Using simple data science techniques, Argive produced a master list of regulations that may be eligible for review under the Congressional Review Act of 1996. Read on for report details and to download a copy of the data.
In 1996, the Congressional Review Act granted Congress the power to review and block new federal regulations within 60 legislative days of a rule’s submission to Congress and the Government Accountability Office (GAO). When Congress issues a “joint disapproval” that passes both houses and is signed by the President, it prevents a regulatory agency from passing a “substantively similar” rule in the future. This year, the Act generated buzz when Congress discovered gaps in past agency submissions, offering an opportunity to restart the review clock on many existing rules.
Argive estimates that from 1996 to present day, 87,147 rules have been submitted to the Federal Register and 68,638 to the Government Accountability Office (GAO), which oversees CRA compliance. Anything entered into the Federal Register but missing from the GAO list would hypothetically be eligible for Congressional oversight*. However, manually comparing these lists would require an extraordinary amount of effort--on top of introducing the possibility of human error.
Argive viewed this problem as a unique opportunity to demonstrate how data science techniques can be applied to the regulatory review process. To generate a list of CRA eligible rules, we retrieved *federal rules submitted by agencies to the Federal Register using the Federal Register dataset from the Mercatus Center’s open-source QuantGov project. Next, we filed a FOIA request with the GAO for a file of all rules submitted under the CRA between 1996 and 2015 (Note: the GAO is still compiling 2015-2017 data).
Relying on a combination of the Regulatory Identification Number (RIN) and rule titles, we compared the two datasets to produce a master list of regulations that were potentially never submitted for Congressional review.
Due to limitations in the raw agency data, we recommend users check our work by searching the Federal Register to confirm a rule's eligibility (see our "Report Documentation" below). The report has been sorted to highlight Federal Register rules at the top that are most likely to be CRA eligible.
Argive’s mission is to improve efficiency and transparency in regulatory management by serving as a data resource to the public. This is just one example of how existing technologies can be deployed to tackle large-scale regulatory data problems.
For more information on our reporting logic, or to try your hand at matching using the original sets, please refer to the below:
- Argive CRA Report Documentation
- Report Citation: If you found our report helpful, please cite Argive and link to this page!
- Special thanks to our data wizard, Yihwan, and FOIA expert, Maleka, for their contributions to this report.
- QuantGov Federal Register Dataset
- GAO Data
See how others are using our data:
- The Cause of Action Institute used Argive's data in its own analysis of CRA eligible rules. Learn more here.
- The Brookings Institution cited Argive's report in a comprehensive article on the CRA's recent revival. Learn more here.
- The CRA requires only that reporting “to Congress” occur in order for the 60-day window for a disapproval resolution to open. Reporting "to Congress" does not directly reference the GAO, and there is a possibility that rules unreported to the GAO were properly reported to the House and Senate, and thus would not be vulnerable to a CRA disapproval resolution.
- It has come to our attention that we may have double-counted some rules by counting all "Final Rule" entries in the Federal Register, which include procedural actions like requests for public comments or announcing dates of effectiveness.