ECONOMIC IMPACT OF THE 2020 CENSUS UNDERCOUNT IN TEXAS

In partnership with Dr. Manuel Reyes-Loya from the University of Texas at Tyler, this research brief estimates the economic impact of the 2020 Census undercount of Texas in terms of output, labor income, GDP, and jobs.

By: Dr. Francisco A. Castellanos-Sosa, Texas Census Institute, Senior Research Associate
Dr. Manuel Reyes-Loya, University of Texas at Tyler, Acting Director and Senior Research Analyst

Texas was one of six states with an undercount in the 2020 Census, with the second-largest statewide undercount (547,968 people were not in the census estimates, 1.92% of the Texas population).¹ In September of 2023, the Project on Government Oversight (POGO) estimated that Texas received $150.3B in Fiscal Year 2020, in funds distributed through federal programs that relied on census data to allocate resources geographically.² However, little is known about how the undercount impacted the statewide distribution of these funds and the overall impact on the Texas economy when using these recent updates.

To our knowledge, there is only one related economic impact analysis, published by The Perryman Group in June 2022.³ Due to the timing of its publication, we infer their analysis used an estimate of federal funds that is different from that of POGO (published in September 2023) or from a related exercise published by the U.S. Census Bureau in June 2023.^2,4

In this study, the Texas Census Institute estimates the economic impact of the 2020 Census undercount in terms of jobs, labor income, value added (gross domestic product), and output by industry. To have this level of disaggregation, we use the list of 338 federal programs obtained by POGO. In particular, we match these programs to their closest 2-digit industry classification from the North American Industry Classification System (NAICS).⁵ To estimate the economic impact of the 2020 Census undercount, we use the Input-Output (I-O) expanded model from IMPLAN.⁶

We undertook this study for two reasons. First, data analysts and researchers, particularly those with industry and sector knowledge, can use our estimates to provide a more detailed analysis of the economic impact of the 2020 Census undercount. Second, in the absence of updated information, our estimates for each of the 20 NAICS industries can inform policymakers and stakeholders about the potential gains for Texas if there is a complete count in the 2030 census.

This study uses three main data sources: POGO’s list of 338 federal programs that relied on census data to allocate $150.3B in resources to Texas in FY 2020, the 2020 Census Post-Enumeration Survey Results, and the 2-digit NAICS classification.^1,2,5

To establish a consistent analysis and consider it applicable to the years between decennial censuses, four COVID-related programs are excluded, reducing the total federal funds we analyze from $150.3B to $130.7B.

From the $130.7B, $116.9B are distributed to 167 federal programs related to the Health Care and Social Assistance industry. However, 6 programs (like Medicare, Medicaid, and SNAP, among others) absorb a relatively high share of federal funding, so we set them in a separate industry category to provide a clearer picture. For instance, $102.7B corresponds to Medicare Parts A, B, D, Medicaid, Medicaid Part D-Clawback, and SNAP. Collectively, this funding is 87.9% of the $116.9B of the total amount distributed to the  Health Care and Social Assistance industry. The remaining $14.1B represents just 12.1% and includes 161 programs (referred to as Health**). Figure 1 shows the allocation of federal funds to Texas in Fiscal Year 2020 for selected industries. The remaining 167 federal programs receive the other $13.8B that fall into 13 industries.

This study builds upon the work of Dudley Poston and Rogelio Saenz.⁷ In mid-2022, they estimated the federal funding losses for Texas—or direct economic impact—due to the 2020 Census undercounting by approximating the per capita amount each person should be entitled to and estimated a 10-year loss of $19B for Texas. However, more detailed and updated data on federal funding was recently published by the U.S. Census Bureau and POGO in 2023, making their estimate less than it should be.^2,4

Poston and Saenz’s per capita approach is limited because some programs do not serve the entire population. Still, it is equivalent to the scenario in which the undercount rate affects proportionally the federal funds received by states. While changes in the population might have larger or smaller effects than the undercount rate, this approach remains conservative by not making further assumptions.

Therefore, to remain conservative in our calculations, we estimate the direct economic impact for each of the twenty 2-digit NAICS industries as a 1.92% reduction of federal funds distributed to each (which would suggest the undercount, on average, impacts proportionally to all the industries). Then, the statewide direct economic impact is measured as the sum of all the industry-level direct economic impacts.

To estimate the total economic impact of the 2020 Census undercount, we use the Input-Output (I-O) expanded model from IMPLAN.⁶ It “…expands upon the traditional Input-Output approach to also include transactions between industries and institutions and between institutions themselves, thereby capturing all monetary market transactions in a given time period. IMPLAN can more accurately be described as a Social Account Matrix (SAM) model…”.⁸

As calculated by IMPLAN, the total economic impact is the sum of direct, indirect, and induced economic impacts. The direct effects are those we estimated first to observe the funding reduction by industry. The indirect effects are purchases cross-linked across industries in the supply chain that originated from the direct economic impact. The induced effects are the spillovers from employees’ labor income and spending behavior. The Total Economic Impact is estimated in terms of Output, Value Added (GDP), Labor Income, and Jobs.

IMPLAN illustrates how the impact in one industry affects others. So, by adding up how each industry was affected by others, we can estimate how the total economic impact in terms of Jobs, Labor Income, and Value Added (GDP) is distributed across 2-digit NAICS industries.

This estimation helps identify the industries that are more vulnerable to inaccurate census counts and more likely to benefit from an accurate 2030 Census.

