FTC Enforcement: The Federal Record of Consumer-Protection and Antitrust Actions

When the Federal Trade Commission decides a company has lied to consumers, rigged a market, or sold a child's data, the public almost always learns of it the same way: a press release. The complaint, the settlement, the proposed rule, the merger it is moving to block— each is announced from ftc.gov, and the accumulated announcements form something the agency never quite designed but anyone can read: a dated, searchable log of nearly every consumer-protection and antitrust action the United States' principal market-conduct regulator has taken. Roughly 10,700 of those records, one row per release, are the subject of this piece—the press archive read not as news but as data.

This article covers what the FTC press-and-enforcement record is and why a press archive functions as an enforcement dataset; the statutory frame—Section 5 of the FTC Act and the prohibition on unfair or deceptive acts or practices—and the agency's dual mission across consumer protection and competition; the structure of the data and the columns that make it searchable; the consumer-protection programs the releases document—deceptive advertising, fraud and the Do Not Call registry, the COPPA children's-privacy rule, and the data-security caseline that produced landmark penalties; the antitrust side, from merger challenges to monopolization suits brought alongside the Department of Justice; how the record reveals shifting enforcement priorities—privacy, junk fees, noncompetes, and big-tech competition; the consumer-redress and civil-penalty amounts the agency obtains; a Python workflow that pulls releases from ftc.gov and tallies them by year, topic, and company; and the caveats of treating a press feed as a complete and neutral record of enforcement.

What the dataset is

The Federal Trade Commission does most of its public communicating through press releases. When it files a complaint, announces a settlement (which the FTC calls a consent order or, in court, a stipulated judgment), proposes or finalizes a rule, returns money to consumers, or challenges a merger, it publishes a release describing the action. Read one at a time, these are news items. Read together—ordered by date, tagged by matter, and made searchable—they become the closest thing there is to a public ledger of FTC enforcement activity over time. That is the dataset: the agency's press and enforcement record, drawn from the news feeds at ftc.gov, comprising roughly 10,700 records.

In our database this record is stored as the table ftc_press, with the grain of one row per release. A single enforcement matter often generates several rows—the initial complaint, an amended filing, the proposed settlement, the finalized order, and a later announcement of refunds mailed—so the table is a log of announcements, not a deduplicated docket of distinct cases. The columns are deliberately spare, which is what makes the archive easy to query: each row carries when the release was issued, its headline, the matter it concerns, and a link back to the full text:

date     -- the date the press release was issued (the primary key axis)
title    -- the headline of the release (the searchable text payload)
matter   -- the case, rule, or program the release concerns
link     -- the URL of the full release on ftc.gov

The date is the organizing axis: because every release is timestamped, the table supports the most natural FTC question—how does enforcement volume, and its composition, change over time? The title is the substantive payload. FTC headlines are unusually informative; the agency writes them to name the company, the conduct, and frequently the dollar figure (“FTC Returns Millions to Consumers Harmed by …”), which means a great deal of analysis can be done on the title text alone—classifying a release as a privacy case, a fraud case, or a merger challenge by the words it uses. The matter ties releases that belong to the same case or program together, and the link is the escape hatch to the authoritative full text—the complaint, the order, the rule—for any release that warrants reading in full rather than counting. The spareness is a feature: it makes the archive trivial to search by topic or company, which is precisely how most people use it.

Section 5 and the FTC's dual mission

The Federal Trade Commission was created by the Federal Trade Commission Act of 1914, in the same Progressive-era wave that produced the Clayton Act, to police business conduct in the national marketplace. Its founding charge is compact and famous: Section 5 of the FTC Act declares unlawful “unfair methods of competition”and “unfair or deceptive acts or practices”in or affecting commerce. Those two phrases divide the agency's work into its two great halves—competition (antitrust) and consumer protection—and almost everything in the press record traces back to one of them.

The consumer-protection prong rests on the second phrase. A practice is deceptive when it involves a representation, omission, or practice likely to mislead a reasonable consumer to their detriment; a practice is unfair when it causes or is likely to cause substantial consumer injury that consumers cannot reasonably avoid and that is not outweighed by benefits to consumers or competition. This unfair-or-deceptive standard, abbreviated UDAP, is the workhorse authority behind the great majority of the agency's consumer cases—false advertising, deceptive marketing, hidden fees, and, increasingly, data-privacy and data-security practices treated as unfair or deceptive. The agency administers it through two bureaus: the Bureau of Consumer Protection brings the deception and unfairness cases, and the Bureau of Competition brings the antitrust cases, with the Bureau of Economics supporting both.

