Are Place-Based Policies a Boon for Everyone?

Despite the widespread implementation of locally targeted "place-based" interventions, their optimal design and effectiveness in addressing regional economic disparities remains open. Proponents argue that they can leverage powerful agglomeration economies and benefit underserved areas. Critics highlight potential pitfalls like inefficiency, gentrification and negative spillovers. Evaluating their effectiveness is challenging due to policy heterogeneity and data limitations, as existing evidence presents a mixed picture: Some programs seem to contribute to job creation and poverty reduction, while others exhibit negligible or even detrimental effects. Ultimately, a balanced approach is recommended. Place-based policies should be considered alongside other interventions within a comprehensive strategy addressing regional disparities.

Consider an economy in which all factors of production can freely move across locations and regions. In this environment, we would not expect to see systematic economic disparities across places.¹ In the U.S., however, socioeconomic conditions vary greatly across regions, cities and even neighborhoods. Segregation — measured in many ways, such as income, educational attainment and race — has been steadily increasing. Moreover, such differences tend to persist over time.

The geographic variation in socioeconomic conditions can be attributed to a wide range of factors. Thus, examining the underlying factors and constraints that influence household and firm decisions regarding where to live and work is crucial.

Limited local access to resources, infrastructure and market opportunities may impose significant economic and financial barriers on residents. These constraints would prevent them from, among other things, following job opportunities to different places. Residents in those areas would, therefore, very likely experience higher unemployment rates than those who are more mobile.

Understanding the nature of the barriers and constraints is essential when formulating effective policy interventions aimed at addressing geographic disparities.

Characterizing Local Economic Development Policies and Incentives

Local economic development policies are economic incentives that include direct financial benefits incentivizing a firm's opening, expansion or retention in certain geographical areas. These policies are designed to selectively offer incentives to individual businesses, aiming to stimulate investments that would not have taken place otherwise.²

Incentives are varied: Some are intended to encourage hiring new workers, while others aim to offset the investment costs of new plants and equipment. The list of incentives includes job creation tax credits (JCTC), payroll withholding tax rebates, investment tax credits, property tax abatements/deferrals, research and development (R&D) tax credits, and subsidized construction costs. Some locations offer specific incentive benefits such as "deal-closing funds," relocation assistance, subsidized worker training, in-state tuition rates (such as immediate in-state residence to employees and their children) and other ad-hoc financing programs (such as discretionary tax incentives and low- or no-interest loans to the companies).

In many cases, local development policies are zone-based, which means that the benefits are targeted to certain geographical areas designated as economic development zones. Various zone-based programs have been introduced over the years in the U.S., with enterprise zones (EZ) and opportunity zones (OZ) standing out as the most common ones:³

• EZs have been used by state and federal governments to target resources to areas typically characterized by high unemployment rates, poverty and other socioeconomic difficulties.
• OZs are part of a similar policy created by the 2017 Tax Cuts and Jobs Act.

Other policies — such as large-scale infrastructure investment programs in targeted areas and community-led economic development initiatives aimed at revitalizing communities — are also generally classified as place-based.⁴

Community-led development initiatives are often enacted by providing tax incentives to real estate developers and other businesses. Examples include the IRS's low-income housing tax credit initiative, offering tax credits to developers constructing affordable housing, the federal New Markets Tax Credit program, facilitating affordable capital for economic developers, and various other initiatives promoting economic development and redevelopment, such as tax increment financing.

Why Invest in Specific Places

What are the underlying reasons for using regionally targeted policies? Several considerations support their design and implementation.

Some households — particularly those facing economic hardship — are less mobile and may find themselves constrained to specific locations. A 2018 paper provides evidence that geographic mobility in the U.S. has been falling considerably. Its authors suggest that the country has evolved into "durable islands of wealth and poverty."⁵

Furthermore, migration by low-skilled and mid-skilled individuals has become less directed toward high-income areas. Higher-skilled individuals, however, still move toward higher-skill/higher-wage areas, making skilled areas more skilled over time. Convergence of incomes across states and MSAs has declined as a result.

