The Long-Run Impact of COVID-19 on the U.S. Population
The Long-Run Impact of COVID-19 on the U.S. Population
The COVID-19 spike in mortality is the pandemic’s most direct demographic consequence, but not the only one. Factoring in changes in fertility, disruptions to immigration, and indirect demographic spillovers, we estimate that the pandemic reduced the U.S. population 0.5 percent over the long term.
Key Points
-
The pandemic led to at least 1.4 million additional deaths. It also produced a decline in the number of births in 2020 with an offsetting rise in births in 2021 and sharp fall in the number of noncitizens coming to live the U.S from abroad.
-
Changes to the population such as deaths, births, and immigration or emigration have spillover effects on others in the population through changes to family structure, marriage, fertility, and migration. These spillovers are not captured in estimates of the pandemic’s immediate impact but are reflected in our estimates.
-
We estimate that about two thirds of the 0.5 percent long-term reduction in population is attributable to increased mortality. The other one third is attributable to the fall in immigration during the pandemic years.
The COVID-19 pandemic produced a dramatic rise in mortality, leading to 1.4 million additional deaths in 2020, 2021, and 2022. These deaths are the pandemic’s most immediate demographic consequence, but not the only one. The number of births declined sharply in 2020 and then rebounded in 2021, interrupting the long-term decline in birth rates. The pandemic also disrupted nearly every aspect of the U.S. immigration system, with lasting effects on the foreign-born population.
More broadly, the immediate effects of COVID-19 do not fully capture its long term consequences for the U.S. population. The deaths that would have been avoided in a world without COVID-19 have demographic implications beyond the decedents themselves, affecting anyone they have or would have interacted with. A single death may have many secondary effects reflected in changes in family structure, marriage, fertility, and migration. The same is true of immigrants who could not come as a result of the pandemic; immigrants who stayed when they would otherwise have emigrated; the births that didn’t occur; the births that did; and many others. Each individual change in the population has ripple effects that may manifest over decades.
In this brief, we present estimates of COVID-19’s long-run population impact that reflect this comprehensive view of the pandemic’s effects. Using PWBM’s detailed microsimulation model, we simulate how the population would have evolved if the pandemic had not occurred. To generate this counterfactual, we estimate the pandemic’s first order effects on mortality, fertility, and immigration. With these estimates, we use the microsimulation to produce a comprehensive estimate that accounts for interactions across the population.
To measure the full impact of COVID-19 on mortality, we estimate “excess” deaths associated with the pandemic. Excess all-cause mortality is a more complete measure of the pandemic’s effects than reported deaths attributed to COVID-19. It describes the deviation of actual all-cause mortality from expected all-cause mortality, which is based on recent trends. Excess mortality captures COVID-19 deaths that were misattributed to a different cause of death (due to a lack of testing or inconsistent reporting practices) as well as deaths from other causes that are indirectly attributable to the pandemic.
As we discuss in more detail in a companion brief, we estimate monthly excess deaths by detailed socio-demographic group. We deem an observed count of monthly deaths to be in “excess” if the number of observed deaths is statistically different (in either direction) from expected deaths under pre-pandemic trends. Otherwise, we use the observed value as the expected number of deaths. We then define the difference between observed and expected deaths to be the number excess deaths.
Source: Penn Wharton Budget Model estimates from National Vital Statistics System and Census Bureau data.
Figure 1 plots our estimates of total monthly excess all-cause deaths from March 2020 through December 2022. Over the full period, we estimate nearly 1.4 million excess deaths occurred. Around 1.2 million of these deaths occurred in the first two years of the pandemic, with the first year (March 2020-February 2021) containing the greatest number of excess deaths.
To estimate the pandemic’s impact on births, we adopt the same approach as for mortality, described above and in a companion brief. We estimate monthly excess births, defined as statistically significant differences from expected births given recent fertility patterns.
Source: Penn Wharton Budget Model estimates from National Vital Statistics System and Census Bureau data.
Figure 2 shows cumulative monthly excess births during this period. Births began to drop consistently below expectations in August of 2020 and then plummeted in late 2020 to early 2021, bottoming out with 21,000 fewer births than expected in February 2021. However, from April 2021 onwards, the number of births leapt back up and maintained or outpaced expectations for most of the year.
In total, negative excess births in 2020 and early 2021 reduced the number of births by 70,000 through February 2021, relative to expectations. The boom of positive excess births in 2021 reversed this decline, however, and by the end of 2021 had largely offset the previous year’s shortfall.
