Introduction

For decades, Congress has used place-based policies to address regional socioeconomic disparities.¹ Many recent federal economic development policies—including rural development policies—have focused on economically distressed areas. Economic development policies are generally designed to develop the resources or capabilities of state, local, and regional stakeholders, or to address obstacles that hinder development. These policies are often implemented through community-level interventions rather than individual-level interventions (i.e., people-based policies). Place-based economic development programs may support a range of activities such as infrastructure development, entrepreneurship, business development, technical assistance, planning, and related activities.

Congress has used economic data and benchmarks to determine which areas are economically distressed and therefore eligible for place-based economic development programs. These economic measures and benchmarks vary by agency and program. Many criteria use survey data available at the census tract or county level, but others allow for discretion to be applied by federal agencies who implement the program. Measures may vary, in part, due to different policy objectives (e.g., economic development, housing, health, equity) or the structure of the policy tool (e.g., tax incentives, grants, credit programs).

This report analyzes the types of economic distress measures used within three policy interventions to illustrate the impacts of the definition of economic distress on the direction of federal economic development spending. It does not represent a comprehensive review of all potential measures of economic distress.² In two of the interventions discussed in this report, Congress did not use the exact phrase economic distress to characterize the areas targeted; however, in all the interventions discussed below, Congress did target funding to areas using measures that indicated local economic conditions in those areas were worse than a specified benchmark. This report uses the phrase economic distress to characterize such areas.

The report further defines key economic measures and geographic concepts, shows how they are used, and compares how they differ in which geographic areas they identify as economically distressed. It uses counties as the unit of analysis; however, as discussed, various programs allow for the potential eligibility of other types of areas or geographies.

This report does not address underlying capacity challenges that may be involved in applying for, receiving, and managing federal awards. While federal policies may geographically target resources toward certain communities, analysts observe that capacity limitations may hamper the ability of these communities to access, manage, and implement available grant programs.³ Analysts also note that economic distress measures could be used to target technical assistance to further address capacity challenges, in addition to their use in determining program eligibility.⁴

Economic Distress: Defining Commonly Used Measures and Criteria

The term economically distressed area lacks a standardized definition and may vary by agency, program, or executive branch initiative. Federal programs that provide assistance to economically distressed areas generally use socioeconomic measures, such as median household income, as criteria for eligibility.⁵ Data for a potential project area are generally compared to a benchmark, such as a fixed number or national-level data.

Some definitions of economic distress may specify the geographic areas or the unit of analysis to be measured (e.g., counties or census tracts), while others allow the specific geographic area to be determined by the applicant(s).

Economic Measures

Congress has used a variety of measures to identify certain areas that may be economically distressed. The measures provide a framework for comparing key socioeconomic data across geographic areas. Four selected measures are discussed below: median household income, poverty rate, unemployment rate, and employment-to-population ratio.⁶

Median Household Income

To identify areas of economic distress using median household income, three concepts are applied: the definition of income as it is measured in surveys, an analysis of the median income, and an analysis of the geographic area to be used when making comparisons (particularly when area median income is referred to as a specific eligibility criterion). Each of these concepts is described below.

Income

In household surveys conducted by the Census Bureau, income refers to money that comes into the household on a regular basis, before payments for personal income taxes, Social Security, union dues, Medicare deductions, and the like.⁷ Because it is based on income before taxes, it also does not include refundable tax credits such as the Earned Income Tax Credit or the Additional Child Tax Credit. This money income concept also excludes noncash benefits such as those from the Supplemental Nutrition Assistance Program (SNAP), health benefits, and subsidized housing, and does not include resources such as food or other goods produced and consumed on one's own farm or in-kind benefits from employers, such as the use of company cars.

The reference period, or time span for which a survey asks households about the money they received, is important for defining income; typically, this is a year-long period. Assets (such as amounts held in savings accounts, the equity in one's home, or the unrealized values of investments) are not considered to be income because the money did not literally come in (to the household) during the reference period. The money used to purchase the investments or to deposit into savings would have been considered income at the time the household member received it. However, the dividends from investments and the interest earned on savings are income because they represent new money received during the reference period.

Programs that use income in their economic distress criteria might not necessarily define it in the same manner as Census Bureau surveys. Likewise, they may not necessarily define a household in the same way as surveys (which typically define a household to include all persons residing together in the same housing unit—and thus household income from survey data typically includes the income from all persons residing together, as opposed to some other grouping such as persons who both live together and share costs and resources as a single economic unit). When a program's definitions for income or household differ from those used by the Census Bureau to publish reports or tables, alternative estimates might be obtainable by programming a custom tabulation using a survey data file, so that the estimates more closely reflect the definitions used in Congress's program of interest. However, the accuracy of such estimates would depend on the level of demographic and economic detail, geographic detail, and statistical accuracy afforded by the survey.

Median Income

In a group of numbers, such as the income amounts of all households in an area, the median is the number that divides the group into two equal parts: one-half is less than the median, and one-half is greater. The median can be thought of as the middle of the income distribution. Medians are useful for characterizing income data because they are less sensitive than the mean (arithmetic average) to large outliers, which occur frequently in income data.

