The Evolution of Housing Data in the United States, Part 1: From Tax Rolls to Surveys

Introduction

The history of housing data in the United States extends from the colonial era to the present day. Housing-related data collection has assumed many forms over the years, including state and local governments recording property tax records in the 18th century, surveyors mapping the built environment in the 19th century, and more recent efforts by the federal government and private sector to capture a detailed picture of housing nationwide. Today, housing surveys, databases, and maps help researchers address complex housing questions. HUD collects housing data to inform its work “to foster strong communities by supporting access to quality, affordable housing, expanding the housing supply, and unlocking homeownership opportunities for the American people.” Although not an exhaustive account, this two-part series will cover some important moments in the historical development of housing data as well as noting HUD’s role in adding to and improving the nation’s housing data.

Pre-1800: The Colonial Era 

Long before federal housing agencies gathered comprehensive information through national surveys and other systematic methods, local governments in the 13 colonies collected housing-related data primarily through property and tax assessments. In 1634, the Massachusetts Bay Colony, by order of the General Court, required each town to conduct tax assessments of “every man according to his estate,” meaning that local officials had to document the size, uses, and value of each household’s land and buildings. Similar practices in Virginia emerged by the mid-1600s in which all landowners paid the King an annual “quit rent” of one shilling for every 50 acres. County sheriffs prepared “quit rent” rolls: lists of landowners and the number of acres each owned. After landowners paid these taxes, they would be freed (“quit”) of other annual feudal obligations. In the mid-1700s, Philadelphia also began collecting property valuations and street directories and recording the location and ownership status of houses. These efforts constituted some of the earliest urban housing inventories in colonial times. 

Early housing data often consisted of manuscript lists or ledgers rather than the detailed tables researchers use today. Maintained by local clerks or tax assessors, these data typically included handwritten lists of names, acreage, number of buildings, and number of free and enslaved inhabitants for tax classification that were updated as properties changed hands or as values shifted. Although data collection and record-keeping systems were not uniform throughout the colonies, these rolls and valuations offer some early glimpses into the housing stock and density for what would later become the United States. Many of these materials are still available in state archives, historical societies, and digital collections.

The first nationwide effort to collect data was the 1790 decennial census, as required by the newly ratified constitution; an accurate population count was essential for determining the makeup of the U.S House of Representatives. The first census questions were narrow, focusing on counting total population (free and enslaved) and number of households. In 1790, the United States had a population of 3.4 million and a total of 0.6 million households. The decennial census evolved over time, adding questions and growing along with the United States. For example, the 1890 census began collecting homeownership data.

Early 19th Century: Local Mapping

Another precursor to the federal government’s large-scale efforts to collect housing data was the publication of local data in map form by private entities. For example, starting in 1867, the Sanborn Map Company produced fire insurance maps for more than 12,000 cities and towns across the United States. Covering commercial, industrial, and residential sections of these cities and towns, these maps supplied fire insurance companies and underwriters with accurate data about the buildings they were insuring. The maps, which the company produced for nearly a century, were color coded and depicted factors that would affect fire insurance assessments — building materials (such as brick, stone, concrete, and iron) used for each structure as well as symbols depicting other building-level characteristics such as elevators, stairs, fronts, windows, height and number of stories, dwellings, openings, skylights, and special hazards. Figures 1 and 2 show Sanborn’s fire insurance maps of Boston and Bethlehem Steel Corporation, respectively. Sanborn’s insurance map of Honolulu, Hawaii, published in 1914, also includes a detailed map key. This collection of historical maps paints a clear and valuable picture of urban and industrial growth throughout the United States. The Geography and Map Division at the Library of Congress has more than 700,000 Sanborn fire maps in its collection. 

