Newspaper, Periodical, Book, and Directory Publishers (U.S.) — NAICS 5131

A Histometrics rollup primer for public-market and private investors. This is a NAICS industry group (4-digit) that gathers five child industries. It synthesizes the five already-written child primers plus our ground-truth federal statistics for this level. Figures with citations are reported facts; statements about the future are labeled as judgments. Federal statistics are the ground truth; company, retail, and third-party figures are flagged as estimates.

NAICS = North American Industry Classification System, the U.S. federal standard for sorting businesses by activity. A 4-digit code is an "industry group"; the 5-digit codes beneath it are "NAICS industries," and the 6-digit codes beneath those are "national industries."

---

1. Overview

NAICS 5131 is the home of traditional publishing — the businesses that create original written and visual content and sell it, in print or electronic form, on a repeating basis. It bundles five very different trades under one statistical roof: newspapers, magazines and journals, books, directories and mailing lists, and a catch-all of greeting cards, yearbooks, calendars, posters, catalogs, and maps. What unites them is a business model, not a product: each is a rights-and-audience business — the publisher bears the financial risk of the content, owns or licenses the underlying intellectual property (IP), and captures the value, while the physical printing (NAICS 323, Printing) and the selling (bookstores, newsstands) usually happen elsewhere [9].

Federal data put the whole group at roughly $99.4 billion of receipts in 2022 across about 13,300 firms [1]. That makes it a large but mature slice of the media economy — and a deeply uneven one. Two of its five children are in outright structural decline (newspapers, directories); one is flat with pockets of growth (books); one is sharply split between a shrinking half and a thriving half (periodicals); and one is a low-growth, cash-generative niche collection (other) [2][3][4][5][6]. The single most important fact for an investor is that the group average tells you almost nothing — the children differ so much on size, growth, concentration, ownership, and how you would actually buy exposure that the only useful analysis is child-by-child. That contrast is the subject of this page.

Three themes run through all five children and are worth holding in mind from the start:

• Repeating vs. recreated revenue. Everywhere in this group, the prize assets earn money that repeats on its own — a newspaper subscription, a library journal license, a book's backlist, a data subscription, a yearbook contract — while the eroding parts must be recreated with every edition — a print ad page, a newsstand issue, a one-off catalog, a paper map. Repeating beats recreated at every level.
• Print-to-digital transition. Each child is somewhere on the same journey from a physical, postage-and-paper cost base to a near-zero-marginal-cost digital one — but at very different stages and with very different success.
• Artificial intelligence (AI) as a two-edged force. Generative AI is simultaneously a new customer (licensing archives and content to train models) and a new competitor (answer engines and "zero-click" search that satisfy readers without a visit to the source). This single force now touches the economics, regulation, and risk profile of every child in the group [4][13].

---

2. What's inside — the five child industries and how they differ

The group splits into five NAICS industries. They are genuine siblings, not overlaps — each NAICS definition explicitly excludes the others [9]. The contrast below is the distinctive value of this page: read it before any group-level number.

(CR4 is the combined revenue share of the four largest firms; CR8/CR20/CR50 extend that to the top 8, 20, and 50. HHI = Herfindahl-Hirschman Index, the sum of every firm's squared market share, which U.S. antitrust agencies treat as "unconcentrated" below 1,500 [10]. Every HHI in this group sits far below that line.)