While low-income workers would still receive significant wage gains from migrating to high-income areas, the costs of moving to these areas have grown. Various factors may contribute to the high costs associated with mobility:

• Local policies that excessively restrict housing supply — particularly in more productive locations — make housing more expensive and disincentivize household mobility.
• Insurance against local shocks (such as earthquake insurance in high-risk areas) and credit constraints may be prohibitively expensive.
• State-specific licensing requirements have emerged as an additional impediment to interstate migration.⁶
• Some areas may experience declining job opportunities, particularly for low-skilled workers or workers with specific skills in sectors subject to negative shocks (such as the manufacturing sector). Persistent unemployment and the lack of participation in the labor market could reduce human and social capital and, consequently, reemployment in other sectors and regions.

Poverty tends to be geographically concentrated and imperfectly documented. In this context, investing in a place with a high concentration of poverty can reduce the costs of identifying and targeting the poor. Regionally targeted policies may provide in this context more "bang for your buck."

Regionally targeted policy interventions may also help overcome specific market imperfections and correct for externalities that have a geographic dimension. The latter include:

Agglomeration Economies

The concentration of firms and individuals in specific geographic locations generate significant economic advantages, which are generally referred to as agglomeration economies. This clustering promotes positive local externalities and enhances productivity, innovation and growth. In the presence of such powerful economic forces, regionally targeted investments may "jumpstart" the forces underlying agglomeration to the degree that the benefits outweigh the costs.

Regional Spillover Effects

Spending in some areas may generate larger responses than in others. Investing in small cities, for example, may generate positive spillover effects on other surrounding areas, including rural communities within commuting distance.

Human Capital Externalities and Knowledge Spillovers

There is a rationale for place-specific labor subsidies when knowledge spillovers take place across individuals with different skill levels. Market outcomes may lead to too much concentration of high-skill workers (and wage inequality) when such external effects are not considered.

Network Effects

Social and production networks may amplify the effects of place-based policies through network connections, such as when residents help other residents find jobs. However, network effects may also increase mobility costs. For example, when individuals consider moving to other locations, they may anticipate it will be costly to establish a new social network.

Pro-Employment Programs

The 2018 paper "Saving the Heartland: Place-Based Policies in 21st Century America" advocates for the geographical targeting of a specific type of policy: pro-employment programs.⁷ Its analysis is centered on the spatial heterogeneity of nonemployment rates. The authors claim that nonemployment can be reduced more in places where nonemployment is currently high. But why the focus on employment and not on, for instance, income?

First, the authors note that nonemployment rates among prime-age (ages 25-54) men have been increasing at certain locations. There is a positive relationship between initial nonemployment rates and the growth rate of nonemployment rates over the period 1980 to 2015.

Second, several measures of well-being suggest that not working is a far worse outcome than low-income employment. Joblessness is positively correlated with higher levels of dissatisfaction (unhappiness), mental health problems, higher suicide rates, opioid use, disabilities, and physical problems. Moreover, persistent nonemployment may reduce reemployment and the ability of individuals to move to other places for better employment prospects.

Third, the paper finds that the response of employment rates across regions to policy interventions is not homogeneous. Employment elasticities differ across regions: A shock to local labor demand has more impact on the nonemployment rate in places where nonemployment has been historically high. For instance, employment elasticity is 0.05 in Wyoming (the state with the lowest nonemployment rate in 1980 at 6.5 percent) and 0.26 in West Virginia (the state with the highest nonemployment rate in 1980 at 16.5 percent).

In light of the previously outlined considerations, the paper suggests that a geographically targeted pro-employment program would be more effective than a geographically uniform one. The rationale is that $1 spent fighting nonemployment in an area with a high nonemployment rate will do more to reduce nonemployment than $1 spent fighting nonemployment in an area with a low nonemployment rate. Accordingly, pro-employment programs relative to nonemployment benefits should therefore be higher in West Virginia than in Wyoming.

