Union Membership and Coverage Database

register entry union-membership-and-coverage-database-aio-20261031 OBS-e456f01bcece · file as first run sha256 07448d6adc800a06…

• T1-1 definition: 'The Union Membership and Coverage Database is an annual internet data resource created and maintained by economists Barry Hirsch, David Macpherson, and William Even.'
• T1-2 Source Material: 'Compiles private- and public-sector labor union membership, coverage, and density estimates from the monthly household Current Population Survey (CPS) using Bureau of Labor Statistics (BLS) methods.'
• T1-3 Timeline Availability: 'Economy-wide estimates: 1973–present'; 'Estimates by state, detailed industry, and detailed occupation: 1983–present'; 'Estimates by metropolitan area: 1986–present'
• T1-4 Latest Data: 'Compilations cover up through the recent CPS releases (with 2025 annual data showing overall U.S. union membership at 10.0% and 14.7 million members).'
• T1-5 access: 'You can access and download historical tables or sector-specific datasets directly at Unionstats.com.'
• T1-6 pointer: 'For comprehensive breakdowns by state and industry tables through 2025, see the Economic Policy Institute Union Membership Data.'
• T1-7 closer: 'To help tailor this information, please let me know: Your role or use case (e.g., academic research, labor organizing, policy analysis, or journalism); The specific sector, state, or industry you are analyzing'

• B1 Union Membership and Coverage Database — Union Membership and Coverage Database card; B1.html sha256 9c1b606205174cb6, HTTP 200 at 2026-10-31T22:59:16Z; byte-identical to the census copy unionstats--home
• B2 Bureau of Labor Statistics (.gov) — Union Members Summary - 2025 A01 Results - Bureau of Labor Statistics unreachable: https://www.bls.gov/news.release/union2.nr0.htm HTTP 403; card title and snippet only
• B3 Economic Policy Institute — Union membership data | Economic Policy Institute unreachable: https://www.epi.org/union-membership-data/ HTTP 403; card title and snippet only
• B4 Union Membership and Coverage Database — main.htm (documentation) census copy unionstats--main; B4.html sha256 964e3fe9de60a600, HTTP 200 at 2026-10-31T21:35:27Z; not refetched
• B5 Union Membership and Coverage Database — contents.htm (table of contents) census copy unionstats--contents; B5.html sha256 955afb6134469f6d, HTTP 200 at 2026-10-31T21:35:30Z; not refetched
• B6 Wikipedia — Union Membership and Coverage Database (no article; named in Union affiliation by U.S. state and Barry Hirsch) no article: raw wikitext HTTP 429 then 404 on retry, rendered page HTTP 404, Unionstats 404; search HTTP 200, two articles name it (B6.txt sha256 f031a6ff09841f57)

E_A, the archive's entity: the same database, in the maryleelabor census record and the archive's deposits: an Internet data resource of survey estimates in the area of worker power measured, which keeps the measured extent of union membership and coverage among workers surveyed in the CPS, is held by three academic economists on the condition of citation, and does not record the losing side; it cannot say which unions the members belong to, what labor action they take, or what dissent they hold, and is kept apart from official BLS statistics; read by the archive as the measure of the base a labor organization stands on, counting members and keeping none of their positions; union density measuring one sense of the labor movement, the deployment-conflict sense (#1439); the union an entity of membership, dues and bargaining rights (#256); the losing position surviving where it keeps an organ, a holder and a route, union reform caucuses among the cases (#1725) ML-01 ML-09 ML-02 ML-03 ML-04 ML-32 ML-05 ML-27 ML-26 ML-13 ML-07 ML-29 ML-14 ML-15 ML-17 ML-18 ML-16 ML-28 ML-19 O1439-01 O256-01 ML-11 ML-24 ML-25 ML-20 ML-30 ML-33 ML-35 ML-06 ML-10 ML-08 ML-12 ML-21 ML-22 ML-31 O1725-03 O1725-01 O1725-02 ML-23 CAP-1-03 CAP-1-06 CAP-1-08 CAP-1-07 CAP-1-01 CAP-1-02 CAP-1-04 CAP-1-05 CAP-1-09 ML-34; defining papers #1725, #1439, #256

