Part A
JUSTIFICATION
The Department of Housing and Urban Development, with the assistance of contractors, has collected basic data on Low-Income Housing Tax Credit (LIHTC) projects placed in service in the years 1987 through 2004. A contractor is currently gathering data on projects placed in service in 2005. These data are available over the World Wide Web at www.huduser.org/lihtc/. The data have been, and will continue to be, used for analytical purposes, such as studies of the characteristics of tenants. Contingent on OMB approval, the Department plans to continue collecting basic data on Low-Income Housing Tax Credit (LIHTC) projects as these projects are placed in service.
The Low-Income Housing Tax Credit (LIHTC) was created by the Tax Reform Act of 1986 to increase the supply of affordable rental housing for low-income families. The states were authorized to issue federal tax credits for the acquisition, rehabilitation and construction of such housing. Investors in tax credit properties take credits against income over a 10-year credit period. The amount of the credit is based on the qualified basis of the property (typically, development costs, less land, times the percentage of units reserved for low-income use) and the applicable credit percentage (approximately 9 percent or 4 percent, depending upon whether federal financing subsidies are used).
Subsequent legislation has modified the credit. Major changes between 1986 and 1993 include:
Making the credit a permanent part of the federal tax code;
Allowing states to carry over for one year unused credit authorizations;
Allowing projects which have not been completed in the year originally scheduled to retain their credit allocations, provided that a minimal amount of progress is made in the year of credit allocation and the project is placed in service within two additional years;
Requiring the states to make certain that no more credit is allocated to a project than is necessary for financial viability;
Eliminating the acquisition-only credit;
Requiring the developer to extend the commitment of the project to low-income tenant occupancy from 15 years to 30 years with certain exceptions; and
Allowing 30 percent additional credit for projects built in Qualified Census Tracts or Difficult Development Areas.
Since 1987, the LIHTC has become the principal mechanism for supporting the production of new and rehabilitated rental housing for low-income households. Although HUD is not formally responsible for the allocation or the use of credit, the Department has monitored and analyzed the credit since its inception because of the important role of the LIHTC in providing for the housing needs of the poor and because of its operation in conjunction with, and in the context of, ongoing HUD programs. Under the current contract, the contractor has completed collection of data on projects placed in service between 1995 and 2002. This contract includes options to collect data on projects placed in service in 2003.
There continues to be great interest in the LIHTC program at HUD, the Department of Treasury, and many others sources. Unfortunately, however, since the program is independently administered by more than 50 different state housing agencies, there would be no centralized source of data about the units that have been developed with this federal subsidy without this data collection effort.
In recognition of the Department’s responsibilities regarding the supply of low-income housing, the Department plans to continue collecting data on LIHTC units. To update the database, the Department plans to request from the states an extremely limited set of the data about each tax credit project that has been placed in service with LIHTC and have not been previously reported to HUD. The data are limited to approximately 15 major data elements, including basic identifiers, minimal descriptive data, and data items that are likely to be used to stratify samples in future research on the LIHTC. All of the items should be readily available to the state allocating agencies from which they will be collected. All of the data should be considered public information.
The number of data items is strictly limited in order to increase the chances if obtaining 100 percent reporting by the state agencies. In this data collection effort, Freedom of Information Act (FOIA) requests will be completed as needed. The data will also be accepted in a variety of formats including computerized data files, hard copy listings that will be abstracted by the contractor, and a one-page data form that can be completed by state agency staff for each project. In addition, allowances have been made to go on-site where absolutely necessary to abstract the data directly from state agency records.
The LIHTC program uses benefits available through the tax code as an incentive to produce qualified low-income housing units. The program is administered by 59 (predominantly state-level) allocating agencies. While the IRS has responsibility for issuing LIHTC regulations and ensuring compliance with the code for individual taxpayers, states administer the program and have recently been made responsible for monitoring compliance at the project level.
