RENT REFORM DEMONSTRATION
SUPPORTING STATEMENT – PART A
THE RENT REFORM DEMONSTRATION
SUPPORTING STATEMENT – PART A
OMB CLEARANCE PACKAGE
TASK ORDER 2
June 22, 2015
Submitted to:
U.S. Department of Housing and Urban Development
Part A: Justification
This supporting statement provides information on the data collection activities associated with the Rent Reform Demonstration administered by the U.S. Department of Housing and Urban Development (HUD). The alternative rent policy was developed by a design team, which includes: MDRC and its subcontractors at the Urban Institute, Quadel Consulting, the Bronner Group, and independent and academic consultants; HUD; and participating housing agencies. MDRC is leading the evaluation effort covered under Task Orders 1 and 2.
This information collection request is a revision of existing collection OMB #2528-0306. Revisions include 1) interviews with PHA staff to understand early implementation of the alternative rent model, 2) interviews with study participants to understand their experience with the alternative rent model, 3) administrative records that will be collected to assess the impacts of the intervention being tested, and 4) cost study data collection for the research activities under Rent Reform Demonstration, Task Order 2 (TO2).
A final data collection request, expected in 2018, will cover longer-term research activities under TO2.
A.1 Circumstances Necessitating Data Collection
The primary goal of the project is to test the effects of an alternative rent policy on voucher holders and housing authorities that issue them. HUD authorized the evaluation of the Rent Reform options following the Office of Policy Development and Research statement of policy objectives and initiatives for 2011 (“2011 report”) (http://hud.gov/offices/cfo/reports/2011/cjs/Research_and_Technology_2011.pdf).
Under Title V of the Housing and Urban Development Act of 1970, as amended, the Secretary is authorized and directed to conduct demonstration projects, evaluations, and research that relate to the mission of the Department. The policy objectives for 2011 set forth the Office’s proposal to support the government-wide initiative to strengthen program evaluation by underwriting rigorous evaluations of three demonstrations, including the Rent Reform option demonstration (2011 report). The Office’s proposed support would supplement funding from the Transformation Initiative.
The report referenced the approval by the Office of Management and Budget of specific evaluation proposals for 2011, which included the study of Rent Reform options. Further, it underscored HUD’s commitment to “strong, independent evaluations that can inform policy and program management decisions” (2011 report, page 0-12).
The evaluation of Rent Reform strategies was designed to test, and “systematically observe,” methods for creating incentives, and removing related disincentives, for tenants to increase their earnings (2011 report, page 0-13).
A.1.1 Background and Policy Context
Overview
Since the passage of the Brooke Amendment in 1969, rents in subsidized housing have been directly tied to income. A variety of stakeholders have advocated reform of the rent system, some emphasizing its adverse impacts on beneficiaries (and eligible non-beneficiaries) and others emphasizing its impacts on public housing authorities (PHAs) and their finances. Rent reform has been elusive, however, because moving to a new system involves some fundamental tradeoffs around which agreement has been hard to achieve. For example, simplifying the rent structure may make it more difficult to ensure that tenants with the greatest need receive the most assistance. At the same time, offering deep subsidies for an unlimited term makes it difficult to serve equally needy families on waiting lists -- given a fixed appropriation level. Also, the advantages of standards and protections built into a common federal approach must be weighed against the benefits of allowing local agencies to set rent rules based on local conditions. And finding the right balance in the mission of housing assistance between a strict focus on providing decent affordable housing versus other objectives, such as family self-sufficiency, is another source of tension among stakeholders.
In September 2012, HUD commissioned a national evaluation of alternatives to the current rent structure for households receiving vouchers from Moving-to-Work (MTW) PHAs, which have the statutory authority to change rent rules. While the current income-based system protects most tenants against excessively burdensome rents (in general, the portion of rent that the voucher holder is responsible for is 30 percent of the household’s adjusted income), it may also encourage underreporting of income and a reduction in work effort leading to lower earned income.
The key goals considered in the design of the alternative policy include: increasing work effort and earnings for voucher families, simplifying administration of vouchers, and cost-neutrality for Moving-to-Work housing authorities via administrative efficiencies or restructuring (simplification) – all while sustaining the voucher program’s important role in reducing homelessness and serving more families with housing assistance.
A. Origins and features of the current rent system
Initially, public housing construction was funded through federally guaranteed, tax-free bonds while rent revenues from tenants covered public housing operations. In the 1960s, however, operating costs began to rise, and at the same time public housing authorities were pressured to serve even lower-income households. These changes contributed to growing financial stress both among the residents of public housing developments and the PHAs that managed them. In response, Congress established federal operating subsidies for PHAs and new rules were instituted governing the amount households could be charged in order to protect them from having to pay excessive rents. The so-called “Brooke amendments” limited tenant payments to a specified percentage of their adjusted household incomes. In the 1970s, when the Section 8 program was enacted, it also utilized the same percent-of-income rent policy for households receiving and utilizing rental certificates, which later become the Housing Choice Voucher (HCV) program.
The current HCV rent system is complicated. For example, (a) PHAs may require minimum rents (except in cases of hardship), (b) payment standards allow HCV recipients to pay more than 30 percent of their incomes for rent, and (c) a variety of other rent system variations apply to a small number of Moving to Work (MTW) demonstration agencies and some HOPE VI developments. These complications aside, the vast majority of public housing residents and HCV beneficiaries operate under a percent-of-income rent system.
The rent system for voucher holders essentially limits their rent payments to no more than 30 percent of household income adjusted for deductions and disregards. However, households are allowed to pay up to 40 percent of their incomes for housing units with rents above the established payment standards for their area (although the subsidy does not cover gross rent above the applicable payment standard). Since enactment of the Quality Housing and Work Responsibility Act (QHWRA) in 1998, PHAs can set minimum rents of up to $50.
