Fifty Years of Formula Fairness
Contributors: Harold Bunce (PD&R: 1975 to 2007); Todd Richardson (PD&R: 1991 to 1997; 2000 to present); Rob Collinson (PD&R: 2009 to 2016); Greg Miller (PD&R: 2022 to present)
“While working for the Office of Policy Development and Research (PD&R) in my early 20s, one of my first assignments was to assess how well the Community Development Block Grant (CDBG) formula targets community development needs.”
Each author of this article could have made this statement at some point in his career. For Harold Bunce, it would have been in 1976; for Todd Richardson, in 1995; for Rob Collinson, in 2014; and for Greg Miller, in 2023.
Although the CDBG formula, which in fiscal year 2023 allocated $3.3 billion to 1,239 jurisdictions, may be HUD’s best-known formula for allocating federal funds, it is far from HUD’s only allocation formula. Roughly 87 percent of HUD’s annual discretionary budget authority, including housing contract assistance renewals, is allocated through funding formulas. Throughout PD&R’s 50-year history, our collective experience with studying the CDBG formula has played a critical role in the creation and operation of many of HUD’s funding formulas.
The CDBG study has served as something of a rite of passage for us. Collectively, we have worked on many HUD formulas, but we learned our craft through our research on the CDBG formula. The successful practice of that craft depends on understanding why the federal government uses funding formulas. Certainly, using formulas is usually a faster and more efficient way to get funding to grantees than other methods, such as competitions. More importantly, both Congress, which authorizes the funding, and grantees must believe that the method used to allocate the funds is impartial and equitable.
The CDBG formula is impartial, but does it equitably distribute funds according to the objectives of the grant? Our approaches to answering this question — how we addressed the intersection of allocation formulas and policy — is how we learned our craft with the CDBG formula.
CDBG was established to “promote decent housing and suitable living environments and expand economic opportunities,” and CDBG funds can be used for numerous activities toward those ends. To target the funds fairly requires agreeing on what constitutes community development need. Harold Bunce established the standard by creating a measuring stick — a “community development needs index” — from which to measure the extent to which the CDBG formula succeeds at allocating funds fairly. Each of our six subsequent studies have done the same.
- Bunce’s 1976 study spent considerable time combining several city distress indicators, such as poverty and job loss, to develop a community development needs index that provided a framework for evaluating different CDBG formulas based on readily available decennial census data. That study used 1970 census data to show that the eventually implemented (and now current) dual formula was more equitable than the formula that was then in use and gave backing for Congress to implement the dual formula system in 1978. Notably, the addition of a second formula (formula B) accounting for pre-1940 housing and slow population growth or population loss (growth lag) captured the needs of older cities losing population — an important component of community development need not captured by the original formula (formula A), which included population, poverty, and overcrowding.
The new dual formula had the following variables:
Formula A: Population, Poverty, Overcrowding
Formula B: Pre-1940 Housing, Poverty, and Growth Lag (growth lag allocates more funding if the entitlement’s population growth is slower than the growth in population for all metropolitan areas since 1960.)
Entitlement cities and counties have their grant calculated using each method and are awarded the larger of the two results, with a pro-rata reduction to bring the overall amount allocated back in line with the amount appropriated.
Bunce’s 1979 study with Robert Goldberg demonstrated the impact of the congressional change to the formula (still using 1970 census data). Harold conducted a follow-up study with Sue Neal and John Gardner in 1983, when 1980 decennial census data became available. This study found that with 1980 data the dual formula still targeted to the more needy entitlement cities such as Newark, Buffalo, and Detroit even though there was a slight decline in targeting relative to its targeting with 1970 data. The 1983 study also highlighted some funding anomalies that resulted in similarly needy communities receiving significantly different per capita funding (a horizontal equity issue). Some older, declining suburban cities that benefited from the formula’s housing age and growth lag variables were not always distressed in terms of poverty and income.
Richardson’s first study, with Kevin Neary in 1995 (using 1990 decennial census data), found that the same horizontal equity problems noted in Bunce’s 1983 study had worsened. Richardson and Neary offered a new formula that maintained the dual formula structure but corrected some of its problems by reducing the amount of funds allocated based on population and growth lag, removed full-time college students from the poverty count, and tied the age of housing and growth lag factors in formula B to income so that well-off areas that were old and slow growing would no longer be overfunded relative to their need. This alternative formula sought to improve the targeting within formula A and formula B but mostly did not correct the growing imbalance between them.
Richardson’s follow-up study in 2005 (using 2000 decennial census data) reinforced that the equity issues identified in the 1983 and 1995 studies continued to worsen — in particular, the growing disparity between formula A and formula B awardees with similar needs — and presented four alternative formulas that would improve targeting to community development need. Richardson discussed the tradeoffs those options entail in terms of formula simplicity, amount of funds reallocated, and the types of community development need that would be prioritized. To address the imbalance between formulas A and B and between entitlements and nonentitlements, one of Richardson’s options was creating a single formula for all. He also introduced the option to use another factor to measure fiscal distress — the per capita income ratio between a city or county and its metropolitan area — rather than growth lag.