Location Decisions

In the presence of externalities, the observed regional allocation of productive resources may not be optimal. With agglomeration externalities at play, the establishment of a new firm at a given location may result in external productivity benefits for existing firms.

A 2010 paper quantifies such agglomeration spillovers by estimating the impact of the opening of a large manufacturing plant (a "million dollar plant" [MDP]) on the total factor productivity (TFP) of incumbent plants in the same county.⁸ The paper quantifies the effect by contrasting incumbent plants in a given county where the new plant chose to be located (the "winning county") with incumbent plants in the runner-up county ("losing county"). The paper finds that, five years after the MDP opening, TFP of incumbent plants in winning counties is 12 percent higher than TFP of incumbent plants in losing counties. The estimated productivity gains are, however, very heterogeneous:

• The average county-level TFP increase is very large in some instances, small in some other cases and even negative for a nonnegligible number of counties.
• The effects are larger for incumbent plants that share similar labor and technology pools with the new plant.
• Winning counties show a relative increase in skill-adjusted labor costs, meaning the impact on firm profits is smaller than the direct increase in productivity.

Considering these results, a place-based policy that subsidizes the location of this type of firm may enhance efficiency from the locality perspective. However, from an aggregate perspective, the efficiency gains depend on whether or not the benefits of attracting a new plant for the receiving county are the same everywhere. Additionally, the large variability in spillovers may affect the decision to offer the incentive and the magnitude of the subsidies. Since the estimated impact of MDPs is negative 40 percent of the time, risk-averse local governments may be unwilling to provide tax incentives with such potential outcomes.

A 2020 paper examines another specific source of externalities that may justify the implementation of corrective place-specific taxes or transfers: skill heterogeneity and spillovers across different types of workers.⁹ According to this work, larger U.S. cities exhibit higher concentrations of high-skilled workers, potentially contributing to higher wage inequality relative to the efficient outcome. Enhancing geographic efficiency would involve increased redistribution to low-wage cities and reduced skill-based sorting compared to the existing data, ultimately resulting in decreased wage inequality in larger cities. The ideal spatial policy — which would enhance efficiency — should aim to promote a more diverse integration of high-skill and low-skill workers in low-wage cities.

A 2021 paper focuses on the spillover effects observed among workers engaged in different occupations with varying task requirements.¹⁰ Occupations are broadly categorized into those involving "cognitive" and "nonroutine" tasks (CNR occupations) and the rest (non-CNR occupations). The authors note that the geographical distribution of these occupations is not uniform: CNR occupation workers tend to concentrate in large cities, while non-CNR occupation workers are typically found in smaller, often declining cities. The paper shows large productivity spillovers within CNR occupations but negligible spillovers among non-CNR workers. Building upon this evidence, the analysis subsequently assesses the ability of a specific set of spatially targeted policies to achieve an optimal allocation of resources across regions.

Economic Development Anchors

Universities, hospitals and other large employers have the potential to contribute significantly to the economic development and stability of a locality and surrounding areas. Due to their size and stability, these institutions can serve as "anchors" for local economies.

Empirical evidence suggests that higher education institutions (HEIs) function as "anchor institutions," generating localized productivity spillovers within their geographical area of influence.¹¹ These positive externalities are particularly pronounced for industries with direct technological linkages to university research and that are actively employing graduates from the HEI. In such instances, the synergistic interaction between academic knowledge and industry-specific demands demonstrably enhances regional innovation, workforce skill development and overall economic growth.

Most of the evidence, however, is from long-established educational institutions (such as universities). So, it is unclear if creating new educational institutions would generate the same effects or how long it would take to see benefits from those investments.

