How advanced analytics can find outdated federal regulations

What if there was a way to find overlapping or outdated regulations and target those first?

Advanced analytics, as outlined in this study, can help agency leaders both expand their view beyond what they and their staff are familiar with and look more specifically into sets of regulations with a higher potential opportunity for improving efficiencies, without reducing protections.

One example is a regulation from 1905 which pertains to a specific method of log rolling on a river in Minnesota. No one has rolled a log on this river since 1937.

Do we still need this regulation on the books?
To identify regulations that could be candidates for reform, Deloitte analyzed all 217,000+ sections of the 2017 Code of Federal Regulations (CFR). Our results suggest that advanced analytics could provide practical assistance in identifying targets of opportunity for regulatory reform.

Advanced analytics: An answer to regulatory reform?

Two-thirds of all federal government regulations currently on the books have never been updated.

If agency executives only examine recent regulations, they may miss an opportunity for reform that could help eliminate significant, unnecessary burden. It could be similar to the classic mistake of “looking for your keys underneath a street lamp because it’s easy to look there.”

In this paper, we’ll walk through common elements agencies can use to help improve the quality of the regulations they have on the books—and ways leaders might start to tackle potentially old, outdated, and overlapping regulations they may no longer need.

What is text analytics?

Text analytics involves the use of computers and software algorithms to process large volumes of text with the intent of extracting quality insights. It is widely used both within and outside of the federal government to analyze documents (like the CFR) which are too voluminous for humans to efficiently handle.

It often makes it possible to detect trends and patterns that, due to the size of the underlying text, humans would generally have trouble identifying.

Within the Code of Federal Regulations, more than 17,000 sections have extremely similar text as another section. Most of these matches are exact matches where the words are precisely the same in both sections (often legal definitions or standard penalties and definitions).