Neither public housing nor vouchers are entitlements. Hence, only a fraction of eligible households receive assistance while other, equally eligible households do not. In recent years, with no expansion in housing assistance, the only way that a new household can be assisted is through turnover. Thus, the longer current recipients receive assistance, the longer that assistance is unavailable to others. Many PHAs report long waiting lists, with the waiting period exceeding three to four years for the large PHAs (for example, in 2008, the reported waiting period for vouchers in cities like Chicago, New York, and San Diego exceeded 50 months). There is tremendous need, therefore, for reforms that can help stretch housing assistance dollars further.
B. Rationale for and criticisms of the current rent system
Protecting low and very-low income households from paying excessive rents is the primary rationale for, and benefit of, the percent-of-income system. It is staunchly defended on these grounds by advocates of low-income housing. Previous efforts to modify the system, such as by allowing PHAs to establish minimum rents, have required them also to establish policies exempting those for whom minimum rents would pose severe hardship.
The percent-of-income system has been criticized by public housing industry groups and others as allegedly having one or another unintended negative consequence on issues such as: (a) labor force participation, (b) resident/beneficiary turnover, which can affect horizontal equity, (c) tenant invasiveness, (d) reported income, (e) inclusion on the lease of additional working adults, (f) housing agency finances, and (g) housing agency administrative complexities and costs. PHAs have reported that the current system sometimes (a) confuses voucher applicants and recipients with respect to what rents they are expected to pay from year to year, (b) has adverse effects on PHA staff morale as a result of having to (invasively) collect and verify incomes, and (c) makes it difficult to justify to stakeholders a different approach than that used by other federal programs such as TANF, Medicaid, and SNAP. Many of these criticisms are documented in HUD’s Study of Rents and Rent Flexibility (2010).
A variety of stakeholders have advocated reform of the rent system, some emphasizing its adverse impacts on beneficiaries (and eligible non-beneficiaries) and others emphasizing its impacts on PHAs and their finances. Policy reform has been elusive, however, because moving to a new system involves some fundamental tradeoffs around which agreement is hard to achieve. For example, simplifying the rent structure may make it more difficult to ensure that tenants with the greatest need receive the most assistance. At the same time, offering deep subsidies for an unlimited term makes it difficult to serve equally needy families on waiting lists -- given an annual appropriation that is set by Congress each year. Also, the advantages of standards and protections built into a common federal approach must be weighed against the benefits of allowing local agencies to set rent rules based on local conditions. And finding the right balance in the mission of housing assistance between a strict focus on providing decent affordable housing versus other objectives, such as family self-sufficiency (FSS), is another source of tension among stakeholders.
C. Ideas for reforming the current rent system
Alternative rent systems that have been contemplated and, in some instances, already adopted by MTW agencies include (a) modifications to the percent-of-income calculation (either to income adjustments involving deductions or disregards, or to the percentage itself such as a stepped/tiered percentage system), (b) modifications to the payment standard system, (c) various versions of stepped/tiered subsidies, such as a stepped-down subsidy with a subsidy floor, (d) a flat subsidy, (e) establishing ceiling rents, (f) varying the amount of the minimum rent, and (g) various hybrid models such as applying a flat subsidy until a household reaches an income threshold and then applying a percent-of-income rent.
Discussion of rent reform also has considered the use of time/term limits either in conjunction with other rent reforms or FSS initiatives, or not. Yet other types of options have also been contemplated, such as the proposals (in Section 8 Voucher Reform Act (SEVRA) and in successor bills) involving a three-year recertification for families on fixed incomes where rent is based on the prior-year’s income except at initial eligibility, or involving limiting interim income recertifications for income changes of more than 10 percent. Finally, still other models include variations on FSS escrow accounts such as by limiting them to earnings above a minimum wage level to encourage more substantial wage increases that could feasibly result in families leaving housing assistance.
With respect to program simplification to deal with the issues of administrative burden and costs, a variety of ideas have been suggested. These include less frequent collection and verification of income given the fact that rent calculations are time consuming and complex (even more so when HCV recipients’ incomes increase or decrease between annual recertifications), and reconsidering allowances for medical deductions, earned income disregards, disabled care deductions, etc. Recent legislative proposals – the SEVRA, and its successor, the Section 8 Savings Act (SESA), and now the Affordable Housing and Self-Sufficiency Improvement Act of 2012 (AHSSIA) – have sought changes in both of these dimensions of rent reform for Section 8. They also have included provisions for more rigorous evaluation.
At least one MTW agency proposed changing requirements for Housing Quality Standards (HQS) inspections and relying, instead, on city building code inspections to minimize the costs involved. Some housing professionals also advocate having different administrative requirements for smaller PHAs than for larger ones.
D. Lessons from MTW agencies and the rent flexibility study
In terms of actual experimentation with respect to rent reform, MTW demonstration agencies have modified aspects of the rent system, although these generally have not been tried in ways that allowed for rigorous impact assessment. In some cases, such reforms encompass changes in rent formula; in others, they are limited to procedural reforms that could reduce administrative costs -- such as extending the time between recertifications.
The research undertaken as part of HUD’s Study of Rents and Rent Flexibility suggests a few broad observations:
HA officials have different ideas about how to deal administratively with hardship, say with respect to (a) whether households must request hardship consideration or hardship reviews should be done proactively and routinely, or (b) whether hardship reviews should be performed by agency staff or others. Each option has unknown cost implications.
With respect to a different aspect of administrative simplification, some PHA staff are skeptical that reforms can produce cost savings while introducing multiple program variations with different rules or procedures. Similarly, they suggest that agencies may have to compensate for reducing one benefit (such as increasing the percent of income charged for rent based on number of years of voucher receipt) with other benefits such as enhanced case management or second-stage reviews of initial hardship request denials.
