August 15, 2017

Philadelphia FSRDC: A New Tool for Urban Research

By: Iourii Manovskii

This map of Philadelphia uses 2010 Census data to map racial and ethnic geographical divisions. (Key: Red is White, Blue is Black, Green is Asian, Orange is Hispanic, Yellow is Other, and each dot is 25 residents.) Courtesy of Eric Fischer via Flickr.

A new center at Penn aims to give researchers a place to access and overlay rich sets of federal confidential data concerning everything from healthcare and labor to housing and urban demographics. The center, which will open in September 2017, is called the Philadelphia Federal Statistical Research Data Center (FSRDC). It is the newest of about 30 such centers across the United States, and offers a valuable new resource for researchers across a variety of disciplines, including urban researchers. 

The new Philadelphia FSRDC will provide Penn faculty and students with access to non-public, confidential, microdata collected by a growing list of federal statistical agencies, including the Census Bureau, Bureau of Labor Statistics, the National Center for Health Statistics and the Agency for Healthcare Research and Quality. In addition, researchers can link public or proprietary outside data to the data provided by federal statistical agencies, thus adding additional possibilities for data comparison and analysis.

The data available through the FSRDC are far more granular and specific than publicly available aggregated data or public-use microdata samples and, notably, are geolocated which will enable spatial and, specifically, urban research.  Importantly, this geolocated microdata enables research on spatial dimensions of inequality and poverty within urban areas and multiple related factors. The available data include administrative and survey records on specific businesses, households, and individuals that have persistent identifiers, so that researchers can link observations over time and across most datasets. The data contain precise geographic information for each observation, including addresses, which is far more specific than what might be available in public versions. Researchers get access to full datasets covering entire population or complete survey samples rather than subsets. Moreover, they can make use of unmodified original data and the full range of response items.

Given the detailed and sensitive nature of the FSRDC’s available data, protecting the confidentiality of respondents’ identity is of central importance. All research proposals must undergo a formal approval process with the agency that owns the data. Researchers with approved projects go through a background investigation that qualifies them for “Special Sworn Status” as if they were a Census Bureau employee. All analysis has to be conducted in a secure lab located in the Federal Reserve Bank of Philadelphia. The lab contains thin terminals that are securely connected to a cluster in the Census Bureau on which the data reside. The data can be analyzed via these terminals using a variety of statistical software packages and programming languages. No results can leave the lab before undergoing a formal review for non-disclosure of personally identifiable information.

The FSRDC makes it possible to overlay information from different surveys to study complex urban phenomena such as the effects of neighborhood change on resident health, employment or education. Urban researchers, including those at Penn, are among the users of the FSRDC data. The latest Census Bureau’s Center for Economic Studies and Research Data Centers Annual Report offers a glimpse at the wide range of research projects urban researchers are pursuing using FSRDC data. These projects include modeling the dynamics of cities and households down to a property and household level; studying the dynamics of suburban and urban areas using rich migration data; and using geocoded data with location information to study a range of topics including the origins of housing segregation and consequences of local public housing policies.[1] [2] 

Of course, this brief overview only scratches the surface of the current urban research that is enabled by the access to FSRDC data. Having local access to these data will likely greatly benefit Philadelphia researchers and offers the potential for significant advances in urban research.More information on Philadelphia FSRDC, descriptions of available data, examples of research using these data, access procedures and contact information for interested researchers can be found on the Philadelphia FSRDC website.

Iourii Manovskii is a Penn IUR Faculty Fellow, an Associate Professor of Economics at the University of Pennsylvania, a co-Director of the Federal Statistical Research Data Center of Philadelphia, and a member of the Executive Committee of the FSRDC Program of the Federal Interagency Council on Statistical Policy. 

[1]Exploring How Transportation Infrastructure Affects Commuting Behaviors of Individuals and Locality Decisions of Business Establishments: A Longitudinal Quasi-Experimental Study in the United States (Marlon Boarnet, University of Southern California; Wei Li , Texas A&M University; Nathanael Rosenheim, Texas A&M University; Haotian Zhong, Texas A&M University) seeks to understand how transit improvements affect the commuting behaviors of individuals and the location decisions of business establishments. Many U.S. cities are making substantial investments in expanding their public transit systems and promoting transit-oriented developments in hopes of reducing vehicle miles driven and in making neighborhoods more compact, economically vibrant, and transit accessible. The researchers will perform before-after, experimental or quasi-experimental analyses that can better illuminate the causal impact of transportation infrastructure investments on the economy and society. This is a two-phase process. In phase one, they analyze patterns of commuting behavior before and after the opening of new light rail transit (LRT) stations. They construct a quasi-experimental setting by using the American Community Survey (ACS) microdata and the propensity score matching technique. In phase two, they study the location behavior of business establishments due to LRT openings by analyzing microdata from the Longitudinal Business Database (LBD) for three metropolitan areas: Dallas, Los Angeles, and Charlotte.

[2] Estimating the Impact of Local Economic Development Incentives Under Fiscal Competition (Ben Hyman – University of Pennsylvania) analyzes the effects of production subsidies on local and aggregate economic development outcomes. Exploiting a seven-year $800 million subsidy lottery in California’s film industry that sorted hundreds of establishments across cities of varying industrial concentration in film, the project studies how production incentives affect establishment location decisions and their associated impact on local wages, employment, municipal revenue, and productivity. Placing these estimates of subsidy benefits in a spatial equilibrium framework, the second part of this study examines the cost-side of attracting employers with local incentives by characterizing the welfare effects of fiscal competition between municipalities. Finally, the project leverages a policy change in 2014 that reformed the subsidy allocation mechanism from a lottery system to a “jobs-impact” ranking formula, providing an ideal laboratory for examining whether subsidy lotteries generate misallocation costs, compared to deliberate employment-based allocation mechanisms.


Bring our latest initiatives, publications and events to your inbox.