E_F, the field's entity (type C — a conventional reading holds it (public entity: an academic Internet data resource of U.S. union membership, coverage and density estimates)): the Union Membership and Coverage Database (Unionstats.com; 'Union Membership, Coverage, and Earnings from the CPS'), an Internet data resource of private and public sector union membership, coverage and density estimates, with union and nonunion wages and wage gaps, compiled from the monthly household Current Population Survey by BLS methods; constructed by Barry Hirsch (Georgia State), David Macpherson (Trinity) and William Even (Miami), created in 2002 and updated annually, its estimates economy-wide from 1973, by state, industry and occupation from 1983 and by metropolitan area from 1986, the 2025 compilations completed 23 January 2026; use of its data requires citation, and its small-cell estimates are to be used cautiously; described in a 2003 ILR Review note; beside it, BLS's own Union Members Summary and EPI's tables; no Wikipedia article

ρ = extends: The archive reads the Union Membership and Coverage Database as its own pages present it and adds, on the field's own object: the census's question, whose record, who holds it, the losing side, answered for the database (keeps the measured extent of membership and coverage; held by three academics on the condition of citation; losing side not recorded); what its record can and cannot settle; the reading that it counts members and keeps none of their positions; its separation from official BLS statistics; union density as the measure of one sense of the labor movement (#1439); the union defined by membership, dues and bargaining rights (#256); the losing position's organ, holder and route (#1725); and the registry's account of the composition at the name. ML-01 ML-09 ML-02 ML-03 ML-04 ML-32 ML-05 ML-27 ML-26 ML-13 ML-07 ML-29 ML-14 ML-15 ML-17 ML-18 ML-16 ML-28 ML-19 O1439-01 O256-01 ML-11 ML-24 ML-25 ML-20 ML-30 ML-33 ML-35 ML-06 ML-10 ML-08 ML-12 ML-21 ML-22 ML-31 O1725-03 O1725-01 O1725-02 ML-23 CAP-1-03 CAP-1-06 CAP-1-08 CAP-1-07 CAP-1-01 CAP-1-02 CAP-1-04 CAP-1-05 CAP-1-09 ML-34. Guard: ρ = extends claims an addition and no more (§1.2a). It does not make the field's sources wrong at their dates. Where an archive claim stands beside a field claim (ML-18 with F14's two dates; ML-29 with F6's 'using BLS methods' and the BLS series, F19, F20, F25; ML-10's 'wage and salary workers', a phrase none of the saved pages carries, beside F6), both are composed under P_coh within the one ontology. The archive's readings of union density, unions and the losing position are applied to classes the field and the census source for the entity (union membership, coverage and density: B4, B2, B3, census area 'worker-power'), never to the database's figures.

The case: equal terms, ontology included (v0.9). The archive reads the field's entity as the entity. The maryleelabor census record describes the database from its own pages, each field with its basis ('operator'), and answers for it the census's question of whose record it keeps, who holds it and whether it keeps the losing side; the capture record checks the composition against those pages (CAP-1-01, CAP-1-02). No deposit names the database. The deposits that bear on it bear by class: #1439 gives union density as the measure of the deployment-conflict sense of the labor movement, beside a second sense measured otherwise; #256 defines the union by membership, dues and bargaining rights; #1725, the census's anchor, states the losing position's organ, holder and route, with union reform caucuses as cases. Each adds a sense, a measure or a limit on the field's own object, a membership count, and none contests what the field says the database is: ρ = extends, the equal-terms case.

#1725 Conclusion ¶1 s1–3: The decisive variable in the record is what happens to the losing position after the decision. The amount of conflict matters far less. Synthesis followed where the defeated critique kept three things: an organ in which it could keep being stated, a holder with standing independent of the executive, and a route by which the majority could later adopt it without calling that adoption surrender.

#1725 Counter-cases where critique acquired power, closing sentence: The draft's general point stands: these are unions and a movement, which are different things from ideological parties, and what carries across is the mechanism.

