Regulations.gov: The Federal Database Behind 25 Million Public Comments on US Rulemaking

Regulations.gov is the federal government's unified portal for proposed and final rulemaking—hosting dockets for every significant federal regulation from 170+ agencies, 25 million public comments, and supporting documents including economic analyses and scientific studies, making it the most comprehensive public record of how federal rules are made and who influences them.

What Regulations.gov is

Regulations.gov launched in January 2003 as a direct implementation of the E-Government Act of 2002, which directed federal agencies to accept public comments electronically and make rulemaking dockets available online. The site is operated by the Environmental Protection Agency as the host agency on behalf of the entire federal government through the eRulemaking Program—an interagency collaboration managed by the EPA and funded through a cost-sharing arrangement among participating agencies. The choice of EPA as host agency reflects the agency's early leadership in electronic docket management, not any substantive role in overseeing other agencies' rulemaking.

Approximately 170 federal agencies participate in the system, ranging from major cabinet departments (EPA, Department of Transportation, Department of Labor, FDA, USDA) to independent regulatory commissions (FCC, CFPB, SEC, FERC) to smaller specialized bodies (the Surface Transportation Board, the Pipeline and Hazardous Materials Safety Administration, the Bureau of Alcohol Tobacco Firearms and Explosives). Every agency that publishes proposed rules in the Federal Register and accepts public comments is expected to route those comments through Regulations.gov, though a small number of agencies—including some components of the intelligence community and certain national security-related bodies— maintain separate docket systems for classified or sensitive proceedings.

The cumulative scope of the database is substantial. As of 2027, Regulations.gov hosts more than 25 million public comments across hundreds of thousands of documents organized into tens of thousands of dockets. These span every category of federal regulatory activity: environmental standards, financial regulation, workplace safety, food safety, transportation rules, telecommunications policy, drug approval processes, and agricultural programs. The database also holds the supporting materials agencies must disclose as part of the rulemaking record: regulatory impact analyses, environmental impact statements, scientific studies, economic modeling, interagency correspondence, and post-comment response documents.

Regulations.gov is not simply a comment submission portal. It is the official public docket for the administrative record that courts review when rulemaking is challenged. Under the Administrative Procedure Act, judicial review of an agency rule is conducted on the basis of the administrative record that was before the agency when it made its decision. The documents posted to Regulations.gov—public comments, supporting studies, agency responses, and the final rule preamble—constitute substantial portions of that record. The completeness and integrity of the Regulations.gov docket is therefore a legal as well as a transparency matter.

The rulemaking process

The legal framework for federal rulemaking is the Administrative Procedure Act of 1946. The APA establishes “notice and comment” rulemaking as the standard process for agencies issuing binding legislative rules. The procedural sequence is defined by statute and has been elaborated through decades of executive orders and judicial decisions.

The process typically begins with an Advance Notice of Proposed Rulemaking (ANPRM), which agencies use when they are considering regulatory action but have not yet developed a specific proposal. An ANPRM invites public comment on whether regulation is needed, what form it might take, and what data or considerations the agency should factor in. ANPRMs are common for complex or politically significant rulemakings where agencies want to test public and industry reaction before committing to a specific rule design.

The central document in the rulemaking process is the Notice of Proposed Rulemaking (NPRM), published in the Federal Register. An NPRM must include the text of the proposed rule (or a description of its substance), the legal basis for the agency's authority to issue the rule, a description of the problem the rule addresses, and an opportunity for public comment. The comment period is typically 30 to 60 days, though significant or complex rules frequently carry 60- or 90-day periods, and agencies sometimes grant extensions in response to stakeholder requests. During this period, anyone—individual citizens, corporations, trade associations, academic researchers, state governments, foreign governments, or other federal agencies—may submit a comment.