Reformers and scholars made their own contributions toward collecting cartographic data on the built environment. Among the early examples of these data maps were the Hull-House Maps and Papers, which were published in 1895. Florence Kelley and members of the U.S. Bureau of Labor collected these series of essays and statistical data products as part of a national study of “The Slums of Baltimore, Chicago, New York and Philadelphia” that Congress commissioned in 1892. Researchers who collected the data lived in Chicago’s Hull House and went door to door, visiting homes and sweatshops to ask residents and workers in the city’s neighborhoods about ethnic origins, wages, and the number of people living in their household. These color-coded maps of Chicago’s working-class immigrant neighborhoods displayed nationality, income, and housing conditions. Figure 3 shows a Hull House map of a Chicago neighborhood that color codes each housing unit by weekly earnings. Northwestern University has collected other Hull House nationality and wage maps. These maps represent some of the first housing data gathered and visualized for the purpose of studying neighborhoods, people, and housing conditions in a social science context rather than for taxation or recording land ownership.

In an 1899 study, scholar W.E.B. Du Bois mapped Philadelphia’s Seventh Ward to the block level, documenting the physical attributes of houses and the socioeconomic characteristics of the inhabitants (see figure 4). His research was among the first of its time to combine interviews, surveys, directly observed data, and existing maps across a small geographic area. His hand-drawn maps, coded just like the Hull House maps were, are another integral part of the early history of urban and housing data visualization. See an interactive map of Philadelphia’s seventh ward here.

These local maps provided valuable information on the demographics, family size, economic standing, and building characteristics of households and neighborhoods for some of the largest cities in the United States. Although limited in scope, these maps marked a step forward in generating localized housing and neighborhood data to better understand and visualize our nation’s housing stock and explore societal problems related to housing, filling the gaps left by the early census and tax records. The work done by the Sanborn Map Company, Hull House, and W.E.B. Du Bois gave policymakers much-needed data on density, neighborhood composition, wages, demographics, and building materials that were not available in tax rolls and the early census. Although limited to a few specific cities, these new methods of block-level comparisons and spatial layering would later become key aspects of housing research and the precursors to the modern geographic information systems we use today. 

20th Century: Nationwide Efforts to Create Housing Data

The 20th century marked a turning point in the creation of housing data in the United States. For the first time, housing data would be standardized — measured using the same questionnaires, measurement techniques, and definitions throughout the country. During the Great Depression of the 1930s, policymakers recognized the need for national datasets that described the U.S. housing stock. Researchers had no access to consistently measured national data on mortgages and financing, housing quality and conditions, or renter and homeowner statistics. The variation in local administrative data and independent research was not sufficient in supplying the information needed to create policies to pull the United States out of the Great Depression. In response the federal government prioritized the collection and creation of a robust ecosystem of national datasets that policymakers and researchers could use to study and understand the nation’s housing issues. Congress created agencies such as the Home Owners’ Loan Corporation (HOLC) and the Federal Housing Administration (FHA) to stabilize the housing market, refinance defaulted mortgages, and foster homeownership. For both agencies, the collection and use of consistently measured empirical data available on a national scale was essential to achieving their assigned policy goals.

Congress established HOLC through the Home Owners’ Loan Act of 1933 in response to mortgage instability. During this period, many homeowners were unable to refinance their loans, leading to mass defaults. Because no consistent data on housing finance and mortgages existed, these issues were difficult to address and quantify. HOLC started compiling data on housing finance and neighborhood characteristics using their administrative data on loan performance, local interviews, and city records. HOLC used these data to build “residential security maps” that graded neighborhoods using housing conditions, building ages, and lending risk. These maps served as a risk assessment tool and as an attempt to create uniform appraisal practices and summarize data that had not yet been captured. These maps, which HOLC created between 1935 and 1940, represented one of the federal government’s first attempts to collect consistent national housing finance and lending data. These maps were used internally and were not available to the public until decades later. Digitized versions of the HOLC maps are now online at the University of Richmond.

Congress established FHA in 1934 as part of the National Housing Act of 1934. The agency’s main tasks were to introduce federal mortgage insurance to encourage homeownership and homebuilding and collect accurate housing and mortgage data. Unlike HOLC, FHA covered new and existing loans and was not limited to refinanced and distressed loans. Together, HOLC and FHA created some of the first national datasets that collected detailed information on property values, home construction, and mortgage performance that were used to study housing quality and risk. These data are instrumental to the federal government’s understanding of the housing market to this day.