Geographic Area and Area Median Income

Area median income is a term used when comparing median income amounts across geographic areas. Depending on the program, the term may refer to specific types of geographic areas (such as Metropolitan Statistical Areas [MSAs]).⁸ For instance, one metropolitan area may have a higher median income than another, but also a higher cost of living. To the extent that differences in median income may reflect differences in average wages and cost of living, comparing the median income of a census tract, county, or town with that of the metropolitan area can help to reduce geographic disparities in program eligibility by metropolitan area. For small areas not in metropolitan areas, a program might use median income for a different geographic type; for example, the statewide median income or the statewide nonmetropolitan median income may be used. Or, as noted previously, some programs may allow applicants to define their own geographic areas as part of their application.⁹

Poverty Rate

Poverty is defined as a measure of economic hardship using income data, family size data, and specified dollar amounts called poverty thresholds that are intended to represent the costs associated with a low level of basic needs. Poverty thresholds vary by family size and composition to represent that the cost of basic needs varies accordingly.¹⁰ Families and single individuals that have income less than their respective poverty threshold are classified as poor. Poverty thresholds are not adjusted for geographic variation in the cost of basic needs, but they are updated annually to adjust for inflation.¹¹

A poverty rate for an area is the percentage of the area's population that is classified as living in poverty. Poverty rates are based on adjusted population totals known as the poverty universe, or persons for whom poverty status is determined. Some persons are not classified as either poor or nonpoor; these include persons living in institutions (such as nursing homes or prisons), children under age 15 who are not related by birth, marriage, or adoption to the householder of their residence (such as foster children), and persons serving in the Armed Forces and living in military barracks. These persons are excluded from the poverty universe when computing poverty rates. As a result, for areas with military barracks or large institutional populations, the totals may seem lower than would otherwise be expected, and the poverty rate may be either higher or lower than expected.

Unemployment Rate

The unemployment rate is the number of unemployed persons as a percentage of the labor force. Unemployed persons are those not working but want to work and are looking for work. The labor force is the sum of employed and unemployed persons. This does not include the entire population: those not working and not looking for work, such as the retired population, stay-at-home parents, those who want to work but have stopped looking for it, and those who are unable to work are classified as not in the labor force. Because the unemployment rate uses the labor force as its denominator, it is viewed as a measure of an economy's excess labor supply—in which more people are available and willing to work than there are jobs available.¹²

Employment-to-Population Ratio

The employment-to-population ratio (EPOP) is the percentage of the civilian noninstitutionalized adult population that is employed. In contrast with the unemployment rate, the denominator for the EPOP includes adults who are not in the labor force.¹³ That is, the EPOP includes persons "who have stopped looking for jobs and left the labor force, not just those who are currently unemployed,"¹⁴ and as a result can help to uncover areas where individuals have stopped looking for work because local opportunities are limited, or where they lack reliable transportation, certain job qualifications, or adequate child care.¹⁵ Areas with low EPOP may also reflect areas with populations that are not working because of other reasons, such as they are in school or retired.

Geographic Areas

The economic measures discussed above are computed for many types of geographic areas. To use a measure to determine whether an area is economically distressed, the geographic area first needs to be specified: which households or persons does the economic measure need to reflect? The choice of geographic area type could be driven by the population intended as a target for funding (e.g., small communities, rural communities, tribal communities, communities of any size with severely entrenched poverty); by the type of project or activity to be funded, which involves determining the project area for the activity (e.g., wide but sparsely populated rural areas that need infrastructure development, parts of urban areas that need more business and jobs); or both.

Congress could specify the types of geographic areas to be used, or it may allow applicants to determine the project area and thereby not require that a specific type of geographic area be used to determine economic distress criteria. For example, applicants self-define project areas for the Distressed Area Recompete Pilot Program (Recompete Program) administered by the Department of Commerce's Economic Development Administration (EDA); project areas may contain one or more census tracts, counties, or parts thereof. Additionally, Congress may provide federal agencies with discretion to determine the geographic areas that are applicable and the standards used to gauge economic distress.¹⁶ Nevertheless, Census Bureau data are commonly used in determining whether an area is economically distressed; thus, it may be convenient for policymakers to refer to the geographic areas defined by the Census Bureau and used in their tabulations. Table 1 summarizes these main geographic area types, and notes that some terms (especially rural) are not defined the same way by everyone. These geographic areas are discussed in further detail in Appendix B.

In addition, policymakers sometimes consider definitions of urban and rural areas. Urban and rural designation is usually not associated with any specific geographic type, but rather characterizes an area based on population density, land usage, and (for rural areas) distance from population centers. Policymakers may also use their own definitions of urban or rural, which may not align with the Census Bureau's definitions.

While geography can play a role in determining where to direct program funding, federal economic development programs are generally community-driven, meaning that projects are planned and implemented by state or local stakeholders, such as a city, county, or other eligible entity. Units of local government, in particular, can vary greatly by state in terms of the geographic areas they serve and the types of services they provide¹⁷—both of which have bearing on where and how federal economic development programs are implemented.

Selected Examples of Federal Policies Using Economic Distress Criteria

Below are descriptions of three economic development policy interventions and the criteria they use to identify economically distressed areas. Two are grant programs—the Emergency Community Water Assistance Grants (ECWAG) program and the Recompete Pilot Program—and the other, the 10-20-30 provision, allocates a percentage of annual appropriations to persistent poverty counties (PPCs), a type of economically distressed area. The definitions used by these programs are further compared, contrasted, and analyzed in the "Comparing Areas of Economic Distress Under Selected Policies" section, and Table 2 provides a summary of the three policy interventions.

Emergency Community Water Assistance Grants Program (ECWAG)

The U.S. Department of Agriculture's (USDA's) Rural Utilities Service provides grants through the ECWAG program to help eligible communities prepare for or recover from an emergency that threatens the availability of safe, reliable drinking water. Emergencies can be natural disasters (e.g., droughts, floods, tornadoes) or man-made disasters (e.g., chemical spills). A federal disaster declaration is not required to participate in the program. Applicants must show a major decline in the water quality or water quantity within two years of submitting their application. Applicants compete on a national basis for available funding. Water transmission line grants of up to $150,000 are available to construct waterline extensions, repair breaks or leaks in existing water lines, and address maintenance issues. Water source grants of up to $1 million are available to construct new wells, reservoirs, transmission lines, and treatment plants.