With respect to yet another potential impact of rent reform, the hypothesis has been that a percent-of-income, as opposed to a flat, subsidy system serves as a disincentive to marriage or cohabitation of multiple wage earners, resulting in mother-only households. The rent flexibility study suggests that rent reform may not have as large an effect on marriage and cohabitation as some have hypothesized.
A.1.2 Overview of the Rent Reform Demonstration
Features of the Alternative Rent Model
The MDRC team has consulted extensively with HUD program and research staff, advocacy organizations, Moving to Work (MTW) public housing agencies (PHAs), and consulting organizations that support the activities of PHAs. Following on this process, an alternative rent model is taking the shape of a framework with several fundamental features while leaving some room for PHA discretion in adapting those features to local conditions.
The alternative rent policy will apply only to HCV recipients1 and include the following key features:
Simplifying the calculation of the household’s total tenant payment (TTP) and subsidy amount by:
Eliminating deductions and allowances,
Changing the percent of income that a household pays for its share, from 30 percent of adjusted income to 28 percent of gross income,
Ignoring a household’s income from assets when the total value of its assets is $25,000 or less, and
Simplifying the policy for determining utility allowances.
Using retrospective income in setting a household’s TTP and housing subsidy (to discourage intentional reductions in income)
Establishing a minimum TTP of at least $50 and requiring that all households pay a minimum amount of rent directly to the landlord, to mirror the landlord-tenant relationship in the non-subsidized rental market
Conducting income recertifications triennially rather than annually, so that earnings gains do not increase TTP for three years (thus creating a strong work incentive by eliminating, for an extended period, the implicit housing-subsidy-related “tax” on increased earnings )
Limiting household-requested interim recertifications to a maximum of one per year, to protect households when their income drops while limiting the burden to the housing agency
Establishing a suitable hardship policy that identifies a standard set of hardship conditions and remedies to protect households from excessive rent burdens.
To increase the likelihood that the alternative rent policy encourages tenants to increase their work efforts, it will be important to inform them clearly and periodically of the implicit incentives associated with not having to report any earnings gains for three years. They must also understand the safeguards in place to protect them from excessive rent burdens if their incomes fall. The MDRC team, as part of its technical assistance role, will help housing authorities develop appropriate materials and strategies for communicating these incentives and safeguards.
Local variations allowed
As previously indicated, the housing agencies that are participating in the demonstration helped to develop and support this common framework. However, they also saw a need for some local adaptions of the model in response to local considerations. For example, the need to set the minimum TTP at different levels. The process for determining hardship remedies (although not the conditions defining a hardship or the remedies themselves) may also vary. The design team attempted to strike an effective balance between the extent of standardization that would be required in a national policy and the need to permit some local flexibility that may also be reasonable to include within a national policy.
Eligibility
The alternative rent policy will apply only to HCV recipients. Eligible sample members will include voucher holders with vouchers that are administered under the (MTW) demonstration. Non-MTW Vouchers (i.e., Veterans Assisted Special Housing, Moderate Rehabilitation, and Shelter Plus Care), Enhanced Vouchers, and Project-Based Vouchers are excluded from the study. Households that have ported out are excluded. Additionally, the study is focused on work-able populations and will not include elderly households, disabled households, households that will become elderly during the course of the long-term study, or households where at least one member does not have legal status in the U.S. Households receiving a child care deduction at the time of random assignment will also be excluded from the study. Households currently participating in the PHA’s Family Self-Sufficiency, homeownership programs, or any of the PHA’s special programs with partner agencies will not be included in the study. Lastly, households who have 0 HAP at the time of random assignment will not be included in the study.
Evaluation design and components
The demonstration will use a randomized controlled experiment to compare the current rent subsidy policy for Housing Choice Vouchers to an alternative rent policy. Four MTW PHAs will participate in the study. The demonstration is being guided by a comprehensive research agenda structured around three study components: impacts, implementation processes, and cost analysis.
In order to evaluate the impacts of the alternative rent policy, all households that meet the eligibility requirements for the demonstration are being randomly assigned to either a new policy group or an existing rent policy group (i.e., the control group). Households will not have an opportunity to switch their assignment from one rent policy group to the other. This approach is consistent with the MTW Demonstration’s policy of authorizing PHAs to implement and test innovative rent policies to try to help voucher holders become self-sufficient and to reduce administrative costs. When MTW PHAs that are not part of the Rent Reform Demonstration have implemented their own rent reforms, they have not been expected to limit these new policies to tenants who volunteer for them.
A research design that includes the broader eligible population fits the compelling need for HUD and Congress to understand the effectiveness of the new rent policy for the full eligible population, not just for a subset of volunteers recruited for a special demonstration project. For a variety of reasons, volunteers may not adequately represent the full eligible population. The new rent policy itself offers all tenants the possibility of becoming economically better off, while also including a number of safeguards intended to help prevent them from becoming economically worse off. (As described in a later section in this document, all tenants will be given an opportunity to withdraw from having their personally identifiable data shared with the researchers if they wish that it not be disclosed.)
The evaluation plan includes an exception for one of its sites, the Louisville Metropolitan Housing Authority (LMHA). Households in this PHA that are randomly assigned to the new policy will be given the option to opt out of the new rent policy group in addition to withdrawing from having their data disclosed to the researchers. HUD agreed to this exception to be responsive to concerns raised by LMHA and the local community. It is hoped that few tenants will choose to opt out of the new rent policy group; if that is the case, a valid comparison of outcomes between the new rent policy group and the control group will be possible. However, if a substantial proportion of tenants from the new rent policy group opt out (i.e., choose to switch to the existing rent rules), a bias may be introduced into the impact estimates. MDRC will conduct an analysis of the potential bias (using techniques similar to those used in a survey response bias analysis) to try to assess the extent of this problem. If the level of bias is deemed unacceptably high, the overall evaluation will rely more heavily on the results from the other three demonstration sites where a more robust research design is being implemented.