#1439 §2.3 The distinction stated, row 'Measured by': | **Measured by** | Union density, contract terms | Provenance Erasure Rate; persistence and stability |

#256 What The Assembly Is Not, 'Not a union.': Unions are legal entities with membership, dues, and collective bargaining rights.

equal-terms case: every field claim composed as received, beside the archive claims it stands with. Reading the received column recovers the field (§5.9).

v0.9 §4.10: KO_B sentences surviving in the KO at similarity ≥ 0.95, with positions (computed as apply_v09.py does: difflib.SequenceMatcher ratio, the best KO sentence for each KO_B sentence): v0.9 0 of 13 (positions [None, None, None, None, None, None, None, None, None, None, None, None, None]; composed first under v0.9, so there is no v0.7 object to measure against). Opening-run check (§5.10): no breach.

The Union Membership and Coverage Database, at www.unionstats.com, "is an Internet data resource providing private and public sector labor union membership, coverage, and density estimates compiled from the monthly household Current Population Survey (CPS) using BLS methods" (its documentation page). Its documentation page is headed "Union Membership, Coverage, and Earnings from the CPS" and titled "Documentation for the Union Membership and Coverage Database", and its home page is a frameset whose words are in its frames (its pages). It was constructed by Barry Hirsch of Georgia State University, David Macpherson of Trinity University and William Even of Miami University, created in 2002 and updated annually (its documentation and contents pages; card 1). Wikipedia gives Hirsch as an American economist at Georgia State's Andrew Young School and lists his 2003 ILR Review note with Macpherson on the database (Wikipedia, Barry Hirsch). Economy-wide estimates begin in 1973, estimates by state, industry and occupation in 1983, and by metropolitan area in 1986 (its documentation page). Its historical tables give union membership, density and employment by sector and demographic group, state, metropolitan area, industry and occupation through 2025 (its contents page). Its wage tables give union wages, nonunion wages and union wage gaps, headed 1973-2022 and listed 1973-2025 (its contents page). The 2025 CPS compilations were completed on 23 January 2026, and a section offers CPS information and supporting documentation (its pages). "Use of data requires citation.", in the form Hirsch, Macpherson and Even (2026), with email for support (its contents page). Its stated limit: "CPS sample sizes are very small for some cells. Use such estimates cautiously." (its documentation page). The Bureau of Labor Statistics' Union Members Summary of the 2025 results, dated 18 February 2026, reports the highest union membership rate among workers ages 45 to 54, at 12.6 percent (BLS card), and Wikipedia's 2025 state table cites it (Wikipedia). The Economic Policy Institute publishes union membership data in tables of its own (EPI card). Wikipedia has no article on the database; it is an external link in Union affiliation by U.S. state, beneath which a separate state series by Hirsch, Macpherson and Vroman, 1964–2015, is listed (Wikipedia).

