These larger areas are in a much better position to exploit what we typically refer to as agglomeration economies. This is a concept that economists use to refer to the benefits of the spatial concentration of activities and people in certain areas. To the extent that residents that live in smaller urban areas or rural communities can interact with these other larger areas, they can really benefit not only from access to larger labor market opportunities, but also to a wider range of local goods and services that they couldn't have otherwise received.

These benefits don't necessarily go in one direction. Larger urban areas can also benefit from their interactions with smaller surrounding areas. These urban areas become even more productive, and they can increase even further their opportunities once they get integrated with these other areas.

Sablik: Hm-hmm. I imagine when it comes to trying to measure this kind of activity, that's probably a little bit tricky. How do researchers go about trying to gather data on this topic?

Pinto: This is where we really get to the idea to use commuting flows to measure this kind of relationship between communities.

In principle, the notion of connectivity tries to capture the levels of economic cohesion or integration between different geographic areas within a region. By studying commuting patterns, we can have an idea of the economic linkages between the regions. Our underlying assumption is that these flows contain information on how labor markets spread across geographies and provide employment opportunity to residents in neighboring jurisdictions. When we see workers commuting to a location for work, that also indicates that regular travel to those locations — and for other reasons such as commerce or recreation — is feasible.

Overall, I would say that these commuting patterns reflect two basic things. On one hand, they show the degree of economic interactions of connectivity across locations. But on the other hand, they signal the ability of these communities to share the benefits of belonging to a much larger regional economy.

Sablik: Turning to what you found in your Economic brief, how connected are rural and urban counties in the Fifth District?

Pinto: That's definitely the question to ask. But before getting into it, let me briefly tell you what we do in our analysis. What we do is we look at the commuting patterns using the 2019 Longitudinal Employer-Household Dynamics data, and this is for all counties in the Fifth District.

Then, we consider three different types of commuting flows, of commuting behavior. One is the intercounty commuting flow — this is the commuting that actually takes place within a county. The outflows — these are the workers that commute from their county of residence to another county. And then the inflows are workers that commute into a county from other places. We compare those flows by calculating what we call outflows or inflows rate.

Putting all this together, what we find is that the median outflow rate in the Fifth District is about 69 percent and the median inflow is about 56 percent, which [are] pretty high values relative to other places in the country.

When we look at inflows and outflows, there are a lot of differences across the states within the Fifth District. For instance, Washington, D.C., as you may expect, has very low outflows, indicating that most residents of the city work in the city. But then it has a lot of inflows, really high inflows from the surrounding areas. The median county outflows in the remainder of the Fifth District range from 60 percent in Maryland to 74 percent in Virginia. Virginia is a place where you see a lot of people commuting to work to other places.

Moreover, the percentages differ across urban and rural communities, which is an important aspect that we are studying in our research. Overall, outflows are 72 percent in urban areas versus 65 percent in rural areas. And, inflows are 60 percent in urban areas and 53 percent in rural areas. This means that rural residents, especially compared to urban residents, more often live and work in the same county, and are therefore less connected to other counties.

Sablik: You mentioned that commuting flows vary across the [Fifth] District, but are there some common patterns that you see in the data?

Pinto: What we do next is we take those commuting patterns and we classify counties according to four different categories.

We consider a group of "connected counties." This is a group that includes counties that have very high inflows and high outflows at the same time. These are potentially counties that benefit from increasing productive capacity and integrating their labor markets with those areas.

We then have a second category that we call "bedroom communities." Bedroom communities are places where you see high outflows of residents leaving their country of residence to work somewhere else. They have, at the same time, low inflows, not many people commuting to those communities.

Then we have a third category that we call "employment centers," counties that have high levels of inflows but low levels of outflows. Finally, we have this category that we call "mostly intracounty commuters" and these are counties whose inflows and outflows of people are very, very low. This is a very interesting category. It might be hiding a lot of behavior inside this category. So, you can have counties here that are at a disadvantage because they are not interacting with any other places. But also, these can be other counties that are self-sufficient. They don't need anyone else.

What we observe is urban counties are more likely to be connected or employment centers. Rural counties are more likely to be the intra-county commuter or the bedroom communities. The level of interaction of rural counties is actually much lower than urban counties, reinforcing the results that I was mentioning before.

Sablik: Was there anything unexpected that surprised you when you looked at the commuting data for the [Fifth] District?

Pinto: When we examined closely the commuting flows, many rural areas are just as connected as some other urban counties. Rural areas are not completely the same. They look different and they face different restrictions and constraints. This difference within the rural counties is extremely important from a policy standpoint.