Quantifying State and Local Economic Incentives

Quantifying the scale of state and local economic incentives is not straightforward. Researchers have developed different methodologies to assess the dollar amount of the benefits.¹² A 2017 report estimates that state and local incentives to firms in export-based industries amounted to approximately $45 billion in 2015.¹³ In present value terms, this figure represents an average of 1.42 percent of business value-added and approximately 30 percent of average state and local business taxes. According to the author's data, the largest incentives — which accounted for about 70 percent of total incentives in 2015 — were led by JCTC, followed closely by property tax abatements. Incentives to businesses have more than tripled since 1990, but they have slowed down in more recent years.

Data from a 2020 paper — which uses a different methodology and covers only state-level initiatives — reveals that incentives vary greatly across states. In 2014, per capita spending on incentives ranged from $5 to $216.¹⁴ Michigan, West Virginia, New York, Vermont and New Hampshire were the top spenders, and incentives were on average about 40 percent of (average) state corporate income tax revenues. States with higher corporate income tax rates were also more likely to offer larger incentives. However, even states with no corporate income tax — such as Nevada, South Dakota, Texas, Washington and Wyoming — still spent (on average) about $44 per capita on incentives. Prominent beneficiaries of incentives often included large, profitable firms in the manufacturing, technology and high-skilled service sectors.

Evaluation and Effectiveness of Local Economic Development Policies

Evaluating the effectiveness of local economic development policies is (as with pretty much every other policy) a very challenging task. In the context of place-based policies, a few specific issues should be taken into consideration.

Causal Effects

Determining the causal effect of a policy is not trivial. To address this issue, several studies rely on clever identification strategies, such as taking advantage of the quasi-experimental nature of the policy design. One commonly used statistical approach is called "regression discontinuity." This approach essentially compares the relative performance (along several dimensions) of two groups of observations (or individuals): a treatment group subject to the policy and a control group.

The analysis focuses on observations that fall into either of the two groups but are similar in every other dimension and very close to a predetermined threshold. This threshold establishes program eligibility and could be geographically determined or defined by the policy itself (such as the poverty rate). Multiple papers, for instance, compare employment growth in areas within an EZ but very close to the boundary (treatment group) to the growth observed in areas just outside the zone (control group).¹⁵

Other work has exploited different types of discontinuities. The analysis of EZs in Texas performed by a 2013 paper relies on the fact that census tracts are automatically designated as EZs if the poverty rate is 20 percent or higher.¹⁶ Thus, the study compares the performance of census tracts at each side of the 20 percent cutoff.

Enterprise Zone Designations

Related to the previous point, the selection of the targeted areas (in the case of EZs, the designation of EZ status) may not be completely exogenous. For instance, if an area was granted EZ status based precisely on its ability to respond to the policy, then the estimates would be subject to endogeneity bias. To proxy (or instrument) for zone applicant success, a 2009 paper evaluating federal EZs, uses the political influence of the zone's representative (representation of the area on the House Ways and Means Committee).¹⁷

Effects on Neighboring Areas

For a full assessment of the impact of a targeted regional policy, it is necessary to determine the extent to which the policy affects neighboring areas ("spillover effects" of the policy). In a geographic context, it is relevant to consider the redistribution of businesses and/or net employment changes across all those areas.

Data Limitations

Data limitations may restrict the ability to perform a careful evaluation of the policy. Targeted areas often don't align neatly with census tracts, ZIP codes or other standard geographical boundaries used for collecting data.

Competing Programs

Another issue that complicates the individual evaluation of the policy is that a variety of different programs typically coexist at the local level at a given point in time. For instance, at some locations, state EZs and empowerment zones (EMZs) overlap. Moreover, some cities or states concomitantly implement other locally targeted policies (for example, incentives to develop or redevelop certain neighborhoods or areas). A clean and pure assessment of each policy becomes very challenging in this setting.

Potential Pitfalls to Consider

When designing local economic development policies, it is also crucial to recognize that geographically targeted benefits may generate unintended effects.