A defining feature of the FTC is that it is both a law-enforcement agency and a regulator. It can sue—in its own administrative tribunal or in federal court—and it can write rules. Its remedial toolkit is what gives the press record its dollar figures: the agency can obtain injunctive relief stopping the conduct, consumer redress(returning money to harmed consumers), and, for violations of rules and certain orders, civil penalties paid to the Treasury. The scope of one of these tools—the agency's power to obtain monetary relief directly under Section 13(b) of the Act—was sharply narrowed by the Supreme Court in 2021, which is why much of the agency's recent strategy emphasizes rule violations and order violations that carry their own penalty authority. The press releases are where each of these remedies is announced, which makes the archive a running account not just of what the FTC challenged but of what it recovered.

The consumer-protection programs

The consumer-protection side of the record is the larger and more varied half, organized around a set of long-running programs that recur in the headlines year after year.

Deceptive advertising and marketing is the oldest line of work and the most diverse. It spans false or unsubstantiated health and weight-loss claims, deceptive “made in USA” labeling, misleading earnings claims by business-opportunity and multi-level-marketing schemes, and undisclosed paid endorsements and fake reviews—an area the agency has pressed hard as influencer marketing and online reviews came to dominate purchasing decisions. The common thread is the deception prong of Section 5: a claim likely to mislead a reasonable consumer. Fraud and the Do Not Call program form a second pillar. The FTC operates the National Do Not Call Registry and enforces the Telemarketing Sales Rule, and a steady stream of releases announces actions against illegal robocallers, telemarketing fraud, imposter scams, tech-support scams, and deceptive subscription and “negative option” billing—the auto-renewing charges consumers cannot easily cancel.

Children's privacy is a distinct and consequential program built on a statute the FTC enforces directly: the Children's Online Privacy Protection Act (COPPA), which restricts the collection of personal information from children under thirteen and requires verifiable parental consent. COPPA cases recur prominently in the record because they carry civil-penalty authority and because they reach the largest online platforms when those platforms collect data from minors. The broadest and most modern program is data security and privacy. Treating inadequate data security and deceptive privacy promises as unfair or deceptive practices under Section 5, the agency has built a body of cases against companies that failed to protect consumer data or misrepresented how they used it—the area that produced the agency's largest penalties and, increasingly, scrutiny of surveillance advertising, data brokers, and the handling of sensitive location and health information. These programs are why a topic classifier run over the title text—privacy, children, fraud, advertising—captures so much of the consumer-protection record.

Landmark privacy and data-security penalties

No part of the record illustrates the agency's reach better than its data-privacy penalties, and the landmark is the 2019 Facebook settlement. After alleging that the company had violated an earlier FTC privacy order—the conduct that surfaced in the Cambridge Analytica episode—the FTC imposed a $5 billion penalty, by a wide margin the largest the agency had ever obtained and one of the largest privacy penalties anywhere in the world. The settlement did more than levy a fine: it restructured the company's internal privacy governance, imposing board-level oversight and individual accountability requirements that became a template for later orders. The case is the clearest single demonstration of how a Section 5 privacy matter, announced through a press release, can carry a remedy of a scale that reshapes corporate behavior.

The Facebook penalty sits atop a deeper caseline. The FTC's data-security cases reach back decades—to retailers and data companies whose lax security exposed millions of consumers' records—and the agency has used consent orders to require comprehensive security programs and independent assessments lasting many years. On the privacy side, the record includes substantial penalties and order modifications against the largest technology and social-media companies for collecting or sharing data in ways the agency alleged were unfair or deceptive, including a notable COPPA matter resolving allegations that a major video platform collected children's data without parental consent. The throughline is that, lacking a comprehensive federal privacy statute, the United States has regulated commercial privacy substantially through FTC enforcement under the general unfairness and deception authority—which makes this press archive, in effect, the de facto casebook of American privacy law, assembled one release at a time.

