Economic Intelligence

Enhancing the Federal Statistical System to Support U.S. Competitiveness

Andrew D. Reamer looks at ways the federal statistical system can better support the nation's economic competitiveness.

Download this report (pdf)

Download the introduction and summary (pdf)

Read this report in your web browser (Scribd)

See also: Overview: Series on U.S. Science, Innovation, and Economic Competitiveness, Rewiring the Federal Government for Competitiveness by Jonathan Sallet and Sean Pool, Universities in Innovation Networks by Krisztina “Z” Holly, Building a Technically Skilled Workforce by Louis Soares and Stephen Steigleder, and Immigration for Innovation by Marshall Fitz.

In discussing about policy options for promoting U.S. economic competitiveness, it’s unusual for anyone to consider producing better statistics. Grants, tax credits, regulation, agency reorganizations, yes. But numbers? The federal data system sits like a large black box in a dark shadow. We know a few high-profile stats shed light on how we’re doing economically, such as GDP and unemployment, but most everything else is opaque. We don’t quite understand what else the system contributes to economic policy or, to be honest, how it works. And so it may not be immediately obvious how the statistical system could better support the nation’s economic competitiveness.

Federal economic statistical agencies—particularly the Census Bureau, the Bureau of Labor Statistics, and the Bureau of Economic Analysis—produce the data that guide federal economic policy. For many decades, the primary focus of federal economic policy has been managing the business cycle, that is, preventing recession and inflation. Statistical agencies’ explicit priority, therefore, is to provide the macroeconomic data to work the levers of fiscal and monetary policy. Even though U.S. competitiveness has become progressively more vulnerable since the early 1980s, the federal government yet to construct a coherent, well-integrated approach to addressing the global challenges to the nation’s economic structure. Consequently, the statistical system hasn’t been asked to come up with the numbers that would support intelligent competitiveness policy.

What would those numbers look like? Federal competitiveness policy, if one existed, would systematically identify and address barriers to the efficient functioning of markets. These barriers contribute to what economists call “market failures” that impede the ability of traded industries to successfully respond to market issues and opportunities. Distinct from the broad “top-down” orientation of macroeconomic policy, federal competitiveness policy would involve more “bottom-up” efforts aimed at improving the likelihood that market actors make decisions that enhance competitiveness. These market participants include business, research and education institutions, workers, and students, as well as thousands of public purpose organizations at all geographic levels, such as regional economic and workforce development agencies.

Identifying and addressing market failures requires sufficient data on the structure and competitiveness of key traded industries (number of jobs, productivity, international trade), the building blocks of competitiveness (innovation, entrepreneurship), the basic factors of firm operations (workforce, finance, research and development expenditures), and the impacts of public programs that support firm competitiveness. The private sector does not have the capability to provide the numbers needed to assess and enhance the nation’s competitiveness. But the federal government does, at a remarkably low cost. Its current annual cost to track the workings of a $14 trillion economy is about $1.7 billion.2 Additional statistical funds needed to support competitiveness policies would bring the total closer to $2 billion.

And yet the U.S. statistical system doesn’t produce the numbers needed to assess and guide national competitiveness. In the absence of a coherent approach to federal competitiveness policy, statistical agencies continue to give priority to the data needs of macroeconomic policy. The production and analysis of competitiveness-relevant data are further crimped by inadequate congressional funding, lack of understanding of data user needs, lack of coordination and collaboration among agencies, insufficient encouragement to be innovative, and outside analysts’ difficulties in gaining access to and working with the data.


  • The Census Bureau and the Bureau of Labor Statistics cannot provide measures of industry size that are consistent with one another in terms of jobs and earnings.
  • BLS says it overestimates growth in manufacturing productivity, a key dimension of competitiveness, because it has difficulty properly accounting for shifts in obtaining manufacturing inputs from domestic to foreign sources.
  • The federal government lacks comprehensive, useful measures of innovation activity.
  • It also does not have adequate data on entrepreneurs’ access to capital.
  • The Bureau of Economic Analysis no longer publishes detailed data by state on foreign direct investment. As a result, states, and the president’s own SelectUSA initiative, are working in the dark as they try to attract foreign-owned firms to these shores.
  • Regional development agencies cannot see business R&D activity by metropolitan area.
  • Educational institutions can’t find regional jobs data that would let them offer credential programs that match employer demand.


To provide the data needed for competitiveness policy, the federal economic statistical system should adhere to five principles:

  • Be demand-driven: The federal economic statistical system should be responsive to the decision-making needs of the wide array of actors that influence economic competitiveness, including those in the Federal government, at the state and regional level, in industry, and in the workforce.
  • Be innovative: Federal statistical agencies should develop innovative cost-efficient approaches to producing data needed for competitiveness policy.
  • Be collaborative: The White House Interagency Council on Statistical Policy should create an interagency working group on competitiveness statistics to develop and implement a common agenda.
  • Be accessible: Federal agencies should construct web-based platforms that allow easy access to, and customized building of, data tables. Further, agencies should provide researchers’ access to microdata (individual records), while protecting confidentiality, so that they may explore the factors that influence competitiveness.
  • Be sufficiently funded: If federal agencies are to obtain adequate funding, they must better educate policymakers about the importance of their data products to the economy.


