Approaches to Redesigning a Quality Jobs Tax Incentive

On March 11, 2020, The Pew Charitable Trusts submitted a memo to a Philadelphia working group about a proposal that would replace the city’s job creation tax incentive program. Pew experts recommended that the city prioritize industries that would expand its economy overall, consider alternatives to geographic targeting of benefits, clarify and reassess the award processes, and ensure that the city collects data to monitor compliance and evaluate effectiveness.

To: City of Philadelphia Working Group on Quality Jobs Program
From:  The Pew Charitable Trusts (Alison Wakefield, Thomas Ginsberg, Josh Goodman, Logan Timmerhoff, Shane Benz)
Date: March 11, 2020
Subject: Philadelphia Quality Jobs Program 

The City of Philadelphia’s pilot program to incentivize job creation, the Quality Jobs Program (QJP), contains many of the features recommended by evaluators and the city’s own consultant. Philadelphia is one of the first U.S. cities to codify an evaluation requirement, and this pilot shows the city is committed to making use of evaluations for the benefit of all residents and businesses.

In response to questions from the Quality Jobs Program (QJP) working group, this memo reflects on elements of incentive design and administration that the group may consider as it develops two new business incentives. There are four sections of the memo: (1) key recommendations; (2) other policy considerations; (3) responses to questions from the working group, and (4) additional resources.

1. Key Recommendations

Based on Pew’s research, we recommend the working group consider the following for the QJP:

1. Prioritize industries that would grow the city’s economy overall;
2. Consider alternatives to targeting benefits geographically;
3. Clarify and reassess the award processes; and
4. Ensure the city collects data to monitor compliance and evaluate effectiveness.

Prioritize industries that would grow the city’s economy overall

Research has shown that when states and cities provide incentives to some businesses, other businesses can suffer as a result. For instance, increased hiring at a business receiving an incentive may come at the cost of a business that isn’t receiving an incentive if the overall market for their goods or services is not increasing. The result is that a city or state’s economy does not grow.

This is an important consideration when providing incentives to businesses that have local competitors, such as restaurants and retail, because they tend to recycle dollars locally and are less likely to generate new economic activity.¹

To minimize displacement, many incentives target industries where both the customers and the competitors of incentive recipients are spread nationally or internationally, known as “traded” activity as opposed to “locally serving” or “non-traded”. Award decisions under the Minnesota Job Creation Fund must consider “the projected sales of the business that will be generated from outside the state of Minnesota” and “the effect of financial assistance on industry competitors in Minnesota.” Similarly, Nebraska’s primary incentive program requires businesses in certain economic sectors derive at least 75 percent of their sales from out-of-state to be eligible.

Both the Small Business Quality Jobs (SBQJ) grant and the High Impact Quality Jobs (HIQJ) loan are open to all businesses. An analysis by the Pew Philadelphia Research and Policy Initiative found that a large percentage of Philadelphia’s businesses are in “local” sectors, including its small businesses, with somewhat slower growth.² Incentivizing businesses without considering their sectors or customers may unintentionally hurt other employers competing locally.

• SBJQ and HIQJ are available to businesses in any sector. The programs do not appear to consider whether businesses’ customers or competitors are primarily local or non-local. The working group should consider how this may affect other city businesses and consider prioritizing businesses in traded sectors.

Consider alternatives to targeting benefits geographically

The framework for the QJP indicates a desire to expand economic activity to underserved areas and ensure pathways to employment for disadvantaged Philadelphians. To achieve these goals, the working group is considering targeting SBQJ eligibility to 1) businesses with “economically disadvantaged” owners and 2) businesses located in low-income census tracts. Regarding the second criterion, focusing job creation efforts on specific neighborhoods may not be the most effective strategy to increase economic opportunity for residents of those neighborhoods.

Economic research shows that job creation programs are more cost-effective if they are directed to distressed areas. This research also suggests targeting job creation programs to large service areas such as commuting zones or local labor markets. Encouraging job creation in specific neighborhoods may not be necessary or effective. When a business locates in a distressed neighborhood, it may hire workers from outside that area. If a neighborhood lacks employment opportunities, that doesn’t preclude residents from finding good jobs within their commuting zone.

With these considerations in mind, the working group could consider alternatives to targeting SBQJ to distressed neighborhoods that would increase opportunity for low-income residents.