The antitrust side: mergers and monopolization

The competition half of the record rests on the first phrase of Section 5—unfair methods of competition—and on the FTC's shared enforcement of the core antitrust statutes. Uniquely, the United States splits federal antitrust enforcement between two agencies: the FTC and the Antitrust Division of the Department of Justice. The two divide work by industry and by a clearance process, and they jointly issue the merger guidelines that frame how both review deals. The press record therefore captures the FTC's share of national competition enforcement, with the understanding that the DOJ handles the rest.

Merger review is the highest-volume competition work. Under the Hart-Scott-Rodino Antitrust Improvements Act, parties to large transactions must notify the agencies and wait before closing, giving the FTC the chance to investigate and, if it concludes a deal would substantially lessen competition under the Clayton Act, to challenge it—by seeking to block it in court, by negotiating a settlement that requires divesting assets, or by litigating to unwind a consummated deal. The releases announce each stage: a second-request investigation, a complaint to block, a consent agreement with divestitures, or a court victory or loss. Monopolization and conduct cases are the rarer but more consequential strand—suits alleging that a dominant firm illegally maintained or abused its market power. These include high-profile actions against the largest technology platforms, which have made big-tech competition one of the defining themes of the recent record, as well as conduct cases in pharmaceuticals, hospitals, and other concentrated industries. Because antitrust litigation unfolds slowly, a single matter can generate a long sequence of releases across years.

Shifting enforcement priorities

The single most powerful use of a dated press archive is watching priorities move. Because the FTC is a commission whose direction is set by its chair and the sitting commissioners, its emphasis shifts with administrations and with the issues of the moment, and that shift—invisible in any single release—is exactly what the time-ordered record makes legible.

Several recent priorities stand out in the headlines. Privacy and data security has grown from a niche program into a central one, with rising counts of releases on data brokers, surveillance advertising, sensitive-data handling, and proposed privacy rulemaking. “Junk fees”—hidden and mandatory charges added late in a transaction—became an explicit campaign, pairing enforcement with a rulemaking aimed at requiring all-in pricing, and the same era pressed hard on deceptive subscription and negative-option practices that make cancellation difficult. Noncompete rulemaking marked an ambitious move beyond case-by-case enforcement: the agency proposed and finalized a rule that would broadly ban employee noncompete agreements as an unfair method of competition—a sweeping use of its competition-rulemaking authority that drew immediate legal challenge and illustrates the limits, as well as the reach, of that authority. And big-tech antitrust—merger challenges and monopolization suits against the largest platforms—has been a persistent theme. Tallying releases by these topics over time turns the archive into a quantitative record of where the agency chose to spend its attention, year by year.

Penalties, redress, and the money the agency obtains

A recurring feature of FTC releases is a dollar figure, and learning to read those figures correctly is essential to using the archive quantitatively. The agency obtains money in two principal forms, and they are not the same thing.

Consumer redress is money returned to the consumers who were harmed—refunds mailed to people who bought a misrepresented product or were billed for a subscription they could not cancel. Redress flows back to the public, and the agency frequently issues separate releases when it actually distributes the funds, announcing how many people received checks and the total mailed. Civil penalties, by contrast, are paid to the US Treasury and are available chiefly when a party violates a rule (such as COPPA or the Telemarketing Sales Rule) or violates an existing FTC order—which is why the agency, in the wake of the Supreme Court's narrowing of its direct monetary authority under Section 13(b), has leaned on rule and order violations to preserve a penalty hammer. For an analyst, two cautions follow. First, the headline number in a release may be a penalty, a redress amount, a disgorgement figure, or merely the alleged consumer harm, and these must not be summed indiscriminately. Second, an announced figure is what was ordered or agreed, which can differ from what is ultimately collected—a settled judgment may be partially suspended based on an inability to pay. The press record is excellent for tracking the announced magnitude and cadence of monetary relief; it is a starting point, not the final word, on what the Treasury and consumers actually received.

Analytical uses

A dated, titled, searchable archive of nearly every FTC action supports a distinctive set of analyses that no single release reveals.