The study will examine the effects of the alternative policy from two perspectives: that of the housing agencies and that of the voucher holders (also summarized in Table 1 below).
A. Research questions concerning PHAs
The goal of reducing the burden and costs that PHAs incur in administering the current rent rules is one of the primary motivations driving interest in rent reform. Indeed, it was a major reason why the MTW PHAs wanted to join MTW in the first place. For this reason, the demonstration will assess to what extent the alternative rent model actually simplifies the administration of rent subsidies, and improves PHA finances, without placing undue burdens on residents. Related goals concern PHAs’ ability to stretch their budgets to serve more residents in need of housing assistance, such as by reducing average subsidy levels and the duration of subsidy receipt.
The evaluation will address these important issues in the following ways:
Document the alternative rent model as implemented in practice. Once the alternative rent model and implementation strategy are developed for the demonstration sites, the four participating PHAs will implement this model, including any adaptations of the model to their local circumstances. The PHAs’ experiences in operating the model are also likely to vary, given their different administrative systems, organizational capacities, and local contexts. Thus, what may seem like a uniform model on paper, may take on distinctive features as PHAs translate it into practice and incorporate it into their existing organizations and communities. Consequently, a thorough understanding of the features of the alternative rent model as implemented will be an important component over the course of the full evaluation.
Three components of the evaluation will document how the new rent policy is operationalized in each HA: (1) an implementation research; (2) ongoing technical assistance monitoring efforts that began as part of Task Order 1 (TO1); and (3) a cost study. Data collected through these methods (which combine direct observations, interviews, and more standardized measurement) would be used in computing the costs of operating the rent systems and in determining whether and how much the alternative system yielded costs savings. The data collection would also include explorations of the choices PHAs make, for example, with regard to hardship exemptions, and what procedures they establish to approve or deny exemption requests. The research will also document the kinds of changes in administrative processes, including data management systems and software, income verification procedures, and staff deployment, all of which are crucial for understanding whether the new policies simplify or complicate rent administration for the PHAs, and whether strategies adopted by some PHAs are more efficient and accurate than others and worthy of emulation. This documentation began under TO1 through the information we are collecting through our technical assistance to help PHAs implement the new policies and set up random assignment. Future data collection efforts will be expected to generate richer data for systematic comparisons of site implementation experiences and practices.
Measuring effects on tenant turnover and the availability of vouchers. Changes in the rent rules could change tenant turnover in a number of ways. First, they may increase earnings and income and, in turn, increase or hasten exits from vouchers. But, second, and in contrast, the new rent policy could reduce tenant turnover if more voucher holders come to view voucher receipt as more attractive than unsubsidized housing on the private market than they would otherwise view it. As described later, we would compare lengths of stay and reasons for exit for the program and control groups, and we would try to discern the relative influence of different factors on that impact.
Effects on tenants’ housing-related hardships. Changes in the rent rules affect tenants’ rent burden and thus their likelihood of being evicted or having their utilities shut off. For example, families at the lower end of the income distribution may strain to afford a high minimum rent, or those with higher incomes may fall into arrears if their income drops, unless adequate hardship protections are included in the rent policy. As described below, we would assess the effects on housing hardships and rent burden in the impact analysis, comparing rates of several dimensions of material hardship and also eviction rates for families in the program and control groups. We will do this for all families and for certain subgroups of families thought to be most at risk for these hardships.
Effects on PHAs’ costs and ability to serve more eligible families. The effect of rent reform on tenant turnover is a key question because it concerns, not only the well-being of families with vouchers, but also the number of families PHAs can afford to serve with a given budget. In theory, a new rent policy may allow PHAs to stretch their budgets and fund subsidies for more families by causing tenants’ earnings to grow and/or more directly by reducing the amount of subsidy offered over time.
Alternatively, some policies could end up increasing rather than reducing PHAs’ costs. That might be the case, for example, if the policy causes some tenants to work less than they might have otherwise (e.g., because of a fear of the implications of the loss of deductions such as child care), despite the fact that they have a greater economic incentive to work in the following three years, while also having no effect on other tenants’ work rates. If that’s the case, (relative to the 30-percent rule) the aggregate amount of tenant contributions would be lower at the end of the three-year rent-freeze period and heading into the subsequent period, thus costing the PHA more in subsidies. Although this result is not anticipated, it cannot be ruled out. The evaluation will determine whether it in fact occurs, or whether, as hoped, the new policy produces increases in tenant work effort. Whether or not a new policy is budget-neutral, and whether it achieves the broader goals of rent reform, will depend to a very important extent on the changes it causes in participants’ labor market and housing decisions.
Administrative reforms and PHA cost savings. The alternative rent policy should reduce the administrative burden on the housing agency because it should be simpler to administer and require fewer tenant-staff interactions. This should produce administrative cost savings. It will thus be important to document how the changes in rent policy affect a variety of administrative processes and the extent to which there is any offsetting increase in the administrative burden of dealing with hardship cases under the new policy.
Assessing administrative efficiencies with an eye toward “scaling-up.” In any demonstration project, one must be concerned that the ways of operating a program or policy as a special research initiative may not mirror the ways it would operate as a scaled-up policy. In the Rent Reform Demonstration, the PHAs will be required to operate dual rent systems – the current income-based system plus an alternative system. At the very least, this means that the PHAs will not be able to achieve the same efficiencies and economies of scale with the hybrid system as they could if the new policy were operated at full-scale for all voucher recipients.
Although it is not possible to avoid this problem in the context of the demonstration, we propose to include as part of a longer-term implementation study an assessment of where further operational efficiencies could be achieved if the rent policy were implemented at scale. For example, we would look for ways in which everything from staff deployment to information systems could be modified or consolidated if the new policy were adopted wholesale. In addition, as part of our proposed cost-benefit analysis (see below), we would include sensitivity tests that build in reasonable assumptions about those possible economies of scale.