Target tradeable sectors that offer noncollege workers good jobs or pathways to good jobs

Governments often prioritize attracting and retaining businesses—such as manufacturers, high-tech firms and corporate headquarters—that offer high-paying jobs that help grow the local economy. But, in addition to considering which industries offer the best jobs, the working group could analyze which industries are most likely to offer opportunities for the residents of distressed neighborhoods. A recent Brookings Institution report identified “opportunity industries” as an industry which disproportionately provides “good” or “promising” jobs to workers without college degrees. A “good job” in this analysis aligns with what the working group calls a “quality job” in the QJP framework (i.e., full-time, high wage, with health benefits). A “promising job” is defined by the report’s authors as an entry-level job which enables workers to get a “good” job within 10 years.

According to the report, industries that provide “good” jobs are not necessarily the same industries that provide “promising” jobs—although some do both. Given scarce resources, the working group could target the QJP to industries in tradable sectors that tend to provide good jobs and pathways to good jobs.

Encourage hiring of target residents

Requiring QJP businesses to hire from target populations would be a straightforward way to encourage businesses to hire low-income Philadelphians, residents of distressed neighborhoods, the long-term unemployed, or other possible priority groups. For hiring requirements to work well, however, residents of distressed areas need to have the skills to fill available jobs. Further, businesses may face barriers to hiring from target populations, and incentives may offer insufficient motivation to overcome these barriers.

A “first-source” hiring requirement is one potential solution to these challenges. With this approach, businesses must first consider job applicants from target groups, like those mentioned above, but then may look elsewhere if they cannot find qualified candidates. Some research indicates that first-source requirements tend to be more successful when government agencies take an active role in identifying candidates and when the policies are connected to workforce training initiatives. Cities that currently have some form of first-source hiring requirements include San Francisco, Washington D.C., and Los Angeles.

Address barriers to low income residents filling jobs

Residents with low-income often face barriers that make it harder for them to fill and keep jobs, including limited job skills, a lack of child care options, or difficulty affording transportation to work. Addressing these barriers could enable more workers to fill existing jobs.

For example, a recent report from Pew’s Philadelphia Research and Policy Initiative on the state of transportation in Philadelphia found that while the rail system serves low-income parts of the city relatively well, low-paying jobs are dispersed throughout the region. This dispersion means that many low-wage earners pay additional fees as they transfer multiple times when commuting. The working group could assess whether helping residents of distressed neighborhoods access existing jobs—such as through improved or subsidized transportation options—would be a more effective approach to helping residents access high-quality jobs, rather than trying to draw employers to struggling neighborhoods.

Use other neighborhood-focused programs

While targeting economic development resources to disadvantaged neighborhoods is a common way to try to improve economic outcomes in those neighborhoods, economic development may not be the most effective strategy to do so. It’s possible that working to improve education, public health, or social services in these neighborhoods would have a bigger impact. For example, a recent study by Harvard-based researchers identified four factors correlated with whether neighborhoods offer strong economic opportunities to children. Of these factors, only one was explicitly economic (a low neighborhood poverty levels), while the others related more to social conditions and education (more stable family structure, better school quality, and greater social capital).

Clarify and reassess the award processes

The working group has continued to refine the administration and application procedures for the Quality Jobs Programs. With additional clarifications, and some possible reconsiderations, the working group could increase alignment of program design and goals.

Award Processes

Both QJP incentives appear to be designed as “by-right” programs, meaning that any business that meets the qualifications would receive an award. The HIQJ loan seems to give the Department of Commerce a bit more discretion than SBQJ.

Incentives typically use either an entitlement or discretionary approach, and there are advantages and disadvantages to each. Entitlement programs are simpler to administer because they require less involvement by administrators. Discretionary programs require clear processes and capacity for determining awards, but gives officials leeway to target businesses that may provide a stronger return on investment. Even programs with a degree of discretion set parameters for which businesses would be accepted. These parameters lay out the types of businesses that are eligible or the performance standards businesses must meet to receive incentives, among other criteria.

• SBQJ is designed as an entitlement grant, where the Department of Commerce would approve applications that meet program eligibility requirements. This practice is uncommon with grants and may be worth reconsidering if the working group wants to be able to target awards. That said, such discretion over the smaller of the two programs may be less important to the working group than ease of administration under the entitlement approach. At a minimum, the working group should explore additional parameters, such as those described in sections A and B.
• HIQJ appears to give the department greater discretion but would benefit from more precision in how those decisions are made. For instance, HIQJ qualifications include “high economic impact due to hiring plans, sector/industry, local investment, or other criteria.” The working group would need to provide definitions (such as what sectors or industries qualify) or thresholds (such as for “high economic impact”) so decisionmakers can accurately assess the applications.