Enforcement volume and composition over time is the most immediate. Counting releases by year shows the overall tempo of activity, and classifying titles into topics—privacy, fraud, advertising, merger, junk fees— shows how the mix shifts, making visible the priority changes described above. This is the analysis the time axis was made for. Company and matter tracking is the second: because titles name companies, an analyst can pull every release mentioning a given firm or matter and reconstruct the arc of an enforcement action—complaint, settlement, order, refunds—from the headlines alone, or build a watchlist that surfaces new actions against entities of interest.

Cross-referencing with other enforcement records is where the archive becomes most powerful. A company named in an FTC consumer-protection or antitrust release may also appear in the CFPB's complaint and enforcement data, in SEC actions, in state attorneys-general settlements, or on federal exclusion lists—and joining the FTC record to those datasets on company name builds a fuller picture of an entity's regulatory history than any one agency's record provides. Finally, redress and penalty tracking uses the dollar figures in the titles, parsed carefully and bucketed by remedy type, to estimate the announced magnitude of monetary relief over time and by program—always with the caveat, developed above, that announced is not the same as collected. Together these uses turn a press feed into a quantitative instrument for studying how the agency works.

Python workflow: pulling and tallying the press record

The script below pulls press releases from ftc.gov, tallies them by year and by topic, and searches the archive for a company or matter. It works against the agency's paginated press feed (the same content the site exposes as RSS), requires no API key, and uses a small keyword map to bucket releases into topics from their title text. Because the exact JSON path and field names on ftc.gov shift between site releases, the script resolves the title and date columns defensively rather than hard-coding them, and any production use should be validated against the current ftc.gov news endpoints and paged through the full archive.

import requests, re
import pandas as pd
from collections import Counter
from datetime import datetime

# The FTC publishes its enforcement and press record at ftc.gov.
# Press releases are exposed as a paginated JSON feed under the
# /news-events/news/press-releases path; the same content is also
# available as RSS. No API key is required for public data.
#
# This script pulls press releases, tallies them by year and by
# topic, and searches the archive for a company or matter. The exact
# JSON path and field names shift between ftc.gov releases, so the
# field accessors are kept defensive rather than hard-coded.
BASE = "https://www.ftc.gov"
FEED = BASE + "/news-events/news/press-releases"


def fetch_releases(pages=40, per_page=50):
    # ftc.gov serves a JSON view of the press list when asked.
    # Each item carries a title, a date, a matter, and a URL.
    rows = []
    for p in range(pages):
        params = {"page": p, "items_per_page": per_page, "_format": "json"}
        r = requests.get(FEED, params=params, timeout=60)
        if r.status_code != 200:
            break
        try:
            batch = r.json()
        except ValueError:
            break
        items = batch if isinstance(batch, list) else batch.get("rows", [])
        if not items:
            break
        rows.extend(items)
    return pd.DataFrame(rows)


def _pick(frame, *names):
    # Return the first column whose name matches (case-insensitive).
    low = {c.lower(): c for c in frame.columns}
    for n in names:
        if n.lower() in low:
            return low[n.lower()]
    return None


# Topic buckets keyed to the language the FTC uses in its titles.
TOPICS = {
    "privacy / data security": r"privacy|data security|breach|surveillance",
    "children (COPPA)":        r"coppa|children|kids|teen",
    "deceptive advertising":   r"deceptive|misleading|false claim|endorsement",
    "fraud / scams":           r"scam|fraud|robocall|telemarket|do not call",
    "merger / antitrust":      r"merger|acquisition|monopol|antitrust|compet",
    "junk fees / billing":     r"junk fee|hidden fee|billing|subscription|negative option",
    "noncompete":              r"noncompete|non-compete",
    "penalty / redress":       r"penalty|refund|redress|settle|disgorge",
}


def analyze(df):
    if df.empty:
        print("No releases returned.")
        return
    title_col = _pick(df, "title", "name", "headline")
    date_col = _pick(df, "date", "created", "published", "field_date")

    df["_title"] = df[title_col].fillna("").astype(str)
    df["_date"] = pd.to_datetime(df[date_col], errors="coerce")
    df["_year"] = df["_date"].dt.year

    # --- 1. Releases per year ------------------------------------------
    print("Press releases by year:")
    for yr, n in df["_year"].dropna().astype(int).value_counts().sort_index().items():
        print(f"  {yr}  {n:>5,}")