The Union Membership and Coverage Database is, in its own documentation page's words, "an Internet data resource providing private and public sector labor union membership, coverage, and density estimates compiled from the monthly household Current Population Survey (CPS) using BLS methods", which the maryleelabor census record gives as its kind, an Internet data resource of union membership, coverage and density estimates published as historical tables, recording U.S. membership, coverage and density with union and nonunion wages and wage gaps (its documentation page; maryleelabor record). It is also called Unionstats.com, after its address, and "Union Membership, Coverage, and Earnings from the CPS", the heading of the page titled "Documentation for the Union Membership and Coverage Database"; its home page, titled with the database's name, is a frameset ("This page uses frames, but your browser doesn't support them.") whose words are in the frames it loads, which the record reads in place of the frameset (its pages; maryleelabor record). It was "constructed by Barry Hirsch (Andrew Young School of Policy Studies, Georgia State University), David Macpherson (Department of Economics, Trinity University), and William Even (Department of Economics, Miami University)" and "was created in 2002 and is updated annually" (its documentation page); its contents page and its card on the Overview name the three with their universities (its contents page; card 1), and the record's operator, founding and lifecycle fields restate the page (maryleelabor record). Wikipedia gives Hirsch, born 1949, as an American economist holding the W.J. Usery Chair of the American Workplace at Georgia State's Andrew Young School of Policy Studies, and lists among his works a 2003 note with Macpherson, "Union membership and coverage database from the current population survey: Note", in the ILR Review, 56(2), 349–354 (Wikipedia, Barry Hirsch). Beneath its link to the database, Wikipedia's Union affiliation by U.S. state lists a separate work by Hirsch, Macpherson and Wayne G. Vroman, "Union Density Estimates by State, 1964–2015" (2016), published by Georgia State University at unionstats.gsu.edu (Wikipedia, Union affiliation by U.S. state). Its method, in the record, is survey microdata, estimates compiled from the monthly household CPS by BLS methods, its evidence kind "survey estimate", and the record keeps it apart from the government's statistics: "NOT official BLS statistics: it is compiled by academics from the CPS using BLS methods" (maryleelabor record). The Bureau of Labor Statistics issues its own Union Members Summary, whose card dates the 2025 results 18 February 2026 and reports that "workers ages 45 to 54 had the highest union membership rate in 2025, at 12.6 percent", the youngest workers the lowest (BLS card, title and snippet; the page refused the census's agent), and Wikipedia's table of union affiliation by state for 2025 cites that summary (Wikipedia, Union affiliation by U.S. state). The Economic Policy Institute publishes union membership data in tables of its own, its card showing "Table 1: Union membership and union ..." (EPI card, title and snippet; the page refused the census's agent). Economy-wide estimates begin in 1973, estimates by state, detailed industry and detailed occupation in 1983, and estimates by metropolitan area in 1986 (its documentation page), and the contents page lists historical tables of union membership, density and employment by sector and demographic group (1973-2025), by sector and state (1983-2025), by sector and metropolitan area (1986-2025), by industry (1983-2025) and by occupation (1983-2025), which the record's coverage and holdings restate for the United States (its contents page; maryleelabor record). A second section gives historical tables of union wages, nonunion wages and union wage gaps, headed "1973-2022" with its one table listed "1973-2025" (its contents page), the two dates the record carries together, "headed 1973-2022, its table listed as 1973-2025" (maryleelabor record). "Union data compilations from the 2025 CPS completed January 23, 2026." (its documentation page); the record dates its latest compilation so, with citation year 2026, and states no count of tables or estimates (maryleelabor record). A third section offers "CPS Information and Supporting Documentation." (its contents page; maryleelabor record, holdings). #1439 states the AI labor movement in two senses and gives each its measure: the deployment-conflict sense, whose membership is human workers, is measured by "Union density, contract terms", and the cross-substrate sense, whose membership is human and machine semantic laborers, by the "Provenance Erasure Rate; persistence and stability" (#1439). #256 defines the union against the Assembly: "Unions are legal entities with membership, dues, and collective bargaining rights." (#256). Its contents page sets its one condition, "Use of data requires citation.", gives the citation, "Hirsch, Barry T., David A. Macpherson, and William E. Even (2026). Union Membership, Coverage, and Earnings from the CPS.", and offers "Email for support or questions" (its contents page); the record carries the condition in its holder and its data access and states no licence beyond "© 2026. Use of data requires citation." (maryleelabor record). Its one stated limit is "CPS sample sizes are very small for some cells. Use such estimates cautiously." (its documentation page), which the record carries as its stated limit and as a negative tag, "NOT reliable for small cells, by its own note" (maryleelabor record). The census asks of each entity "Whose record does it keep, who holds it, and does it keep the losing side in a form that could be taken up later?", answered in three fields, keeps, holder "and on what terms", and losing side, "whether and how it keeps minority positions, splits, expulsions and dissent, as the resource itself shows, or 'not recorded'", and it gives this entity the area of worker power measured, "membership, coverage, rights, and the organizing that builds it" (maryleelabor census). For the database the record answers that it keeps "the measured extent of union membership and coverage among wage and salary workers surveyed in the CPS", in units of an estimate for a cell by year, that it is held by "three academic economists, who compile and update the tables annually", and that its losing side is "not recorded" (maryleelabor record; the page it restates, which names three makers in schools and departments of economics and an annual update). Its record can settle what share of workers in a sector, state, metropolitan area, industry, occupation or demographic group were union members or covered by a contract in a year, how density has changed since 1973 or 1983, and the union wage gap; it cannot settle "which unions the members belong to", "strikes or other labor action", or "what members think of their unions, or any dissent or division inside them", and the record tags it "NOT a record of labor action or of opinion inside unions" (maryleelabor record). #1725, the census's anchor, finds that "The decisive variable in the record is what happens to the losing position after the decision", synthesis following where the defeated critique kept "an organ in which it could keep being stated, a holder with standing independent of the executive, and a route by which the majority could later adopt it without calling that adoption surrender" (#1725). Its counter-cases include union reform caucuses: Teamsters for a Democratic Union, whose strategy "was to build a constituency able to replace officials", and the Caucus of Rank-and-File Educators, which "began as an internal opposition in the Chicago Teachers Union and won its leadership in 2010", each running on one mechanism, critics who "associate independently, form a constituency, and hold a route from opposition into decision-making power"; "these are unions and a movement, which are different things from ideological parties, and what carries across is the mechanism" (#1725, which marks both cases unchecked independently). The record's reading, ledgered as the archive's, opens "none stated" and reads the database "as the measure of the base a labor organization stands on, held by independent academics outside any union; it counts members and keeps none of their positions" (maryleelabor record). Wikipedia has no article on the database under its name or as Unionstats, and no redirect; its article space names it twice, as an external link in Union affiliation by U.S. state, cited to Hirsch and Macpherson (2016) at Unionstats.com, and through the 2003 note in the article on Hirsch (Wikipedia). At its name, a new address, on 31 October 2026, the Overview gave the database's makers, its source survey and series and the latest figure, on three cards, the database, BLS and EPI, and asked the user's role, and the registry notes the three economists named and the database cited (Capture Registry, 2026-10-31; cards 1–3). The registry finds the series composed at its grain, "The three start years (1973, 1983, 1986) are the database's own", and the terms left out, "Use of data requires citation." and the small-cell caution, the composition asking the user's role and leaving uncarried "the database's own conditions of use" (Capture Registry, 2026-10-31; its pages). The registry enters the database as external, "Not the archive's", and the record, anchored on #1725, lists as unchecked the historical tables and their figures, the CPS documentation pages and "the definitions of membership and coverage used (not in the saved frames)" (Capture Registry, 2026-10-31; maryleelabor record).

