Longitudinal Business Database

Longitudinal Business Database
Name	Longitudinal Business Database
Established	1970s
Managed by	United States Census Bureau
Country	United States
Type	database
Subject	firm-level longitudinal data

Longitudinal Business Database The Longitudinal Business Database (LBD) is a firm-level panel resource maintained by the United States Census Bureau that links establishment and firm records across time to support research on U.S. economy-related topics. It underpins empirical work by providing persistent identifiers for businesses, enabling longitudinal analysis used by scholars at National Bureau of Economic Research, Harvard University, Massachusetts Institute of Technology, and policy analysts at Bureau of Labor Statistics and Federal Reserve Board. The LBD is central to studies informing programs at the Department of Commerce, Congressional Budget Office, and international comparators such as Statistics Canada and Office for National Statistics.

Overview and Purpose

The LBD was developed to trace business dynamics over multiple years, supporting analysis of firm entry and exit, employment flows, productivity, and industry evolution for stakeholders including Small Business Administration researchers, Brookings Institution fellows, and academics at Stanford University and University of Chicago. Its purpose aligns with efforts at RAND Corporation, World Bank, Organisation for Economic Co-operation and Development, and International Monetary Fund to build comparable microdata for policy evaluation. The database enables linking to datasets like the Business Register, Census of Manufactures, and administrative tax records used by Internal Revenue Service analysts.

Data Sources and Coverage

The LBD integrates information drawn from administrative censuses and surveys such as the Economic Census, the Annual Survey of Manufactures, and employer tax filings administered by the Internal Revenue Service and Social Security Administration. Coverage spans private sector establishments and firms across North American Industry Classification System sectors, capturing attributes relevant to industries tracked by Bureau of Economic Analysis and U.S. Department of Labor. Panels typically cover decades, enabling comparisons across periods associated with events like the Great Recession, Dot-com bubble, and regulatory changes from acts such as the Tax Reform Act of 1986.

Variables and Structure

Core variables include firm identifiers, establishment identifiers, employment counts, payroll totals, industry codes (NAICS), geographic markers (county, metropolitan area), and time stamps used in analyses at institutions like Columbia University and Yale University. The LBD’s structure accommodates hierarchical links between establishments and parent firms, supporting studies by researchers at Princeton University and University of California, Berkeley on productivity concentration and market structure. Supplemental variables allow integration with patent data from the United States Patent and Trademark Office and trade data associated with U.S. International Trade Commission investigations.

Methodology and Linkage Techniques

Linkage techniques combine administrative identifiers, name and address matching, and probabilistic record linkage methods pioneered in work by scholars at Carnegie Mellon University and Johns Hopkins University. The LBD employs algorithms to resolve mergers, acquisitions, splits, and reestablishments, drawing on corporate events documented in filings with the Securities and Exchange Commission and registration records from state-level offices such as the California Secretary of State. Panel construction mirrors methods used in longitudinal cohorts like those at National Center for Health Statistics but tailored to firm dynamics and establishment churn.

Uses and Applications

Researchers use the LBD to study job creation and destruction, wage dynamics, firm-level productivity, and the effects of trade shocks, with findings cited in publications from American Economic Review, Quarterly Journal of Economics, and policy briefs by Urban Institute and Peterson Institute for International Economics. Applications include evaluation of programs administered by Small Business Administration, analysis of regional development in reports by Federal Reserve Bank of San Francisco and Federal Reserve Bank of Cleveland, and scholarly work on innovation linking to National Science Foundation funding patterns. Cross-linkage with longitudinal household data such as the Panel Study of Income Dynamics enables joint labor-market and firm-level research.

Limitations and Quality Issues

Quality issues include undercoverage of informal establishments, misclassification of industry codes during transition seasons, and measurement error in payroll reported to Social Security Administration. Linkage errors can arise in complex corporate restructurings, obscuring analyses of mergers reviewed by Federal Trade Commission and Department of Justice antitrust divisions. Temporal discontinuities occur when classification systems like the transition from Standard Industrial Classification to North American Industry Classification System are implemented, complicating trend analysis used by scholars at University of Michigan and Northwestern University.

Access, Privacy, and Governance

Access to LBD microdata is tightly controlled under disclosure protection standards enforced by the United States Census Bureau, with secure access arrangements similar to restricted data enclaves used by National Center for Health Statistics and Bureau of Labor Statistics research data centers. Privacy protections align with statutory authorities like Title 13 and procedures coordinating with Internal Revenue Service confidentiality rules and Privacy Act of 1974 obligations. Governance involves interagency agreements with entities such as Bureau of Economic Analysis and oversight by research review boards used by institutions including Yale University Institutional Review Boards.

Category:Databases