Threats to Government Data Are Threats to Democracy

The federal government bears few responsibilities as weighty as its obligation to be a good steward of data and information. From shaping our congressional districts, to understanding students’ academic progress; from informing budget priorities and their impact on future generations, to simply knowing how many people live in the country; the quality of our policies and policymaking are contingent upon the data they’re based upon.

As an organization that relies on data, researchers at the Brookings Institution are paying close attention to the Trump administration’s management of data. Here are a few of the data sources our scholars will be watching in 2018 and beyond:

Census Bureau

If data is the lifeblood of policy, then the Census Bureau is the heart that pumps it. Not only is Census the federal government’s largest statistical agency, the data collected touches all other departments within the federal government, primarily through its official count of residents in the country.

Brookings expert Randall Akee says that in addition to the decennial Census, “Other Census Bureau products like the annual American Community Survey and the Economic Census, which provides comprehensive data on output and businesses in the U.S. every five years, should be of primary importance to everyone. These wide-ranging data products serve as the foundation for governance, policy advocacy, and business decisions. And the depth of these data provide insight for small populations that are not often included in meaningful numbers in other national surveys.”

The Census Bureau has already canceled some crucial preparations for the upcoming decennial census in the face of persistent underfunding that threatens the accuracy of the 2020 count. Concerns about funding have also prompted the Bureau to delay the next Economic Census.

Complicating matters, Census Bureau Director John Thompson resigned abruptly in May 2017, and a replacement has yet to take the helm. Meanwhile, some experts worry that anti-immigrant and anti-Muslim rhetoric from the White House could make it more difficult to get responses from marginalized groups, who may worry that their personal information will be used against them—especially if the agency adds a question about citizenship as the Justice Department has requested.

Federal Election Commission

Molly E. Reynolds and Curtlyn Kramer of the Governance Studies program at Brookings know that data restrictions inhibit transparency around election spending. Between 1984 and 2014, the total amount spent by U.S. House candidates roughly doubled in inflation-adjusted terms. Contributions by political action committees in House races saw similar growth over the same time period. Reynolds and Kramer find that independent spending on behalf of House candidates by non-party organizations saw an even more dramatic increase—spending was approximately 100 times greater in 2014 than in 1984.  

“Monitoring these trends, which are included in Brookings’s Vital Statistics on Congress thanks to a partnership with the Campaign Finance Institute, requires accurate, timely data releases by the Federal Election Commission,” note Reynolds and Kramer. It is not just campaign finance data, however, that the FEC collects and disseminates. The Commission is also the source of an invaluable compilation of the official, certified federal election results provided by the official election office in each state. Those data underlie several other data tables in Vital Statistics, which is one of Brookings’s longest running products and helps researchers, students, and the public monitor the work of the first branch of government.

“The past year has seen both encouraging and troubling developments for users of FEC data,” write Reynolds and Kramer. They note that the Commission rolled out a new website last May, demonstrating its commitment to modernizing its digital infrastructure. But the 2016 biannual election results were not released until December 2017, approximately five months after the document’s typical publication date between 2007 and 2013, and one month later than its delayed 2015 distribution.

“Quality federal election data is central to the work that we do on the health of the U.S. Congress, and we hope the FEC continues to provide it in timely and accessible way,” Reynolds and Kramer add.

Centers for Disease Control and Prevention

Like most of the agencies on this list, the CDC is facing steep budget cuts: Trump has proposed reducing the agency’s budget by 17 percent, or $1.2 billion. Late last year, there was an uproar over the agency’s alleged ban on words like “transgender,” “diversity,” and “fetus” in the CDC’s budget request. This is probably a self-imposed exercise in linguistic gymnastics to avoid provoking CDC’s funders. But wherever the pressure to vet vocabulary is coming from, Brookings experts hope that health data will not bend to political pressure.

Carol Graham and her team at Brookings have been using CDC data to investigate rising rates of premature mortality among non-Hispanic whites in the U.S. due to preventable causes such as alcohol poisoning, drug overdose, and suicide. Graham worries about what the politicization of the CDC’s language means for populations like the LGBTQ community, who face specific health threats that are currently documented by the CDC.

