Projecting the long-term effects of the COVID-19 pandemic on U.S. population structure – Nature.com

Our study provides first results on how the COVID-19 pandemics reshaping of the U.S. population is expected to repercuss into the future. Despite the general perception that the COVID-19 pandemic mainly affected old populations, our projections show that population pyramids will exhibit consequences of the pandemic until at least 2060. These rippling effects are expected when modeling the consequences of COVID-19 pandemic-induced changes in all three processes: mortality, fertility, and migration. We highlight three of the most important results from our study.

First, among the three demographic processes, the loss of net migration during the COVID-19 pandemic years is expected to have the biggest long-term impact on the size of the U.S. population. In light of concerns about below-replacement fertility and baby boomer cohorts reaching retirement age, migration represents one important mechanism for slowing down population aging. The number of resettled people in the U.S. has been declining since 1980 but declined even more dramatically after the Trump administrations 2017 Executive Order titled Protecting the Nation from Foreign Terrorist Entry into the United States6,28. Then, after the enactment of Title 42 in March 2020, immigration and resettlement to the U.S. reached the lowest level of the past forty years. Title 42 was harmful for hundreds of thousands of people and ultimately resulted in the expulsion of over 1 million migrants and asylum seekers at the U.S. border, a decision that had no clear statistical relationship with reducing COVID-19 cases19,29. Our results show that the decline in migration resulted in the loss of U.S. population at all ages, but especially at working and reproductive ages. This result highlights that the COVID-19 pandemics effect on migration is more consequential for population size than its effect on mortality, a finding that is consistent with a similar study on Spain8. Government policy responses during crises can have profound effects on the population, through entirely different channels than their desired effect.

Second, in the next four decades there is projected to be fewer reproductive-aged (1549 years old) people in the U.S. This is a result of fewer migrants in childbearing ages, as well as, to a lesser extent, COVID-19 pandemic deaths and second-order implications of migration and mortality for never-born children. Our estimates are likely conservative, as the effects of long COVID, or the prevalence of COVID-19 symptoms long after infection, remain to be seen. Long COVID is similar to other post-acute infections in its ability to cause health complications and disabilities30,31. While less is known about its mortality consequences, it stands to reason that long COVID will be a future contributor to premature deaths.

Third, the high mortality rates of the older age population during the COVID-19 pandemic have led to a small reduction in the U.S. dependency ratio. The magnitude of this reduction is attenuated by missing migration, which by itself would likely increase the dependency ratio. In 2025, almost one half of the reduction in the dependency ratio due to mortality is projected to be balanced out due to missing migration. The balance between population health and national economic stability remains a point of discussion in the U.S.32,33,34. The economic stimulus for COVID-19 pandemic relief and public health policies were important for alleviating the individual economic burden brought on by the COVID-19 pandemic and for aiding in the reduction of COVID-19 cases and mortality, but also placed extraordinary fiscal burden on the U.S. Our dependency ratio projections provide indicators for how demographic changes brought about by the COVID-19 pandemic might continue to affect public finances in the long-term. It should also be noted that, while the dependency ratio is projected to remain slightly smaller as an effect of the COVID-19 pandemic, dependence on working-age individuals may increase due to higher healthcare needs among the older population following the COVID-19 pandemic. Additionally, we note that our calculations of dependency ratio are relatively simplistic. More nuanced calculations of dependency ratios (e.g., the non-working-aged dependency ratio) necessitate estimations of the number of working vs. non-working people at each age35,36,37, and this data are not available in projected form from the UNWPP.

Although the UNWPP data represent a gold standard in terms of population projections, our counterfactual analysis is subject to three limitations. First, our findings are based on UNWPPs medium scenario, i.e., not the most aggressive or the most conservative estimate. As the baseline mortality, fertility, and migration rates and counts represent forecasts themselves, they are subject to uncertainty, which is carried over to our counterfactual estimates. We attempt to mitigate this by focusing on the difference between baseline and counterfactual scenarios. Thus, because mortality, fertility, and migration conditions are set to equal after 2024, there is little room for forecasting errors to compound over time, as these will mostly cancel out. Moreover, the published UNWPP forecasts for the year 2022 correspond well with preliminary estimates of mortality, fertility, and migration4,6,18, generating further trust in our baseline and counterfactual estimates for the COVID-19 pandemic period. Additionally, due to the nature of counterfactual analyses, it is not possible to truly know what observed rates and counts would have been in the absence of the COVID-19 pandemic. While we estimate these to the best of our ability, all analyses must be considered with this limitation in mind.