Statutory authority for the ECWAG program is in the Consolidated Farm and Rural Development Act, as amended, Section 306A (7 U.S.C. §1926a). Regulations for this program are codified at Title 7, Part 1778, of the Code of Federal Regulations.

The Agriculture Improvement Act of 2018 (2018 farm bill; P.L. 115-334, §6407(a)) reauthorized appropriations of $50 million per fiscal year for FY2019 through FY2023 for the ECWAG program. Congress appropriated $15 million for the program for each fiscal year from FY2019 through FY2023.

Eligible Areas

Eligible areas include rural areas with populations of 10,000 or fewer people, tribal lands in rural areas, and colonias (unincorporated communities with poor or no infrastructure, near the U.S.-Mexico border). The area served must also have a median household income that is less than the state's median household income for nonmetropolitan areas. Applicants may use geographic units of measurement other than counties, such as towns, incorporated areas, or census tracts.

Distressed Area Recompete Pilot (Recompete Pilot) Program²¹

EDA administers the Recompete Pilot Program, which was established to address persistent economic distress and support long-term, comprehensive economic development.²² Recompete Pilot funding may be used to address local and regional circumstances to create jobs or connect workers to jobs. EDA administers the program with two types of funding awards through two award phases. EDA launched the first round of Recompete funding in FY2023. For Phase 1, EDA instructed applicants to choose either a strategy development grant,²³ approval of a Recompete Plan, or both. ²⁴ In Phase 2, EDA plans to fund implementation awards to support a range of economic development activities based on regional priorities (e.g., workforce development, business and entrepreneur development, infrastructure, planning, technical assistance activities).²⁵

Eligible applicants include

• political subdivisions of a state;
• tribal governments;
• U.S. territories;
• the District of Columbia;
• nonprofit organizations working in cooperation with a political subdivision of a state;
• economic development districts; and
• a coalition of any of the entities listed above.

P.L. 117-167, commonly referred to as the CHIPS and Science Act, established the Recompete Pilot Program in August 2022,²⁶ and authorized the appropriation of $1 billion to fund it over five years (FY2022 through FY2026). The Consolidated Appropriations Act, 2023 (P.L. 117-328) appropriated $200 million for the program.²⁷

Eligible Areas

The Recompete Pilot Program targets areas where prime-age (25-54) employment falls below the national rate.²⁸ This is referred to as the prime-age employment gap (PAEG) and is calculated as the difference (expressed in percentage points) between (1) the national five-year average prime-age employment rate and (2) the five-year average prime-age employment rate of the eligible area.²⁹ In this context, the term employment rate is synonymous with employment-to-population ratio because it includes persons outside of the labor force (including those who have stopped looking for work) instead of measuring only the unemployed.³⁰ Members of the Armed Forces are excluded.³¹

Among other requirements, Recompete projects must be located in one of two eligible area types and meet certain PAEG requirements. The two eligible geographic areas are the following:

• 1. Local Labor Markets (LLMs)—an MSA, a Micropolitan Statistical Area (μSA), a commuting zone, or tribal lands. LLMs with a PAEG of at least 2.5 percentage points are eligible. Tribal lands and Pacific Ocean territories are considered eligible LLMs.
• 2. Local Communities (LCs)—an area served by a general-purpose unit of local government that is located within, but does not fully cover, an ineligible LLM. In addition, one of the following conditions must apply:
• the entire area served by the unit of local government on average has a PAEG of at least 5 percentage points and a median annual household income of no more than $75,000; or
• a subset of the area served by the unit of local government has five or more contiguous census tracts that each have a PAEG of at least 5 percentage points and median annual household income of no more than $75,000. In this instance, the applicant's service area must be contained within the identified census tracts.³²

The data required to calculate PAEG were unavailable for certain tribal lands and some Pacific Ocean territories. As such, the EDA's guidance indicates that Pacific Ocean territories and all tribal lands are eligible to participate in the program.³³

Persistent Poverty Counties (PPCs) and the 10-20-30 Provision

The term persistent poverty county has been defined in appropriations laws for a variety of programs since the passage of the American Recovery and Reinvestment Act of 2009 (P.L. 111-5.).³⁶ PPCs are defined as those counties that have had poverty rates of 20% or greater for at least 30 years, as measured using data collected at three points in time.

One funding set-aside provision is commonly referred to as the 10-20-30 provision because a program subject to it is required to set aside no less than 10% of funds for counties that have had poverty rates of 20% or greater for at least 30 years. The 10-20-30 provision has been applied by the USDA's Rural Business-Cooperative Service in its Rural Business Program Account, Rural Economic Development Loans Program Account, Rural Cooperative Development Grants, and Rural Microentrepreneur Assistance Program; and applied in the EDA's Public Works and Build to Scale programs, among others.³⁷ The Infrastructure Investment and Jobs Act (P.L. 117-58) provided a set amount ($20 million) for certain PPCs through two programs rather than the 10% used in the 10-20-30 provision.³⁸ P.L. 117-58 also provided the Secretary of Transportation with discretion to prioritize PPCs to receive technical assistance under the Thriving Communities Initiative.³⁹

In this report, PPCs will be identified using poverty rates from the 1990 Census (which, while collected in 1990, recorded the incomes and by extension the poverty statuses of persons in calendar year 1989), Census 2000 (which, similarly, provided poverty data for 1999), and the 2021 Small Area Income and Poverty Estimates (SAIPE) (for calendar year 2021).⁴⁰

Comparing Areas of Economic Distress Under Selected Policies

The three approaches to identifying economic distress discussed in the policy interventions above have distinct objectives. The ECWAG program addresses water infrastructure in rural areas, the Recompete Pilot Program addresses economic development and jobs, and the PPCs provision mandates that economically distressed counties receive a minimum amount of funding through certain existing programs. The interventions identify areas of economic distress in different ways, and as a result, different sets of geographic areas are classified as distressed. The data and figures discussed below are focused at the county level. However, as noted previously, researchers and agencies observe that areas of economic distress may be found in certain counties not designated as distressed.⁴¹ Tables with numerical detail are available in Appendix A.