Negative Spillovers

Spillovers from place-based policies could be negative. Subsidizing one region may divert resources from other areas and lead to a net loss in aggregate productivity.

Migration Pattern Distortion

By smoothing income differences across geographic areas, place-based policies can distort migration patterns, raise housing costs in low-income areas (gentrification) and potentially concentrate poverty by inducing the poor to stay in poor places.

Unintended Beneficiaries

A large portion of the gains may go to groups not intended as the primary beneficiaries of the policies. For instance, instead of locals benefitting, landowners or workers from neighboring areas may receive the bulk of the benefits.

Consider a location-specific policy that increases the relative attractiveness of the targeted area in a context with no market failures. Assume initially that individuals are perfectly mobile and that the housing supply is inelastic. The policy may then result in an inflow of workers that would increase the demand for housing in the area and, therefore, housing prices. Thus, landowners (rather than existing residents) would fully capitalize on the benefits of the policies. Imperfect mobility may mean that residents may benefit, but it would then be necessary to assess the benefits of the policy against the costs to non-targeted areas and the deadweight loss of the tax.

Agglomeration Economies

Understanding how agglomeration economies work more globally and quantifying their aggregate effect is not straightforward. The rationale for place-based policies based on agglomeration economies is that overall benefits are larger than overall costs, which is difficult to quantify. While these economies may drive local economic growth, what happens elsewhere? Subsidizing agglomeration economies in one region may divert resources from other areas, leading to a net loss in aggregate productivity. Firms and workers relocating to one area may reduce agglomeration economies in the areas from which they move.

Competition to Lure Businesses

The decentralized implementation of place-based policies may trigger wasteful processes of strategic competition among local jurisdictions to attract businesses.

Benefit Comparison Murkiness

It is not entirely clear which combination of place-based policies can deliver long-run benefits. The design and implementation of place-based policies vary enormously across space and the details matter in explaining their effectiveness.

Empirical Evidence on the Effectiveness of Local Economic Development Policies

To what extent do geographically targeted incentives influence a company's decision to locate, expand or retain operations in the specific area providing these advantages? A pair of papers by Timothy Bartik summarizes existing evidence in this area.¹⁸

One paper distinguishes between firms that would have naturally selected to operate in a specific area and those where incentives influenced the decision (that is, the decision that would not have occurred "but for" the incentive).¹⁹ The work finds that the "but for" cases range from 2 to 25 percent. In other words, the incentive is decisive in steering a location, expansion or job retention decision toward that state or local area in only 2 to 25 percent of the cases.

The question is whether, under these conditions, the benefits are greater than the costs. Ultimately, the answer depends on, among other factors, the job multiplier effects of the policy. With multipliers ranging from 1.5 to 6 (as estimated in related literature) and a "but for" percentage of 12 percent, the benefit-cost ratio would range from 0.4 to 4.

A previously cited report from Bartik also notes that incentive costs can be significantly reduced by restricting refundability and shortening incentive periods.²⁰ Moreover, net benefits may increase substantially when tax incentives are replaced with customized services (such as customized job training).

One of Bartik's papers also highlights the importance of considering not only how the incentive policies are designed but also how they are financed.²¹ The opportunity costs of how incentives are paid for — such as what taxes are increased or what spending is cut — are important. For instance, financing incentives by cutting back on productive services (K-12 education) have substantial negative effects on local incomes and highly regressive effects on income distribution.

A 2021 paper finds that, overall, incentivized establishments experienced lower employment gains than the control group of non-incentivized establishments.²² The outcome, however, differs by firm size: In terms of job creation, small and medium-size firms (less than 250 employees) that are incentivized tend to perform better than larger ones. The type of policy also matters: In states where incentives are distributed more evenly between attracting external businesses and fostering the growth and retention of local businesses, larger incentivized firms demonstrate comparable employment effects to their non-incentivized counterparts.