Due to the decentralized nature of the LIHTC program, there are few data available on the units that are currently being developed with this federal tax subsidy. The Department of Housing and Urban Development, while not responsible for administering tax credits, has special responsibilities in understanding and evaluating credit usage, both because the LIHTC helps provide for the housing needs of low-income persons and because credits work in conjunction with HUD subsidies in some units.
Absent this data collection, HUD will not have at its disposal the most current, comprehensive LIHTC data, rendering HUD unable to determine the types of areas in which the units are located, the concentration of such units geographically and with respect to other subsidized housing types, or whether incentives to develop LIHTC units in a set of HUD designated Difficult Development Areas has been effective. In addition, without these data, both HUD and private researchers will be unable to conduct sample-based studies on the LIHTC due to the difficulty of constructing a valid sample without a complete data set on the universe of LIHTC projects.
In summary, this data collection and analysis effort will provide both the latest data on the locations of tax credit units and contribute to a sampling frame for future research. The desired data are relatively few and comprise basic data about these properties. At the same time, they will make a significant contribution to knowledge about the use of the LIHTC and enable future research into its effectiveness. Absent these data, HUD and other federal and private researchers will be unable to evaluate the effectiveness of the credit.
The statutory authority for PD&R’s research activities is found in Title V of the 1970 Housing Act, which authorizes programs of “research, studies, testing, and demonstrations relating to the missions and programs of the Department.”
A2 How and By Whom the Data will be Used
The updated information to be collected in the LIHTC database will be used initially by HUD to develop descriptive statistics about the current tax credit projects and to conduct an analysis of the locations of units produced in recent years.
HUD also intends to make the database available to other federal and private researchers who may use the database to support sample-based studies of tax credit projects. HUD intends to support future research studies into selected aspects of the credit, such as an analysis of whom the program serves (tenant characteristics) and an analysis of the financing and subsidy characteristics of tax credit units.
Absent the revision and expansion of the existing National LIHTC database, HUD will not have the information needed to analyze the locations of recent tax credit projects, the interactions of tax credit units with other HUD supported low-income housing, the beneficiaries of the program (low-income tenants) or the costs and effectiveness of this method for subsidizing low-income development projects. As noted above, researchers, both within federal agencies and from academia, are hampered in their efforts to evaluate the credit due to the lack of a reliable, up-to-date sampling frame.
The primary objective of this project is to update the National LIHTC Database. The database contains an extremely limited set of variables about tax credit projects that can be used both for initial research and to support future sample-based studies. Limiting variables to those that are both readily available to states and necessary to stratify potential future samples is an explicit strategy designed to minimize the burden of the information collection and to improve response rates among states. Key features of the data collection plan are outlined below.
Exhibit 1 lists the variables to be collected as part of this study. The items were selected based on their availability, their usefulness as potential stratifiers, and the need to limit the data set to public information, which should pose no issues of confidentiality to the states. All of the items should be readily available to states, as they are among the most basic descriptive data concerning these federally subsidized properties. Moreover, virtually all of the data are reported by the states on IRS Form 8609 for each building at the time it is placed in service. Thus, states are required to collect and maintain these data as part of their federal tax reporting responsibilities.
The data to be collected will cover the universe of tax credit projects placed in service. Data will be collected from 59 administering agencies, which will be asked to fill out the LIHTC data request.
Based on HUD’s experience with a prior LIHTC data collection, it has been determined that the best time to collect the desired data is in the early spring. This is because states are busy during the end of the year completing the allocation cycle and during the beginning of the year meeting IRS deadlines for reporting on new projects placed in service (due at the end of February). Data collection by the end of March will give states time to prepare the data request after the completion of their IRS submissions.
The data collection request will be mailed as early as possible (scheduled for mid-January) in order to give states adequate time to prepare to meet the data request. During the period between the mailing and the due date for data, contractor staff will work with the individual state agencies to understand the nature and organization of their record-keeping systems and to ensure that the data request can be fulfilled in a timely manner.
