by August Benzow

This research brief explores how the prime-age employment gap (PAEG) and the official poverty measure (OPM) differ in identifying areas of economic hardship for federal aid allocation. While high-poverty counties often have high PAEGs, many high-PAEG counties do not have high poverty rates. This distinction results in different geographical patterns and identifies demographically distinct groups of distressed counties. As implemented in the Recompete program, the PAEG’s expansive eligibility criteria could pose challenges in the competitive grant allocation process due to limited funding.

Introduction

Enacted in August 2022, the CHIPS and Science Act created a new place-based policy specifically designed for deeply distressed areas: the Recompete Pilot Program. The Recompete program uses the prime-age employment gap (PAEG)—a new measure in federal place-based policy—to identify communities where employment among prime-age workers (25-54 years old) lags behind the national rate. With its distinctive focus on employment, Recompete breaks from the traditional place-based policy approach of using the Official Poverty Measure (OPM) as the primary diagnostic criteria.

Policymakers are familiar with the OPM and generally know what types of places it identifies and why, but are much less familiar with the PAEG and how this lens filters the country’s economic map. This brief will help guide federal decision-making by laying out how these two measures of geographic distress overlap and where they differ. It offers insights for EDA program administrators as they make near-term decisions on where to target scarce funding and guidance for Congress as it considers how to expand and improve upon place-based policy in the United States in the long term.

Despite its Shortcomings, the Official Poverty Measure has Dominated Federal Place-Based Policy for Over 50 Years

Developed in the mid-1960s, the Official Poverty Measure (OPM) has guided geographic targeting for most federal place-based policies. Its simplicity and ease of measurement make it an attractive option for policymakers, but it is not without its detractors. The federal poverty threshold is an inflation-adjusted benchmark based on the cost of food in 1963 and does not reflect how our understanding of poverty has evolved over the past 60 years. Importantly, the OPM:

  • Does not consider cost of living differences between communities
  • Only reflects pre-tax income
  • Excludes some of the largest transfers to lower-income families
  • Does not take into account relative changes in the experience of poverty over time[1]

Despite its many shortcomings, the OPM has persevered as the targeting mechanism for many federal programs because it provides an easy way to quantify economic distress and is closely correlated with other indicators of economic disadvantage. For instance, regions of the country that are synonymous with concentrated disadvantage like the rural Deep South and Appalachia have persistently high poverty rates. Consequently, policymakers have historically used the OPM as a shorthand for a lack of economic opportunity.

The first large-scale use of the OPM to target federal funding to specific communities was in the 1990s, when Congress established the Empowerment Zone (EZ) program under the Empowerment Zones and Enterprise Communities Act of 1993. The EZ program leveraged tax incentives and block grants to drive economic investment in distressed areas of the United States, and used poverty rates alongside unemployment and general metrics of distress to identify eligible census tracts. As with many federal programs, a 20 percent poverty rate was the minimum benchmark a census tract needed to meet to qualify.

Authorized by the Community Renewal Tax Relief Act of 2000, New Market Tax Credits (NMTCs) deployed similar eligibility criteria with the OPM at its core. Designed to incentivize economic development through the use of tax credits, the policy targeted “low-income” census tracts, defined as a 20 percent poverty rate or a median family income below 80 percent of the surrounding area. Nearly one-quarter of the county’s 75,000 census tracts qualify under this criteria, but only a few thousand have received an investment. More recently, Opportunity Zones, which became law as part of the 2017 Tax Cuts and Jobs Act, pulled from the same criteria in designating communities in which certain private sector investments would carry beneficial tax treatment.

In 2009, the federal government took a different approach to addressing long-term economic distress with the designation of persistent-poverty counties through the so-called 10-20-30 provision as part of the American Recovery and Reinvestment Act. This provision requires certain federal agencies to direct 10 percent of their funding to counties with a poverty rate of 20 percent or more for at least 30 years, and marked a step forward in how the federal government thinks about economic distress by prioritizing persistent poverty; however, its use of counties instead of census tracts excluded most urban areas. Furthermore, it was only a targeting requirement, with no new programs created to address the specific economic challenges of distressed communities.[2]

The newly-established Recompete program shifts the focus of place-based policy to labor force participation and does not use the OPM at all in its targeting criteria (although a household income threshold is required for certain geographies). By focusing on participation, the PAEG captures not only those receiving formal unemployment benefits, but also those working-age adults who have exited the labor force entirely. In this sense, it addresses some of the root causes of poverty and economic distress; however, while poverty and PAEG are related, they are also distinct. The choice of indicator meaningfully changes the map of places that will be eligible for any given federal program.

Why the Choice of a Distress Metric Matters

Poverty and prime-age employment rates often diverge

On an individual level, the absence of gainful employment is causally linked to poverty, as breaking free from poverty often proves highly challenging without steady work.[3] Both indicators can signal larger structural issues within a community or region, including weak labor markets, low median wages, and other economic disparities. The complexities of these structural issues mean that prime-age employment and poverty rates are not always aligned and may point to different challenges.

