Sections

Commentary

It is impossible to make sound policy without high-quality data

A general view of the U.S. Capitol building in Washington

The President’s 2019 budget contains a variety of proposals that would reshape policy and reallocate spending across various agencies and policy programs. As the nation debates the pros and cons of these proposals, it is imperative that they be informed by reliable data—data that is often collected and made freely available by the federal statistical agencies.

Since the nation’s founding, our federal statistical agencies have produced data that is vital to both business and policy decision-making. In an effort to underscore the need for continued federal support of this essential government function, The Hamilton Project and the American Enterprise Institute published a joint report in 2017 exploring the many valuable uses of government-collected data.

Unfortunately, the President’s 2019 Budget does not make federal data infrastructure a priority. While proposed FY 2019 funding for the Bureau of Labor Statistics (BLS) and other agencies remains constant, cuts have been proposed to funding levels for other key statistical agencies including the Bureau of Economic Analysis (BEA), Bureau of Justice Statistics (BJS), and Economic Research Service (ERS).

Given the importance of the decennial census to our democracy, federal policymakers should be acting now to ensure that the 2020 census is well-managed and sufficiently funded. While the FY 2019 budget proposal actually recommends a significant funding increase for the U.S. Census Bureau, it remains unclear whether the spending ramp-up will prove sufficient for the Bureau to accomplish its constitutionally mandated responsibilities.

While the data-collection activities of the federal statistical agencies—the U.S. Census Bureau, the BEA, the BLS, among others—do not require a particularly large investment in the context of the federal budget, as indicated in the figure below, they are not without cost. Accordingly, to ensure that our data is accurate and reliable, it must be carefully collected and analyzed by skilled personnel, and this requires our government to allocate adequate funding within the federal budget.

this figure shows the amount of federal spending by purpose

Failing to adequately fund our nation’s statistical agencies threatens to harm the quality of the evidence necessary to inform many of our nation’s policy decisions. National-level statistics collected by the BLS, such as the unemployment and labor force participation rates, or statistics collected by the BEA, such as GDP, are essential to the functioning of both the public and private sectors. Without this information, policymakers like those at the Federal Reserve would be unable to fulfill their mandate of generating maximum employment and stable prices.

Our economy’s increasing complexity necessitates ever more fine-grained analysis by businesses, much of which can only be accomplished with the help of government data. Suppose that a retailer wants to tailor its product mix to the needs and preferences of consumers in specific local communities: the Census Bureau’s American Community Survey (ACS) helps it to do so. If a university would like to know how many college-age students live in its region, or a local government wants to formulate a regional transportation plan, it would use the same dataset. The figure below, showing the elderly share of the population across the United States, provides an example of this sort of detailed demographic data provided by the ACS.

this figure shows the share of the elderly population as it is distributed across the United States.

Without sufficient funding, these statistical activities are in danger of yielding less-useful and less-timely data. As the U.S. economy becomes more reliant on data, we should not be reducing our financial commitment to the statistical agencies; instead, we should be exploring new opportunities for federal investments that are valuable to the private sector and policymakers alike.

Authors