Degree of Geographic Overlap Among the Definitions

Of the nation's 3,143 county areas, 1,073 (34%) met one or more of the three definitions of economic distress used by the programs or approaches described above (see also Table A-2). Of those, 158 counties (15.4%) met two or more of the distress definitions.⁴² This indicates that these three sets of criteria mostly do not overlap, but rather identify different counties with different aspects of economic distress.

The three following maps (Figure 1, Figure 2, and Figure 3) geographically illustrate economic distress specified in the ECWAG program, the Recompete Pilot Program, and PPCs, respectively. The number of counties by region that meet the distress criteria under each program, and the regional shares (percentages) of counties under each definition, may be found in Table A-1. Counties defined as distressed under multiple programs are not broken out separately.

Figure 1 displays counties identified as economically distressed for the ECWAG program, many of which are in the Great Plains, western mountain areas, northern Michigan, and northern Wisconsin. However, distressed counties also appear in Appalachia, the Mississippi Delta, and the Southeast. ECWAG limits eligibility to rural areas with populations of 10,000 or fewer people. As such, many of the areas classified by ECWAG criteria as economically distressed appear in rural areas and in low-population counties. As noted earlier, applicants set the project area. They may use geographic units of measurement other than counties such as towns, census tracts, or villages.

Figure 2 displays counties with high rates of PAEG, subject to the other criteria applied in the Recompete Pilot Program (as described in Table 2). As in the other figures, the data in Figure 2 are shown at the county level only. As noted earlier, applicants self-define project areas, and they may use geographic units of measurement other than counties. The counties identified as distressed in Figure 2 are more widely dispersed throughout all regions of the country, including along the east and west coasts and in the more populous portions of the Midwest, when compared to counties in Figure 1. This is likely due to the lack of criteria related to rural status or capping population size (as is done in ECWAG). The criteria for the Recompete Pilot Program, as discussed earlier, includes specifications for labor market areas based on metropolitan areas, as well as for local communities that may include nonmetropolitan areas. Both the labor market areas and local communities use employment gaps (though local communities must have a median household income not greater than $75,000 in addition to an employment gap of 5 percentage points, as opposed to solely an employment gap of 2.5 percentage points for labor market areas). These two mutually exclusive specifications of geographic entities help explain the dispersion of counties in Figure 2, in that both densely populated and sparsely populated counties meet the specifications in the Recompete Pilot Program language.

In Figure 3, PPCs are found predominantly in Appalachia, the Southeast, the Mississippi Delta, some urban areas, and tribal lands, as well as along the southern border. In classifying a county as a PPC, the specification that the county have persistently high poverty (and not just temporarily high poverty) suggests historical, structural, and other reasons for the classification of such counties. Because these areas named above have histories that are distinct from each other and from the rest of the country politically, economically, and socially, policymakers seeking to address poverty in these areas may benefit from a thorough reading of their histories, with emphasis on the similarities and differences by which their residents have experienced poverty. Selected references discussing these areas are presented in Appendix C.

Figure 1. Map of Counties Using Population and Income Specifications in Program Language for the Emergency Community Water Assistance Grant Program (ECWAG) Comparison of the county's median household income with that of the nonmetropolitan portion of the state, among counties with 10,000 or fewer persons

Source: Congressional Research Service, using data from U.S. Census Bureau, American Community Survey, 2021 Five-Year Estimates, Table S1903, "Median Income in the Past 12 Months (in 2021 Inflation-Adjusted Dollars)," for counties, county subdivisions, places, and state nonmetropolitan areas; U.S. Census Bureau, Population Estimates Program, population estimates base for April 1, 2020, "Subcounty Total Resident Population Estimates (Vintage 2022)," SUB-IP-EST2022-POP.csv, and "County Population Estimates (Vintage 2022)," CO-EST2022-ALLDATA.csv.

Figure 2. Map of Counties Using Specifications on Age, Employment-to-Population Ratio, Labor Market, and Median Income, Used in the Recompete Pilot Program

Source: Congressional Research Service, using data from Recompete Eligibility Mapping Tool, Argonne National Laboratory, https://disgeoportal.egs.anl.gov/Recompete/. Notes: The map displays county-level data only; the Recompete Pilot Program allows applicants to use other geographic areas, such as census tracts and tribal lands, to define their project area. "Eligible" and "Partially Eligible" areas as shown above are categorizations used by the Recompete Mapping Tool: they do not take into account any further criteria that the Secretary of Commerce may establish as set forth in the Recompete Pilot Program legislation, nor should they be considered guarantees of receiving funding; conversely, areas in gray that apply might be considered eligible for a similar reason. "Partially Eligible" indicates areas in which the Mapping Tool may show census tracts that fit the eligibility criteria even though the entire county area does not meet them. The figure does not display U.S. territories or the boundaries of tribal lands, which EDA has determined to be eligible to participate in the program. For details on the Recompete Pilot Program specifications, see Table 2.

Figure 3. Persistent Poverty Counties Using Poverty Rates in 1989, 1999, and 2021

Source: Congressional Research Service, using data from U.S. Census Bureau, 1990 Census, Census 2000, and 2021 Small Area Income and Poverty Estimates.