After the comment period closes, the agency must review all submitted comments. The APA does not require the agency to accept any particular comment or change the proposed rule in response to public input. It does require the agency to consider the comments and respond substantively to significant ones in the preamble of the Final Rule. “Significant” in this context means comments raising new data, legal arguments, or policy considerations the agency did not already address in the NPRM—not merely comments expressing opposition or support. A form letter submitted by thousands of identically worded signatories counts as one substantive comment (or possibly no substantive comment at all, depending on its content). An agency that ignores a well-documented technical critique from a scientific institution risks having the final rule vacated by a reviewing court on arbitrary-and-capricious grounds.

The Final Rule is published in the Federal Register with a minimum 30-day lead time before it takes effect, allowing regulated entities to prepare for compliance. Rules that are economically significant—defined by Executive Order 12866 as rules with an annual economic effect of $100 million or more—are subject to additional requirements. Agencies must prepare a Regulatory Impact Analysis quantifying estimated costs and benefits, and the rule is subject to review by the Office of Information and Regulatory Affairs (OIRA) within the Office of Management and Budget before publication.

Executive Order 12866, issued by President Clinton in 1993, remains the governing order for regulatory review. It requires cost-benefit analysis for significant rules, mandates OIRA review, and requires agencies to use the least burdensome regulatory alternative that achieves the statutory objective. Executive Order 13563, issued by President Obama in 2011, supplemented 12866 by requiring retrospective review of existing regulations and emphasizing consideration of non-quantifiable benefits such as human dignity, fairness, and equity. Later administrations have issued their own supplemental orders modifying how cost-benefit analysis is conducted and which costs must be offset by regulatory eliminations, but EO 12866's core framework has persisted across administrations.

Information Collection Requests (ICRs) represent a distinct category of regulatory document in Regulations.gov. Under the Paperwork Reduction Act, any federal form or data collection that imposes a burden on 10 or more respondents must receive OIRA approval before the agency can use it. ICRs submitted for OIRA review are posted to Regulations.gov for a 60-day public comment period. ICRs are not rulemakings in the traditional sense, but they are significant regulatory actions that generate public comment and are archived in Regulations.gov alongside conventional rulemaking dockets.

Docket structure

Each rulemaking on Regulations.gov is organized around a docket. A docket is a collection of all documents associated with a single regulatory action: the initiating ANPRM or NPRM, supporting studies and analyses, public comments received, agency responses, and the final rule or withdrawal notice. The docket is the official administrative record for the rulemaking and is the starting point for any legal challenge to the resulting rule.

Every docket has a docket ID, a structured identifier that encodes the issuing agency and a sequential number. The format is agency abbreviation, headquarters or program office code, program office or rulemaking topic code, year, and sequence number. For example, EPA-HQ-OAR-2021-0257 identifies a docket from the EPA (EPA), headquartered in Washington (HQ), issued by the Office of Air and Radiation (OAR), initiated in 2021, with sequence number 0257. Docket IDs are stable identifiers used across the Federal Register, Regulations.gov, OIRA's reginfo.gov database, and the eCFR—the same docket can be tracked across all four systems using this identifier.

The Regulation Identifier Number (RIN) is a complementary identifier assigned by OIRA when a rule enters the Unified Regulatory Agenda. A RIN takes the form XXXX-AAYYY where XXXX is a four-digit agency code, AA is a two-letter prefix, and YYY is a three-digit sequence number (e.g., 2060-AV06 for an EPA rule). The RIN tracks a rule from its Agenda entry through proposed and final rule stages, allowing analysts to link Federal Register documents, OIRA review records, and Regulations.gov dockets to a single regulatory action across its entire lifecycle.

Within each docket, documents are classified into four types. Proposed Rules are the NPRMs and ANPRMs that initiate the public comment process—the regulatory text the agency is proposing to adopt or amend.Rules are the final rules that emerge from the notice-and-comment process and carry the force of law once effective. Supporting & Related Material encompasses the evidentiary and analytical foundation of the rulemaking: economic analyses, scientific studies, environmental impact statements, technical background documents, and interagency correspondence. This document type is where the substantive analytical record is concentrated.Public Submissions are the comments received from external parties during the comment period.

