The Census Bureau took the Residential Finance Survey (RFS) as part of the decennial census from 1950-2000. The RFS is the only survey designed to collect and produce data about the financing of nonfarm, privately-owned residential properties. The RFS is a unique survey for several reasons:

• It collects, tabulates, and presents data for properties, the standard unit of reference for financial transactions related to housing. In most other demographic surveys, the unit of reference is the person, household, or housing unit.
• It is the only source of information on property, mortgage, and financial characteristics for multi-unit rental properties. Information on multi-family loans and properties is particularly difficult to obtain, but is important to understand if progress is to be made in the development of standards for underwriting multi-family mortgages.
• It conducts interviews of property owners and mortgage lenders, resulting in more accurate information on property and mortgage characteristics. The RFS is the only survey which is able to provide a comprehensive view of mortgage finance in the USA, by providing information not only about the loan itself from the lender, but also information about the property owner's demographic characteristics.
• As part of the decennial census, it is mandatory. This is important in collecting information from mortgage lenders. The RFS is exempt from statutes prohibiting release of financial records by financial institutions.
• It is able to subdivide the industry into relevant components. Different parts of the industry have excellent information on their own loans and clients, but not that of the industry as a whole. Information on lending by individual investors or small groups of investors such as pension funds is collected only by the RFS.

We undertook several steps to make sure that the content for the 2001 RFS addressed the changes that have occurred in the mortgage industry in the 1990's. This included bringing together a group of primary RFS data users from in and out of government to develop the questionnaire content. We also conducted focus groups with property owners and mortgage lenders to get their input into the RFS content and collection procedures. Finally, we did cognitive testing of the proposed questionnaires to improve the validity of the questions asked.

We selected the sample for the 2001 RFS from Census 2000 address records and stratified based on the number of units at the address and other characteristics of the units as recorded in the Decennial Census 2000. Approximately 68,000 addresses were selected.

In the RFS, we determined if the property is a homeowner property or rental property. A homeowner property was defined as one having fewer than five units where the owner of the property lives in one of the units. A rental property was one with five or more units or a property of fewer than five units with none being owner occupied. Condominium apartments were homeowner properties if the owner lives in the unit and rental properties if the owner does not live in the unit.

We used the following four questionnaires to collect data for the 2001 RFS.

• An owner seeker letter. This was mailed to units at basic street addresses having two to nineteen housing units in order to identify the name and address of the owner or the owner's agent and to determine if the property was a homeowner property or a rental property. For addresses with twenty or more units, a field representative from one of the Census Bureau's twelve Regional Offices collected the same information.
• A homeowner questionnaire. This was mailed to all addresses with only one unit, to all mobile homes, and to units identified on the "owner seeker" letter (contact) as having fewer than five units, one of which was owner-occupied. If the property was mortgaged, the respondent was asked to report the person or institution to whom mortgage payments are made.
• A rental and vacant property questionnaire. This was mailed to owners or agents of properties with five to nineteen housing units or those with fewer than five units when none were owner occupied as indicated by responses on the "owner seeker" letter (contact) and homeowner questionnaire. For larger properties, with twenty or more units, questionnaires were sent directly to our Regional Offices and field representatives collected the information. As with the homeowner questionnaire, if the property was mortgaged, the respondents were asked to whom mortgage payments were made.
• A lender questionnaire. This was mailed to financial institutions, government agencies, firms or individuals to whom mortgage payments were made, as indicated on the homeowner and the rental and vacant property questionnaires. The questionnaire was designed to collect data on both first and junior mortgages, including home equity lines of credit.

We conducted a pre-survey contact with mortgage lenders in the summer of 2000. The purpose of this contact was to alert lenders of the survey and to establish a contact person within each organization.

We mailed the "owner-seeker" letters (or send them to our Regional Offices) and the first round of homeowner questionnaires in April 2001. A follow-up mailing of the homeowner questionnaires was done in June 2001. The rental and vacant property questionnaires were mailed or sent to our Regional Offices in May 2001, with a follow-up mailing in June 2001. The lender questionnaires were mailed in two primary cycles, the first in July 2001 and the second in September 2001, with possible additional smaller mailings resulting from follow-up operations.

The 2001 RFS was both a centralized mail-out/mail-back operation conducted by the Census Bureau's National Processing Center (NPC) in Jeffersonville, Indiana, and a field data collection operation conducted by our twelve Regional Offices. The enumeration of property owners took place mainly in the months of April through August 2001, and that of the lenders in August through December of that year.

Field interviewing for property owners occurred for cases in which the owner was either not identified or failed to respond to the original mailed questionnaire and follow-up letters, and for those cases sent directly to the field. Follow-up interviewing of lenders occurred in two stages. For financial institutions, that is, banks, insurance companies, and so on, follow-up telephone inquiries were made. If the lenders had not received the initial mailout questionnaires, new documents were mailed to them. Follow-up for mortgages held by individuals involved a personal visit by field representatives from the 12 census regional offices.

• A published summary report
• Unpublished summary tabulations
• A public use microdata file
• An electronic release of data over the Internet through the Census Bureau's homepage

For general assistance on obtaining Census Bureau data, please contact the Census Call Center at 1-800-923-8282 (toll free) or visit ask.census.gov for further information.

For more information on the Residential Finance Survey, please contact (301)763-3199.