Distribution of Distressed Counties, by Census Region

Analyzing the distribution of economically distressed counties by census region can help in summarizing how different definitions of distress target different parts of the country. The census regions are defined by four groups of states listed in the text box below, as categorized by the Census Bureau, and reflect broad historical patterns of settlement and economic organization. The analysis below compares the number of economically distressed county areas for each region with the region's total number of counties to determine, for each measure of economic distress, whether the region has a disproportionately larger or smaller share of counties compared with the region's share of counties overall. This analysis focuses on each region's share of counties (i.e., the geographic areas themselves) as opposed to shares of the population.

Figure 4. Regional Shares of Counties by Selected Economic Distress Criteria Regions as defined by the Census Bureau

Source: Congressional Research Service, using the following data: ECWAG: U.S. Census Bureau, American Community Survey, 2021 Five-Year Estimates, Table S1903, "Median Income in the Past 12 Months (in 2021 Inflation-Adjusted Dollars)," for counties, county subdivisions, places, and state nonmetropolitan areas; U.S. Census Bureau, Population Estimates Program, population estimates base for April 1, 2020, "Subcounty Total Resident Population Estimates (Vintage 2022)," SUB-IP-EST2022-POP.csv, and "County Population Estimates (Vintage 2022)," CO-EST2022-ALLDATA.csv. Recompete Pilot Program: Recompete Eligibility Mapping Tool, Argonne National Laboratory, https://disgeoportal.egs.anl.gov/Recompete/. Persistent poverty counties: U.S. Census Bureau, 1990 Census, Census 2000, and 2021 Small Area Income and Poverty Estimates. Notes: U.S. totals do not include Puerto Rico or U.S. Territories. Percentages may not sum to 100.0% because of rounding. ECWAG: U.S. Department of Agriculture, Rural Utilities Service, Emergency Community Water Assistance Grant Program. Recompete: U.S. Department of Commerce, Economic Development Administration, Recompete Pilot Program. Persistent poverty counties: counties that have had poverty rates of 19.5% or greater in 1989, 1999, and 2021 (according to the 1990 Census, Census 2000, and 2021 Small Area Income and Poverty Estimates, respectively). For lists of states by region, see the "States in Each Census Region" textbox.

ECWAG Program

In comparing each region's share of economically distressed counties with its share of all counties (Figure 4), under the definition used in the ECWAG program, disproportionately greater shares of counties were defined as distressed in the Midwest (42.9% compared with 33.6%) and West (24.1% compared with 14.3%), while disproportionately lower shares were found in the Northeast (0.9% compared with 6.9%) and South (32.2% compared with 45.2%).

These differences can be explained by the prevalence of low-population counties by region (see Table 3). As noted, the ECWAG definition includes a requirement that the area population not exceed 10,000. Among counties of 10,000 persons or fewer, the Midwest and West have disproportionately greater shares (meaning the percentage of small-population counties in column D is greater than the percentage of all counties in column B) and the Northeast and South have disproportionately smaller shares. These differences are also reflected in the median populations by region (column E): the median populations among counties in the Midwest (19,872) and West (22,734) are smaller than the overall U.S. median (25,697), while in the Northeast and South they are larger (105,744 and 25,905, respectively). These differences underscore that the ECWAG program targets rural counties, which the program defines by population.

Recompete Pilot Program

Using the Recompete Pilot Program's criteria as a proxy for economic distress, the results are nearly reversed from ECWAG's results (Figure 4). The Northeast and South have comparatively larger shares of distressed counties than all counties generally, while the Midwest has a smaller share (25.2% compared with 33.6%) and the West's share is close to being proportional (14.2% compared with 14.3%).

As noted, the Recompete Pilot Program uses two primary ways of classifying potentially eligible areas—LLMs and LCs. Both approaches include criteria for median income and employment-to-population ratio thresholds (via the program's PAEG). However, LLMs also use the metropolitan area status as the geographic basis for the eligible area. The Recompete Pilot includes more counties in the Northeast than does ECWAG, in part because of how the Recompete Pilot includes MSAs in its definition of LLMs. While Northeastern states tend to have relatively higher incomes compared to other parts of the country, they also have many MSAs (as evidenced by the median county population of 105,744 in the Northeast—well above the 50,000 population mark for urban areas used in delineating MSAs).⁴³

Because MSAs are more abundant near the coasts and inland waterways than in the country's interior,⁴⁴ a greater share of the land in the Midwest and West lies outside of any MSA as compared with the Northeast and South.

Densely populated areas may be considered eligible under the Recompete Pilot Program but not ECWAG because ECWAG's eligibility criteria include a maximum population size (both to define rural areas and to focus on them).⁴⁵ Because of the relative abundance of MSAs in the Northeast and South, these regions have disproportionately greater shares of eligible counties under the Recompete Pilot Program, compared with both their shares under ECWAG and their shares of all U.S. counties. Additionally, the EDA's program guidance identifies an entire MSA (which may be made up of multiple counties) as eligible if its PAEG is at least 2.5 percentage points. Further, smaller areas that are within nonqualifying MSAs might also be considered eligible if the smaller area has a PAEG of 5 percentage points or greater and a median household income at or below $75,000.