“How can we accurately document the key drivers of our crisis of premature mortality if we cannot explicitly identify particular demographic cohorts?” asks Graham. “As a scholarly and scientific community, we must be vocal in the dangers that these trends pose.”

EPA, Interior, NOAA, and NASA

When the United States withdrew from the Paris Agreement, it sent a global message about the administration’s stance on climate change. It remains to be seen how this will affect the collection of federal climate data.

Trump appointees seemingly are steering the Environmental Protection Agency and the Interior Department away from discussions of, or even references to, climate change. The White House has proposed cuts to the National Oceanic and Atmospheric Administration’s climate research programs and future satellite launches, as well as to NASA’s Carbon Monitoring System. And ProPublica reported in December that more than 700 people—including over 200 scientists—have left the EPA since Trump took office, leaving the agency with a dire shortage of expertise.

But even amid these developments, climate research is ongoing. In November, NOAA and NASA launched JPSS-1 (now called NOAA-20), a polar satellite that will extend a continuous 30-year data record on climate change, though the Trump administration has advertised it solely as a weather-forecasting system. It is worth keeping an eye on these two non-policymaking agencies to see if they continue to collect and discuss climate data in the stead of the reigned-in EPA and DOI.

Housing and Urban Development

The American Housing Survey is funded by HUD and conducted by the Census Bureau. Brookings expert Jenny Schuetz explains why the AHS is crucial for researchers, policymakers, and industry members: “The AHS provides the most reliable and historically consistent estimate of U.S. homeownership rates over time. It follows individual housing units and allows users to track changes in price, quality, and residents of the same house over time. Short topical components provide more details on timely, relevant topics, such as food insecurity and the financial situation of homeowners after the Great Recession.”

HUD has requested a constant level of funding for research and technology, which would include support for the AHS. But research is not performed in a vacuum: as other parts of HUD show signs of decay, we at Brookings hope that this vital data source is properly maintained.

Bureau of Labor Statistics

BLS employment, productivity, and inflation statistics are probably the most-cited indicators of U.S. economic performance. Unfortunately, the Bureau has been underfunded for years, resulting in the elimination of some data products, such as statistics on mass layoffs and green jobs, and cutbacks in hiring and various research and development projects. The White House has proposed a 21 percent cut to the Department of Labor, which houses the BLS, though the Wall Street Journal reported last May that the Department does not plan to cut any BLS programs or surveys.

Erica Groshen, who served as the Commissioner of Labor Statistics from 2013 to 2017, warned in an op-ed that "even holding the BLS budget flat at its 2016 level would provide $25 million less than needed to continue current activity . . . leaving the BLS short-handed risks serious errors or delays in its statistics. It hasn’t happened yet, thanks to the BLS’s dedicated staff, but sooner or later it will."

Federal Reserve

On a more positive note, the Federal Reserve (which funds itself to maintain independence) is using data to keep up with innovations in financial technology. Aaron Klein, the policy director of the Brookings Center on Regulation and Markets, is watching how technology is radically changing how Americans use financial services.

The term “FinTech” captures this merger: existing transformations such as online and mobile banking, and future changes that could transform all aspects of finance. This new and rapidly evolving world has few measures of data to allow the public to see what is changing, how rapidly, and who is being impacted. The Federal Reserve’s Consumers Use of Mobile Financial Services data set is one of the only pieces of such information. It began recently in 2011, but has already captured significant changes in how technology is changing the use of money.

“Did you know that smartphone usage is uncorrelated to income? Or that among people with smartphones, African Americans and Latinos are more likely to engage in mobile banking than whites?” asks Klein. These data have been critically valuable in shaping how entrepreneurs and policymakers think about FinTech and achieve shared goals of improving financial services for all Americans. While annual collection may seem more frequent than other data sets, the rapid nature of technological change means that real world realities change quickly.

Andre Perry is a David M. Rubenstein Fellow in the Metropolitan Policy program at the Brookings Institution. Katherine Guyot is a research assistant in the Economic Studies program at Brookings. Also contributing to this piece are Brookings scholars Randall Akee, Molly E. Reynolds, Curtlyn Kramer, Carol Graham, Jenny Schuetz, and Aaron Klein.