Putting these principles into action, the federal government should improve economic statistical programs to better support competitiveness by facilitating analysis of traded sectors, the intermediate outcomes that determine competitiveness (innovation, entrepreneurship), the underlying factors of competitiveness (R&D expenditures, workforce), and program impacts. The relative cost of these recommendations is minimal compared to their substantial potential long-term impacts on jobs, wages, and government revenues in a $14 trillion economy.

Improving traded sector analysis

Problem: Economic analysts lack access to data needed to assess the competitiveness of key traded industries.


  • In order to make economic data consistent across the government, Congress should pass a law allowing the Census Bureau to share IRS-derived business data with BLS and BEA.
  • The Office of Management and Budget should revamp the North American Industrial Classification System, or NAICS, so that analysts can see data on all the establishments involved in a particular industry sector. At present, for example, General Motors headquarters is classified in the “management” industry, not the “automobile manufacturing.”
  • The Department of Labor should request, and Congress should approve, funds for BLS to create an input price index to more accurately measure manufacturing productivity.
  • The Census Bureau’s Center for Economic Studies should implement the low-cost, high-impact planned expansion of its Longitudinal Business Database, or LBD, by incorporating new datasets from other sources, including those on patents, foreign direct investment, imports and exports, and management and organizational practices.
  • The Labor Department’s BLS and Employment and Training Administration should work together to implement a new classification typology, business processes, to all business establishments.
  • The Securities and Exchange Commission should implement its contemplated bulk download tool that would allow researchers access to the full set of firm SEC filings.
  • Congress should fund BEA to organize data by legal organization of firm (i.e., C corporations, S corporations, partnerships, and sole proprietorships), at a cost of about $1 million or less.
  • Congress should provide $3 million for BEA to improve its collection of foreign direct investment data.
  • Congress should take steps that allow BEA to carry out its proposed improvements in trade in services statistics.
  • The Department of Labor should request, and Congress should provide, the small amount of funds required by the BLS to build import and export price indices that fully cover traded services.
  • Federal statistical agencies should collaboratively explore ways of building a technology balance of payments measure for the U.S.
  • The Department of Labor should request, and Congress should provide, the small amount of funds that would allow BLS to create a foreign currency price index.
  • Congress should see that BEA has the funds to produce price indices that allow comparison of the costs of doing business across U.S. regions.

Improving measures of intermediate outcomes that influence competitiveness

Problem: Federal statistical agencies do not provide adequate measures of intermediate outcomes that influence competitiveness, particularly innovation, entrepreneurship, and relationships between organizations.


  • BEA and the National Science Foundation’s National Center for Science and Engineering Statistics should co-lead an interagency working group to coordinate disparate efforts to develop and implement measures of innovation activity.
  • The Census Bureau and BLS should take the steps necessary to make their longitudinal business databases—Business Dynamics Statistics and Business Employment Dynamics—more useful to researchers and policymakers in identifying the role of entrepreneurship in national and regional competitiveness.
  • The Commerce Department’s Economic Development Administration should take the lead in seeing that the federal economic statistical system enables regional economic development organizations to identify and analyze intra-cluster relationships and determine the implications of these for regional competitiveness.
  • Congress should provide the Census Bureau with funds sufficient to conduct the 2012 Economic Census.

Improving factor analysis

Problem: The federal statistical system does not provide sufficient data regarding the factors that influence competitiveness, including R&D expenditures, workforce, education and training, business finance, and energy.


  • The National Center for Science and Engineering Statistics should provide a current, complete, detailed picture of R&D expenditures for the nation, states, and regions.
  • NCSES also should expand its analysis of science and engineering workforce and so-called STEM (science, technology, engineering, and math) education to include workers with less than a baccalaureate degree.
  • The Department of Labor should request, and Congress should approve, funding that would allow BLS to improve state and regional labor market data, particularly on occupations and labor turnover, and retain the International Labor Comparisons program.
  • The Department of Labor should request, and Congress should approve, funding that would allow the Employment and Training Administration to improve the accuracy and detail of the O*NET occupational classification database and support state-level workforce information programs and decisions tools.
  • The Education Department’s National Center for Education Statistics should seek, and Congress should approve, continued funding for data programs that track post-secondary educational activity, credential attainment, and workforce outcomes.
  • The Census Bureau should implement its planned Local Employment Dynamics program job-to-job flows tool.
  • The Secretary of Labor should expand Workforce Information Council membership to include all federal statistical agencies providing workforce and education data and representatives of their state counterparts. Members of the new WIC should collaborate to identify and address labor market information failures.
  • The Small Business Administration should convene and lead an interagency working group to identify and address the needs for small business finance data.
  • Congress should provide $1.2 million to BEA to create a satellite account that measures the role of the energy industry in the economy and its impacts on economic activity.

Improve evaluation of competitiveness programs

Problem: Federal, state, and regional competitiveness-related program agencies lack data on the impacts of their programs.

Solution: The Census Bureau should create a program to use the Longitudinal Business Database to assess the impact of program support to individual firms in terms of survival, revenues, jobs, wages, exports, innovation, and other outcomes related to competitiveness.

Andrew D. Reamer is Research Professor at the George Washington University Institute of Public Policy.

Download this report (pdf)

Download the introduction and summary (pdf)

Read this report in your web browser (Scribd)

See also:

The positions of American Progress, and our policy experts, are independent, and the findings and conclusions presented are those of American Progress alone. A full list of supporters is available here. American Progress would like to acknowledge the many generous supporters who make our work possible.