B. Research questions concerning individuals and families
A premise behind many rent reform proposals is that the reforms will benefit assisted families as well as PHAs. Thus, drawing on the available data sources (which does not include a participant survey), the evaluation is designed to assess whether this is true, using the following approach:
Assessing voucher recipients’ understanding of rent reform incentives. Tenants’ understanding of the new model and its implicit incentives will inform how they make labor market and housing choices. Using qualitative research methods, the evaluation will explore whether tenants understand the new rules, and the “frames” they use in interpreting them, such as whether they believe that “extra work is penalized.” The MDRC team will conduct a small number of interviews to get an early read on voucher recipients’ understanding of these.
Measuring tenant outcomes: We identify the following clusters of tenant outcomes.
Household composition and structure: To explore effects on household composition and structure, the analysis would rely on information collected about all household members, including names, ages, employment status (if appropriate), and relationship to the head of household through the HUD 50058 form.
Work behaviors: Unemployment Insurance wage records will include information on employment and earnings.
Income, assets, and rent burden: If rent reform increases tenants’ disposable income, it may help them accumulate assets. Data on income (for housing agency data), rent, and utilities payments would be used to construct measures of rent burden.
Homelessness: We will measure effects on homelessness using Homeless Management Information Systems (HMIS) data.
Other benefits: TANF and SNAP data will be used to examine effects on other subsidy use, since changes in the receipt of these public benefits may flow from any impacts that rent reform has on tenants’ earnings.
Voucher use: Using the HUD 50058 data, the study would examine the effects of alternative rent strategies on the duration of voucher receipt and exit reasons.
Knowledge and perceptions of rent rules: In-depth qualitative data would be used to examine voucher recipients’ perceptions, understanding, and awareness of the rent rules, and their attitudes toward the PHA and frontline staff.
C. Counterfactual and service context
Knowing not only the control group’s outcomes, but also the “treatment” they receive is crucial to making sense of impact findings in any random assignment trial. Put differently, what a randomized trial tests is not simply the effects of an intervention for the program group, but, rather, the difference in treatment between the program and control groups, which is what produces the difference in outcomes (or program “impacts” or effects) between those two groups. This is critical because it will influence what the experiment actually tests.
For this demonstration, how much the MTW sites have already implemented some features of rent reform was a subject of initial exploration. Our scan of MTW plans and reports indicated that MTW PHAs have taken some steps in this direction, and some have moved quite far down the road. Thus, the benchmark against which the sites’ impacts would be judged differ, making the intervention (the alternative rent policy) a bigger change or a qualitatively different change in some sites than in others (to limit this problem, we have not selected sites that have already instituted wholesale rent reform). For instance, we know that in the sites we have selected, the counterfactual will differ in terms of length of the current recertification period: some are annual, but one is already biennial, and we are considering conducting the analysis separately for the full sample and for only the three sites with the 3 years recertification versus 1 year recertification comparison.
The control group benchmarks could also differ across sites in terms of the intensity of services available to study participants. For instance, the control group might receive more employment services in certain sites simply because those sites are located in “service-rich” environments, which might influence the effectiveness of the incentives. These circumstances of the control group will have very important implications for interpreting the impact findings and drawing lessons for policy.
The MDRC team would address this issue as part of its implementation research, drawing on information obtained in the PHA site selection discussions and visits in which the alternative rent rules are discussed as part of TO1; through later interviews with HA staff; through PHA data on participation in self-sufficiency programs (where appropriate); and through the participant interviews, which would ask respondents about the extent to which they receive relevant work-related services.
D. Measuring the effects on place-based outcomes
Rent reform may affect the residential choices that voucher holders make. For example, because the tenants’ rent share will remain fixed for each three-year recertification period, some tenants may choose to move to higher-cost apartments or neighborhoods if they increase their earnings and keep more of their extra income.
For voucher holders, we would rely on HUD 50058 data to track the location of both voucher holders in treatment and control groups. We would then compare overall mobility rates for treatment and control groups.
A.2. How, by Whom, and for What Purpose Are Data to be Used
How the Information will be used?
The findings from the study will be used to inform the Federal government, public housing agencies, and other stakeholders about the effectiveness of the alternative rent policy.
Who will collect the information?
Under the first task order, baseline data are currently being collected through interviews administered by housing agency staff during recertification and from housing agency or HUD data. Study participants were also provided with a brief information sheet that ensures that they understand the demonstration, as well as their role and rights within the study. This form was also approved under the TO1 OMB submission. Under Task Order 2, the longer-term evaluation will also collect follow-up data through interviews with key informants and access to housing agency / HUD data and other administrative records (employment records, for example). As with the baseline data, and in an effort to limit data collection burden on participants, the participant interviews will focus on information that is not readily available in available data sources (or not available in the format required for the evaluation).
Data sources
Table 1 provides a summary of the impact analysis outcome clusters described above and the associated data sources. As shown, these data will come from a few data sources, including UI wage records, HUD data, and interviews. Table 2 provides more specific information about the data source for each quantitative data type.
