What the Government Shutdown Means for Federal Information Collections and Data
a discussion between Beth Jarosz and Christopher Marcum
Beth: Here at dataindex.us we are always keyed into the heartbeat of the federal data collection process. However, even our team found ourselves wondering what would happen during the government shutdown to federal information collections and data. To that end, we have brought in our friend and colleague, Chris Marcum to answer a few questions. I’ve known Chris since we were both graduate students at the University of California, Irvine and he has had a fascinating career in public service. Most salient for our discussion today, Chris was a senior statistician and scientist at the Office of Information and Regulatory Affairs (OIRA) at OMB, where the Information Collection Request (ICR) process is governed through the Paperwork Reduction Act (PRA). (This conversation focuses mainly on data collected through forms and surveys and subject to the PRA, not other types of data like those collected through streamwater sensors or satellites.)
Chris, thank you so much for helping us out here. Tell us, what happens to federal information collections and data when the government shuts down?
Chris: The government shutdown has significant impacts on our national data infrastructure by freezing the flow of critical data used by researchers, businesses, and public agencies alike. The shutdown can disrupt data collections, delay report releases, and hamper the public’s ability to access federal data. OIRA halts its review of ICRs and the PRA process stalls out pausing new collections and revisions to surveys.
Beth: So let's start with Information Collection Requests. What happens to those during a government shutdown?
Chris: Without getting into too much technical detail, OIRA – which is part of the Office of Management and Budget – governs the ICR process. OIRA is just like any other federal agency and closes during a government shutdown. Like many agencies, a few career staff will be appointed as “essential” or “excepted” for emergency and continuity purposes (likely, Deputy Administrator Dominic Mancini and maybe a few branch chiefs) but work on ICRs will cease.
This means that:
ICRs currently under review in OIRA are paused
Agencies cannot submit new ICRs, or request changes to existing ones
Nothing new from affected agencies or OIRA appears on reginfo.gov
Work stops on the 60- and 30-day notice periods for ICRs but comments from the public can continue to be submitted
Nothing on ICRs is published in the Federal Register, with extremely limited exceptions
One possible, but unlikely, exception within OIRA is work conducted under the Inflation Reduction Act of 2022 (IRA). OIRA has several career staff whose positions were filled under the special hiring authority of the IRA. I was one of them and was exempt during the last shutdown. Theoretically, some ICR work related to the IRA could continue as funding for many of the IRA programs did not expire on October 1, 2025.
Beth: Can the public still submit comments on ICRs during a government shutdown?
Chris: Absolutely. However, federal agency staff will not be able to receive, review, or respond to comments until after the government reopens. Additionally, should an agency’s comment submission system go offline during the shutdown it is unlikely to be restored until the lapse in appropriations is over. You might recall that reginfo.gov went down for a couple weeks about a month ago? It would be even harder for GSA to bring it back up during a shutdown.
As for comment periods, there are some scenarios where agencies may consider providing an extension or reset on their 30- or 60- day notices once the shutdown ends. For instance, the public may not realize that comments can be submitted and the agency wants to provide the chance, or they may use the opportunity as a reason to justify a longer timeline on an ICR. If this is the case, the agency will huddle with OIRA to make a decision on a case-by-case basis.
Beth: So what about data that are actively being collected in the field? Are there any federal collections that will continue during this shutdown?
Chris: Many administrative information collections will continue, albeit at a slower cadence. For instance, Medicaid will continue to process forms and accumulate data. Funding doesn’t run out for Medicaid until the end of the calendar year. The NIH Clinical Center and VA hospitals remain open as necessary to protect life and I’m sure patient clinical records will continue to be updated in those settings. However, most other statistical and research data collections will pause for the duration.
There are a few notable exceptions within the federal statistical system. For example, Census will continue limited operations using FY25 carryover funds for its mission critical 2030 Decennial Census testing. The Microeconomic Survey Unit (MSU) within the Federal Reserve Board of Governors is not funded through the same appropriations process as other agencies. The MSU remains open during a shutdown and its Consumer Finances Survey collections and release schedule should be unaffected by the current lapse in appropriations. The Department of Energy has also indicated that the Energy Information Administration, a principal statistical agency, has sufficient carry-over funding to continue normal data collection and release activities. Also, the Statistics of Income program at the IRS will likely remain operational for at least the first few days with funding from the Inflation Reduction Act of 2022. Finally, the Department of Transportation’s shutdown plan indicates that programs funded through the Infrastructure and Investment Jobs Act will continue; The Bureau of Transportation Statistics does have funding authorized and appropriated through that law, so it’s possible that some BTS statistical products could still come out during the shutdown.
