Measuring Regulatory Obligation Intensity and Pipeline Pressure, 2016–2026

<p><span>When a proposed rule becomes final, what happens to the discrete obligations it contains” This paper introduces a computational method for tracking how individual regulatory obligations survive, transform, and drift between proposal and finalization. Applying a sentence-first extraction pipeline to 520,244 Federal Register documents across all federal agencies and comparing obligation sets across 954 paired proposed</span><span></span><span>final rules, we find that only 32% of proposed obligations survive unchanged syntactically into final rules—across strict modals this same 32% survival rate is observed.</span></p> <p><span>The net obligation flow across that group is -6,902, meaning final rules shed more proposed obligations than they introduce. Deontic drift is rare (0.6%). Agency-level decomposition reveals wide variation: Treasury obligations survive at 57%, Interior at just 19%. A 2024 spike in obligation production (+47% year-over-year) reflects concentrated agency activity, with EPA accounting for 58% of the net increase. Section (3)(f)(1) major rules survive at 40, versus 26% for non-major rules, suggesting that high-stakes rulemakings undergo more intensive pre-proposal development.&nbsp;</span></p> <p><span>These findings suggest that notice-and-comment materially restructures the binding content of rules, and proposals are poor proxies for final rules. Proposed obligations are more than four times more likely to be eliminated syntactically than modified, suggesting the comment process operates predominantly as a filter rather than an editing mechanism.&nbsp;<i></i></span></p>