In order to minimize burden on states and to take advantage of computerized data systems that may already be in place, data for this study will be collected in several different ways. The data collection strategy will be tailored to the individual needs of the states.
1. Form-based data collection. For states without suitable automated systems, agencies will be requested to complete the one page data form (Appendix B) for each tax credit project. The data collection forms will be collected by mail. Data will be keypunched by the contractor and entered into an automated database.
2. On-site data collection. To the extent this is absolutely necessary, contractor staff will be available to visit selected state agency offices and abstract the data directly from agency files. In this case, contractor staff will enter the data directly into electronic form using laptop computers and a data entry screen that mimics the LIHTC Data Form.
3. Listings and Records. It is likely that many of the states will available (or be able to generate) listings that include the desired variables about tax credit projects. Such listings (with appropriate cross-referencing by project identifiers) constitute a third acceptable means of providing the data. Where this is deemed to be the most suitable (and least burdensome) method, we will request that the agency forward all necessary materials. Contractor staff will then enter the data from various sources using the computerized data entry screen described above.
4. Computerized Data. Based on the previous LIHTC data collection, a large number of state agencies have computerized databases on their tax credit projects. These include in-house systems as well as commercial LIHTC packages. Data entered into these systems tends to include the variables listed in Exhibit 1, and, in fact, availability in electronic form was a primary consideration in developing the list of desired data items.
For states with all or most of the needed data in computerized form, the contractor shall request an electronic file containing the records and variables needed for the study. Since the platforms and formats used by the states vary, the contractor is prepared to accept the data in various forms. The contractor will formulate specific instructions for generating the data aimed at data processing personnel. Computer staff from the contractor will be available by telephone to work with local data processing staff to fill the request.
Process and Instruments
The data collection process will begin with a mailing to all states. The expected timing is early 2008. The mailing will include a letter requesting the data, the LIHTC Data Form and instructions, and a set of detailed instructions designed specifically for computer staff preparing an automated data submission. The package will include instructions to the state housing agency to indicate whether a Freedom of Information Act (FOIA) request will be needed, to provide a contact who will be responsible for the data collection activity, and to specify the form in which the state will provide the data.
The contractor will initiate a FOIA request (if needed) and contact the designated state agency contact by telephone. This conversation will focus on gaining an understanding of how the state maintains its records and determining the most appropriate approach for collecting the data from that state. Contractor staff will be available throughout the data collection period to answer questions and/or assist states as needed. In extreme cases, the contractor shall go on site to collect the data directly from an agency’s files.
As describes above, data collection will be tailored as far as practical to the individual circumstances of the state agency, both to improve response rates and to minimize burden on the states. This may mean that some states will use multiple methods (e.g., a combination of computer files and paper listings or computerized data for some years and the LIHTC Data Form for others). The data collection approach is intended to be as flexible as possible to accommodate the needs of the states.
In the final step of the process, states will deliver the materials or data files. Contractor staff will then abstract, enter, or read in data (as appropriate) in order to create a standardized database containing the above-mentioned variables. All data will be cleaned; a printed version of the data will be generated; and the hard copy report will be returned to the states for verification. Should a state request the computerized files containing their data, these will be provided as well. Once cleaned and verified, the database will be transmitted to HUD.
Once the database is completed, the contractor will undertake initial analysis focusing on the location of tax credit projects. Each property will be geocoded with the appropriate Census Tract code via the use of a commercial geocoding package. The package will also be used to attach Census Tract variables and data for the area within a one-mile radius of the project.
Using this database, the contractor will generate basic descriptive statistics about the universe of tax credit projects. Data will be presented separately for subgroups of interest (depending on data availability), e.g., by tax credit year, by sponsor type, or by receipt of additional subsidy. The locational analysis will address the following questions:
How are tax credit units distributed across central city, suburban or rural locations?
In what sorts of neighborhoods are the projects being developed: Census Tracts with high/low minority concentrations? Census tracts with/without other HUD – subsidized housing?