Table 1 Summary of Data Collection Clusters and Data Sources
|
|
PHA’s perspective |
|
Experiences operating and implementing rent reform; relationship with participants |
Ongoing TA observations Implementation research |
Tenant turnover, availability of vouchers, reasons for exit |
HUD 50058
|
PHAs’ costs and ability to serve more eligible families |
Cost-analysis data Interviews with administrators |
Administrative efficiencies and scaling-up |
Cost-analysis data Interviews with PHA staff |
Tenant’s perspective |
|
Understanding, knowledge, awareness of rent reform; perceptions of and relationship with HA |
Ongoing in-depth interviews / focus groups with tenants
|
Household Composition and Structure |
HUD 50058
|
Employment and earnings |
National Directory of New Hires (NDNH). |
Income and income supports
|
HUD 50058 TANF and SNAP |
Assets and finances |
HUD 50058 |
Rent burden |
HUD 50058, UI, TANF |
Material hardship |
Participant interviews |
Homelessness |
HMIS data |
Neighborhood conditions and safety, and housing quality |
Neighborhood data |
Voucher use |
HUD 50058
|
Frequency of moves |
HUD 50058
|
Through these various data sources, the analysis will address the types of topics listed above. The analysis will take a comparative approach, examining the similarities and differences across PHAs in how they structure and implement the alternative rent policy. (Additional details about the analytic approach are outlined in Supporting Statement B).
Table 2: Data Sources |
||
DATA TYPE |
DATA SOURCE |
|
Baseline Characteristics |
MDRC Baseline Information System |
|
50058 baseline from PHA data |
Housing Authority (DCHA, LHA, LMHA, SAHA) |
|
Housing Authority Data |
Housing Authority (DCHA, LHA, LMHA, SAHA) |
|
HUD Data |
Port out info |
HUD |
All HCV households in the US |
HUD |
|
Data for updated modeling |
HUD |
|
Employment and Earnings from NDNH |
Office of Child Support Enforcement |
|
TANF and SNAP |
DC Department of Human Services; Kentucky Cabinet for Health and Human Services - Division of Family Support; Texas Health and Human Services Commission |
|
Homeless Management Information Systems (HMIS) |
Kentucky Housing Corporation; The Community Partnership for the Prevention of Homelessness (DC); Haven for Hope (San Antonio) |
|
HUD 50058 data and other housing data: MDRC will collect data recorded from HUD MTW 50058 forms directly from the housing agencies. All voucher households enrolled in the study will complete or update a 50058 form as part of their redetermination interview at the beginning of the study. Where possible, we will use 50058 data from 1-3 years prior to random assignment to supplement data collected at random assignment and to describe voucher household characteristics and their monthly rent to owner and estimated TTP. Data from later extracts will be used to track changes in tenants’ reported income, rent to owner, and receipt of vouchers over the course of the follow-up period. In addition to the 50058 data, we will also collect data from the PHA’s internal reporting systems that are not available in the 50058 data, such as total HAP, actual TTP, use of interim recertifications, reason for termination, and hardship exemptions. For the Rent Reform rules group, we will collect additional information on their use of the grace period, as well as retrospective income. MDRC is working with the housing authorities to identify changes that will be required to their existing software. Data will be collected directly from HUD to describe the HCV national population, estimate 50058 data for households who port out, update the national modeling analysis, and possibly to enable geographic analysis.
Non-Housing Administrative Records: HUD’s interest in tracking employment and earnings information of those in the Rent Reform Demonstration can be accomplished directly through agreement with the state labor agency, indirectly through the local housing authority, or via the National Directory of New Hires (NDNH). The NDNH is a national database of wage and employment information that was established pursuant to the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA). As the central location for earnings and employment data, NDNH data offers advantages over seeking data from separate state labor or employment agencies. MDRC and HUD are exploring this preferred data option.
In addition, other data sources, such as administrative data on TANF, SNAP, and homelessness (HMIS data) will also be collected. The data collection methods will vary in form, intensity, and duration, based on (1) how each agency organizes and extracts the data; and (2) whether MDRC has had prior experience in collecting these data for previous and ongoing studies.
PHA financial data: In order to build accurate estimates of the costs incurred by PHAs to administer the alternative rent policy compared with the current policy, MDRC will collect and analyze comprehensive financial and organizational data from PHAs on labor and non-labor costs; salaries; number of staff; staff schedules; recorded hours of employment; and the total number of annual (or triennial) and interim recertifications and other types of action recorded on HUD Form 50058 per month. We will collect these data from the PHAs’ financial statements and other administrative records, as well as Financial Assessment System (FASS) data on total revenues and expenditure, and will combine the data with PIC or PHA management information system data on total households served and types of action each month during the follow-up period.
Process data: Implementation research visits are budgeted as part of the TO2 research activities. PHA and participant experiences will be documented as part of two rounds of implementation research. Specifically, Round 1 would cover the initial recertification process and experiences (initial briefings, income verifications, notices, and early use of the grace period and hardship requests), focusing on interviews with PHA managers and housing specialists. Wave 2 would focus more on the interim recertifications in-between the initial and triennial recertification, and on the triennial recertification. This sequence also aligns with the data requirements for the interim and final reports. During each round, MDRC will conduct structured in-depth interviews with approximately 20 tenants at each PHA to learn about their experiences with and perspectives on the new rent policy, including any hardships that appear to be created by the new policy. The interview protocols for the first round are included in this submission.
Wherever possible, advanced technology will be used in data collection efforts to reduce burden on study participants and on site staff. As it pertains to this OMB submission, where relevant person-level data has been identified as available through an accessible, centralized, and computerized source, that source is being pursued. For example, MDRC is working with HUD to obtain records from the National Directory of New Hires database to analyze study participants’ employment and earnings. While program implementation data collection relies on evaluation staff efforts on-site, we have sought, wherever possible, to minimize overlap between questions we include in implementation research protocols and information that is available through PHA data and other administrative records.
The information collection will not duplicate information that is already available. Where possible, the evaluation will use available data sources, such as employment records.
A.5 Burden on Small Business
We do not anticipate that this study will burden small businesses.
This evaluation represents an important opportunity for the Federal government to build a body of knowledge about the effects of an alternative rent policy. This is consistent with the Administration’s strong focus on evidence-based policymaking. If this study is not conducted and the data not collected, analyzed, reported, and disseminated, Federal and local program or policy decisions will not be informed by high quality evidence upon which to base critical decisions regarding future rent policy.