How to read the contrast. Four things stand out:

1. Size ranking is nearly even. Books, periodicals, and newspapers each take roughly a quarter of group receipts; "other" takes about a sixth; directories are a rounding-error 5%. There is no single dominant child — the group is a genuine portfolio of comparable-sized businesses.
2. Firm count is lopsided the other way. Newspapers and periodicals are each ~29% of the firms but a smaller share by some measures of revenue-per-firm, while directories are tiny in both. Books have relatively few firms (2,191) for the largest revenue pool, because book revenue concentrates in a handful of large houses.
3. Concentration runs opposite to size. The smallest child (directories) is the most concentrated (HHI 841); the second-largest (periodicals) is the least (HHI 201). Concentration here is about how a niche's economics work — data assets and yellow-pages legacies consolidate; magazine titles fragment — not about scale.
4. Ownership is the real dividing line for investors. Only newspapers offer clean listed pure-plays. Everywhere else the durable value is private or foreign — a card duopoly, a book "Big Five" owned by Bertelsmann/KKR/Lagardère, PE-owned data compilers and yearbook houses — and listed exposure comes wrapped inside diversified owners. This is the single most important practical takeaway of the whole group.

---

3. Size (this level's rollup figures)

Our ground-truth federal file for NAICS 5131 reports the 2022 Economic Census figures below [1].

Our ground-truth file for this level carries receipts, firm count, and concentration only. It has no level-wide employment, payroll, establishment, or margin figure, so none is stated here as a group total. The employment and payroll figures below are assembled from the child primers' County Business Patterns (CBP) 2023 data and attributed as such [8].

How the group breaks down. The five children sum almost exactly to the group total. Receipts: $23.2B (newspapers) + $24.4B (periodicals) + $29.2B (books) + $5.1B (directories) + $17.5B (other) ≈ $99.4 billion [2][3][4][5][6]. Firm counts sum to about 13,426 against the group's reported 13,326 — the ~100-firm gap is the ordinary result of some firms operating in more than one child and being counted once at the group level but in each child below. Adding the child primers' CBP employment gives roughly 250,000 paid workers (≈91,000 newspapers + ≈65,000 periodicals + ≈65,000 books + ≈10,000 directories + ≈19,000 other) and on the order of $20 billion of annual payroll across the group [8] — implying a healthy but not extreme ~$100,000 average revenue per firm and a workforce that has shrunk for two decades as print declined [5].

Why the group looks so "unconcentrated" — an HHI of 110 that is lower than every child's. The group HHI of 110.2 [1] sits below all five children's HHIs (201 to 841), and the group CR4 of 14.2% is lower than any child's CR4 (23% to 46%). That is not a contradiction; it is arithmetic. When you pool five distinct product markets — a newspaper does not compete with a scholarly journal or a greeting card — each firm's share of the combined $99.4 billion pool shrinks far below its share of its own market. A firm with 40% of the $23 billion newspaper market holds under 10% of the group. The lesson is the same one the "other publishers" child teaches internally, but sharper here: the group concentration statistic is meaningless as a competition gauge. Real market power lives at the child level and, below that, in local markets (a one-newspaper town) and product pockets (the scholarly-journal oligopoly, the greeting-card duopoly, the yearbook near-duopoly). Never read competition off the 4-digit number.

Undercount caveat (material across the group). The Economic Census and CBP count only employer firms with payroll; CBP excludes the self-employed, nonemployer businesses, and most government publishing [8]. That matters more here than in most industries because several children have enormous nonemployer tails: more than 2.6 million self-published book titles carried an ISBN (International Standard Book Number) in 2023 alone, most from sole-proprietor authors who never appear in the employer count [4][12]; magazine publishing hides a long tail of freelancers and one-person newsletters, plus periodicals produced as a secondary activity by associations and corporations (counted under their main industry) [3]; directory publishing has many solo list brokers classified under advertising or data-processing codes [5]; and the "other" child is numerically dominated by solo card artists, Etsy sellers, and hobby publishers [6]. Conversely, the group's biggest names book much of their money outside it (News Corp's cable and real-estate assets, Hallmark's Crayola and cable channel, the credit-bureau arms of the data compilers). Treat $99.4 billion as the size of the employer, primary-activity core of traditional publishing — accurate as far as it goes, but an undercount of publishing as an economic activity.

---

4. Investable universe — where value concentrates across the children

The defining feature of this group is that clean public exposure is scarce and clusters in just one child. Value and access diverge sharply by child.