The working group could consider alternative designs summarized below, such as using a scoring system, a hybrid scoring and discretionary process, or a tiered approached, for the award process.

Scoring System: Austin’s Business Expansion Portfolio uses extensive scoring systems to determine applicant eligilibty and award amounts. The systems include a variety of economic, regulatory, and community benefit thresholds to which the city assigns a numerical value from 0-100.

Hybrid: The California Competes Tax Credit uses a two-step process to determine which applications to approve. First, the state conducts a quantitative analysis of businesses’ proposals. Then, economic development officials score proposals using a set formula to advance top-scoring businesses to the second phase. In the second phase, economic development officials consider qualitative factors such as “the strategic importance of the business to the state, region, or locality.”

Tier System: San Diego’s Business Incentive Program has four tiers of benefits available to eligible companies. They can earn higher value incentives by meeting “community goals” related to workforce development, educational investment, sustainability, locating in low to moderate income areas, community investment, and transportation/mobility improvements.

Governments may not be able to eliminate instances of rewarding businesses for what they would have done without the incentive, but they can take steps to reduce these occurrences. A possible step could be to exclude activity that is already planned or underway. The Texas Enterprise Fund uses this approach in its pre-award analysis of businesses seeking incentives for moving into the state. A company cannot have signed a lease, purchased land, or hired employees in the state prior to applying for a grant. Doing so would indicate that the business has already chosen Texas and therefore does not need the incentive to aid in its choice.

• The QJP framework should incorporate steps to reduce awards to companies for making investments they already made or planned to make absent the incentive. This could include demonstrating that they have not made the investment prior to applying for QJP.

Company Agreements

Both programs are performance-based, with SBQJ grants paid as jobs are created and HIQJ loans forgiven at the end of five years for each job created. The framework alludes to company agreements, but they are not a clear part of the process. The Quality Jobs Programs include many parameters with which businesses must comply and company agreements would make those requirements explicit.

For instance, the working group would want to clearly define a “new” job to help safeguard against businesses claiming funds for jobs created in the past, reducing employment while claiming credits for new positions, or claiming credits for positions acquired through mergers. Additionally, the agreements could make it clear how wages are calculated, so businesses know whether to exclude benefits (healthcare and pension contributions and paid leave) or not.

Award Frequency and Cost Controls

The working group would also need to decide on the frequency of the awards. Some programs accept applications year-round and others only accept them once a year. The challenge with the latter is that this timing may not work well with a company’s schedule. At least one program holds multiple rounds each year, with a separate cap on the value of credits awarded to make sure the program does not cost more than intended.

Additionally, the working group would want to further clarify how it manages authorizations and claims. For instance, would the amount authorized and claimed be capped each year? Can a business receive funds in future years if the claim cap is reached and they couldn’t claim their award? Further information on caps and how to design them can be found in our report, “How States Use Annual Caps to Control Tax Incentive Costs.”

Ensure the city collects data to monitor compliance and evaluate effectiveness

High-quality incentive data serves dual purposes: compliance monitoring and program evaluation. When program administrators give early thought to data collection, they can determine what data they would need to collect for monitoring and evaluation, that it is in a useable format, and that those in need of the data would have access to it.

Compliance Monitoring and Enforcement

The working group would need to make sure they can verify compliance with program requirements (number of hours worked, job type, wage, provision of health insurance and paid time off, location of the business, ownership, number of jobs created, etc.) and enforce them. The working group may be able to use existing reporting mechanisms to collect some of this information but may need to collect additional information from the business as a condition of receiving the grant.

The working group would also need to define a mechanism to validate key self-reported data. Several states, either through unfortunate experiences or incentive evaluations, have found relying on self-reported data is insufficient, such as a cautionary tale in Hawaii. Philadelphia officials themselves have cited quality issues with self-reported data obtained under its “tax subsidy self-reporting” process.

The amount and frequency of reporting and the documentation required would all contribute to the cost to the business for participating in the program. Some evaluations have found that compliance complexity can reduce the value of an incentive for small businesses.

• The working group should ensure that the reporting requirements are sufficient to verify compliance with program rules. Would aggregate payroll statements and BIRT and Wage Tax receipts indicate the benefits offered, the hours worked or wages paid to individuals hired through these programs, and whether these jobs are new?
• The working group should explore lessons learned about the administration of existing incentive programs from the first incentive evaluation report, such as improving coordination among agencies. This could be supplemented with additional research into how the compliance and enforcement system is currently operating.