We also know that, for the Fifth District, residents in rural areas are more likely to commute intracounty or to another rural county. In the case of urban areas, about two thirds of residents tend to commute to a different urban area.

We also noticed that the behavior across states within the Fifth District is far from uniform. For instance, residents in rural counties in Maryland, central Virginia, and central North and South Carolina are more likely to commute to urban counties. But residents in rural West Virginia, western Virginia and eastern North Carolina counties are more likely to commute to other rural counties.

Sablik: Do highly connected counties have different economic characteristics than less connected ones?

Pinto: We did some very preliminary analysis, and we found that counties in each of the four groups share some interesting characteristics.

But before getting into the findings of that analysis, let me emphasize that at this stage of our analysis, we are simply characterizing or describing those counties. We do not intend with this to establish any sense of causality.

First, if you look at labor force participation, this is highest in communities categorized as employment centers. This is true for both urban and rural counties. Bedroom communities, on the other hand, have the lowest labor force participation rate.

The limited connectivity group has the highest unemployment rate. These are communities or counties that do not interact quite a lot with the rest of counties in the region. For rural areas, the unemployment rate is highest in bedroom communities.

And, we look at median household income. We see that that tends to be higher in both urban and rural areas for communities in the connected group.

Finally, the poverty rates tend to be lower for connected communities in both rural and urban areas.

Sablik: Do we have any sense yet about whether commuting patterns and connections between communities and job markets have changed since the pandemic?

Pinto: That's a good question. Let me make an important caveat of our analysis. We are using pre-pandemic data and commuting flows.

One thing that we know that happens after the pandemic is that the connection between the place of residence and the place of work has become a lot weaker. This is mind blowing for anyone that studies spatial economics because we usually think about the trade-offs of living farther from your work and housing prices.

And, we know that during the pandemic, household, businesses and real estate demand in most large U.S. cities have shifted from high-density to low-density suburbs. Some people responded by just fleeing the metro area altogether, but a lot of those households have moved further out in the existing metropolitan area.

This is what is known as the "donut effect," where the center has been emptied and everybody commutes or lives in the surrounding area. In our District, this has been an important phenomenon in the D.C. area. A lot of people left the D.C. area. However, when you zoom out further, you notice that a large number of households have moved mainly to neighboring counties in Maryland and Virginia and they have not left the MSA.

To the extent that we think that COVID has permanently shifted people's commuting patterns, then it will be necessary to incorporate other information to understand the degree of economic interaction between residents in different communities.

Sablik: That brings up another question about whether there are any policy implications that arise from this research.

Pinto: The analysis we conduct in our research can help policymakers identify regions that could benefit from policies that would improve accessibility to a broader set of economic prospects. And, it would inform these policymakers on how to effectively target place-based policies. It's different to target a policy to a rural area, for instance, that is very well connected to surrounding areas than a rural area that is completely isolated. The policies are completely different.

Moreover, the research also underscores the importance of strong communication networks. These are critical factors that allow households residing in distant locations to take advantage of economic opportunities typically available in more densely populated urban areas.

Finally, when resources are targeted to regions with high levels of connectivity, we expect the policy to have much larger spillover effects and benefit not only the target area, but also those living in the connected area. So, when you want to evaluate the effectiveness of the regional policy and the associated spillover effects, then the degree of local connectivity becomes a crucial factor.

Sablik: Are you hoping to expand your research on this topic in any area?

Pinto: In the very short run, we're already working on trying to understand better what determines and characterizes communities in each of the four groups that we mentioned before, especially those communities that are the intracounty commuters. Remember, these are a group of counties [where] the inflows and outflows of people is very, very low. The other categories, bedroom and employment counties, are more straightforward to understand and explain. For counties with large intracounty commuting flows, it is crucial to determine if these are self-sufficient counties or are more completely isolated.

For connected counties, we also intend to examine another very interesting commuting pattern that has to do with cross-commuting, people from one county commuting to another county. Why do we observe this kind of commuting behavior?

Overall, the present research is part of a much broader research agenda where we attempt to understand where people reside and work and, basically, the role of agglomeration economies. While there are benefits of agglomeration economies, including higher wages and employment growth, I think with distance from the urban core, it is likely that this attenuation will depend on the degree of regional interconnectivity. So, by quantifying these effects, we will be able to understand how certain small areas and rural communities may benefit from their connection to larger regional economies. This topic becomes even more relevant nowadays with the growing ability of individuals to work from home.

Sablik: Santiago, thank you so much for coming on the show to talk about this research.

Pinto: You're welcome.

Sablik: And as always, listeners can head over to the show page for a link to the paper we discussed today. And if you enjoyed this episode, please consider leaving us a rating and review on your podcast app.