HUD and the Importance of Data Sharing
Solomon Greene, Principal Deputy Assistant Secretary for Policy Development and Research.
HUD’s Office of Policy Development and Research (PD&R) is charged with building evidence to inform and strengthen housing and community development policy and practice and accelerate progress on HUD’s priorities. We support evidence-based, data-driven, and community-informed policymaking and program improvements across the department and in the broader field. We do this through research and evaluation, data collection and analysis, demonstrations, and innovation initiatives, and by building the capacity of HUD grantees and community partners. Data and partnership are two key ingredients in everything we do here in PD&R, and I recently had the opportunity to discuss both at the Council of Large Public Housing Authorities’ 2023 Housing Is Summit in a plenary session on “Enhancing Federal Data Sharing to Improve Programming for Low-income Families.”
In today’s world, data is everywhere, and its significance cannot be overstated. While data is essential to decision making within federal agencies, its value increases exponentially when it is shared and linked with other data sources. For example, federal data sharing can improve outcomes for low-income families and communities that have faced the greatest barriers to safe, stable, and affordable housing and economic opportunities.
This is because most federal datasets only reveal a single dimension of an individual or household, a single type of “transaction” with a federal agency, or a narrow slice of a community. Combining and linking datasets can paint a fuller picture of an individual, a household, a neighborhood, a city, or a region, offering a more comprehensive understanding of both needs and assets at all these levels.
Forms of Data Sharing
We can think of federal data sharing through (at least) three pathways. First, federal agencies can — and do — share data with each other. Second, HUD and other agencies often share restricted-use data with researchers, either through research projects we fund or at the request of researchers for independently funded projects. Finally, we make data available to local governments, civil society partners, and the public through open data initiatives.
Of course, data sharing isn’t easy. In addition to complex rules and regulations and important privacy procedures and protections, there are a host of challenges to doing it well. This can include limited resources, technical issues, and the siloed nature of any bureaucracy.
Why Data Sharing Matters
There are many important ways that federal data sharing helps HUD advance our mission. By sharing data, HUD can make more informed choices based on a comprehensive and timely understanding of the issues facing HUD-assisted households and communities. Sharing data across agencies can help to identify trends, patterns, and relationships that might not be apparent when looking at data in isolation. This can lead to better policy decisions, more effective program implementation, and ultimately better outcomes for the American people.
Data sharing can also reduce costs and duplication of effort, verify information, and ultimately make it easier for the public to access our services. Data sharing can help HUD and our partner agencies identify and address inefficiencies and redundancies within the federal government, including in the information we request from beneficiaries. In addition, we can better understand where resources are being duplicated or wasted and identify opportunities to streamline processes and improve our programs and services.
Finally, when it comes to sharing data with researchers outside the federal government, we can better leverage our federal data assets to answer pressing research questions. Funding for research and evaluation within federal agencies is limited, and outside researchers help to fill all the knowledge and research gaps we identify each year through our Learning Agenda and budgeting process. In this way, by sharing data with researchers we can “democratize” both our data and the learning process.
What Are We Doing at HUD?
HUD has several data sharing efforts underway involving other federal agencies, often using personal identifiers to link data. For example, we match HUD administrative data on tenants and Federal Housing Administration data on borrowers to large surveys conducted by the U.S. Census Bureau, including the American Housing Survey, the American Community Survey, the Household Pulse Survey, and the Current Population Survey. This allows us to learn more about those tenants and borrowers to understand their needs and improve our services and programs.
To help ensure that HUD programs are efficiently reaching as many eligible families as possible, HUD shares data on who is receiving housing assistance with the Social Security Administration and the U.S. Department of Health and Human Services to verify important eligibility information.
In another area of HUD’s work, disaster response, HUD provides the Federal Emergency Management Agency (FEMA) with data on HUD-assisted tenants within a disaster area to speed up rehousing; FEMA then sends back to HUD data on the location of tenants that are receiving hotel/motel or rent assistance from FEMA, along with location of the displaced tenant.
When we share restricted-use data on HUD-assisted tenants with independent researchers through a data license process, we leverage outside expertise to increase our understanding of HUD programs. A great example of this was when HUD permitted the use of the Moving to Opportunity (MTO) data by Harvard economist Raj Chetty and his team of researchers to match to Internal Revenue Service data to show the long-term impacts of MTO on children as adults. More recently, we’ve been able to combine consultation with public housing agencies and insights gleaned from researchers who have used our data on HUD-assisted tenants to help us assess next steps in the development of Small Area Fair Market rents.
Finally, HUD often combines interagency data sharing, data licensing, and grant funding to answer key research questions. For example, the U.S. Treasury has shared granular data from states and localities that administered the federal Emergency Rental Assistance (ERA) program with HUD. In turn, HUD is sharing these data with researchers to study the ERA program. HUD also recently awarded $2 million to three research teams to use these data to assess the impact of the ERA program on housing stability.
Even as we await the results of that research, HUD is using another data sharing agreement — this time with the Census Bureau, involving microdata from the Pulse Survey — to look at the impact of the ERA program on HUD-assisted households. Using data sharing, we can mine lessons from this first-of-its-kind program that enabled vulnerable households to pay their rent and remain housed through the pandemic to improve the housing delivery system and safety nets going forward.
Where Are We Heading Next?
HUD is committed to continuing to improve and expand our data sharing efforts. This will help to advance our mission to create strong, sustainable, inclusive communities and quality affordable homes. Data sharing is a complex and ever-evolving space, and we look forward to working with our partners and stakeholders to achieve this important goal through collaboration, partnership, and capacity building.
Read about the panel discussion on federal data sharing initiatives here.