Evaluation

In addition to compliance, the working group should consider designing data reporting requirements with an eye toward evaluation. Doing so can ensure that data would provide useful information for analysts and avoid unnecessarily burdening businesses. This would include thinking about the types and format of the data needed to assess the working group’s evaluation questions and metrics. Reviewing the findings from the first city tax incentive evaluation may be useful, as would speaking with other tax incentive evaluators to learn from their experiences about what type of data should be collected and in what format.

Design the incentive to facilitate evaluation

The working group has described this program as a pilot and plans to evaluate it to gauge whether to phase out existing tax incentives. With careful planning, the working group could design the program to facilitate a robust performance evaluation.

For instance, if the working group chose an eligibility scoring model, it could compare companies that received incentives to those that did not. An evaluation of a Michigan program compared the performance of companies that scored just above the eligibility threshold and received incentives to those that scored just below and didn’t—the idea being that these two groups of companies were basically the same, except that some received incentives and others didn’t. The evaluation found that the incentives did make a difference in improving certain measures of company performance.

However, the working group would need to consider such an evaluation and how to compare it to existing programs (such as the Job Creation Tax Credit) because they differ in their goals, business qualifications, and outcome measures.

• Planning for future evaluation would help the working group think about design choices that could facilitate analysis, the types and format of the data needed, and how to assess program results. Getting input from evaluators would be useful for learning about approaches that could be applied to future evaluations.

Monitor and improve targeting over time

No matter how carefully governments determine which areas should be eligible for place-based programs, they still need to revisit these decisions periodically. Conditions in areas change and it may no longer be appropriate to subsidize job creation in certain areas. They also need to monitor where program activity actually takes place. Often, activity occurs in the least-distressed eligible areas because investors see them as most appealing. As a result, programs that are well-targeted to distressed areas on paper may not be in practice.

To assess how well a program is targeted, government officials need data and analysis showing where it’s available, where it’s used, and who benefits. The working group could include this analysis in future evaluations and revise SBQJ’s targeting as needed. By updating the eligible areas at routine intervals, the working group would reduce the likelihood that SBQJ funds would go to areas that are no longer distressed.

• The working group could require the collection of program data that would help asses targeting, require that the existing evaluation process includes an analysis of which of the eligible areas are receiving investment under QJGP, and regularly update the eligible areas.

Data Access

Access to the right data, at the right time, by the right people is essential for monitoring compliance and to conducting a high-quality evaluation. States employ a variety of mechanisms to ensure that policymakers, evaluators, and compliance monitors have access to the data necessary for them to carry out their duties. These include contractual obligations of the incentive recipient to submit data (sometimes paired with penalties for non-compliance); waivers to confidentiality; publication of taxpayer information as a condition of receiving an incentive; or special arrangements like memorandums of understanding or liaisons to facilitate the exchange of relevant project data with administrators and evaluators. Some governments require agencies to formalize data sharing agreements between them, while others have developed technology solutions to share unified databases with incentive information.

Pew has additional resources on data access and availability that can be found here and here.

• In addition to defining the type of data incentive recipients would be required to submit, the working group may also want to make sure it is in a useable format and is accessible to those who need it (administrations and evaluators).

2. Other Considerations

To what extent would the benefits of the incentive stay inside the city?

Since the goal of QJP is to increase access to jobs for Philadelphia residents, the number of incentivized positions filled by residents would affect the program’s outcomes. Labor is mobile within a region and it is reasonable to expect that some positions would be filled by non-residents. However, the working group may wish to explore how much “leakage” (benefits flowing outside the intended geographic area) it could expect and whether there are steps that could reduce or minimize it. A Maryland evaluation of the state’s Enterprise Zone Tax Credit program demonstrates how publicly available data can be used to assess leakage and one of the causes – skills mismatch.

• The working group should be able to understand the potential amount of leakage by viewing Longitudinal Employer Household Dynamics data from the Census Bureau on the percent of workers who both live and work in the city versus those who live elsewhere.

How can QJP be designed to target specific geographies?

Should the working group decide to include a geographic targeting component, a key challenge would be ensuring that the program is successfully directed to distressed areas. Pew’s research and the research of others points to three steps policymakers can take to ensure the program benefits the intended locations:

•  Use quantitative measures to determine which areas should be targeted
• Identify potential measures that reflect underlying distress
• Monitor and improve targeting over time

Use quantitative measures to determine which areas should be targeted

Any program aimed at targeting disadvantaged areas should rely on clear and objective economic, demographic, and social indicators in determining which areas are eligible. Often programs have strayed from their intended geographic focus when subjective factors are used. For example, while Illinois’ Tax Increment Financing (TIF) program is intended to strengthen blighted areas, Chicago has used the program far more heavily in its prosperous downtown because of Illinois statute’s subjective criteria.

