NAHASDA at 25
Todd M. Richardson, General Deputy Assistant Secretary for Policy Development and Research.
This year is an important one for Native American housing. October 26 marks the 25th anniversary of the signing of the Native American Housing and Self-Determination Act of 1996 (NAHASDA), which created the Indian Housing Block Grant (IHBG).
IHBG is the primary source of federal housing funds for more than 500 tribes. One key aspect of IHBG is that it is a block grant formula allocation that all tribes receive. After NAHASDA passed, I was part of the HUD team that participated in negotiations with tribal leaders that resulted in the new regulations — 24 CFR Part 1000. My specific role was to provide technical support to the committee developing the new allocation requirements for the IHBG formula.
On October 8, I had the privilege of participating in a panel discussion at the National Congress of American Indians conference focusing on how the 2020 decennial census might affect various federal programs, including IHBG. Moderating the panel was Joe Garcia, head councilman of Ohkay Owingeh (San Juan), who served as the co-chair of the committee that developed the IHBG formula in 1997.
Other panelists were Nicholas Jones, director and senior advisor of race and ethnic research and outreach at the U.S. Census Bureau’s Population Division, and Gary LaPointe, chief executive officer of the Sicangu Wicoti Awayankapi Corporation, which coordinates and administers housing programs for the Rosebud Sioux Tribal Housing Authority.
One of IHBG’s special features is that it is a formula negotiated among HUD and the tribes using decennial census and American Community Survey data as the key data sources for the Needs portion of the two-part formula. The other component of the formula is Current Assisted Stock; however, for this post, I want to focus on the census data used in the Needs formula because data products from the 2020 decennial census are beginning to being released.
Data from each new census affect the IHBG formula. This post discusses each of these changes since NAHASDA began and how the 2020 data may affect the IHBG formula.
1990 Decennial Census
Both the 1990 and 2000 census used two survey forms: the short form, which was sent to every American household to get a complete population count, and the long form, which was sent to a sample of households that collected more detailed information in categories such as income and housing.
After many weeks of negotiations in 1997, the tribes and HUD agreed to use 1990 census data to develop the original IHBG formula, not because the tribes thought the census was perfect, but because it was the most comprehensive data source on housing needs that was consistently collected for all tribal areas. The data we used primarily came from the 1990 census long form; however, rather than use the standard tabulation of census data, we used a special tabulation that the U.S. Census Bureau created for a 1993 research study that predated NAHASDA.
As a relatively new analyst in PD&R’s program evaluation division, I helped senior staff design and implement the Assessment of Native American Housing Needs and Programs. My job was to work with the U.S. Census Bureau to create a special tabulation of 1990 census data to support this study. Although we modeled the data request after special tabulations we had developed to support the new Comprehensive Housing Affordability Strategy data requirements of the Cranston-Gonzalez National Affordable Housing Act of 1991, we modified the data request in some important ways. First, we expanded the definition of an American Indian and Alaska Native (AIAN) household for the research. Although regular census tabulations defined an AIAN household by the race of the head of household only, our special tabulations defined an AIAN household as a household in which either the head of household or the spouse of the household head identified as AIAN.
Second, we developed a way to use census data to manage complex tribal geography, including areas that were not defined as Reservation/Trust Land, Alaska Native Villages, or Oklahoma Tribal Jurisdictional Areas but were still areas where tribes provided services to their members. We made sure to use “balance of county” estimates to analyze these near-tribal areas.
We used these special tabulations of 1990 census data for the negotiated rulemaking. The two tribal leaders co-chairing our committee — Jack Sawyers, who represented the Utah Paiute Tribe, and Joe Garcia — led the process in a very deliberate way. After the group agreed on its goals within the statutory requirements, they asked me to explain what data I thought could be used in the formula. This process eventually led to the selection of census data as the foundation of the formula. However, because they recognized that census data were imperfect, they allowed tribes to challenge the census data if they collected their own data in a similar way.
The 1990 census allowed individuals to identify only as a single race. The 2000 census permitted individuals to designate one or more races with which they identified. This change created a situation in which some American Indians and Alaska Natives continued to identify as a single race, but many others identified as AIAN plus another race.
On this topic, Congress, in its annual appropriations bills, has required HUD to run the IHBG formula in one of two ways: for AIAN alone, or for AIAN alone or with one additional race. The method that results in the larger allocation is the one used for a tribe, and a pro-rata reduction brings the formula back into line with appropriation.
The 2010 census eliminated the long form, replacing it with the annual American Community Survey (ACS). Because all but one of the Need variables were derived from the long form, HUD had to engage in another round of negotiations with tribes to include data from the ACS.
Initially, those negotiations did not reach consensus on using the ACS. Some tribes felt that the small sample size of the ACS did not accurately capture their housing needs – particularly in remote areas. Instead, the committee formed a study group made up of tribal data experts and HUD representatives. The study group was tasked with compiling all available data sources, including tribal enrollment data, that could be considered by the negotiated rulemaking committee as an alternative to the ACS data. The committee agreed that if it could not find an alternative to the ACS, it would adopt the ACS starting in FY2018. Ultimately, the committee accepted the ACS data into the formula.
The negotiators did reach consensus on a way to reduce potential volatility in the allocations resulting when new data sources are introduced into the formula, such as the ACS. This ensured that no single Tribe would lose too much funding in any given year as a result of the introduction of ACS data into the IHBG formula, and no single Tribe would gain too much in funding.
After assessing the 2010 Decennial Census, the Census Bureau concluded that there was likely an undercount in Reservations/Trust Land. The Committee agreed to give a corresponding bump up in total AIAN population to Tribes whose formula area included Reservations/Trust Land and did the same for geographies in remote Alaska.
It was challenging and it took years. But we were able to adopt the ACS data into the IHBG formula while also ensuring that we tried to address the concerns of some of the tribal representatives on the negotiated rulemaking committee through volatility control, adjusting for undercounts in certain tribal areas and remote Alaska, and more.
So how will the 2020 census affect the IHBG formula? Our current expectation is the IHBG formula would likely not start being impacted by the 2020 Census until FY 2024 at the earliest, but we’ll have to assess the data first and make that determination when the time is right. This will give the tribes and HUD some time to analyze the new data. Before updating the formula, HUD will determine whether there was an undercount (as happened in the 2010 census) and how new U.S. Census Bureau policies to improve privacy protections might affect the data.
When the 2020 Census data are incorporated into the formula, we will apply the volatility control adjustment similar to the process we used for the implementation of the 2010 Decennial Census/ACS discussed above.
NAHASDA is 25 years old and has proven to be successful. The program — and its allocation formula — also continue to evolve as circumstances change.
Working with tribal leaders and representatives to develop the IHBG allocation formula was one of the highlights of my career at HUD and I was able to make many friends. I look forward to seeing what the next 25 years of NAHASDA will bring.