To what extent are tax credit projects located in lower income neighborhoods, i.e., Qualified Census Tracts where 50 percent of residents have incomes under 60 percent of median? Are there any discernable trends over time?
In what sorts of markets are tax credit units being located? Have incentives for hard to serve markets increased the share of units developed in these locations?
The results of the analysis will be submitted as part of the final report of the project.
The data collection proposed here is specifically designed to take advantage of computerized information technologies. Prior experience suggests that many states have computerized systems containing all or some of the needed data items. A detailed set of instructions has been developed to enable state data processing staff to construct a readable data file using whatever data system they have. A vendor of state LIHTC data systems might also provide system-specific instructions or programs. Given HUD’s desire to update the database annually, development of such programs would be very cost effective.
The data collection to be conducted by the contractor represents the only ongoing federal effort to collect data on tax credit projects (other than IRS reporting which is not available for HUD or public use). A contractor has collected data on projects placed in service from 1987 through 2004. A contractor is currently gathering data on projects placed in service in 2005. However, new data is required for projects placed in service since 2005. Private researchers have conducted some work on the tax credit, but these efforts have either been very limited in scope or do not provide data that is publicly available. Indeed, one of the benefits to states of this project is that HUD plans to make the LIHTC database available to the public, thus eliminating the need for states to fill redundant requests from individual researchers.
No data are available which are similar to the data that will be collected through this study.
Respondents for this data collection include state tax credit allocating agencies. None of these entities is a private business. We have no information on the size of the entities administering the credits in individual states; however, responsibility typically lodges with a state housing finance agency, which would ordinarily not be considered a small entity.
Less frequent data collection places data at HUD’s disposal with longer lags. Hence, the Department’s analysis of an important housing program is rendered less current and less comprehensive at any given time. Annual data collection keeps HUD abreast of any trends as they are developing and are most amenable to policy intervention. A secondary benefit of annual updates is, presumably, they are far less burdensome than the larger requests that would be needed if less frequent data collection were instituted.
The data collection creates no special circumstances of the type listed.
The Notice of Proposed Information Collection for Public Comment was published in the Federal Register on Monday, April 16, 2007, on page 19011. One comment was received. A response was sent to the commentor. The letter is attached.
No payments or gifts will be provided to respondents.
No offers of confidentiality will be made. The data to be collected are basic, descriptive data about publicly-subsidized housing projects, specifically their location, size, and use of specific financing mechanisms. We believe that the data are public information; we are also prepared to initiate Freedom of Information Act requests as needed. HUD plans to make the data base available to the public.
There are no questions that would be considered sensitive in nature. Information focuses on standard variables concerning the location, size, and other characteristics of federally subsidized housing projects.
The amount of time each respondent will spend responding to the survey will vary depending on the selected response medium. Respondents may, at their sole discretion, select either electronic or non-electronic response media. Through HUD’s consultations with its data collection contractor, HUD has ascertained that respondents that select a non-electric response medium generally spend no more than twenty-four person-hours responding to the survey. In contrast, respondents that select an electronic response medium generally spend no more than one-person hour responding to the survey. Since HUD cannot a priori know how many respondents will select each response medium, HUD cannot a priori know the aggregate amount of time that will be devoted by respondents to this Collection. For the purpose of responding to OMB 83-I Item 13(d), HUD has assumed that every respondent will select a non-electronic response medium and that average response time will be twenty-four person-hours. Past experience suggests that some respondents will select an electronic response medium and that the average response time will be below twenty-four person-hours. Thus HUD believes that the total hour amount indicated at OMB 83-I Item 13(d) represents a high-end estimate of the burden associated with this Collection. HUD has instructed its data-collection contractor to make every reasonable effort to work with individual respondents to help them develop response methods that are minimally burdensome. For example, if a respondent requested data transcription assistance, HUD’s data-collection contractor would provide this service.
No additional cost are estimated.
The total budgeted cost to the federal government for the LIHTC Data Collection is $440,000 per annum.