The Health Care and Social Assistance industry had the largest direct economic impact across all variables under study. Approximately 90%, 92%, 91%, and 92% of the total impact on Output, Labor Income, Value Added (GDP), and Jobs were derived from this industry. These shares equal $46B in output, $21.6B in labor income, $27.1B in value added (GDP), and 352K+ jobs during a decade. The next three industries that could benefit the state economy from an accurate 2030 Census are Construction, Educational Services, and Manufacturing.

After the Health Care and Social Assistance industry, the three industries with the largest direct impact are Construction, Educational Services, and Manufacturing. Altogether, $204.9M annually is lost to Texas in federal programs related to these programs, with a potential $2.0B 10-year effect. The remaining 10 industries have a direct economic impact of $56.3M yearly and $563M over 10 years.

By studying how Texas was affected by the 2020 Census undercount, we estimate the potential benefits of a complete count in the 2030 Census. Without a 2030 undercount, Texas would have the following gains over the next decade:

One of the advantages of the Input-Output approach is that we can observe how changes in one industry affect others. Table A1 in the Appendix shows how the industries were potentially affected by the undercount, which could translate into gains if the 2030 Census is accurate.

Our research shed light on the direct and total economic impact of the 2020 Census undercount on the Texas economy. The implications are profound, impacting the allocation of federal funds across various sectors and regions within the state. The Texas Census Institute’s Funding Implications Series has undertaken a comprehensive analysis of the economic impact of the census, serving as a valuable resource for informed decision-making, advocacy, and action, especially as we approach the 2030 Census.

As noted above, Texas incurred a significant financial loss due to the 2020 Census undercount, resulting in a staggering $25.1B loss over the decade after the 2020 Census.

Considering the population projections of the Texas Demographic Center, Texas is expected to have 32,912,882 people in 2030. If action is not taken to reduce the undercount in Texas and it remains the same (at a 1.92% rate), Texas would miss 631,927 people in the next Census. If those residents were counted, Texas would access an additional $28.3B over a decade, resulting in $57B+ in output, $26B+ in labor income, $33B+ in GDP, and 434K+ jobs for the next decade.

Along with the funding implications, the census undercount also brings issue-specific implications. The industries with the largest direct economic impact due to the census undercount are Health, Construction, Education, and Manufacturing. After observing all the inter-industry effects, Health, Real Estate and Rental and Leasing, and Finance and Insurance were the industries most affected by the undercount in 2020 and are predicted to experience the largest benefits of an accurate count in 2030.

The funding implications presented in this brief underscore the critical importance of an accurate Census count.

Additionally, inaccurate data can misallocate resources, hinder effective planning, and risk taxpayer funds. Key industries, including Health Care and Social Assistance, Construction, Educational Services, and Manufacturing, have an outsize impact on economic stability, workforce development, and educational opportunities, with repercussions on the health and well-being of the population, educational quality, and public infrastructure development.

To address the funding implications of the 2020 Census undercount, we offer the following recommendations:

Census Data Quality: Promote initiatives to improve Census data quality, especially at the sub-state level. Encourage cooperation between the U.S. Census Bureau and external stakeholders to ensure more accurate and comprehensive data collection.

Stakeholder Engagement: Engage relevant stakeholders, including industry experts, local governments, and advocacy groups, to raise awareness of the undercount’s impact and work collaboratively to develop strategies for better data collection in the future.

Continued Research and Analysis: Encourage researchers and data analysts to delve deeper into the specific funding implications in various sectors and regions, using the findings from this study as a foundation for further research and detailed analysis.

The potential benefits for the state, projected for the decade following the 2030 census, highlight a compelling opportunity for substantial growth and recovery. By avoiding an undercount, Texas stands to gain over $57.8 billion in output, $26.5 billion in labor income, $33.7B in value added (GDP), and the creation of more than 434,000 jobs.

Notably, the Health Care and Social Assistance industry emerges as pivotal, bearing the largest impact on all examined variables. This industry, accounting for around 90% of the total impact, underscores the critical role that accurate census data plays in shaping economic outcomes. The insights from this study not only serve as a foundation for further in-depth analyses by data analysts and researchers but also offer valuable information to stakeholders, emphasizing the potential gains for Texas in the absence of an undercount in the upcoming 2030 census.

As Texas navigates the path toward the next census, these estimates serve as a crucial resource for informed decision-making, resource allocation, and strategic planning. The prospect of realizing substantial economic benefits underscores the importance of accurate and comprehensive census data in shaping the future trajectory of the Lone Star State.

The 2020 Census undercount in Texas significantly impacts the state’s federal funding, resulting in substantial financial losses across various crucial programs. With an estimated 547,968 people missed in the count, Texas faces a projected shortfall of $25.1 billion in federal funds over the next decade.

The industries with the largest direct economic impact due to the census undercount are Health, Construction, Education, and Manufacturing. After observing all the inter-industry effects, Health, Real Estate and Rental and Leasing, and Finance and Insurance were the industries most affected by the undercount in 2020 and are predicted to experience the largest benefits of an accurate count in 2030.

This research highlights the critical importance of accurate census counts and the wide-reaching consequences of undercounting. The findings advocate for concerted efforts to improve data collection and awareness ahead of the 2030 Census. By engaging stakeholders, policymakers, and the public in understanding and addressing these funding disparities, Texas can better prepare for future censuses to ensure it receives its fair share of federal resources. The Texas Census Institute’s initiative serves as a call to action for robust planning and advocacy to avoid similar undercount repercussions in the next decade.