Since the 1970s, the U.S. Census Bureau has utilized administrative tax data from the Internal Revenue Service (IRS) to improve the precision of migration and demographic estimates. Currently, this partnership is evolving, as the IRS is expanding the variety of tax information it provides for research and statistical studies.
For this analysis, we apply a similar approach at the local level by utilizing D.C.’s income tax records to analyze population and migration trends during and after the COVID-19 pandemic. This approach aims to identify additional data patterns and insights that may be missed in annual survey-based datasets. Our findings show that D.C.’s income tax records add valuable context to Census estimates and pinpoint pandemic-related migration shifts with even greater precision than the American Community Survey (ACS).
Population and Tax Filer Statistics
According to the U.S. Census Bureau, between 2019 and 2021, during the COVID-19 pandemic, the District lost 17,683 residents, a 2.6 percent decline in the District’s population (Figure 1). From 2021 to 2024, the city regained 21,673 residents (an average of 7,224 per year). The most recent census estimates indicate the population exceeded its pre-pandemic level in 2024. By comparison, the District's individual income tax data show that between 2019 and 2022, DC lost 9,953 income tax filers (2.7 percent) and regained 2,743 in 2023, the most recent year for which data are available (Figure 2). As of 2023, DC’s total number of tax filers remained 7,210 (1.9 percent) below its 2019 level.
The number of city income tax filers is lower than the total population because many residents are not required to file, such as children, other dependents, low-income and no-income earners, some college students, and people living out of state. Despite the differences in general trends between the overall population and the number of tax filers, tax data still offer valuable insights into demographic and migratory changes during the pandemic.
Against this backdrop, it’s also important to note that the number of tax filers also declined between 2019 and 2022, due to several particular factors. These include pandemic-related job losses, IRS processing delays caused by office closures and deadline extensions, and expanded federal tax credits, which reduced the need for many low-income individuals to file. More specifically, the American Rescue Plan Act of 2021 temporarily increased the Child Tax Credit and the Earned Income Tax Credit and issued stimulus payments, effectively eliminating tax liability for many households.
Using Income Tax Data to Measure DC Migration Patterns
Of the hundreds of thousands of residents who file city income taxes each year, part-year filers represent a key group for analyzing migration trends. By law, part-year filers must report their first or last day of residency in the city on their tax forms. This information is valuable because it reveals the timing and demographics of residents moving into or out of the city.
In this analysis, “in-migrants” refer to new tax-filing residents who reported their first month and day of residency on their tax form in a given year and were not on the city’s tax rolls in the two previous years. “Out-migrants” refer to former tax-filing residents who reported their last month and day of residency on their tax form in a given year and had been full-year filers in the two prior tax years.
DC Migration Demographics
Figure 3 illustrates that the average number of in-migrants in 2020 and 2021 was only 22,487, the lowest recorded among all the years presented. This signifies that these two years experienced not only a sharp increase in out-migration but also a noticeable in-migration decline. Figure 4 shows an unusual surge of 8,205 in-migrants in 2021, but these likely included individuals who delayed moving in 2020 due to the pandemic. Taken together, these figures show that the net population decline was driven not only by increased out-migration but also reduced in-migration.
Figures 5 and 6 indicate that over 65 percent of annual migration occurs during the second and third quarters of the year, with the third quarter representing the largest portion. Additionally, the data suggests that the pandemic impacted the timing of both in-migration and out-migration in 2020. Specifically, there was a significant decline in both types of migration during the second quarter of 2020, which coincided with the onset of the pandemic.
The notable drop in total migration during the second quarter was likely, in part, the immediate response to the first reported COVID-19 cases in the U.S., as announced by the Centers for Disease Control and Prevention, along with the mandatory “stay-at-home” order issued by the District of Columbia government on April 1, 2020. This order required all residents to stay at home except for essential activities, which may have inadvertently kept some potential out-migrants in the city long enough to meet the 183-day threshold for statutory tax residency. The figures also show that while out-migration rebounded in the third quarter of 2020, in-migration continued to lag behind.
Despite the significant fluctuations in migration levels during the pandemic years, other demographic trends among in-migrants and out-migrants remained largely consistent. Over 75 percent of migrants were single filers (Figure 7), with women making up the majority (Figure 8). In-migrants typically had starting incomes below $50,000 (Figure 9), but many who left the city reported earnings between $50,000 and $100,000 (Figure 10).
Most in-migrants arrive in the city around age 23, while those leaving are typically between 25 and 27 (Figure 11). But during the pandemic, this pattern coincided with a generational shift among residents. As Figures 12 and 13 illustrate, Generation Z, individuals born after 1997, is emerging as a key segment of the city’s working population (see Part 1 of this blog series).
Conclusion
The U.S. Census Bureau uses IRS tax data to improve population estimates and to gain insights into migration and demographic shifts across counties. Following this approach, this study examined D.C. population data in conjunction with city income tax records to better understand migration trends during the COVID-19 pandemic. By examining over 40,000 annual tax filer records for residents who officially migrated into and out of the city, the analysis found that the city's sharp population decline between 2019 and 2021 was driven not only by a significant increase in out-migration but also by a substantial decrease in the average number of individuals moving into the city during that period
Tax data also suggest that the early guidance from the Centers for Disease Control and Prevention, along with the Mayor's “Stay-at-Home” order issued on April 1, 2020, appeared to have prevented some residents from leaving the city during the second quarter of 2020. Such retention could have affected the tax residency of some potential out-migrants because of COVID-19, given that anyone residing in the city for at least 183 days is subject to D.C. income tax. Additionally, the tax data suggest that Generation Z may now represent the largest share of new residents moving into the city each year.
What is this data?
This analysis is based on population statistics from the U.S. Census Bureau and income tax data from the DC Office of Tax and Revenue.
Note: The U.S. Census Bureau’s Population Estimates Program (PEP) produces official annual population estimates for the U.S. and its subdivisions between decennial censuses. Currently, this model-based program uses administrative data from numerous federal and state agencies to inform its models, including IRS individual income tax data, which plays a crucial role in measuring inter-county migration and producing statistics on income and poverty.
The American Community Survey (ACS) is an ongoing U.S. Census Bureau program that provides detailed demographic, social, economic, and housing data annually. It surveys over 3 million households nationwide, offering updated information on more than 40 topics like income, employment, education, and housing for communities from sample based surveys. Unlike the 10-year census, the ACS is conducted continuously every year. The Internal Revenue Service (IRS) is currently expanding the range of tax data shared with the U.S. Census Bureau for use in research and statistical studies.