Second, our finding that changes in migration during the COVID-19 pandemic are projected to exert the biggest long-term effects on population size may partially be driven by the lack of adequate age- and sex-specific migration counts for the U.S. and the application of model migration schedules38 for both the baseline and the counterfactual scenario. We assume a family migration schedule, with migrants concentrated in young and working ages. This also means that the second-order effects of migration through never-born children are particularly large in our study. Immigration to the U.S. has traditionally been concentrated in working ages39 and it is plausible that the largest declines in migration during the COVID-19 pandemic occurred in these age groups. Although it is entirely possible that migration decreased more in other age-groups, including ages older than reproductive ages, existing data on foreign-born immigration to the U.S. indicate that different types of migration (i.e., refugees/asylum seekers, students, work visas, immigrant visas) were similarly affected during the COVID-19 pandemic6. Moreover, the enactment of Title 42 during the COVID-19 pandemic contributed to declines in migration to the U.S. and targeted a broad range of countries19,20,21,22. Thus, our decision to use similar migration schedules for our baseline and counterfactual scenario appears justified. While we are limited by the lack of migration data at smaller temporal windows (e.g., month or week), future work with better data availability might consider analyzing this to gain a more nuanced understanding of how these processes vary across other temporal dimensions.

Third, following UNWPP, we assume that mortality, fertility, and migration return to their pre-COVID-19 pandemic trajectories after a few years. There is inconclusive evidence about what signals the end of a pandemic or epidemic40, so it is possible that the assumptions from UNWPP are incorrect. Should that be the case, and mortality continue to remain higher than expected, and fertility and/or migration continue to remain lower than expected, then our estimates represent an underestimation. The indirect consequences of the COVID-19 pandemic may continue to negatively affect the U.S. mortality, fertility, and migration environments well into the future, and we are not able to measure these indirect consequences here. First, long COVID and unmet healthcare needs during the COVID-19 pandemic may increase the risk of mortality in the long run. Other consequences of the COVID-19 pandemic, such as the loss of next of kin41, learning loss42, or racist and xenophobic behavior against Asians and Asian-Americans43,44 may also exert negative effects on population health and mortality for generations to come. Second, the experience of economic uncertainty and stress related to the balancing of work and childcare obligations during the COVID-19 pandemic may have raised doubts among some couples about having (additional) children in the future45,46. Finally, migration to the U.S. may remain below expected levels in the future, as some individuals who would have migrated to the U.S. may have died during the COVID-19 pandemic, or established families in their country of origin or other countries with less restrictive migration policies. Based on these reflections about the potential long arm of the COVID-19 pandemic, the findings presented in this manuscript, which assume a short pandemic shock, most likely represent a lower bound.

Despite these limitations, our approach is valuable because it considers the interacting effects of changes to population processes. The U.S. will face a variety of public health challenges in the coming years that may have long-lasting effects on the population size and structure, and the COVID-19 pandemic is just one of these challenges. The maternal health and midlife mortality crises are likely to affect the U.S. population through multiple avenues. Demographic predictions warn that a total abortion ban could lead to excess pregnancy-related deaths of nearly 25%47,48, while other work suggests that it may have consequences for in-vitro fertilization rates, contributing to a decline in number of births49. Additionally, if the midlife mortality crisis in the U.S. persists50,51, and if rising mortality rates from the opioid epidemic are not curtailed, then deaths among reproductive-aged people will continue to rise, resulting in fewer people at young adult and midlife ages. Applying the cohort component projection method to these crises will be valuable for understanding the magnitude of their consequences for the U.S. population. It will also be valuable to apply this approach to other countries (beyond Spain and Australia8,9), as the COVID-19 pandemic unequally affected each nation.

The consequences of the COVID-19 pandemic are not over. They ripple beyond immediate, independent changes to mortality, fertility, and migration to affect the population structure of the United States for decades to come. It is thus important to move from process-specific models to a broader and more informative approach that accounts for co-occurring disruptions in mortality, fertility, and migration. As this paper shows, such a design is a powerful tool for quantifying the relative size of different effects of the COVID-19 pandemic and for projecting their effects over time. Because the United States is known for having exceptionally high COVID-19 mortality52, it is important to note that COVID-19 pandemic-induced migration changes are projected to have a comparatively large and longer-lasting effect on population size.

Read the original here:

Projecting the long-term effects of the COVID-19 pandemic on U.S. population structure - Nature.com