Public Submissions themselves vary considerably in source and character. Comment submitters include individual members of the public writing in their personal capacity; corporations and their legal or regulatory affairs staff; trade associations representing industry sectors (the American Chemistry Council for chemical manufacturers, the American Petroleum Institute for oil and gas, the American Medical Association for physicians); environmental and public health advocacy organizations; academic researchers; state and local governments; foreign governments and international organizations; and other federal agencies commenting on rules that affect their own programs or constituencies. The diversity of commenter types on any given docket is itself analytically significant—it maps the stakeholder landscape for a regulatory action.

Mass comment campaigns

The largest dockets on Regulations.gov have received not thousands but millions of public comments, driven by coordinated advocacy campaigns in which organizations encourage their members, subscribers, or the general public to submit comments using pre-written or lightly edited templates. Mass comment campaigns are a recognized feature of notice-and-comment rulemaking, but their scale has expanded dramatically with the internet, and their integrity has become a persistent problem for both agencies and the public record.

The Federal Communications Commission's 2017 net neutrality docket— FCC-2017-0200—is the defining example of mass comment pathology. The FCC received approximately 22 million comments on the proposed repeal of the Open Internet Order. Subsequent investigations by the New York Attorney General, the FCC Inspector General, and academic researchers found that millions of these comments were fraudulently generated using stolen identities. The fraud operated at scale: bot operators used personal information obtained from data breaches to submit comments under real Americans' names and addresses without those individuals' knowledge or consent. The New York AG's investigation found that at least 500,000 comments were submitted by a single fraudulent campaign using real names; broader estimates of inauthentic submissions in the docket ran from 7 to 9 million. Real Americans whose identities had been used contacted the AG to report they had never submitted any comment.

The 2014 FCC Open Internet docket—the earlier net neutrality proceeding—received approximately 3.7 million comments, the largest comment count for any federal rulemaking at the time. While this docket did not face the same systematic identity theft problem as 2017, analysis by researchers found substantial volumes of form letters and coordinated campaign submissions that were near-identical in content. The John Oliver effect— named after the HBO host whose 2014 and 2017 segments drove millions of viewers to Regulations.gov—demonstrated that media campaigns could generate comment volumes that overwhelmed agency processing capacity. After the 2014 segment, the FCC's comment submission system crashed from traffic volume.

EPA climate rules have generated millions of comments in several proceedings. The EPA's 2015 Clean Power Plan, which proposed to reduce carbon dioxide emissions from existing power plants, received approximately 4.3 million comments—the largest comment volume for any EPA rulemaking at that time. The comments were heavily influenced by both industry and environmental advocacy campaigns. The EPA's 2022 New Source Performance Standards for fossil fuel power plants and related greenhouse gas rules generated similarly large comment volumes. Analysis of these dockets shows a clear bimodal distribution: a large number of short form letters from campaign-driven individual commenters, and a smaller number of long, technically detailed submissions from industry groups, environmental organizations, and state agencies that constitute the substantive analytical record.

The legal weight agencies are required to give mass comment campaigns is limited but not zero. The APA requires agencies to respond to “significant” comments—those raising new data, arguments, or analytical perspectives not already addressed in the NPRM. A form letter expressing general opposition to a rule does not compel a detailed agency response; a coordinated campaign of identical submissions does not carry more weight than a single submission with the same content. However, the volume of comments expressing a particular view may be relevant to the agency's assessment of public concern about a rule, even if the individual submissions do not contain unique substantive content. Agencies that ignore large volumes of public opposition without explanation risk appearing arbitrary, even if they are legally required only to respond to substantive unique arguments.

OIRA and OMB have explored technological responses to mass comment authenticity problems. The proposed solutions—CAPTCHA requirements, identity verification, comment deduplication algorithms, and machine-learning-based authenticity scoring—all involve tradeoffs between fraud prevention and public accessibility. Requiring identity verification would deter fraudulent submissions but would also deter genuine comments from individuals who lack or distrust government ID systems. The Regulations.gov interface as of 2027 does not require identity verification for comment submission, though it does require a name and address.