Persistent Poverty Counties (PPCs)

The greatest share of PPCs is in the South (Figure 4). As noted in the discussion of Figure 3 in the "Degree of Geographic Overlap Among the Definitions" section, the persistence of poverty over time, as required by the PPC definition, highlighted specific areas such as Appalachia, the Southeast and Mississippi Delta, tribal areas, and the southern border (see Figure 3) that have historical legacies and economic histories distinct from one another and the rest of the country. Many of these areas are located in the South region, with notable exceptions in other regions, particularly among tribal lands and certain urban areas. Given the persistence of high poverty rates in these disparate areas, a policy tool—a set-aside to dedicate 10% of program funds to be directed to PPCs—was developed to be used across a wide variety of programs, from rural development to brownfield remediation.⁴⁶

In addition to a historical lack of resources in PPCs throughout the South, another factor amplifies their presence in this region: the same poverty thresholds are used throughout the country and are not adjusted for regional variations in the costs of basic goods and services. As stated previously, a poverty threshold is a dollar amount representing a low level of basic needs; families with income below that amount are considered poor. Because the official poverty thresholds are the same nationwide, poverty status under the official measure does not reflect the same purchasing power nationwide. That is, because the official measure classifies a low-income family as poor if its income is below the threshold, it means if they live in an area with low average wages and low prices for basic goods, their purchasing power may be the same as another family with income above the threshold, and not classified as poor, if the second family lives in an area with higher average wages and prices. Prices of goods and services, particularly for housing, tend to be lower in the South and Midwest than in the Northeast and West.⁴⁷

Policy Considerations

Congress may choose to keep the criteria for economically distressed areas as they currently stand; or, if Congress chooses to change the criteria or develop new ones, it may consider the following:

• economic and demographic measures,
• eligibility options based on different units of geography, and
• data sources and their limitations and availability.

Economic and Demographic Criteria

The three examples of economically distressed area definitions featured in this report target different types of communities. Congress may want to consider what types of communities it would want to target and use these examples as models. For example, if Congress is interested in targeting low-income rural communities, it may want to consider establishing a definition of economic distress similar to that used in the ECWAG program, which includes median household income and a maximum population cutoff (such as areas of 10,000 or fewer people). If Congress is interested in targeting historically poor areas in the United States, it may want to consider using measures similar to those used in the PPCs designation. That designation considers poverty rates over time. If Congress is interested in targeting communities with low employment levels, it may want to consider using the Recompete Pilot Program's definition as a model and use labor force statistics, such as EPOP (as is used in Recompete) or the unemployment rate, to be compared either with the national average (as is done in Recompete) or against a fixed benchmark percentage.

Geography

Some definitions of economically distressed area require that the project area be a specific type of geographic area (e.g., PPCs use counties as the project area). Other definitions allow the applicant to determine the project area (e.g., in the ECWAG program). In that case, the project area can range from multiple counties to an unincorporated area. See Appendix B for a list of possible geographic areas that may be used, and for which data are published. Congress may want to consider whether to specify which geographic area type may be used to classify an area as economically distressed. The benefit of specifying the geographic area is that it may be more likely that data would be readily available for the project area from tables published by the Census Bureau. The drawback is a lack of flexibility; for instance, small communities in need within more affluent larger communities may not qualify. For example, if the geographic area were specified to use only county-level estimates, an unincorporated area with a low median household income located within a county with a high median income may not qualify as economically distressed.

Data Source

As previously mentioned, many of the definitions of economically distressed area use household survey data from the Census Bureau to confirm project areas meet the criteria. Household survey data may not accurately reflect current conditions for several reasons, including coverage (e.g., the unhoused population not being surveyed because sampling is performed using housing units), lag time between the collection of data and making the data publicly available, potential for low response rates for small areas or populations, and estimates not being precise for small areas because they are based on a sample and not the entire population. Some definitions of economic distress allow the Secretary for the federal program or initiative to use his or her discretion to allow alternative sources of data to be used, such as administrative records or other sources of federally collected data. Congress may want to consider whether to allow alternative data sources to be used at the discretion of the respective Secretary. A potential benefit in allowing alternative data sources to be used is that these sources, such as administrative records or locally obtained information, may be more accurate than household surveys in determining which project areas should qualify. On the other hand, a potential drawback is that the alternative data sources may not have data that are as accurate as household surveys—the accuracy for which has been thoroughly investigated and quantified wherever possible—and it could potentially take additional time to review each application and evaluate the evidence presented.

Concluding Remarks

This report examines three distinct ways Congress has used various criteria in statute, in three programs or policy contexts, to direct agencies to focus funding and other resources on economically distressed areas. The interventions were designed to serve different purposes. The ECWAG program was designed to provide grants for water infrastructure to sparsely populated rural areas that need it and may not have the resources to build it or clean it on their own. The Recompete Pilot Program was designed to spur economic development in a variety of locations, which may include both rural and urban areas. Funding set-asides for PPCs were designed to limit the degree to which administrators in any number of programs overlook areas with persistently high poverty for funding opportunities, in favor of areas with more resources.

The approaches described in this report were selected in order to illustrate that different economic distress criteria result in distinct regions being eligible for different policies. The criteria used to measure aspects of economic distress in the three examples examined here—the ECWAG program, the Recompete Pilot Program, and the set-asides for PPCs in appropriations legislation—identified different geographic areas with considerable lack of overlap. The differences among the areas deemed eligible for assistance from the three interventions stem from the following:

• the use of a rural definition in the ECWAG program that focuses on counties of 10,000 or fewer persons,
• the inclusion of MSAs in one of the eligibility pathways for the Recompete Pilot Program, which correlate to areas with large populations (unlike the definition of rural used by the ECWAG program in its eligibility criteria), and
• the use of high county-level poverty rates for over 30 years for PPCs, which calls attention to counties where poor economic conditions have been particularly entrenched, and for which the measure (poverty) applies to areas with either large or small populations.

As Congress conducts oversight on existing programs and considers new programs, it may seek to understand how economic distress criteria impact the distribution of program funds.

