HUD’s Research Capacity at 50

HUD’s Research Capacity at 50

Katherine O’Regan, Assistant Secretary for Policy Development and Research.

On September 9, 1965, President Lyndon Johnson signed legislation creating the U.S. Department of Housing and Urban Development as a cabinet-level agency. HUD’s celebration of its 50th anniversary has inspired our department and our Secretary to reflect on the past and the future. As part of this celebration, in October the Office of Policy Development and Research (PD&R) will be publishing HUD at 50: Creating Pathways to Opportunity, a volume that looks both back on HUD’s 50-year history and forward to the policy issues the department will face in the next 50 years.

This anniversary, then, is an appropriate time for PD&R to reflect on its own role at HUD as well as the role of research within the department. In 2008, in response to a congressional request, the National Research Council of the National Academies (NRC) released Rebuilding the Research Capacity at HUD. This report assessed PD&R’s performance and capacity at the time in three core functions: conducting and funding research, performing policy development, and collecting data. Were that report written today, what would it find?

Focusing solely on “conducting and funding research,” the 7 years since the NRC report was published have been incredible for both PD&R and HUD.

Answering big questions

Large-scale studies, employing the rigorous methodologies needed for causal inference, now consume the vast majority of our research funding — a marked change from previous decades. PD&R has initiated seven large-scale, multiyear studies in the past 5 years, each expected to cost more than $5 million: the Evaluation of the Family Self-Sufficiency Program; Pre-Purchase Counseling Demonstration and Impact Evaluation; Family Options Study; Rent Reform Demonstration; Housing Discrimination Study (HDS); Housing Choice Voucher Program Administrative Fee Study; and Assessment of American Indian, Alaska Native, and Native Hawaiian Housing Needs. The first four of these use random assignment to test program impacts. HDS uses the paired testing research methodology pioneered by PD&R in 1977 for measuring housing discrimination nationwide, and the Administrative Fee Study uses random motion sampling techniques.

Combined, these studies represent two-thirds of PD&R’s funding for research and evaluation since 2010. They aim to answer fundamental questions about policy effectiveness and impact, the type needed for long term policy development. As the NRC report stresses, taking this longer view — studying enduring policy questions that bridge administrations — is one of the most important roles that a policy research office within a federal agency, such as PD&R, can assume. Doing so requires the investment of both resources and time, along with the appropriate high-quality methodology, to measure efficacy and impacts. HDS, Family Options, and the Administrative Fee Study have already provided powerful evidence that is informing our policies for fiscal year 2016. Over the next few years, we look forward to more policy-relevant findings from additional projects, such as:

• Rent Reform. Since 1996, a small but growing number of public housing agencies have operated under considerably greater regulatory flexibility as part of the Moving to Work (MTW) demonstration. Many MTW agencies have used this flexibility to experiment with changes in rent calculations, both to reduce administrative burdens and increase work incentives;

• yet we know almost nothing about the impact of these mini-experiments. The Rent Reform evaluation, which MDRC is undertaking at several MTW sites, uses an experimental design to evaluate the most common changes made to rent calculations and income certification for voucher households. This evaluation will provide important insight into a vexing question in the provision of subsidized housing generally and voucher subsidies specifically: how can we better structure subsidies to minimize work disincentives while reducing housing cost burden and providing adequate tenant protection from income volatility? This is exactly the type of enduring question our research should answer to shape better housing policies.

• Pre-purchase housing counseling. Throughout its 50-year history, HUD has supported prepurchase housing counseling as a strategy for managing lender risk while increasing access to credit. Although existing studies have suggested generally positive outcomes for participants and lenders, these studies have not been able to fully address concerns about self-selection; namely, that lower-risk borrowers are selecting into counseling programs rather than becoming lower risk because of the counseling. Using random assignment to address this issue, the Pre-Purchase Counseling Demonstration focuses on potential first-time homebuyers with low, moderate, and middle incomes in 28 cities to answer the question: can housing counseling lower the mortgage delinquency and default rates of borrowers? This evaluation comes at an important time, as the nation continues to recover from the 2008–2011 housing crisis.

The potential impact of such studies is much broader than HUD programs, as demonstrated so well by the recent research of Raj Chetty and colleagues that built on HUD’s Moving To Opportunity demonstration. Their research is just one example of how valuable these studies can be to researchers beyond HUD’s walls, even years after the initial work is completed.

Engaging more broadly to identify those questions

Very much in response to NRC’s findings that HUD’s process for setting its research agenda was too insular and too focused on short-term goals, PD&R completely revamped the process in 2011. Over the course of a year, PD&R reached beyond the agency to engage with a range of stakeholders, including academics, practitioners, advocates, and policymakers, to identify the most policy-relevant and timely research questions in the fields of housing and community development. The result, the Research Roadmap, is our strategic plan for research.

A strategic plan with broad input and vetting is particularly beneficial for a research division of a federal agency, which needs to justify its budget requests and use of public dollars to Congress. The Roadmap process tends to surface those large, enduring policy questions, and it identifies a broader set of stakeholders who want those questions answered. That in itself can affect whether we receive funding from Congress to support the work. We are in the early stages of updating our Research Roadmap and creating a more streamlined, sustainable process for gathering input.

A second means of broadening our research is through our Research Partnerships. Through cooperative agreements, PD&R partners with outside researchers who are addressing questions of high priority to HUD, and where 50 percent or more of the costs are borne externally. This strategy has proven to be a highly productive way to move forward on a larger set of projects, leveraging both the ideas and financial resources of a broader set of stakeholders. To date, we have entered into 21 research partnerships.

Answering smaller questions more quickly

Not all questions require large-scale, experimental studies, and often answers are needed within months, not years. Although most research takes time, the pace of research funded by federal agencies can be glacial. Some smaller, time-sensitive studies can be done in house, but when external capacity or expertise is needed, speed can be a particular challenge for federal agencies. PD&R faces the same challenge.

Our newest addition to the “faster” toolkit is our Multidisciplinary Research Team (MDRT). Conceptually similar to the use of research cadres for research and program monitoring, MDRTs are composed of prequalified research experts who use data from HUD and other sources to answer a set of well-defined policy questions important to HUD. Five such projects have been completed, each in well under a year. We hope to be able to use this vehicle in fiscal year 2016.

We are also building our internal research capacity with a newly created postdoctoral program. Research offices in federal agencies are ideal places for recent graduates of applied doctoral programs to spend a year or two completing their own publications. In our case, postdocs will also work with PD&R staff on policy research using HUD data. In 2015, we brought on board our first class of four postdocs along with several doctoral student interns. This too we expect to continue in fiscal year 2016, budget permitting.

For as slow as government agencies can be, PD&R has shown considerable gains in the 7 years since NRC published its thoughtful report. Those gains were aided tremendously by the existence of the report, a roadmap of its own for PD&R to follow in shaping a more impactful office and against which we can compare our progress, today and 50 years from now.