The proposed data collection activities are consistent with the guidelines set forth in 5 CFR 1320 (Controlling Paperwork Burdens on the Public). There are no special circumstances that require deviation from these guidelines.
A.8 Form 5 CFR 1320.8(d) and Consultations Prior to OMB Submission
Please see Appendix x for a copy of the HUD’s notice in the Federal Register, required by 5 CFR 1320.8(d), soliciting comments on the information collection prior to submission to OMB.
The notice is posted at: https://www.federalregister.gov/articles/2015/04/14/2015-08538/60-day-notice-of-proposed-information-collection-rent-reform-demonstration-task-order-2.
b. Consultations Outside of the Agency
During study design, we have sought the input of nationally recognized experts on public housing and Section 8 housing support, including Professors John Goering (City University of New York) and Ingrid Gould-Ellen (New York University).
As part of TO1, study participants were offered a payment of $25 in the form of a gift card upon study enrollment when the BIF is completed. This payment was intended as a small compensation for study participants for their time to complete the BIF. HUD also provided the participating PHAs with resources to offset costs related to necessary changes to their software systems.
Under TO2, we plan to provide a small payment of $30 to the households who participate in the in-person interviews as part of the implementation research (about 20 participants per site, for a total of 80). As with TO1, this payment is intended to compensate participants for their time participating in these discussions.
Strong protections have been implemented to maintain the privacy of tenants in all four demonstration sites, to the extent permitted by law. All tenants are allowed to withdraw from having their personally identifiable data shared with MDRC or its research partners. Tenants are being notified about this opportunity as part of the normal certification process they undergo to renew vouchers. This process was approved by MDRC’s IRB and by HUD (also approved by OMB Tenants may withdraw from data disclosure by contacting MDRC.
As previously described, MDRC will collect data on each household from HUD 50058 forms prior to random assignment and for a minimum of three full years after random assignment. Employment and earnings data will be sought through state agencies or the National Directory of New Hires (NDNH) data base, and public assistance data will be sought through state agencies. HMIS data will be collected through local Continuum of Care organizations. In all circumstances, MDRC will protect the confidentiality of the data it collects through its regular high-security safeguards and practices. All tenants will be informed that any personal information they provide or allow access to for the study will be used only for the purpose of this. Individuals will not be identified in prepared reports. All research staff working on the project have been trained to protect private information and have signed a pledge stating that they will keep all information gathered private to the extent permissible by law. All papers that contain study participant names or other identifying information will be kept in locked areas and any computer documents containing identifying information will be protected with a password.
The qualitative data collection proposed under TO2 will focus on staff and participant experiences with the alternative and current rent policies. Respondents will be informed prior to their interviews with MDRC staff that they may refuse to answer any question and that their responses will be used in the aggregate (and that individuals will not be identified by name).
The hour burden estimates for data collection for Rent Reform operators and participants is outlined below in Table 2. The estimates included below are based on MDRC’s extensive experience with previous random assignment studies involving similar populations and data collection.
Cost to respondents for collection of information
Households participating in the Rent Reform Demonstration will range widely in employment position and earnings. We have estimated the hourly wage at the expected prevailing minimum wage, which is $7.25 per hour in Kentucky and Texas. The hourly minimum wage in the District of Columbia is expected to be $10.50 by Q3 of 2015.2 Accordingly, we assume an hourly rate across all sites of $8.13 that represents an average of these two rates, weighted by the pledged sample at each site (2,000 pledged participants in Washington, D.C. and 5,400 pledged in the remaining sites.) Moreover, we expect about 50 percent of the participants to be employed at the time of study entry. A recent report by the Center on Budget and Policy Priorities, some 55 percent of non-elderly, non-disabled households receiving voucher assistance reported earned income in 2010. The typical (median) annual earnings for these families were $15,600, only slightly more than the pay from full-time, year-round minimum-wage work.3 Based on this assumption, and the frequency and duration of responses listed in Table 2, the estimated total respondent costs are: Study Participant Interviews and/or Focus Groups: 40 employed sample members (50% of 80) * $8.13 * 1.5 hours = $487.80.
Information Collection |
Number of Respondents |
Frequency of Response |
Responses Per Annum |
Burden Hour Per Response |
Annual Burden Hours |
Hourly Cost Per Response |
Annual Cost
|
Study Participant Interviews and/or Focus Groups |
80 participants (20 participants * 4 sites) |
Once |
One |
90 minutes, on average (1.5 hours) |
120 (80 * 1.5) |
$8.13 |
$487.80 (40 employed sample members * $8.13 * 1.5 hours) |
PHA Staff Interviews |
32 staff (8 staff * 4 sites) |
Once |
One |
90 minutes, on average (or 1.5 hours). |
48 hours (32 * 1.50) |
$24.33 |
$1,167.84 (32 staff * $24.33 * 1.5 hours) |
Housing Authority Database Extraction Activities by PHA staff |
4 staff (1 staff * 4 sites) |
8 responses in the covered period (monthly through January 2015, then annually through 2018) |
Four in 2015, two in 2016, one in 2017, one in 2018 |
60 minutes, on average (or 1 hour) |
16 hours (4 staff * 1 hour * 4 responses in 2015) |
$33.58 |
$537.28 (4 staff * $33.58 * 1 hour * 4 responses in 2015) |
Cost Study Data Collection Activities with PHA staff |
8 staff (2 staff * 4 sites) |
Three times over the covered period |
One |
120 minutes, on average (or 2 hours) |
16 hours (8 staff * 2 hours) |
$33.58 |
$537.28 (8 staff * $33.58 * 2 hours) |
Interviews to Understand Implementation of New Rent Model. Includes meetings with PHA staff for technical assistance purposes |
32 staff (8 staff * 4 sites) |
Four times |
Up to four times |
30 – 60 minutes (or .5 to 1 hours) Incorporated into technical assistance, monitoring visits and follow-up; |
128 hours (4 one-hour meetings * 32 staff) |
$24.33 |
$3114.24 (32 staff * $24.33 * 1 hour * 4 meetings) |
TOTAL |
156 |
|
|
|
328 |
|
$5,844.44 |
Cost burden to PHA staff
As discussed in Supporting Statement B, participation in the evaluation will require PHAs to undertake research activities not normally required for their regular operations. These activities will span the duration of the study, although some activities, especially those involving program staff, will be more intensive in the earlier years. Required tasks will fall across four areas: planning for random assignment launch, BIF data collection and sample build-up, data extraction and transfer, and ongoing management and review.