Newspapers (51311) — the only child with real listed pure-plays. This is where a public investor can actually buy the industry: The New York Times Company (NYSE: NYT), the scaled paid-digital leader; News Corp (Nasdaq: NWSA/NWS), owner of Dow Jones/The Wall Street Journal, the New York Post, and HarperCollins; USA TODAY Co. (NYSE: TDAY, formerly Gannett), the largest local footprint; Lee Enterprises (Nasdaq: LEE); and niche Daily Journal Corp. (Nasdaq: DJCO). But most local-newspaper equity is already private — distressed roll-ups (Alden Global Capital, Chatham/McClatchy), family groups (Hearst, Advance), and billionaire-owned metros [2].

Periodicals (51312) — buy the segment inside a diversified owner. No U.S.-listed company is a magazine pure-play and there is no magazine ETF (exchange-traded fund). Listed exposure runs through consumer digital roll-ups — People Inc. (Nasdaq: PPLI, formerly Dotdash Meredith) and Ziff Davis (Nasdaq: ZD) — and, more durably, the scholarly houses: RELX (LSE/NYSE: RELX), John Wiley & Sons (NYSE: WLY), Informa, Springer Nature, and Future plc. Most consumer titles sit inside private Hearst and Condé Nast [3].

Books (51313) — indirect, via education names or a parent. There is no large U.S.-listed pure-play trade-book stock; the trade "Big Five" are private or foreign-owned (Penguin Random House/Bertelsmann; Simon & Schuster/KKR; Hachette/Lagardère). Listed exposure concentrates in education and professional publishers — Scholastic (Nasdaq: SCHL), McGraw Hill (NYSE: MH), Wiley (NYSE: WLY) — plus News Corp (HarperCollins) and international names (Bloomsbury, Pearson, RELX, Lagardère). The distribution gatekeeper Amazon (Nasdaq: AMZN) captures much of the economics without publishing risk [4].

Directories & mailing lists (51314) — a segment inside a data company. The nearest listed exposures are Thryv Holdings (Nasdaq: THRY), the Yellow Pages descendant pivoting to small-business software, and ZoomInfo/GTM (Nasdaq: GTM), the largest listed business-to-business (B2B) data platform; diversified partial exposure comes via Experian, TransUnion, and Xometry (owner of Thomasnet). The largest capacity is private — Data Axle, Dun & Bradstreet (Clearlake), Acxiom (Omnicom) [5].

Other publishers (51319) — essentially all private. No U.S. pure-play exists in greeting cards or the catch-all. Foreign/adjacent proxies (Moonpig, IG Design Group, Card Factory on the LSE; ACCO Brands and Cimpress for calendars/print) offer only slivers; the real assets — the Hallmark/American Greetings card duopoly and PE-owned yearbook, calendar, and poster franchises (Jostens/Platinum, Herff Jones/KKR, Andrews McMeel, Trends International) — are private [6].

Bottom line for a public investor: the tradable, understandable pure-plays are essentially the newspaper names plus the scholarly-information houses; everything else is a segment inside a diversified owner or a private franchise. For a private investor the picture inverts — the most defensible cash flows in the entire group (scholarly journals, book backlist and rights, data compilers, yearbook contracts, the card duopoly's replacement demand) sit off the public market entirely.

---

5. How the money works

Every child in this group runs on the same three revenue pools, mixed in different proportions: audience-access revenue (subscriptions, single-copy/newsstand sales, library site licenses, data subscriptions), advertising (print and digital, cyclical and structurally shrinking), and rights, licensing, commerce, and events (syndication, brand and IP licensing, affiliate commerce, and — new across the group — AI-training licenses) [2][3][4][5][6].