Evidence on EZs

Several studies have evaluated the performance of EZs implemented in different states.²³ The effectiveness is measured by assessing how designating an area as an EZ affects certain socioeconomic variables, both within the EZ and in surrounding areas. Evaluating state EZ programs as a whole is complex, given that they include different combinations of benefits that are not directly comparable.

A 2015 handbook chapter summarizes the main findings of the academic research.²⁴ The research has focused on changes in a series of variables, including employment (by skill level), unemployment, poverty, household income, wages, rents, home values, vacancy rates, demographic composition and number of establishments or businesses.

In general, the evidence on the effectiveness of these policies is mixed. For instance, while EZs in California and Florida do not seem to increase local employment, EZs increase annual employment growth rates in Texas by 1 to 2 percent, with most of the positive effects concentrated in lower-paying jobs. Similar inconclusive evidence is found concerning the impact of these policies on local poverty and unemployment rates.

In terms of who benefits from the programs, some studies find that housing rents do not increase and there are no changes in local population composition. Other studies find that housing price increases, local population composition changes (because of displacement) and higher-income households (landowners) gain the most.

Some of the work that has also quantified the spillover effects of EZs finds that the local positive effects of the program are often offset by negative effects elsewhere (for example, due to the relocation of establishments). In other words, the program seems to generate negative spillover effects elsewhere, so net benefits are nil.

A 2013 paper performs a thorough examination of EMZs within a general equilibrium framework.²⁵ The paper considers the simultaneous effects of this kind of intervention on wages, employment, population migration, housing prices and housing rents. The authors conclude that, after considering all these effects, EMZs tend to have large positive effects more than compensating the (overall) cost of the program. However, they suggest that due to the lack of precision of their estimates, the conclusions would not necessarily hold for other similar programs.

Several recent papers find that EZs do not have a significant impact on employment in the targeted areas. For instance, a 2023 paper evaluates the impact of New Jersey's Urban EZs on local employment.²⁶ The empirical analysis relies on a synthetic control approach to address the problem of endogeneity in the designation of EZs. The paper shows that areas granted Urban EZ status do not experience significant job growth.

Conclusion

Despite numerous criticisms, place-based policies are the most prevalent interventions used by governments to address local economic disparities across regions. Unfortunately, place-based policies are far from being the silver bullet to solve all sources of economic disparities across regions.

Still, several reasons would justify their use. One of the most frequently used justifications is that place-based policies would jumpstart local economies. The powerful forces of agglomeration economies would imply that even small interventions could have large and sustainable local economic effects over the long run. One word of caution, though: Because agglomeration economies arise organically, it is hard to say what incentives could attract new firms to certain locations.

Evaluating the effectiveness of place-based policies, as with all other policies, is challenging. One problem in evaluating the effectiveness of EZs as a whole is that these policies are very heterogeneous across geography both in terms of their objectives and implementation.

The empirical evidence on the effectiveness of local economic incentives is mixed. Some policies have helped reduce local poverty and increase employment, while others have induced some residents to leave the area, increased rents and housing prices, and generated negative spillover effects in surrounding locations.

In the Richmond Fed's 2016 annual report, we delved into the policy debate between advocates of regionally targeted place-based policies and proponents of people-based or geographically blind policies.²⁷ As we recommended in that report, we advocate for a balanced approach: Place-based and people-based policies should not be seen as mutually exclusive.

Santiago Pinto is a senior economist and policy advisor in the Research Department at the Federal Reserve Bank of Richmond.

To cite this Economic Brief, please use the following format: Pinto, Santiago. (February 2024) "Are Place-Based Policies a Boon for Everyone?" Federal Reserve Bank of Richmond Economic Brief, No. 24-07.

This article may be photocopied or reprinted in its entirety. Please credit the author, source, and the Federal Reserve Bank of Richmond and include the italicized statement below.

Views expressed in this article are those of the author and not necessarily those of the Federal Reserve Bank of Richmond or the Federal Reserve System.