For program staff participating in interviews, the estimate uses the median hourly wages of selected occupations (classified by Standard Occupational Classification (SOC) codes) sourced from the Occupational Employment Statistics from the U.S. Department of Labor’s Bureau of Labor Statistics. Potentially relevant occupations and their median hourly wages are:
Occupation |
SOC Code |
Median Hourly Wage Rate |
Community and Social Service Specialist |
21-1099 |
$19.26 |
Social/community Service Manager |
11-9151 |
$29.40 |
Source: Occupational Employment Statistics, accessed online March 20, 2015 at http://www.bls.gov/oes/current/oes_stru.htm
To estimate cost burden to program staff respondents, we use an average of the occupations listed, or $24.33/hr.
For program staff supporting data extraction activities, the estimate uses the median hourly wages of selected relevant occupations in a manner similar to the above. A standard wage assumption of $33.58 was created by averaging median hourly wage rates for these occupations:
Occupation |
SOC Code |
Median Hourly Wage Rate |
Database Administrator |
15-1141 |
$37.75 |
Social/community Service Manager |
11-9151 |
$29.40 |
Source: Occupational Employment Statistics, accessed online March 22, 2015 at http://www.bls.gov/oes/current/oes_stru.htm
Based on this assumption, and the frequency and duration of responses listed in Table 2, the estimated total PHA staff costs are:
PHA staff interviews: 32 staff (8 staff4 * 4 sites) * $24.33 *1.5 hour * 32 meetings = $1,167.84.
PHA database extraction: 4 staff (1 staff * 4 sites) * $33.58 * 1 hour * 4 responses in 2015 = $537.28
Cost study data collection: 8 staff (2 staff * 4 sites) * $33.58 * 2 hours = $537.28
Interviews to understand implementation of new rent model and to provide technical assistance: 32 staff (8 staff * 4 sites) * 24.33 * 1 hour * 4 meetings = $2,983.
Total: $5,356.64
Note, this estimate does not include fringe benefits or other overhead costs. This estimate also only pertains to staff directly involved in administering study protocols and participating in data collection activities, and does not include effort associated with other housing authority staff whose contribution will be needed to effectively implement the study at the site. Further, for staff included in this estimate, this calculation does not include the time that will be required to check-in regularly with MDRC and to provide other regular status updates throughout the duration of the study.
Total estimated cost: $5,844.44
A.13 Other Cost Burden to Respondents and Record Keepers
Data collection requires housing authorities to modify their existing data systems to capture the newly implemented rent rules, and HUD has made resources available to support system modification.
The total cost to the federal government for TO2 of the study is $3,030,919 – included in this task order are costs associated with administrative data collection, qualitative data collection, PHA financial data collection, continued technical assistance, data processing and analysis, and reporting. The data collection portion of this budget is $1,883,786.
This submission is a revision of an existing collection (OMB #2528-0306). The revision to the existing collection is necessary to obtain approval for data collection that is part of Task Order 2 of the Rent Reform Demonstration: 1) interviews with PHA staff to understand the early implementation,, 2) two waves of study participant interviews to capture their early experiences, 3) administrative records that will be collected to assess the impacts of the intervention being tested, and 4) cost study data collection activities. A second revision will be requested for longer-term data collection activities (currently scheduled for 2018).
A.16 Tabulation, Analysis, and Publication Plans and Schedule
To determine the effectiveness of the targeted programs, the evaluation is designed to collect three categories of data: 1) baseline data, 2) implementation and process data, and 3) administrative records.
The following table shows the report schedule for this project.
Report Schedule for the Rent Reform Demonstration |
||
DELIVERABLE |
SCHEDULE |
|
Task Order 1 |
|
|
|
Report focusing on study design, enrollment, and sample characteristics |
7/11/2016 |
Task Order 2 |
|
|
|
Report focusing on interim impacts |
1/17/2018 |
|
Synthesis paper covering overall findings from the demonstration |
7/30/2019 |
|
Brief working papers on: |
|
|
a. Impacts on families |
7/30/2019 |
|
b. Impacts on PHAs |
7/30/2019 |
|
c. Modeling of effects for the broader national voucher population. |
7/30/2019 |
A.17 Reasons for Not Displaying OMB Approval Expiration Date
The expiration date for OMB approval will be displayed on any forms completed as part of the data collection.
A.18 Exceptions to Certification Statement
No exceptions are necessary for this information collection.
1 See eligibility section below for specific eligibility criteria.
2 (Source: District of Columbia Department of Employment Services, http://does.dc.gov/sites/default/files/dc/sites/does/page_content/attachments/DC%20Minimum%20Wage%20Increase%20-%20DC%20Register%20Public%20Notice.pdf)
3 (http://www.cbpp.org/cms/?fa=view&id=3634). Based on this, we assumed 50% of tenants would be working at the federal minimum wage.
4Number of PHA staff interviews could increase if the housing agency deploys more staff to work on activities related to Rent Reform implementation.
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| File Title | OMB Package Part A |
| Author | nunez |
| File Modified | 0000-00-00 |
| File Created | 2021-01-24 |