U.S. flag

An official website of the United States government

Skip Header

New Data Reveal Continued Outmigration From Some Larger Combined Statistical Areas and Counties

Written by:

Over two thirds (73.1%) of U.S. counties had more deaths than births from July 1, 2020, to June 30, 2021, up from 55.5% during the same period a year earlier.

When there are more people dying than babies being born, this is known as “natural decrease.” While the newly released data may suggest widespread population declines, this was not the case.

On average, smaller counties tended to have increases in net domestic migration this past year, while larger counties (those with populations of 500,000 or more) tended to have decreases in net domestic migration.

In fact, estimates released today by the U.S. Census Bureau reveal that 58.0% of counties experienced population growth, often because of net domestic migration, which means in many instances enough people moved into counties to offset natural decrease.

Domestic migration is the movement of populations that occurs during specified periods of time within the United States.

Between the 2019-2020 and 2020-2021 migration periods, many counties witnessed changes in net domestic migration, with a few populous counties experiencing higher net domestic outmigration (more people leaving than arriving) and many others seeing higher net domestic in-migration (more people moving in than leaving).

Given that the 2020-2021 migration period falls entirely within the COVID-19 pandemic, it is likely that these changes are related to the pandemic’s impact on domestic migration preferences. 

How We Measure Domestic Migration

Every year, the Census Bureau’s Population Estimates Program (PEP) measures population change since the date of the last decennial census using administrative records and other data.

Population changes for a given county result from a combination of natural change (the difference between births and deaths) as well as international and domestic migration.

The net domestic migration for a given geographic area is the difference between in-migration (the number of people moving in) and outmigration (the number of people moving out) during a migration period. We typically measure migration from July 1 of one year through June 30 of the next (e.g., July 1, 2020, to June 30, 2021, represents the 2020-2021 migration period).

Positive net migration or net in-migration indicates more individuals moved in than out over the period. Negative net migration or net outmigration represents the opposite — more individuals moving out than in over the period.

The domestic migration measure represents a net — the difference between in-migration and outmigration. That means an increase in net migration over time can occur because more people are moving in, fewer people are moving out, or a combination of the two.

A decrease in net migration is the opposite: It can be due to increased outmigration, decreased in-migration or a combination.

This shapes the types of conclusions we can draw about the data and trends over time.

Relatively Modest Changes in Regional Net Domestic Migration

Net domestic migration trends remained broadly similar between the 2019-2020 and 2020-2021 periods in all regions except the West, which shifted from slight net domestic in-migration to outmigration (Table 1).

The South’s net domestic migration from July 1, 2019, to June 30, 2020, already the largest among the regions, increased by another 154,000 in the 2020-2021 period.


Combined Statistical Area Migration

Table 2 shows the top 15 Combined Statistical Areas (CSAs) with increases in net domestic migration between 2019-2020 and 2020-2021. A CSA is defined by the Office of Management and Budget as an area that consists of various combinations of adjacent metropolitan and/or micropolitan areas with economic ties measured by commuting patterns.

Of the top 15 CSAs with increases in net domestic migration, seven were in the South and four in the West, while the Northeast and Midwest each had two CSAs.



As noted previously, some of these changes in net domestic migration may be due to fewer people moving into an area rather than to a large increase in the number leaving, contributing to overall higher net outmigration.

The combined magnitude of change for the 15 CSAs with the largest net domestic outmigration between 2019-2020 and 2020-2021 is much larger than for those gaining in net domestic in-migration, suggesting that increased domestic outmigration from some areas is translating into net domestic gains for many others.

New York-Newark, NY-NJ-CT-PA tops the list of the 15 CSAs with the largest decreases in net domestic migration between 2019-2020 and 2020-2021, with a reduction of about 144,600.

San Jose-San Francisco-Oakland, CA came in second with a decrease of over 94,200 in net domestic migration, followed by Los Angeles-Long Beach, CA.

Phoenix-Mesa, AZ, Las-Vegas-Henderson, NV, and Charlotte-Concord, NC-SC were also on the list of largest decreases but showed continued positive net domestic migration which, as previously noted, could occur because fewer people moved in, more people moved out, or a combination of the two (Table 3).


Net Domestic Migration Patterns by Size of County

The share of counties that experienced positive net domestic migration increased from under half (45.8%) during the 2019-2020 period to nearly two thirds (65.6%) in the 2020-2021 period.

While nearly one out of four counties (24.5%) switched from net domestic outmigration to in-migration during the 2020-2021 period, only about 1 in 20 (5.1%) went from net domestic in-migration to outmigration.  

As shown in Table 4, on average, smaller counties tended to have increases in net domestic migration this past year, while larger counties (those with populations of 500,000 or more) tended to have decreases in net domestic migration. 


Changing County-Level Spatial Patterns of Net Domestic Migration

The maps below illustrate the changing patterns of net domestic migration for counties between the 2019-2020 and 2020-2021 migration periods.

A number of areas had increases in the rate of net domestic migration in 2020-2021 (Figure 2) compared to 2019-2020 migration period (Figure 1), showing that more people per 1,000 population entered a county than left.

These areas include much of northern New England, parts of the Catskills in New York and Poconos in Pennsylvania, the Ozarks region of Missouri and Arkansas, northern Michigan and Wisconsin, and much of the interior West.

Some of these areas may have attracted more domestic migration during the pandemic because of their higher numbers of seasonal housing units. Again, it’s important to keep in mind that these increases in net migration rates may reflect more people moving into a county, fewer people leaving the county, or both.

Other counties that saw large increases in net domestic migration are outer counties of metro areas. The Columbus, Ohio, and Indianapolis, Indiana, metro areas are good examples (Figure 2).

There were also clear differences in the patterns of the rates.

In the 2019-2020 period, there was a higher prevalence of negative net domestic migration (outmigration) counties. The prevalence of counties with rates of in-migration over 10.0 migrants per 1,000 people was more apparent in 2020-2021 than in 2019-2020.

Overall, the pattern of high in-migration rates was widespread across the country and particularly strong in the Midwest, Northeast and the South.



It remains to be seen whether these emerging patterns of internal migration will continue or change direction beyond the pandemic.

How We Calculate Net Domestic Migration

PEP calculates overall net rates of movement between counties using several data sources (Internal Revenue Service, Medicare and Social Security).

Those rates are then applied to the household population and added to the group quarters population to create domestic migration values.

Finally, these domestic migration estimates are controlled to sum to zero at the national level. For a full detailed description, see Methodology for the United States Population Estimates: Vintage 2021.

Data Limitations

The Vintage 2021 estimates of net domestic migration for July 1, 2020, released today represent only a quarter of a year (April 1, 2020 – June 30, 2020).

For this story, the July 1, 2020, values from Vintage 2020 estimates series (representing July 1, 2019–June 30, 2020) were used for the computation of differences in net domestic migration over time.

While this seems to compare two consistent time series, the input data are slightly different for each. As a result, some of the differences in total net domestic migration between the 2019-2020 and 2020-2021 migration periods could partially be attributed to methodological differences. For more details, visit Vintage 2021 Release Notes.


Amel Toukabri is chief of the Local Government Estimates and Migration Processing Branch in the Census Bureau’s Population Division.

Crystal Delbé, Esther Miller and Basak Ozgenc are demographers/statisticians in the Local Government Estimates and Migration Processing Branch.


This article was filed under:

Page Last Revised - April 13, 2022
Is this page helpful?
Thumbs Up Image Yes Thumbs Down Image No
255 characters maximum 255 characters maximum reached
Thank you for your feedback.
Comments or suggestions?


Back to Header