Conclusion

From the colonial era until the early 20th century, Americans collected sparse and inconsistent housing data, relying on colonial tax rolls, handwritten ledgers, localized maps, and a few surveys to understand the state of the nation’s housing. These early data sources, although incomplete, gave policymakers a rough measure of the early housing stock and laid the foundation for the data collection methods we use today. The country was undergoing rapid change in the early 20th century, and these data were no longer adequate. Part 2 of this series will pick up in 1940, a crucial and exciting time for our nation’s housing data ecosystem and explore how federal agencies such as HUD expanded and improved the state of housing data we have today. 

Sources

American Society of Civil Engineers. n.d. “City Plan of Philadelphia.” Accessed 20 November 2025.

Stephanie Boddie and Amy Hillier. 2022. “The Making and Re-Making of The Philadelphia Negro,” Digital Humanities Quarterly 16:2. Accessed 20 November 2025.

Delaware Public Archives. n.d. “Assessment and Census Records.” Accessed 2 December 2025.

W.E.B. Du Bois and Isabel Eaton. 1899. The Philadelphia Negro: A Social Study. Philadelphia: Published for the University.  Accessed 20 November 2025.

FamilySearch. n.d. “Blockley Township, Pennsylvania.” Accessed 15 April 2024.

Margaret Fortier. 1970. “Land and Quit-Rent Reform in Virginia during the Administration of Francis Nicholson, 1698–1705,” master’s thesis, College of William and Mary. Accessed 2 December 2025.

John Gibbs. 2025. “Housing at 250,” PD&R Edge, 5 June. Accessed 20 November 2025.

Library of Congress. “Sanborn Maps: Introduction to the Collection.” Accessed 20 November 2025.

Library of Virginia. n.d. “Research Guides & Indexes: Personal Property Tax Records.” Accessed 20 November 2025.

New York State Archives. n.d. “Real Property Tax Records.” Accessed 20 November 2025.

Office of the Secretary of the Commonwealth of Massachusetts. n.d. “Massachusetts Archive Collection Overview: Records, 1629–1799.” Accessed 20 November 2025.

Residents of Hull House. 1895. Hull-House Maps and Papers: A Presentation of Nationalities and Wages in a Congested District of Chicago, Together With Comments and Essays on Problems Growing Out of The Social Conditions. New York, Boston: T. Y. Crowell & Co. Accessed 22 December 2025.

Nathaniel B. Shurtleff, ed. 1853. Records of The Governor and Company of the Massachusetts Bay in New England, Vol. 1: 1628–1641. Boston: Press of William White, p.120.  Accessed 1 December 1, 2025.

U.S. Census Bureau. n.d. “Decennial Census of Population and Housing by Decade.” Accessed 23 July 2025.

U.S. Department of Housing and Urban Development. n.d. “Federal Housing Administration History.” Accessed 20 November 2025.

U.S. National Archives and Records Administration. 2007. “A Discovery: 1798 Federal Direct Tax Records for Connecticut,” Prologue Magazine, 39:1. Accessed 20 November 2025.

U.S. National Archives and Records Administration. 2024. “1790 Census Records.” Accessed 20 November 2025.

University of Massachusetts Amherst. n.d. “Genealogy: 1771–1861 Tax Records.” Accessed 20 November 2025.

University of Richmond, Digital Scholarship Lab. n.d. “Mapping Inequality: Redlining in New Deal America.” Accessed 20 November 2025.

Laura Vaughan. 2021. “Revisiting W.E.B. Du Bois: Mapping Race and Health Inequalities.” Mapping Urban Form and Society, 3 June.

“Virginia Quit Rent Rolls, 1704” (1920), The Virginia Magazine of History and Biography 28:3, 207–18. Accessed December 2, 2025.