The structural inversion is common to the group: as print advertising has collapsed, the winners have shifted their weight from advertising (recreated every issue) to recurring audience revenue (repeating on its own). Newspapers crossed the line where reader revenue overtook advertising around 2020 [2]; magazines and books lean on subscriptions and backlist; directories sell data subscriptions with an application-programming-interface (API); the "other" child prizes yearbook contracts and card replacement demand [6]. Digital delivery has near-zero marginal cost, so once a title, catalog, database, or backlist exists, incremental sales are enormously profitable — which is why scale, retention, and rights ownership decide returns everywhere.

But the best economics differ by child, and this is where the group's internal variance is widest:

• Scholarly journals (in periodicals) are the single most profitable model in the group — unpaid authors and peer reviewers, must-have library subscriptions with little substitution, and the same content reselling worldwide at operating margins near 40% [3].
• Yearbooks (in "other") are the best contract economics — multi-year school contracts, pre-paid by students, negative working capital, and EBITDA (earnings before interest, taxes, depreciation, and amortization — a cash-profit proxy) reportedly above 25% [6].
• Book backlist is a high-margin annuity — older titles supplied 61% of HarperCollins consumer revenue in fiscal 2024 [4].
• Data compilation (directories) looks like software: build the database once, monetize it many times, at very high margin — but contact data decays ~2–2.5% per month, so continuous re-verification is the dominant cost [5].
• Greeting cards carry very high product gross margins (a card costs ~$0.20–$0.50 to make, wholesales near $1.50) spent back on creative labor, licensing, and funding in-store racks [6].

The weakest economics cluster in the recreated, ad-and-postage-heavy pockets: consumer magazines, general newspapers' print advertising, mail-order catalogs (where postage can exceed 65% of cost), and paper maps and low-end calendars [2][3][6]. Across the whole group, three shared levers decide whether high reported margins convert to cash: scale (to absorb fixed creative and plate costs), the print-versus-digital mix, and control of paper and postage inflation.

---

6. Demand drivers

The children respond to different customers, but the drivers rhyme:

• Willingness to pay for content. The pivotal demand question group-wide is whether audiences will pay a publisher directly — for journalism, a subscription magazine, a book, a data feed — rather than take a free substitute. Distinctive, must-have content (financial news, scholarly research, breakout fiction, verified B2B data) commands payment; commodity content does not [2][3][4][5].
• The advertising cycle still moves the ad-dependent pockets (local newspapers, consumer magazines, directories) with GDP and consumer spending — but it is a structurally shrinking base as budgets migrate to Google, Meta, and other platforms [2][3].
• Institutional and R&D budgets drive the least-cyclical demand in the group — library and university spending on scholarly journals, school and district budgets for textbooks and yearbooks, corporate budgets for B2B data and trade titles [3][4][5][6].
• Platform and referral dynamics. Search and social send much of digital publishers' traffic, so algorithm changes are a swing factor across newspapers, magazines, and books alike [3].
• Demographics and generational behavior. An aging print audience is a headwind, but younger buyers have shown willingness to pay for premium, personalized products — subscription news bundles, #BookTok-driven fiction, premium planners and cards [4][6].
• AI answer engines — the group-wide swing factor. Every child names AI as both an opportunity (licensing archives, building AI research tools) and a threat ("zero-click" answers that satisfy demand without a visit) [2][3][4][5][6]. This is the one demand force that now cuts across all five.

---

7. Regulation

Traditional publishing is among the least-regulated industries in America, and the light-touch regime is common to all five children: First Amendment protection for editorial content, no federal license to publish (unlike broadcasting, licensed by the Federal Communications Commission, FCC), and no content mandate or price control [2][6]. The rules that actually bite are cross-cutting and commercial rather than editorial:

• Copyright — now the central regulatory story, group-wide. Copyright is what makes a catalog, archive, backlist, or database an asset, and it has moved to the center of the AI debate. Landmark actions span the children: The New York Times' suit against OpenAI and Microsoft (newspapers) [2]; Bartz v. Anthropic (2025), which held training on lawfully acquired books can be fair use while pirated copies are not, alongside a ~$1.5 billion settlement — the largest U.S. copyright settlement on record [4][13]; and publisher suits over AI search summaries across periodicals [3]. Chain-of-title diligence is the characteristic legal task across the whole group [6].
• Postal regulation. For everything that moves by mail — periodicals (the discounted USPS Periodicals class), catalogs, mailed cards, and directories — U.S. Postal Service (USPS) rate decisions, overseen by the Postal Regulatory Commission (PRC), are effectively industry regulation; a rate hike is a direct cost shock [3][6].
• Antitrust and merger review. The Department of Justice (DOJ) and Federal Trade Commission (FTC) review deals under the 2023 Merger Guidelines [10]. Enforcement is real despite low headline concentration: a federal court blocked Penguin Random House's purchase of Simon & Schuster in 2022 on monopsony (dominant-buyer) grounds [4], and newspaper review turns on local markets even though the national HHI is tiny [2].
• Marketing, privacy, and data law. The FTC's advertising and auto-renewal rules, CAN-SPAM (commercial email), COPPA (children's online privacy), and state privacy laws (California's CCPA) touch several children — and directory/mailing-list publishers face the sharpest front: the Fair Credit Reporting Act (FCRA), the Telephone Consumer Protection Act (TCPA), and fast-moving state data-broker deletion regimes (California's Delete Act and DROP platform) that can shrink the underlying data asset [5].
• Segment-specific rules. Open-access mandates (the 2022 OSTP "Nelson memo") are reshaping scholarly-journal economics [3]; button-battery safety (Reese's Law) governs musical cards [6]; FERPA governs yearbook student records [6].

Net: regulatory risk is low relative to media peers, but two levers — copyright/AI outcomes and postal rates — can reprice whole segments, and each falls on a different subset of the children.

---

8. Consolidation

By the group statistics this looks like an open, fragmented field (CR4 14.2%, HHI 110 [1]) — but as Section 3 warned, that is a pooling artifact, and the real consolidation stories live inside the children, in opposite directions:

• Newspapers are consolidating among the survivors and closing the rest: GateHouse merged into Gannett (2019), Alden bought Tribune (2021), McClatchy passed to Chatham (2020), Lee was recapitalized in 2026 [2]. Most surviving markets are local monopolies.
• Periodicals show consumer roll-ups (Dotdash's $2.7 billion Meredith purchase; Ziff Davis's brand deals) against a stable scholarly oligopoly — roughly five firms control about half of an estimated $20 billion global journal market [3].
• Books blend a tightly held trade oligopoly (the "Big Five" hold ~80% of consumer publishing) with a fragmented education/independent tail; top-tier mergers are now capped by antitrust, while Amazon dominates distribution and self-publishing disintermediates the base [4].
• Directories consolidated hard at the top — Dun & Bradstreet taken private by Clearlake ($7.7 billion, 2025), Infogroup rebuilt as Data Axle, Acxiom folded into Omnicom [5].
• Other publishers run a mature card duopoly (Hallmark + American Greetings, >80% of card sales) alongside pocket-by-pocket PE roll-ups in yearbooks (Jostens/Platinum, Herff Jones/KKR) and posters (Trends International) [6].

The common thread: private equity and strategic buyers are consolidating the cash-generative, contract- or rights-defended pockets across every child, while the commodity print businesses are cut, sold in distress, or closed. The group index hides all of it.

---

9. Risks

The children share a common risk spine, weighted differently by segment:

• Secular print decline — the existential risk, most acute for newspapers, consumer magazines, paper maps, general catalogs, and low-end calendars; ongoing, not one-time [2][3][6].
• AI and "zero-click" substitution — AI answer engines and AI-generated content can divert traffic and undercut undifferentiated work across all five children; the flip side of the AI licensing opportunity [2][3][4][5][6][13].
• Platform and channel dependence — reliance on a few gatekeepers: Google and Meta for ad and referral traffic (newspapers, magazines), Amazon for book distribution (books), and a handful of giant retailers or school contracts for the "other" child [2][4][6].
• Advertising cyclicality and share loss — a shrinking, cyclical base for the ad-dependent pockets [2][3].
• Input-cost inflation — paper, ink, freight, labor, and especially postage are volatile and largely outside operators' control [3][6].
• Copyright/AI litigation uncertainty — unresolved law on training and output cuts both ways for every catalog-rich child [4][13].
• Leverage at PE- and sponsor-owned assets — buyout debt loaded onto mature, slowly declining cash flows (distressed newspaper chains, American Greetings, KKR's Varsity Brands, the private data compilers) [2][5][6].
• Segment-specific hazards — subscriber churn and price resistance (news, magazines); returns and hit-driven volatility (books); privacy-law compression of the usable data pool (directories); seasonal inventory and school-participation decline (other) [2][4][5][6].
• Measurement opacity — employer-only federal data undercount the large nonemployer economy, and the dominant players in most children are private, so top-down sizing is imprecise [4][6].

---

10. How to invest, and the outlook

The group is not a single trade — pick the child, then the vehicle. For a public investor, the practical menu is narrow and clusters in two places: the newspaper pure-plays (NYT the cleanest scaled paid-digital winner; News Corp for premium business news; USA Today Co. and Lee as leveraged turnarounds) and the scholarly-information houses inside the periodical and book children (RELX, Wiley, Informa, Springer Nature) as the most defensible recurring cash flows in the group. Everywhere else, listed exposure is a segment inside a diversified owner — education names (Scholastic, McGraw Hill) for books, People Inc./Ziff Davis for consumer digital magazines, Thryv/GTM for directory data, foreign/adjacent proxies (Moonpig, ACCO, Cimpress) for the "other" child — so the discipline is always the same: isolate the publishing segment's revenue and operating profit from the parent, then judge recurring-vs-transactional mix, renewal/churn, pricing power, rights ownership, digital conversion, and free cash flow after debt, leases, and (for newspapers) pensions.

For a private investor the opportunity set is far richer, because the group's best assets are private: scholarly and B2B/trade titles with durable subscription cash flow; book backlist, catalogs, and rights libraries; data compilers with proprietary, frequently refreshed records; yearbook and calendar/poster franchises; and the replacement-demand base of the card duopoly. Routes run through PE control deals, family-business successions, catalog/royalty investing, niche roll-ups, and — at the lowest capital entry — building or buying an independent line (a title, a data niche, a card or planner brand, a self-published author-business). Diligence weights repeat sell-through over catalog size, rights and chain-of-title over raw audience, and cash generation over accounting profit [4][6].

Outlook (judgment). Expect the group to keep bifurcating, both across and within its children. The commodity, print- and ad-dependent pockets — general local newspapers, mass-consumer magazines, paper directories and maps, low-end calendars and catalogs — should keep eroding, while the defensible, recurring pockets — scholarly journals, premium news subscriptions, book backlist and audio, proprietary data platforms, yearbooks, and the card duopoly's premium lines — should hold up and keep attracting private capital. The shared swing factors to watch are how much durable revenue AI content-licensing delivers versus how much traffic AI answer-engines destroy, the trajectory of postal rates and paper costs, the pace of the open-access transition in scholarly publishing, and the ordinary advertising cycle. No official industry forecast is provided in the federal statistics; the group is best underwritten child by child, and franchise by franchise — never off the deceptively sleepy 4-digit average. These are judgments, not guarantees.

For the full treatment of any child — complete company tables, unit economics, detailed regulation, and per-metric sourcing — read the five child primers: 51311 Newspaper, 51312 Periodical, 51313 Book, 51314 Directory & Mailing List, and 51319 Other Publishers.

---

Sources

This rollup is synthesized from the five child primers and our ground-truth federal statistics for NAICS 5131. Per-metric citations for the children live in their own primers.