Appendix A. Number and Percentage of Economically Distressed Counties by Region

Appendix B. Geographic Area Definitions

Census Tracts

Census tracts are small geographic areas with populations typically between 1,200 and 8,000 (approximately 4,000 on average) that are contained wholly within counties and are designed, where possible, to have recognizable physical boundaries and to have a stable population over time.⁴⁸ Census tracts are sometimes split or merged to accommodate changes in population density that have occurred in an area. For many economic and demographic detailed characteristics, the census tract is the smallest geographic area for which data are published.

Places

Places correspond roughly to cities and villages in general parlance. That is, they have a population center and a name generally recognizable to residents of the area. Places may be incorporated places, meaning a city with a municipal government; or census-designated, meaning a known population center without a municipal government.⁴⁹ Places are not defined by any specific minimum nor maximum population size. Five incorporated places had populations of zero persons as of April 1, 2020 (Census Day for the 2020 census): these are areas that remain incorporated according to the laws of their respective states but had lost their entire populations.⁵⁰ According to the Census Bureau's Geographic Areas Reference Manual, "the Census Bureau includes, in the decennial census, all active incorporated places and inactive incorporated places for which it has certified corporate limits as of January 1 of the census year (the date used to tabulate the census results)."⁵¹ The most populous place in the United States is New York City, which had a population of 8,804,194 as of April 1, 2020, and which encompasses five county areas (Bronx County, Kings County, New York County, Queens County, and Richmond County, which correspond to the boroughs of the Bronx, Brooklyn, Manhattan, Queens, and Staten Island, respectively).⁵²

Despite numerous exceptions and the absence of any definitional requirement to be bound within a single county, places are generally smaller than counties. As of April 1, 2020, among all 19,519 incorporated places in the United States, the median population was 1,132 (half of all incorporated places had fewer than 1,132 persons and half had more).⁵³

County Subdivisions (Townships)

County subdivisions are smaller than counties, and are defined by geographic area rather than by population center. They are "the primary divisions of counties and equivalent entities."⁵⁴ In 20 states, there are general-purpose local governments associated with these geographic areas (specifically, the county subdivisions termed minor civil divisions as opposed to unorganized areas or census-defined areas), which the Census Bureau identifies as township governments.⁵⁵ These are distinct from municipal governments (which may be found in all states) or other entities such as election districts. As explained by the Census Bureau, "[an] incorporated place is established to provide governmental functions for a concentration of people as opposed to a minor civil division, which generally is created to provide services or administer an area without regard, necessarily, to population."⁵⁶

The term town has been used in common parlance in different states to refer either to municipalities or to township governments, or to the geographic areas associated with those governments, and thus the term requires additional clarification when precision is needed.

Counties

The term county may refer either to a geographic area or a type of general-purpose government. There are 3,143 geographic areas that the Census Bureau includes in its county-level statistics for the 50 states and the District of Columbia. County populations as of April 1, 2020, ranged from 63 in Loving County, TX, to 10,014,042 in Los Angeles County, CA⁵⁷; the median population was 25,697.⁵⁸

Of the geographic areas that the Census Bureau includes in its county-level data, not all are called counties and not all have county governments. The umbrella term county area includes parishes in Louisiana, boroughs in Alaska, some areas in which a large city and its surrounding county have been consolidated into one entity (such as Suffolk County, MA, which is administered by the city of Boston, or Honolulu city and county in Hawaii), as well as cities that are outside of any county (such as Washington, DC, and independent cities in Virginia). In Puerto Rico, the areas called municipios correspond to county-level geographic areas. Further, the states of Connecticut and Rhode Island have county areas in a geographic sense but no county governments associated with them, as township governments in those states provide the wide range of general-purpose government services that in some other states would be provided by the county government.⁵⁹

Metropolitan Areas

Metropolitan Statistical Areas (MSAs) and their smaller counterparts Micropolitan Statistical Areas (µSAs)⁶⁰ together constitute Core-Based Statistical Areas (CBSAs), which are delineated by the Office of Management and Budget (OMB). CBSAs are composed of one or more entire counties, and have "a substantial population nucleus, together with adjacent communities having a high degree of social and economic integration with that core."⁶¹ CBSAs are not linked to any one type of local government, and typically multiple local governments exist within their borders.

The fact that CBSAs are composed of one or more entire counties, the borders of which are well-known to most laypersons, is convenient for identifying which entities are inside or outside of CBSAs and determining which areas within a state are nonmetropolitan (i.e., outside of MSAs).

As a concept for geographically based statistics (not necessarily for programmatic purposes), OMB first delineated CBSAs in 2000, modifying the older concept of MSAs to recognize and take account of recent building development patterns and how the spatial distribution of residences and businesses has changed since the mid-20^th century. Program legislation and administrative rules written before 2000, therefore, likely will refer to MSAs instead of CBSAs; it is generally prudent to verify whether the appropriate executive agency reinterpreted the language to include µSAs or if the legislation had been amended in some other way to include them.⁶²

Urban and Rural

The term rural can evoke an image of farms, unsettled lands, and an absence of cities (which are characterized by a large population with many buildings located close to one another). However, specifying that image numerically using measures of population or land usage presents a variety of choices that could be made to classify areas that do not clearly reflect either pole of rural or urban. As a result, the definitions of rural or urban for the purposes of economic development programs vary from program to program, and may not reflect the Census Bureau's classification schema.⁶³ For the USDA Water and Waste Disposal Program established by the Consolidated Farm and Rural Development Act of 1961, rural areas are defined as those with populations of 10,000 or fewer; for the USDA Community Facilities Loan and Grant Program established by the same act, rural areas are those with 20,000 or fewer inhabitants; and at the Census Bureau, the concept is more complicated, with population, land usage, and distance all considered together using a set of rules.⁶⁴

Appendix C. Selected References on the Histories and Economies of Areas with Persistent Poverty Counties