API capabilities

Regulations.gov provides programmatic access through a REST API atapi.regulations.gov/v4. The v4 API, the current version, requires a free API key obtained by registering at api.data.gov—the federal government's shared API management platform. Standard accounts receive a rate limit of 1,000 requests per hour; requests above this threshold receive HTTP 429 responses and must be retried after the reset period. Researchers with demonstrated need for higher throughput can request an elevated rate limit from the eRulemaking Program.

The API exposes three primary endpoints. The /documents endpoint retrieves documents (proposed rules, final rules, supporting materials) across the database. Filtering parameters include filter[agencyId] for agency (using the agency's Regulations.gov abbreviation, e.g., EPA, DOT, FDA, CFPB), filter[documentType] for document type (Proposed Rule, Rule, Supporting & Related Material, Public Submission),filter[postedDate][ge] and filter[postedDate][le]for date range, and filter[searchTerm] for keyword search across titles and content. Results are paginated with a maximum of 250 records per request, and the response includes a meta.totalElements count for iterating through all matching pages.

The /dockets endpoint retrieves docket-level metadata: docket ID, agency, title, docket type (Rulemaking, Nonrulemaking), and associated document counts. Filtering by agency and keyword is supported. The/comments endpoint retrieves public comment submissions and is the most heavily used endpoint for research applications. Comments can be filtered by filter[commentOnId]—the docket ID or document ID the comment was submitted in response to—allowing retrieval of all comments on a specific rulemaking. Individual comment records include the commenter's name, organization (if provided), submission date, and comment text (where available). Not all comments include attached documents, and comment text is sometimes stored as a PDF attachment rather than inline text.

The API's sort parameter allows ordering by postedDate(ascending or descending with a leading minus sign) or by lastModifiedDate. The combination of agency filter, document type filter, date range, and sort order enables systematic ingestion of a specific agency's rulemaking activity over any time period. For comment analysis at scale, the/comments endpoint supports filtering by docket ID and sorting by most recent, allowing incremental synchronization of a local comment database with Regulations.gov as new submissions arrive during an active comment period.

The API is actively used by academic researchers, journalists, advocacy organizations, and industry regulatory affairs teams. Published research using Regulations.gov API data includes studies of industry comment frequency and sophistication across regulatory domains, analysis of geographic patterns in public comment participation, network analysis of trade association commenting behavior across related rulemakings, and natural-language processing studies of comment sentiment and substance. The API's structured metadata makes large-scale quantitative analysis tractable in ways that scraping the web interface would not be.

High-profile rulemakings

The breadth of Regulations.gov's coverage is best illustrated by examining the major rulemakings archived in the system across different policy domains.

EPA Clean Power Plan (EPA-HQ-OAR-2013-0602) was the Obama administration's flagship climate rulemaking under Section 111(d) of the Clean Air Act. Proposed in 2014 and finalized in 2015, it established the first nationwide standards for carbon dioxide emissions from existing power plants. The docket accumulated over 4 million comments and hundreds of technical supporting documents. The rule was stayed by the Supreme Court in 2016 before taking effect, and subsequently repealed and replaced multiple times across administrations—each iteration generating a new docket with additional comment cycles.

CFPB Payday Lending Rule (CFPB-2016-0025) established the first federal underwriting standards for payday loans, vehicle title loans, and certain other high-cost short-term credit products. The rulemaking process took over five years, including an extensive Small Business Review Panel process, and the final rule ran to over 1,600 pages of regulatory text and preamble. The docket contains detailed economic analysis of loan rollover rates, borrower financial distress, and the projected effects of the ability-to-repay requirement on credit availability.