    # --- 2. Releases by topic (a release can match several) ------------
    counts = Counter()
    for t in df["_title"].str.lower():
        for topic, pat in TOPICS.items():
            if re.search(pat, t):
                counts[topic] += 1
    total = len(df)
    print("\nShare of releases mentioning each topic:")
    for topic, n in counts.most_common():
        print(f"  {topic:<24} {n:>5,}  ({n / total:.1%})")
    return df


def search(df, term):
    # Find every release whose title mentions a company or matter.
    hits = df[df["_title"].str.contains(re.escape(term), case=False, na=False)]
    print(f"\n{len(hits)} release(s) mention {term!r}:")
    for _, row in hits.sort_values("_date", ascending=False).head(10).iterrows():
        d = row["_date"].date() if pd.notna(row["_date"]) else "????-??-??"
        print(f"  {d}  {row['_title'][:72]}")
    return hits


releases = fetch_releases()
analyze(releases)
search(releases, "data privacy")

Two practical notes apply. First, the topic classification is keyword-based and therefore approximate: it buckets a release by the language in its title, so a case that the headline describes obliquely will be miscounted, and a release can match more than one topic (a privacy case that also announces a penalty), which is why the topic shares sum to more than one hundred percent. For rigorous classification, the title keywords are a first pass that should be refined against the full release text reached through the link column. Second, the script reads only the titles; the authoritative detail of any matter—the exact remedy, the legal authority, the defendants, whether a penalty was suspended—lives in the linked complaint and order, not in the headline. The archive is built for counting and for navigating to the source; the dollar amounts and legal specifics should always be confirmed against the underlying document before being reported as fact.

Limitations and analytical caveats

The press record is the most accessible public account of FTC enforcement, but it is a press archive, and treating it as a complete, neutral case database invites several errors an analyst must guard against.

A release is not a case, and the grain is announcements.The table records press releases, not distinct enforcement matters, and a single matter commonly generates several releases across its life—complaint, proposed settlement, final order, refunds distributed. Counting rows therefore counts announcements, not cases, and any measure of “how many actions” the agency brought must first deduplicate to the matter level using the matter field and careful title matching. Conversely, not every action gets a release: routine or smaller matters, and a great deal of the agency's pre-complaint investigative work, may never be announced at all, so the archive undercounts total activity even as it overcounts headline-worthy matters.

The agency chooses what to announce and how to frame it.Press releases are advocacy as much as record-keeping: the FTC selects which actions to publicize and writes the headlines to emphasize the consumer benefit and the dollar figure. That framing is useful—it is what makes the titles classifiable—but it also means the archive reflects the agency's communications priorities, not a neutral docket. A surge of releases on a topic can signal a genuine enforcement push or merely a decision to publicize that work more aggressively, and the two are hard to distinguish from the press feed alone.

Dollar figures are heterogeneous and announced, not collected. As the penalties section stressed, a headline number may be a civil penalty, consumer redress, disgorgement, or alleged harm, and these are not interchangeable; summing them produces a meaningless total. Worse, an announced amount is what was ordered or agreed, which can diverge sharply from what is ultimately paid when a judgment is suspended for inability to pay. Any quantitative claim about “money the FTC obtained” built from titles alone should be treated as an upper-bound estimate of announced relief, clearly labeled as such, and confirmed against the orders for anything load-bearing.

Company names are free text, and the FTC is not the whole story. Joining the press record to other datasets relies on matching company names in the title, which carries all the entity-resolution hazards of free-text matching— abbreviations, subsidiaries, doing-business-as names, and common-name collisions—so a name hit is a candidate to verify, not a confirmed link. And the FTC is only one of the agencies that polices market conduct: the Department of Justice handles much of federal antitrust, the CFPB handles much of consumer-financial protection, state attorneys general bring parallel actions, and sector regulators cover their own industries. The FTC press record is an authoritative account of what the FTC announced, not a complete account of consumer-protection or antitrust enforcement in the United States.

Held with these caveats in mind, the ftc_press table is a uniquely useful resource: roughly 10,700 dated, titled, linkable records that together trace how the United States' principal consumer-protection and competition agency has used its Section 5 authority over time—the privacy penalties, the fraud and advertising cases, the merger challenges, and the shifting priorities—read not as a stream of news but as the public log of enforcement it quietly is.