This is a request for an extension of a currently approved collection. The adjustment to Item 13 of the OMB 83-I is because additional Credit Allocating Agencies have come to light.
The data collection and analysis described in Section A2.2 will be completed by a HUD contractor. The data collection is scheduled to begin in early 2008, with the mailing of the initial data request. Data for projects placed in service in 2006 is to be submitted by March 31, 2008, allowing nearly three months for states to prepare and/or begin collecting the data.
Between April and July 2008, the contractor will complete the database, have all projects geocoded with Census Tract identifiers, and begin the locational analysis described above. The first year report of this study, as well as all data files and documentation for 2006 projects, will be submitted to HUD by October 31, 2008. The final database will be submitted in December, 2008. A similar schedule will apply for the collection of data on projects placed in service in subsequent years.
There is no request to not display an expiration date.
No exception to Item 19 of the OMB 83-I are requested.
Part B
The study will collect data for the universe of tax credit projects. We estimate about 5,000 such projects each year. Data concerning these projects will be collected from 59 state allocating agencies.
B2.1 Stratification and Sampling Plans
Not applicable
B2.2 Sampling Procedures
Not applicable
B2.3 Justification of Level of Accuracy
Not applicable
B2.4 Problems Requiring Specialized Sampling Procedures
Not applicable
B2.5 Uses of Less Frequent Data Collection Cycles
See Question A5.
Reporting is the responsibility of the state agencies that allocate credits to individual properties. All of the data should be readily available (in fact, similar data are reported to the IRS, but are not accessible by the public or other federal agencies). Every effort has been made to minimize burden on states. The list of desired variables is limited. The data collection strategy is flexible and will be tailored to the individual circumstances of the states. Maximum use will be made of computerized data systems.
It is essential to the success of the effort that all allocating agencies cooperate with the study. The contractor shall encourage participation through the use of flexible data collection systems, individual contact with each state to determine the most appropriate reporting method, and long lead times so that states can prepare. Finally, in selected (and cost-efficient) cases, the contractor shall go on-site and collect the data directly from agency files, thus eliminating any burden on state agency staff.
To ensure cooperation, the contractor solicit letters and/or other expressions of support from key actors, including the National Council of State Housing Agencies.
The study is designed to be a census of states and their tax credit projects. A 90 percent response from the state agencies is expected.
The forms and methods discussed above have been used by a previous HUD contractor in a prior data collection for projects placed in service for 1987 through 1994 and by the current contractor for projects placed in service from 1995 through 2005.
Not applicable
B5.2 Information Collection Agents
Information will be collected by a contractor to be designated.
EXHIBIT 1
The following information is requested for each LIHTC project under the jurisdiction of a given Allocating Agency:
Allocating Agency State?
Allocating Agency Name?
Project Identifying Number (if any)?
Project Name?
Project Address?
Owner/Owner’s Representative Name and Address?
Annual Amount of Tax Credit Allocation?
Number of Total Units?
Number of Total Units by Size?
Number of Low-Income Units?
Elected Rent/Income Ceiling for Low Income Units?
Are any units set aside to have rents below the elected rent/income ceiling? If Yes, how many?
Year Placed in Service?
Year Project Received Allocation or Bond Issued?
Project Type (e.g., new construction, rehab, existing)?
Credit Percentage (e.g., 9% (i.e., 70% present value) or 4% (i.e., 30% present value) or both)?
Did the project have a non-profit sponsor?
Did the project have an increased basis due to qualified census tract or difficult development area?
Did the project use tax-exempt bonds?
Did the project use a Rural Housing Service (FmHa) Section 515 loan?
Did the project have HOME Investment Partnership Program (HOME) funds?
Did the project have an FHA loan? If yes, loan number?
Did the project form part of a HOPE VI development?
Did the project target a specific population?
| File Type | application/msword |
| File Title | Part A |
| Author | ccarter |
| Last Modified By | Preferred User |
| File Modified | 2007-12-18 |
| File Created | 2004-06-30 |