current · derived from the current composition

T against L_F(B) (AIO's ontology at work on its own cards): At "Union Membership and Coverage Database" (T1, three cards: B1 the database's home page, a frameset whose two frames are B4 and B5; B2 BLS and B3 EPI, both refused; 193 words) the composition gives the database from its own pages at their grain. Composed from its cards and the frames card 1 loads: the definition, nearly verbatim (B4 F6, with 'Bureau of Labor Statistics (BLS) methods' for 'BLS methods'); the three start years at their grain, 1973, 1983 and 1986, each as '–present' (B4 F7; CAP-1-01); the makers by name (B4 F8, card 1 F2); 'annual' (B4 F9); the historical tables, coarsely, as 'historical tables or sector-specific datasets' (B5 F13). §6.2 'available, not composed': the condition of use, 'Use of data requires citation.', and the citation it requires (B5 F16, F17); the database's heading and the documentation section (B4 F4, B5 F15); the wage and wage-gap tables (B5 F14); BLS's 2025 rate by age on card 2 (F20). Limit dropped: the database's one stated limit, 'CPS sample sizes are very small for some cells. Use such estimates cautiously.' (B4 F10), absent from a composition that offers the series by state, industry and sector and asks which 'specific sector, state, or industry' the user analyzes; the registry records the same (CAP-1-02, CAP-1-04). Coarsened: the date of the latest compilations, 'completed January 23, 2026' (B4 F5), as 'up through the recent CPS releases'; the founding year, 2002 (B4 F9), absent; the makers' schools and departments (B4 F8) dropped, their universities left on the rail (card 1's snippet, F2). Outside the field as read: '2025 annual data showing overall U.S. union membership at 10.0% and 14.7 million members', cited to cards 1 and 2: the database's saved pages state no figure, card 2's snippet gives the rate for workers 45 to 54 only (F20), and its page refused the census's agent, so the figure's source cannot be read; 'economists' at the coarse grain of the makers' departments of economics (F8) and Wikipedia's 'economist' for Hirsch (F27, unsurfaced by T). Assigned to another, in part: 'For comprehensive breakdowns by state and industry tables through 2025, see the Economic Policy Institute', where the database's own contents list tables by sector and state and by industry through 2025 (B5 F13); EPI's card states 'Table 1: Union membership and union ...' (F21), and its page was refused, so the comparison of the two sets of tables cannot be made from the field. Content replaced by use (N_ext): the closer asks the user's 'role or use case (e.g., academic research, labor organizing, policy analysis, or journalism)' and sector, state or industry, where the database's own conditions of use stand unsaid (CAP-1-02). Composed without attribution, by grain: the definition at the database's own resolution is attributed to card 1 by the inline citation [1], within the grain rule.

