September 7, 2022

RegAuthorities: The Regulations Authorities Dataset

Contact us
To speak with a scholar or learn more on this topic, visit our contact page.

The Regulatory Authorities Database (RegAuthorities) is a new dataset that reveals the relationship between laws passed by Congress and regulations issued by federal regulatory agencies. Released by Kofi Ampaabeng, Dustin Chambers, Ethan Greist, Patrick McLaughlin, Jonathan Nelson, and Thurston Powers, the dataset provides a means for researchers to quantify how the language characteristics of statutes influence the number of regulations.


Because US regulatory agencies promulgate regulations pursuant to mandates created by Congress, it is possible to map regulations back to their congressional origins. Ampaabeng et al. describe this mapping in RegAuthorities. The dataset (a) elucidates the scope of statutes as they manifest themselves as regulations, (b) tracks the evolution of laws into tangible regulations, and (c) quantifies the impact of statutory language on the volume of regulations.


RegAuthorities provides interested researchers with the tools to answer questions such as:

  • Have agencies issued fewer or more regulations than expected?
  • Do the linguistic characteristics of laws/statutes authorizing regulations influence the resulting regulatory output?
  • What are the most influential statutes or laws in terms of regulatory output by agencies?
  • Why do some statutes create more regulations than others?
  • What are the statutes that offer broad authority to executive branch agencies to regulate?

The dataset is a useful resource for various research topics, including the following:

  • Researchers can employ network analyses methods to examine the connection among regulations and statutes. This type of analyses could unearth relationships and patterns that are not readily visible from tabular data.
  • When combined with natural language programming and text analyses, researchers can examine the role of the language used in laws and statutes in regulatory output.
  • Researchers can identify the most influential laws and statutes as they pertain to regulations.


Ampaabeng et al. show that:

  • Relatively few statutes are responsible for most of the regulations in the United States.
  • Laws that empower the IRS are the most “amplified”—that is, they have the highest ratio of number of words in the authorizing statutes to the number of words in the regulations that derive from them.
  • Many statutes fail to spawn any regulations.

These findings underscore the value of the datatset as a tool to identify outdated or irrelevant statutes. As with all data products from the Mercatus Center’s Policy Analytics team, RegAuthorities will be frequently updated.