1. U.S. Census Bureau, 2022 Economic Census — Concentration by Largest Firms, NAICS 5131 — our ground-truth file for this level (receipts $99,410,725 thousand; 13,326 firms; CR4 14.2%, CR8 23.3%, CR20 40.6%, CR50 58.0%; HHI 110.2). https://www.census.gov/programs-surveys/economic-census.html
2. Histometrics child primer, NAICS 51311 Newspaper Publishers (receipts $23.2B; 3,891 firms; CR4 41.0%; HHI 522.4; investable universe, economics, regulation, consolidation, risks), synthesizing U.S. Census Economic Census 2022 and CBP 2023, Northwestern/Medill State of Local News, and company 10-K filings.
3. Histometrics child primer, NAICS 51312 Periodical Publishers (receipts $24.36B; 3,805 firms; CR4 22.8%; HHI 201.3; scholarly-journal economics, open-access, postal), synthesizing U.S. Census Economic Census 2022 and CBP, and company filings.
4. Histometrics child primer, NAICS 51313 Book Publishers (receipts ~$29.15B; 2,191 firms; CR4 35.2%; HHI 461.5; backlist/frontlist economics, Amazon distribution, self-publishing, AI-copyright), synthesizing U.S. Census Economic Census 2022 and CBP 2023, Association of American Publishers StatShot, Bowker self-publishing data, and Bartz v. Anthropic.
5. Histometrics child primer, NAICS 51314 Directory and Mailing List Publishers (receipts $5.146B; 469 firms; CR4 45.6%; HHI 841.4; data-asset economics, privacy/data-broker regulation, consolidation), synthesizing U.S. Census Economic Census 2022 and CBP 2023, Dun & Bradstreet First Research, and BLS employment series.
6. Histometrics child primer, NAICS 51319 Other Publishers (greeting cards 513191 + all-other 513199; receipts $17.5B; 3,070 firms; CR4 35.4%; HHI 461.3; card duopoly, yearbook/calendar/poster economics, postal), synthesizing U.S. Census Economic Census 2022 and CBP 2023 and company/ownership filings.
7. U.S. Census Bureau, 2022 Economic Census — Concentration by Largest Firms (underlying child-level receipts, firm counts, CR ratios, HHI). https://data.census.gov/table/ECNSIZE2022.EC2200SIZECONCEN
8. U.S. Census Bureau, County Business Patterns 2023 (child-level establishments, employment, annual payroll; employer-only universe). https://www.census.gov/programs-surveys/cbp.html
9. U.S. Census Bureau, 2022 NAICS Definitions — 513110, 513120, 513130, 513140, 513191, 513199 (scope and exclusions). https://www.census.gov/naics/?year=2022
10. U.S. Department of Justice and Federal Trade Commission, Merger Guidelines, 2023 (HHI thresholds). https://www.ftc.gov/system/files/ftc_gov/pdf/2023_merger_guidelines_final_12.18.2023.pdf
11. Association of American Publishers, StatShot Annual Report (U.S. book-publishing revenues). https://publishers.org/news/aap-statshot-annual-report-publishing-revenues-totaled-32-5-billion-for-calendar-year-2024/
12. Publishers Weekly / Bowker, Self-Publishing's Output and Influence Continue to Grow (2.6M+ self-published ISBN titles, 2023). https://www.publishersweekly.com/pw/by-topic/industry-news/publisher-news/article/96468-self-publishing-s-output-and-infuence-continue-to-grow.html
13. NPR / The Authors Guild, Anthropic settles with authors in first-of-its-kind AI copyright case; Bartz v. Anthropic (2025); and The New York Times Co. v. OpenAI/Microsoft — AI-and-copyright litigation across the group. https://www.npr.org/2025/09/05/nx-s1-5529404/anthropic-settlement-authors-copyright-ai
14. Northwestern University / Medill Local News Initiative, The State of Local News 2025 (newspaper closures and consolidation). https://localnewsinitiative.northwestern.edu/projects/state-of-local-news/2025/