The scatterplot below shows this divergence with the prime-age employment gap for all U.S. counties along the y-axis and poverty rates along the x-axis. Counties with a high PAEG are highlighted in blue, those with a high-poverty rate are highlighted in gold, and those that meet both criteria are highlighted in gray. It clearly shows that most high-poverty counties (87 percent) also have a PAEG of 5 percentage points or greater. But the inverse is less true: only 42 percent of counties with a PAEG gap of five percentage points or more are also high poverty. Among the 1,207 counties with a PAEG of at least five percentage points (nearly 40 percent of all counties), just 507 have a poverty rate above 20 percent. Although high-poverty counties typically experience high PAEGs, the same cannot be said for high-PAEG counties—a majority of high-PAEG counties are not high-poverty.

The important takeaway from the large group of counties in the top left quadrant is that even setting a higher PAEG threshold—10 or 15 percentage points, perhaps—would still pull in many counties that are not high poverty. This raises questions as to whether economic or non-economic factors are driving elevated PAEG levels locally. There’s a strong case to be made that no place with a high PAEG is meeting its full economic potential, but whether the gap is responsive to policy intervention will depend heavily on whether it’s driven by more social/cultural factors (perhaps single-earner households are the norm) or a clear lack of economic opportunity (or something else). Consequently, policy interventions should carefully consider the driving factors for low prime-age employment when other indicators of economic distress are absent.

Pinal County, Arizona, near Phoenix, is one such country from the top left quadrant. Just 67 percent of adults in the country are employed, a gap of 12 percentage points compared to the national rate. Despite this high worklessness, the county’s poverty rate was 11.4 percent in 2021, higher than the typical county but well below the high-poverty definition embraced by the federal government. Forest County, Pennsylvania, is an even more extreme example—just 25 percent of adults are employed, a gap of 53.1 percentage points compared to the national rate. Despite this extremely high worklessness, the county’s poverty rate was 16.7 percent in 2021.

These examples demonstrate that a county’s PAEG can capture a type of distress overlooked by poverty rates alone. In some cases, and especially in rural areas, non-labor sources of income can reduce poverty rates[4] even if employment opportunities are scarce. Other factors such as informal employment[5] and declining male labor force participation can complicate the relationship between prime-age employment and poverty.

Different measurements of distress, different demographics

At the county level, both the PAEG and a high poverty rate capture a subset of communities that are more demographically diverse than the nation as a whole, and also more distressed on several key metrics. Educational attainment rates are equally low for both subsets of counties, and both are losing establishments on average while having significantly fewer high- and moderate-income jobs per capita than the nation.

On other characteristics, however, high-poverty counties diverge more significantly from the nation than those with a high PAEG. Whites comprise 51.4 percent of the population of the average high-poverty county versus 65.3 percent for the average high-PAEG county, above the national share of 59.5 percent. Not only is the poverty rate seven percentage points higher in the average high-poverty county compared to the average county with a high PAEG, but prime-age employment is also slightly lower. The slightly lower overall distress level in the high PAEG cohort is partly due to the larger number of counties in this group.

These statistics do not make the case for one indicator over the other as a more accurate measurement of economic distress, especially since a PAEG gap of five percentage points pulls in a much larger number of counties. Instead, each indicator identifies unique subsets of distressed communities that may require different policy interventions. Many questions about how these two indicators interact will have to be answered with additional analysis. It is enough to emphasize here that each of these indicators is intended to address a distinct economic challenge. While a lack of access to good jobs often operates in tandem with a lack of income for community residents, this is not always the case and different interventions may be needed depending on the goals of a particular program.

The Recompete Eligibility Map is Expansive

Nearly one-third of the U.S. population lives in a Recompete eligible community. This broad eligibility puts the onus on program administrators to identify the most worthy recipients to receive funding from a pilot program that has only received $200 million out of the $1 billion authorized by Congress.

It is important to note that there is an implicit arbitrariness in the selection of any targeting threshold. A PAEG of 5 percentage points or higher as an indicator of economic distress originated in research conducted by the economist Tim Bartik. He arrived at this number by estimating the per capita expenditure needed to halve that gap, which he found to be comparable to what was spent by the Tennessee Valley Authority (TVA), a massive New Deal program that began in the 1930s with the goal of transforming the job markets of distressed rural area.[6] Additionally, a PAEG of 5 percentage points was seen as the right threshold to create a map of distress that was sufficiently expansive to generate broad political support and allow a large number of places to be eligible for participating in the grant competition while still targeting places that were unlikely to improve without support.

As Recompete made its way through Congress, the bar was lowered even more for certain geographies. The eligibility criteria for Core Based Statistical Areas (CBSAs) and Commuting Zones (CZs) was set at 2.5 percentage points. This decision alone expands the map of eligible communities even further. Ultimately, the most important test of a targeting metric is whether it is likely to support the goals and scope of a policy or program. The Recompete model was designed for depth/concentration—providing large grants to a small number of places in order to really move the dial—but it was saddled with an eligibility map more appropriate for breadth/participation.

