Housed within the National Academies of Sciences, Medicine and Engineering (National Academies), for more than 50 years the Committee on National Statistics (CNSTAT) has helped strengthen the federal statistical system.
But as the National Academies have recently lost $40 million in federal contract funding, they have begun consolidating and reducing staff. Reporting indicates that up to one-fifth of the roughly 60 top-level advisory boards that produce much of the National Academies’ output may be eliminated.
Eliminating CNSTAT would remove a key contributor to the quality of federal statistics. Its independent evaluations, methods guidance, standards, and cross-agency problem-solving are essential to credible government policy development, including tax analyses, as well as business and household decision-making.
A highly decentralized statistical system needs an independent quality engine
The US Office of Management and Budget (OMB) recognizes 16 principal statistical agencies and almost 100 additional statistical units across the federal government. Together, they inform public- and private-sector decision making, academic research, and public understanding. For example, businesses use population statistics to determine where to locate and use information on economic conditions to determine when and how to expand.
The Office of the Chief Statistician of the United States, appointed by the OMB director, provides necessary coordination but is perpetually underfunded given its broad mandate to coordinate, guide, and oversee US federal statistical agencies and activities.
Fortunately, CNSTAT fills gaps with hands-on assistance. As former Director of the IRS Statistics of Income Division (SOI), I saw CNSTAT play a central role in improving the data and statistics produced by SOI and encouraging responsible uses of tax data to improve tax and other policies at all levels of government.
Drawing on the best scientific expertise from academia, industry, nonprofit, and government sectors, CNSTAT has released over 300 publications for approximately 50 sponsor agencies, including SOI, all aimed at helping agencies be more effective in meeting user needs. Through CNSTAT’s efforts, business owners get more timely and detailed local markets and workforce data, helping them choose sites, set prices, negotiate leases, and plan inventory. Consumers’ paychecks, benefits, and cost-of-living pay adjustments more closely track reality.
Consensus Reports change and improve practice
Consensus Reports—CNSTAT flagship products—take a deep dive into a specific topic and share actionable recommendations. They give the US statistical system the foundation to address emerging issues, such as how to measure and report on the gig economy and ways to modernize the Consumer Price Index, the most widely used measure of inflation.
A recent series of Consensus Reports challenges the statistical system to embrace data and technology innovations to create a modern 21st Century data infrastructure that would rely heavily on non-traditional data sources, including administrative government (such as tax data) and commercial data.
This approach would also facilitate the Committee’s recent recommendations for creating an integrated data set on income, consumption and wealth using data from multiple sources and grounded in information provided by the IRS to the Census Bureau. Such data would address current data gaps and reduce misalignment among statistics released by different agencies, which will better support policymakers.
Public workshops offer cost-effective training and professional development
CNSTAT speeds adoption of proven methods across agencies through its public workshops. Experts deliver technical presentations that CNSTAT often publishes to provide important references.
Some workshops provide feedback to a specific agency, such as a workshop on 2020 Census data products. Others provide valuable educational opportunities to a wide audience, such as webinars on using non-survey data for statistics and evidence-building, understanding and fostering public trust in official statistics, the importance of reproducibility and transparency in statistical research, and ways to improve measurement of poverty.
When I led SOI, these workshops provided my team with low-cost, high-impact training that could be directly applied to their daily work. Two memorable in-person events held at the National Academies’ headquarters – Big Data Day and AI Day – convened staff from multiple statistical agencies to showcase examples of how large, non-traditional datasets, such as commercial data collected from grocery store scanners or satellite images, were transforming official statistics and how then-emerging AI (artificial intelligence) tools were making the statistical system more efficient.
These types of low-cost, peer-to-peer sharing experiences jumpstart innovation and build capacity across the statistical system. This is exactly what a decentralized statistical system requires, and it is uniquely possible thanks to the CNSTAT mission to continually advance federal statistics.
CNSTAT sets the standard to protect independence, privacy, and trust
No discussion of the importance of CNSTAT to American statistics would be complete without mentioning Principles and Practices for a Federal Statistical Agency. This publication, updated every four years and shared with each administration’s incoming staff, outlines the expectations of a well-functioning federal statistical agency, including the requirement that it produce timely, useful, reliable statistics in a manner that fosters public trust.
Now in its eighth edition, this publication was the backbone of the Foundations for Evidence-Based Policymaking Act and its regulations, and is cited in other OMB Statistical Policy Directives. Periodic updates ensure that it remains relevant as technology and public expectations evolve. I regularly shared Principles and Practices with new SOI staff to help them understand their role, and with incoming IRS leadership to help them to better understand and support SOI’s mission.
CNSTAT exemplifies the National Academies’ charge to “provide independent, objective advice to inform policy with evidence, spark progress and innovation, and confront challenging issues for the benefit of society.” And CNSTAT is helping the National Academies achieve its vision for “a nation and a world that rely on scientific evidence to make decisions that benefit humanity.”
Tight budgets at federal statistical agencies elevate the CNSTAT’s importance. Its role as an active and engaged standing committee has never been more important to safeguard the vitality and viability of the federal statistical system. CNSTAT’s continued operation supports the integrity and utility of federal statistics for policymaking and decisions by businesses and households alike.