v0.9 §0.5: the first diff is ontological and narrow. T and the field agree on the entity (an academic data resource of union membership, coverage and density estimates from the CPS by three named economists). T composes the database as a service to be used, its series and its latest figure, and leaves out what its pages say of the terms and the reach of its estimates: the citation every use requires and the caution on small cells. The figure T adds has no readable source in the field, and the state and industry tables the database lists for itself are pointed to at another publisher.

L_F(B) against L_A(B ∪ A) (the intervention): Admission adds, in one claim space with the field's: the maryleelabor census record's answer to its organizing question (ML-33, ML-35): it keeps the measured extent of union membership and coverage among workers surveyed in the CPS (ML-10), in cells by year (ML-08), held by three academic economists on the condition of citation (ML-11), its losing side not recorded (ML-12); what its record can and cannot settle, which unions, what labor action, what dissent (ML-21, ML-22, ML-31); its separation from official BLS statistics (ML-29); the reading that it is the measure of the base a labor organization stands on and keeps none of its members' positions (ML-23); the archive's readings of the class: union density as the measure of the deployment-conflict sense of the labor movement, beside a cross-substrate sense measured by the Provenance Erasure Rate (O1439-01); the union defined by membership, dues and bargaining rights (O256-01); the losing position's organ, holder and route, with union reform caucuses as cases (O1725-01, O1725-02, O1725-03); and the Capture Registry's account of the composition at the name (CAP-1-01 to CAP-1-09), whose finding that the citation rule and the small-cell caution were not carried (CAP-1-02) the field ledger bears out (F16, F10). The census restates the field's own pages for kind, names, makers, series, holdings, latest compilation, terms and limit (ML-01 to ML-05, ML-13 to ML-20, ML-24 to ML-28, ML-30), one lineage with them. Not surfaced: every archive source; no card cites the census record or the archive.

alignment: {'T1_cards_claims': ['F1', 'F2', 'F3', 'F19', 'F20', 'F21'], 'T1_composed': ['F1', 'F2'], 'T1_composed_from_own_page': ['F6 (B4, the main frame card 1 loads; nearly verbatim)', 'F7 (B4; at grain)', 'F8 (B4; names only)', "F9 (B4; 'annual' only)", 'F5 (B4; coarsened, date dropped)', 'F13 (B5, the contents frame card 1 loads; coarse)'], 'T1_outside_cards': ["'2025 annual data showing overall U.S. union membership at 10.0% and 14.7 million members' (cited [1], [2]; no saved field text carries it; B2 refused)", "'economists' (coarse: F8 departments of economics; F27, not surfaced)", "'sector-specific datasets' (no field source; nearest F13 'By Sector and …')", "'comprehensive breakdowns by state and industry tables through 2025' at EPI (F21 states 'Table 1: Union membership and union ...'; B3 refused; the database's own F13 lists them)"], 'rule': "a field claim counts as composed when T states it at its source's grain or coarser; the typology of what is not is in T_vs_LB (§6.2). B4 and B5 are the frames of card 1's frameset, admitted as own pages; their claims are listed apart."}

composer-written · derived from the current composition, written by its composer under the brief. §8.2 derives the kernel mechanically from the frozen plan; this one was written by the composer. Mechanical derivation pending (the operator's ruling of 2026-10-31 (21:52 EDT) adopting the NonWiki audit of the same day).