To avoid such outcomes, governments often use measures from the U.S. Census Bureau and other official sources. Common measures include poverty rates, unemployment rates, and per capita income. In addition to avoiding the pitfalls of subjective criteria, these measures are updated regularly—which is important because governments should refine the targeting of place-based programs over time.

Identify potential measures that reflect underlying distress

It’s not enough to use quantitative measures; governments need to select appropriate criteria to measure underlying distress. This can be challenging. Measures which at first glance seem to measure distress may not actually do so. For example, governors were able to designate thriving university areas as federal Opportunity Zones because the program used poverty rates as an eligibility criterion—many students earn little income and therefore are nominally in poverty.

Through careful analysis, governments can identify criteria that would effectively measure distress. One example of this analysis is New Jersey’s 2017 update of their Municipal Revitalization Index (MRI), a tool that ranks municipalities from most to least distressed to target a range of programs. The department looked to similar indexes across the country to identify common criteria and assessed which indicators made the most sense for New Jersey.

Philadelphia could conduct similar analysis to determine what criteria to use to target the QJP to neighborhoods. In targeting programs at the neighborhood level, however, special care is required. Much of the relevant data on neighborhoods comes with trade-offs between reliability and timeliness.

In addition to considering quantitative factors, it can be valuable for governments to seek input from the communities the programs are designed to benefit. For example, in selecting its Opportunity Zones, the Colorado Office of Economic Development and International Tourism facilitated conversations with stakeholders at the local and regional levels, and incorporated feedback from the public. These conversations supplemented the data the office assembled.

3. Response to Questions

a) Non-profits

The working group asked whether grant programs typically allow or exclude non-profits from claiming grant benefits. Our scan did not find a definitive answer; a few programs exclude them, but others do not. It was common, however, for programs to exclude some business types or industries, such as retail or utilities.

b) Award amount

Our research has not found a standard or universal way to set the award amount. Studies by Dr. Tim Bartik of the W.E. Upjohn Institute for Employment Research have concluded that larger incentives may bring larger benefits but they also cost more and that these roughly balance out, meaning that there is no clear optimal size of an incentive.^{3 4}

4. Resources

• What Works to Help Manufacturing-Intensive Local Economies? A report by Dr. Tim Bartik of the W.E. Upjohn Institute for Employment Research
• Who Benefits From Economic Development Incentives? How Incentive Effects on Local Incomes and the Income Distribution Vary with Different Assumptions about Incentive Policy and the Local Economy A report by Dr. Tim Bartik of the W.E. Upjohn Institute for Employment Research
• Opportunity Industries: Exploring the industries that concentrate good and promising jobs in metropolitan America, from The Brookings Metropolitan Policy Program
• Michigan Business Development Program Effectiveness Study: a recent evaluation of Michigan’s job creation grant program that has information about alternative strategies to calculate return on investment, methodologies for assessing “but for”, and findings from reviewing job creation grants in other states.
• The Cost of Commuting for Philadelphians • How States Can Gather Better Data for Evaluating Tax Incentives
• Better Incentive Information
•  How States Use Annual Caps to Control Tax Incentive Costs

1. Bartik, Timothy J. 2018. "What Works to Help Manufacturing-Intensive Local Economies?" Upjohn Institute Technical Report No. 18-035. Kalamazoo, MI: W.E. Upjohn Institute for Employment Research.
2. Thomas Ginsberg, “Philadelphia's Midsize and Small Businesses” (The Pew Charitable Trusts, 2020), http://www.pewtrusts.org/philaresearch. (Publication expected in mid-2020)
3. Bartik, Timothy J. 2018. "Who Benefits From Economic Development Incentives? How Incentive Effects on Local Incomes and the Income Distribution Vary with Different Assumptions about Incentive Policy and the Local Economy." Upjohn Institute Technical Report No. 18-034. Kalamazoo, MI: W.E. Upjohn Institute for Employment Research.
4. See Footnote 26, and Appendix D “Assumptions about How Incentive Effects Vary with Scale, Local Characteristics, and National Incentive Competition: Their Implications for the Benefit-Cost Analysis of Incentives”