Appalachia

• James Ziliak, "Introduction: Progress and Prospects for Appalachia," in Appalachian Legacy: Economic Opportunity after the War on Poverty, Brookings Institution Press, 2012, https://www.brookings.edu/wp-content/uploads/2016/07/appalachianlegacy_chapter.pdf
• Appalachian Regional Commission, Appalachia Then and Now: Examining Changes to the Appalachian Region Since 1965, 2015, https://www.arc.gov/report/appalachia-then-and-now-1965-2015/
• Ronald D. Eller, "The War on Poverty in Appalachia," keynote address at "A Region Reflects: Seeking Truth, Seeking Vision" conference, Ohio University, Athens, OH, 2014, https://irjci.blogspot.com/p/a-look-back-at-war-in-poverty-in.html
• Shelley Smith Mastran and Nan Lowerre, "Federal Development of the Southern Appalachians, 1960-81," Chapter VII in Mountaineers and Rangers: A History of Federal Forest Management in the Southern Appalachians, 1900-81, Forest Service publication FS-380, U.S. Department of Agriculture, 1983, http://www.npshistory.com/publications/usfs/region/8/history/chap7.htm
• Trish Kahle, "Overview: Coal Mining and Labor Conflict," Energy History Online, Yale University, 2023, https://energyhistory.yale.edu/coal-mining-and-labor-conflict/.

Southeast and Mississippi Delta

• James Giesen, "Sharecropping," New Georgia Encyclopedia, last modified September 29, 2020, https://www.georgiaencyclopedia.org/articles/history-archaeology/sharecropping/
• Ronald L.F. Davis, "Sharecropping," Mississippi Encyclopedia, University of Mississippi Center for the Study of Southern Culture, last modified April 15, 2018, http://mississippiencyclopedia.org/entries/sharecropping/
• Bradley L. Hardy, Trevon D. Logan, and John Parman, The Historical Role of Race and Policy for Regional Inequality, The Hamilton Project, 2018, https://www.hamiltonproject.org/publication/paper/the-historical-role-of-race-and-policy-for-regional-inequality/
• Heather O'Connell, "The Impact of Slavery on Racial Inequality in the Contemporary U.S. South," Social Forces, March 2012, vol. 90, no. 3, pp. 713-734, https://www.jstor.org/stable/41682675
• Regina S. Baker, "The Historical Racial Regime and Racial Inequality in Poverty in the American South," American Journal of Sociology, 2022, vol. 127, no. 6, pp. 1721-1781, https://www.journals.uchicago.edu/doi/10.1086/719653

Southern Border Colonias

• U.S. Department of Housing and Urban Development, "Colonias History," HUDExchange website, https://www.hudexchange.info/cdbg-colonias/colonias-history/ (accessed December 27, 2023).
• Jane E. Larson, "Free Markets Deep in the Heart of Texas," Georgetown Law Journal 1995, vol. 84, pp. 179-260, https://heinonline.org/HOL/P?h=hein.journals/glj84&i=199
• Richard Delgado, "Rodrigo's Twelfth Chronicle: The Problem of the Shanty Chronicle," Georgetown Law Journal, 1996, vol. 85, pp. 667-690, https://scholarship.law.ua.edu/fac_articles/409

Urban Areas

• Francesco Andreoli, Arnaud Mertens, Mauro Mussini, and Vincenzo Prete, "Understanding trends and drivers of urban poverty in American cities," Empirical Economics, 2022, vol. 63, pp. 1663-1705, https://doi.org/10.1007/s00181-021-02174-5
• Patrick Sharkey, Bryan Graham et al., Mobility and the Metropolis, The Pew Charitable Trusts, 2013, https://www.pewtrusts.org/~/media/legacy/uploadedfiles/pcs_assets/2013/mobilityandthemetropolispdf
• August Benzow and Kenan Fikri, The Persistence of Neighborhood Poverty: Examining the Power of Inertia and the Rarity of Neighborhood Turnaround Across U.S. Cities, Neighborhood Poverty Project, Economic Innovation Group, 2020, https://eig.org/neighborhood-poverty-project/neighborhood-turnaround/
• Margery Austin Turner et al., Tackling Persistent Poverty in Distressed Urban Neighborhoods: History, Principles, and Strategies for Philanthropic Investment, Urban Institute, 2014, https://www.urban.org/sites/default/files/publication/22761/413179-tackling-persistent-poverty-in-distressed-urban-neighborhoods.pdf

Tribal Lands

• Narayana Kocherlakota, "Persistent Poverty on Indian Reservations: New Perspectives and Responses," remarks at the 2015 Federal Reserve System Community Development Research Conference, Washington, DC, 2015, https://www.minneapolisfed.org/speeches/2015/persistent-poverty-on-indian-reservations-new-perspectives-and-responses
• Sara Usha Maillacheruvu, The Historical Determinants of Food Insecurity in Native Communities, Center on Budget and Policy Priorities, 2022, https://www.cbpp.org/research/food-assistance/the-historical-determinants-of-food-insecurity-in-native-communities
• K. Whitney Mauer, "Indian Country Poverty: Place-Based Poverty on American Indian Territories, 2006-10," Rural Sociology, 2017, vol. 82, no. 3, pp. 473-498, https://onlinelibrary.wiley.com/doi/abs/10.1111/ruso.12130
• Donna Feir, Rob Gillezeau, and Maggie E.C. Jones, "The Slaughter of the Bison and Reversal of Fortunes on the Great Plains," CICD Working Paper 2019-01, Federal Reserve Bank of Minneapolis, Center on Indian Country Development, 2019, https://www.minneapolisfed.org/research/cicd-working-paper-series/the-slaughter-of-the-bison-and-reversal-of-fortunes-on-the-great-plains