OSHA COVID-19 Healthcare ETS (OSHA-2020-0004) and Vaccination and Testing ETS (OSHA-2021-0007) represent emergency temporary standards issued under OSHA's authority to address grave dangers to workers. The vaccination and testing mandate for large employers generated intense public and industry commentary before the Supreme Court stayed its application to most employers in January 2022 in NFIB v. OSHA. The dockets contain the scientific and epidemiological record the agency relied on, as well as comments from major employer associations, healthcare systems, unions, and public health organizations.

FCC Net Neutrality Rules (FCC-2014-0117, FCC-2017-0200, FCC-2023-0086) represent three separate rulemaking cycles on the same fundamental policy question, each generating millions of comments and each reversing or partially reversing the preceding rule. The 2023 docket, issued under a reconstituted FCC majority, proposed to restore the Open Internet Order framework repealed in 2017. The three dockets together constitute the most commented-on regulatory controversy in Regulations.gov history.

USDA School Nutrition Standards (FNS-2011-0010) establishing the Healthy Hunger-Free Kids Act nutrition requirements for school meals generated substantial comment from school food service operators, food manufacturers, nutrition advocates, and parents. The rulemaking required increasing whole grain, fruit, and vegetable content while limiting sodium, saturated fat, and calories—changes that affected billions of meals served annually and generated significant industry opposition from food companies whose products did not meet the new specifications.

FDA Tobacco Regulations (FDA-2014-N-0189) extending the FDA's regulatory authority to cover electronic cigarettes, cigars, pipe tobacco, and hookah tobacco under the Family Smoking Prevention and Tobacco Control Act generated extensive industry and public health comment. The docket includes detailed submissions from e-cigarette manufacturers arguing their products should be regulated differently from combusted tobacco, and from public health organizations arguing for harmonized standards.

SEC Climate Disclosure Rule (SEC-2022-0072), proposed in 2022 to require public companies to disclose climate-related risks and greenhouse gas emissions in SEC filings, received over 16,000 comments—an unusually high comment volume for a financial regulatory proceeding. The comments spanned institutional investors supporting disclosure, industry groups arguing the requirements were too burdensome, environmental organizations arguing they were insufficiently comprehensive, and academics debating the materiality standard. The rule was finalized in 2024 in a scaled-back form and immediately challenged in federal court.

Research applications

Regulations.gov has become a primary data source for empirical research on the regulatory process. Political scientists have used comment metadata to study how industry and public interest groups differ in the timing, sophistication, and strategic targeting of their comments across related rulemakings. Studies have found that industry comments are systematically submitted closer to the deadline (suggesting more deliberate preparation), are longer and more technically detailed, and are more likely to cite specific data and legal arguments rather than general policy preferences.

Research on comment influence—whether substantive public comments actually change rule content—has produced nuanced findings. Studies comparing proposed and final rule text across several hundred EPA and FCC rulemakings found that rules receiving more technically detailed industry comments were more likely to be modified between proposal and finalization, but that the direction of modification was not uniformly favorable to industry positions. Rules receiving substantive scientific comments from academic researchers were also associated with greater modification rates, particularly when the comments identified methodological problems with the agency's own supporting analysis.

Journalism applications have focused heavily on mass comment campaigns and organized lobbying. Investigations using Regulations.gov API data have documented coordinated comment submission patterns across different industry groups—cases where thousands of near-identical comments were submitted within short time windows, suggesting automated submission tools rather than genuine individual engagement. Comment network analysis has mapped which trade associations comment on the same dockets, revealing informal coalitions and regulatory opposition structures not visible from lobbying disclosure data alone.

Public health researchers have used FDA dockets to analyze how the medical and scientific community engages with proposed drug, device, and food safety rules. Academic medical organizations (the American Medical Association, specialty societies, academic medical centers) submit detailed clinical and epidemiological comments on FDA proposed rules that often provide the most technically rigorous input in a given proceeding. Tracking these submissions across dockets reveals which professional organizations are most active in regulatory advocacy and how their positions align with or diverge from industry comments on the same rules.