COVID-19 virtually shut down much of the U.S. immigration system in 2020 and 2021, and significant components of the immigration policy response to the pandemic remained in effect through 2023. As we discuss in a companion brief, the pandemic led to sharp fall in the number of noncitizens admitted to live in the U.S. on a permanent or temporary basis.
Figure 3 plots the number of resident nonimmigrants admitted to the U.S. year, which includes nonimmigrants who reside in the U.S. temporarily for purposes like employment or education and excludes non-resident visitors such as tourists. The number of nonimmigrants entering the U.S. declined more than 50 percent from 2019 to 2021, falling below 2 million.
Source: Penn Wharton Budget Model estimates based on data from the Office of Homeland Security Statistics.
Notes: The number of nonimmigrant individuals admitted each year is estimated based on annual nonimmigrant admissions and the number of times each nonimmigrant is admitted per year.
Resident nonimmigrants are temporary workers, students, exchange visitors, diplomats and other representatives, and their families. It does not include temporary visitors for business or pleasure or other classes of nonimmigrants who do not remain the U.S. long enough to be considered residents.
Years are fiscal years.
Resident nonimmigrants are authorized to live in the U.S. temporarily and most return to their home country within a few years of their arrival. However, a temporary nonimmigrant visa is often a first step toward permanent immigration, and many nonimmigrants transition to a different temporary status to remain in the U.S. after their initial visa is no longer valid (for example, an international student on an F1 visa transitioning to an H-1B visa).
The number of resident nonimmigrants entering the U.S. has rebounded since 2021 and in 2023 rose to nearly 4.2 million, more than 10 percent higher than the pre-pandemic peak of around 3.75 million in 2019. This rebound likely reflects some degree of pent-up arrivals from the pandemic years, but it is highly unlikely that the 2020-2022 shortfall of roughly 4 million nonimmigrants arrivals will ever be made up.
Some features of the U.S. immigration system helped mitigate the impact of disruptions to travel and to the government’s operations abroad. Figure 2 shows that the number of new lawful permanent residents (LPRs) arriving from abroad with family-sponsored green cards fell sharply in 2020 and 2021, as one of the most common paths to permanent immigration from outside the U.S. was closed off.
Source: Office of Homeland Security Statistics.
Note: The dashed line shows the linear trend from 2014 to 2019, extrapolated through 2023.
Years are fiscal years.
When the number of family-based green cards issued in a year is less than annual statutory limit, they are added to the limit on employment-based visas the following year in a process referred to as rollover. Figure 2 plots pre-pandemic trends as simple proxy for typical green card issuance, given statutory numerical limits. The decline in family-sponsored new arrivals in 2021 left about 141,000 green cards unused in that year. In 2022, employment-based new arrivals rose by 16,000 and adjustment of status rose by 125,000, totaling 141,000 additional employment-based LPRs. New employment-sponsored LPRs remained elevated in 2023, reflecting the continuing shortfall in family-based arrivals in 2022. Hence, one consequence of the pandemic was a temporary shift in the balance of permanent immigration away from family preferences and toward employment preferences.
See the companion brief for a discussion of immigration outside of lawful processes during the pandemic.
To estimate the pandemic’s long run population impact, we simulate the U.S. population beginning in 2020 under three counterfactual assumptions: 1) the excess mortality related to COVID-19 does not occur; 2) the excess births related to COVID-19 do not occur; 3) the disruptions and policy changes to the immigration system related to COVID-19 do not occur. We then compare that scenario with historical estimates and projections of the actual U.S. population.
Figure 3 plots the estimated percentage difference in the size of the U.S. population over time, broken down into contributions from mortality and immigration. Because the number of excess births was small (around 30,000 in 2020 and 2021) and positive and negative excess births largely offset each other over the period, their effect is too small to distinguish from normal random variation in the population from year to year, so no contribution is shown.
Source: Penn Wharton Budget Model
We estimate that the percentage reduction in the population due to COVID-19 grew from 0.13 percent in 2020 to a peak of -0.64 percent in 2023 as the pandemic persisted the effects of previous years’ losses began compounding. Over the following decade, it shrunk to less than half that, largely because the pandemic accelerated the deaths of many people who would otherwise have died in those years. We estimate a long-run effect of just over 0.5 percent on the U.S. population, about two thirds about which reflects the impact of COVID-19-related mortality (past and ongoing) and one third of which reflects the absence of immigrants who would have settled in the U.S. and their descendants.
This analysis was produced by Alex Arnon under the direction of the faculty director, Kent Smetters. Mariko Paulson prepared the brief for the website.