Collinson’s 2014 study, the first to use data from both the 2010 decennial census and the American Community Survey, provided insight into the changing needs of specific communities, showing that many previously distressed communities were now less distressed. His study was the first to construct a needs index across three decades to demonstrate the formula’s decreasing ability to target to need over time. Collinson offers several alternatives, including tweaking the dual formula in a way similar to that suggested in the 1995 study, creating a single formula as the 2005 study suggested, and a new idea that factors in neighborhood distress by having variables that count only needy people in high-need neighborhoods (that is, neighborhoods with high rates of poverty or housing vacancy).
Miller’s current study builds on previous research and will offer a new approach for the 2020s and beyond.
Only Bunce’s research resulted in changes to the CDBG formula. Because the CDBG formula is specified in detail in statute, changing it requires an act of Congress, and no changes have been made since 1981.
In the meantime, the knowledge and experience each of us built by studying the CDBG formula has had a significant impact on several other formulas:
Bunce leveraged his CDBG needs experience to join Office of Economic Development staffers Randall Scheessele and Sue Neal to develop a definition of underserved neighborhoods (census tracts) that has been used in numerous studies of mortgage originations and to set affordable housing goals for Fannie Mae and Freddie Mac. Robert (Goldberg) Benjamin, Bunce’s coauthor for the 1979 study, developed the Public Housing Capital Fund formula currently in use.
Richardson’s experience directly informed his work supporting the development of the Indian Housing Block Grant, CDBG-Disaster Recovery, Neighborhood Stabilization Program (NSP) 1, Support Act, CDBG-Coronavirus (CDBG-CV), Emergency Solutions Grant-Coronavirus (ESG-CV), and the Emergency Housing Voucher programs as well as the recent Housing Choice Voucher fair share formula, among others.
Collinson developed the NSP 3 formula.
We are looking forward to what Greg will contribute to this work.
Improving the CDBG formula presents a twofold challenge. First, although each of these studies shows that equity under the current CDBG formula has considerably worsened, these studies also show that, on average, the formula still provides more funding per capita to higher-need communities than to lower-need communities. In other words, the formula is a bit like a 50-year-old car that still runs: a newer car would be better, but this one still gets us close to our destination most of the time.
The second issue is that over the course of CDBG’s nearly 50-year history, the combination of a growing population and inflation have reduced its buying power. A 2006 U.S. Government Accountability Office report noted that between 1978 and 2006, “real per capita funding for the CDBG has declined by almost three-quarters from about $48 to about $13 per capita.” That trend has continued to the present day.
Although the loss of buying power is a reason to advocate for improving the targeting of the funds, it also makes changing the formula more difficult because no community wants its allocation reduced. A formula is easiest to change when funding amounts are increasing. Changing the CDBG formula, if the opportunity arose, would require coordinating the formula change in the authorizing statute with the funding change in the annual appropriation. These two laws would need to be linked so that a change in the formula would be associated with an increase in appropriation. The PD&R studies offer a roadmap on what Congress could use for a new formula if that moment occurs.
Although the base formula for the core CDBG program has remained unchanged since Bunce’s research, we hope that more recent research is carefully considered for other programs that rely on the CDBG formula (for example, the Emergency Solutions Grant (ESG) allocation, which is intended for homeless funding, uses the CDBG formula for allocating resources) and supplemental appropriations (most recently, the U.S. Department of the Treasury’s American Rescue Plan Coronavirus State and Local Fiscal Recovery Funds program for metropolitan cities used the CDBG formula to allocate $45.6 billion in 2021). Congress did allow HUD to modify the CDBG formula for funds allocated under the Coronavirus Aid, Relief, and Economic Security Act for CDBG-CV and ESG-CV, and entirely new formulas were developed for NSP and CDBG–Disaster Recovery to reflect the specific emergency needs of these programs that also use some of the CDBG statutory authorities.
The allocation of funds for any government program is a policy question, and it is the role of data-oriented policy specialists to provide accurate and reliable evaluations and recommendations to inform the formulas that guide allocation. PD&R has filled this role through various HUD programs over its 50-year history.
H.L. Bunce. 1976. “An evaluation of the Community Development Block Grant formula.” Washington, DC: U.S. Department of Housing and Urban Development.
H.L. Bunce, and R.L. Goldberg. 1979. “City need and community development funding.” Washington, DC: U.S. Department of Housing and Urban Development.
H.L. Bunce, S.G. Neal, and J.L. Gardner. 1983. “Effects of the 1980 census on community development funding.” Washington, DC: U.S. Department of Housing and Urban Development.
K. Neary and T. Richardson. 1995. “Effect of the 1990 census on CDBG program funding.” Washington, DC: U.S. Department of Housing and Urban Development.
T. Richardson. 2005. “CDBG targeting to community development need.” Washington, DC: U.S. Department of Housing and Urban Development.
R. Collinson. 2014. “Assessing the Allocation of CDBG to Community Development Need,” Housing Policy Debate 24:1, 91–118, DOI: 10.1080/10511482.2013.854945
Richardson notes that Kevin Neary, who died in 2018, was his mentor and longtime supervisor during his early years at the Office of Policy Development and Research and taught him everything he knows about formula allocations. ×