Unlike other federal place-based policies, Recompete further expands eligibility by allowing communities to qualify for the program at multiple geographic levels. In addition to the generous CBSA and CZ qualification pathway, counties, cities, and groups of census tracts are all potentially eligible for the programs, albeit with a higher PAEG of 5 percentage points and a median household income (MHI) cutoff of no more than $75,000 tacked on, as well. All tribal lands qualify regardless of their PAEG.

In total, 1,420 counties are eligible for the program through 415 eligible CBSAs and 313 eligible CZs. Another 141 counties with a PAEG of at least five percentage points and a low MHI also qualify for the program, which means around half of U.S. counties are eligible. They are joined by 1,891 eligible municipalities and 4,308 census tracts.

The table below shows how a significantly higher number of every geography except census tracts and cities qualify for the Recompete program than would under an OPM-based measure of high poverty.

How expansive are the Recompete geographies compared to other place-based programs?

More communities are eligible for Recompete than almost any other place-based tool for targeting economic distress. The only other federal place-based program that casts as wide of an eligibility net as Recompete is the New Markets Tax Credit (NMTC), under which any census tract qualifies as a low-income community (LIC) if it has a poverty rate of 20 percent or higher or a median family income of 80 percent or less than the area it is benchmarked against. The chart below shows that 125.9 million people live in an LIC compared to 114.6 million in a Recompete-eligible geography. Opportunity Zones, which governors selected from the universe of LIC tracts along with a small number of contiguous tracts, only represent 31.5 million people. The most scoped targeting approach is through EDA-defined persistent-poverty counties, which must have had a poverty rate of 20 percent or higher for 30 years or more and cover just 8 percent of the population.

The Recompete eligibility map captures 31 percent of the country’s population. If a goal of the program is geographic diversity, then the Recompete targeting criteria achieves this. Nearly every state has at least one county that qualifies for the Recompete program, with much more extensive coverage in the western United States than is achieved with poverty-based measures. Not shown on the map are the contiguous groups of urban census tracts and cities that also qualify for Recompete funding, which expands eligibility for the program even further. While the Recompete program is among the few federal initiatives specifically targeted toward distressed communities, its expansive eligibility criteria encompasses areas that may not be considered distressed by other criteria.

In practice, the Economic Development Administration (EDA), which is tasked with administering Recompete, will evaluate applications and presumably award grants to communities that demonstrate a high need. However, this expansive map makes it more difficult to quickly determine which communities should receive limited funds from the pilot program. A well-scoped eligibility map can play an important role in filtering the communities that can apply for funding. The Recompete map does not accomplish this, which places more pressure on later stages of the selection process to grant a finite number of awards to places that truly have a large and meaningful PAEG to close.

Furthermore, the pitfalls of a map of high-poverty communities have been well researched (e.g., high poverty rates around college campuses because students generally don’t earn wages, other populations with little labor income, and the need for fine-grained, longitudinal poverty data). Referring back to the scatterplot above, researchers and program administrators are fairly knowledgeable about the types of places falling into the bottom right quadrant (high poverty rates but low PAEGs); we know a lot less about those in the top left. The PAEG/Recompete map is wholly new, in other words, and it will take a careful review of eligible communities to determine which ones stand to benefit the most from the program.

Conclusion

Neighborhoods, cities, and entire regions without adequate employment opportunities hold back national growth and perpetuate economic hardship for millions of Americans. Low prime-age employment indicates either a lack of jobs in a community or a high number of residents whose skills do not match those jobs, signaling a need for policy interventions that stimulate private-sector job growth and connect residents to employment opportunities.

The Recompete Pilot Program was designed to do exactly this. It marks an important milestone in the formulation of federal place-based policy: a new tool (flexible grants) designed to take on a specific challenge (boosting work) with awards guided by a custom-selected measure (prime age employment gap) intended to align the intervention from diagnosis to prescription. If adequately funded and executed, Recompete has tremendous potential to help the unemployed and those who have exited the labor force get back to work and serve as a model for future federal economic development policies. However, the funds allocated so far are disproportionately small relative to its ambitious goals and geographic breadth.[7]

The current challenge is ensuring the success of a pilot program with limited resources and an expansive map of eligibility. Ultimately, the process of awarding grant money will be competitive, with only a small number of these distressed communities receiving Recompete dollars. In order to deliver on its mission and prove itself as a pilot, Recompete funds must be carefully allocated to communities that demonstrate both a high degree of need and a clear potential for success.

Notes

  1. For further discussion of issues with the OPM see Blank, 2008, Kolesnikova and Liu, 2012, and Meyer and Sullivan, 2012.[]
  2. Benzow and Fikri, 2023[]
  3. Gorman, 2006[]
  4. Lawson, 2014[]
  5. Nightingale and Wandner, 2011[]
  6. Bartik, 2020[]
  7. Bartik and Muro, 2023[]

Community Development Geographic Trends  Spatial Inequality

Related Posts