Beth: Pivoting to the output of those information collections: Will federal datasets still be available during a shutdown?
Chris: It depends. Some open government data assets will remain available to download from the internet as long as an agency’s website is live. For example, public use files from the Census Bureau’s Center for Economic Studies will remain available, as will the genomic open datasets from the National Library of Medicine’s dbGaP. However, open government data products hosted by third parties could go down if the underlying contracts run dry, although that would be a rare circumstance and I can’t think of any examples at the moment. Mainly, the biggest risk to open government data during the shutdown is that if a website or server hosting the data goes down, there may not be anyone there to fix it and bring it back up. In that scenario, the public might check non-governmental archives of data for back-ups like the Internet Archive or the data liberation project.
Restricted data, on the other hand, will invariably be inaccessible during a shutdown. Review of requests to access controlled-access datasets will pause, as will disclosure review of ongoing projects that already use those data. That means economic studies using confidential statistical data requested through the standard application process, or infectious disease research needing access to restricted clinical data, will not move forward until the government reopens.
Beth: If I FOIA’d a bunch of data under an expedited request, would I still get it?
Chris: FOIA offices are closed during a government shutdown. You will not get your data until the government reopens, assuming your request actually gets cleared for release.
Beth: What about the monthly jobs report or other data scheduled for release by statistical agencies?
Chris: I expect that there will be very few data updates or reports released by statistical agencies during a shutdown.
As part of their orderly shutdown procedures, each statistical agency head notifies the US Chief Statistician at OMB of any impact the shutdown will have on its release schedule. OIRA grants exceptions to the release schedule due to the lapse of appropriations and approves any reasonable proposed modifications requested. There are two exceptions that would authorize an agency to follow through with a scheduled release during a shutdown and neither are likely to transpire during this one: a statutory requirement or a court order.
Frustratingly for us data nerds, OMB has decided not to consolidate agency shutdown plans and share them with the public on its website. It’s hard to reconcile how this decision conforms with the Administration’s concentration of centralized power under its unitary executive theory views. Luckily, outside stakeholders have put together useful lists of these plans, including this one from Federal News Network. Still, the plans are not standardized. Some are very specific, delving into the suboffices of each agency like the Department of Commerce does and some, like the Department of Health and Human Services, are vague and lack any meaningful programmatic data. I put together a summary table for the 16 OMB recognized statistical agencies and units that dataindex.us readers may find helpful.
What’s more, is that it will take time after the government reopens to get the release schedule back on track. Under normal circumstances, the days ahead of a scheduled release can be quite intense for statistical agency staff: data preparation, curation, finalizing documentation, and quality control checks all have to be completed prior to staging a statistical product. This means that the shutdown creates a backlog of work that will need to be completed once the government reopens. Depending where along the release timeline the shutdown occurred, this could delay the posting of data and reports by as little as a day, at best, or as long as the duration of the shutdown, at worst, after reopening.
Beth: You mentioned statutory requirements as one of the situations in which data might still be released. The “big ones” I can think of (like the date required to release census apportionment counts), aren’t relevant right now. But I can think of a few, like the monthly jobs numbers, that seem like they should count as required. Do we know what will happen to those?
Chris: What constitutes a “requirement” is notoriously vague, and agencies and OMB can take a lot of latitude in interpretation. Talking with colleagues about past shutdowns, there have been times when work on data with Congressionally mandated deadlines was deemed essential. I agree that the monthly labor statistics do seem like they could rise to that level, but BLS has already stated that they won’t publish during the current shutdown.
Beth: Is there anything else you wish people knew about the effect of a shutdown?
Chris: We’ve had shutdowns in the past - some short, and some lengthy. This one, though, feels different. OMB has raised the stakes with rhetoric about making furloughs and program cuts permanent during the shutdown and agencies engaging in unprecedented political speech about who to blame for it. I think it’s important to remember that career staff in the federal statistical system - including those coordinating it in OIRA - need to feel supported during this uncertain time. They are the life force of federal data and this shutdown pulls them away from being able to do the work we all rely on for research and analysis.
Christopher Steven Marcum formerly served as Senior Statistician and Senior Scientist in the White House Office of Management and Budget and as Assistant Director for Open Science and Data Policy in the White House Office of Science and Technology Policy.