Economic research on the regulatory burden has used Regulations.gov document metadata—particularly the Regulatory Impact Analyses attached as supporting material—to build systematic datasets of projected regulatory costs and benefits across agencies and administrations. The variation in how agencies conduct cost-benefit analysis, what discount rates they apply to future benefits, and how they quantify non-market benefits (such as statistical lives saved or ecosystem services protected) is visible in the supporting documents archived in Regulations.gov and has been the subject of ongoing methodological debate among regulatory economists.

Python: querying the Regulations.gov API

The following script queries the Regulations.gov v4 API to retrieve recent EPA proposed rules and fetch public comments on a specific docket. A free API key from api.data.gov is required; the DEMO_KEYplaceholder used below works for testing at reduced rate limits. The script illustrates the two most common research operations: enumerating proposed rules for an agency, and retrieving comments on a specific docket.

import requests

# Regulations.gov API -- free key from api.data.gov
API_KEY = "DEMO_KEY"  # register at api.data.gov
base = "https://api.regulations.gov/v4"
headers = {"X-Api-Key": API_KEY}

# Search for recent proposed rules from EPA
params = {
    "filter[documentType]": "Proposed Rule",
    "filter[agencyId]": "EPA",
    "sort": "postedDate",
    "page[size]": 10,
}
resp = requests.get(f"{base}/documents", headers=headers, params=params, timeout=20)
data = resp.json()
docs = data.get("data", [])
print(f"Recent EPA proposed rules: {data.get('meta', {}).get('totalElements', 0)} total")
for d in docs[:5]:
    attrs = d.get("attributes", {})
    print(f"  {attrs.get('postedDate', '')[:10]}  {attrs.get('title', '')[:70]}")
    print(f"    Docket: {attrs.get('docketId', '')}  Comments: {attrs.get('commentCount', 0)}")

# Get comments on a specific docket
docket_id = "EPA-HQ-OAR-2022-0985"  # example: EPA greenhouse gas rule
params2 = {
    "filter[commentOnId]": docket_id,
    "page[size]": 5,
    "sort": "-postedDate",
}
resp2 = requests.get(f"{base}/comments", headers=headers, params=params2, timeout=20)
comments = resp2.json().get("data", [])
print(f"\nMost recent comments on {docket_id}:")
for c in comments:
    attrs = c.get("attributes", {})
    org = attrs.get("organization", "Individual")
    print(f"  {attrs.get('postedDate', '')[:10]}  {str(org)[:50]}")

The response structure for documents follows the JSON:API specification: each item in the data array has a type,id, and attributes object. Key attributes for proposed rules include title, agencyId,docketId, postedDate, commentStartDate,commentEndDate, and commentCount. ThecommentCount field reflects the count at the time the document metadata was last updated and may lag real-time submission volumes during active comment periods. For large dockets receiving millions of comments, the comment count is updated periodically rather than in real time.

The filter[commentOnId] parameter for the /commentsendpoint accepts either a docket ID (returning all comments submitted to the docket) or a specific document ID (returning comments submitted on a particular document within the docket). For most research purposes, filtering by docket ID is appropriate, as submitters may direct their comments to the docket generally rather than to the specific NPRM document. The maximum page size of 250 comments per request means that retrieving all comments from a large docket requires iterating through hundreds or thousands of pages—a process that must be throttled to stay within the 1,000-request-per-hour rate limit. At 250 comments per request and a 1-second delay between requests, retrieving 100,000 comments takes approximately 7 minutes; retrieving 22 million comments from the FCC net neutrality docket would take on the order of 25 hours at standard rate limits.

For analysis requiring the full text of submitted comments rather than just metadata, individual comment records include either inline text or references to attached PDF documents. The attached documents must be downloaded separately using the document URL in the response. Comment text quality varies considerably: mass campaign submissions are typically short plain text, while industry and advocacy organization submissions are often multi-page PDFs with technical appendices. A complete analysis of a major docket's substantive content requires both metadata analysis (to characterize the comment distribution by submitter type) and document retrieval (to extract and analyze the text of substantive submissions).