Purpose, mission, and institutional legitimacy

A declared purpose produced a bounded refusal

On February 5, 2014, CVS Caremark announced that more than 7,600 U.S. CVS/pharmacy stores would stop selling cigarettes and other tobacco products by October 1. The company called tobacco sales inconsistent with its purpose of helping people toward better health. It also estimated approximately $2 billion in annual revenue associated with the tobacco shopper, equal to about 17 cents per share, while saying that 2014 operating-profit and earnings guidance would not change and that it had identified opportunities to offset the profitability effect.¹

CVS reported completing the removal on September 3, nearly a month early, on the same day it changed the corporate name from CVS Caremark to CVS Health. Its release placed the decision beside 7,700 retail pharmacies, 900 walk-in clinics, a pharmacy-benefit manager, and specialty-pharmacy services.² The forecast therefore supports a consequential revenue refusal, not a measured $2 billion reduction in profit or a purely sacrificial act.

The public-health premise was stronger than the company's own assertion. The 2014 U.S. Surgeon General's evidence review describes the continuing disease and death burden caused by smoking and evaluates causal evidence across cancer, respiratory, cardiovascular, reproductive, and other outcomes.³ That evidence establishes the harm of smoking; it does not establish how much one retailer's withdrawal changed smoking behavior.

The pharmacy contradiction was an attributed argument

I01-N01, the pharmacy-counter contradiction, comes from the tension between delivering pharmacy and clinic services and selling tobacco. In a contemporary JAMA viewpoint, CVS chief medical officer Troyen Brennan and UCSF smoking- cessation leader Steven Schroeder argued that expanding clinical roles made tobacco sales in pharmacies increasingly untenable.⁴ One author was the senior medical officer of the company announcing the decision. The article records a participant justification joined to a public-health argument, not an independent evaluation of CVS's motives or effects.

I01-N02, CVS's tobacco refusal, is the documented operating response. I01-E01 is Grade A for the narrow relation: CVS announced and then completed the nationwide removal.² Its qualification matters. Ending one product category does not validate every product, labor practice, health service, or corporate claim.

The refusal made the stated purpose more demanding in one domain. Tobacco could no longer be treated as an ordinary retail option inside CVS stores. That is evidence that purpose constrained a choice; it is not by itself evidence that everyone affected accepted the purpose, that the implementation distributed benefits fairly, or that the whole institution became legitimate.

Purpose and strategy were presented together

I01-N03, purpose and strategy reinforce one another, does not assign a single private motive. CVS's February release said both that the decision was right for customers and that it would distinguish the company and position it for growth in a changing health-care market. The September release paired tobacco removal with the CVS Health name and a broader service portfolio.¹²

I01-E02 is therefore Grade B, not proof of altruism or opportunism. The company publicly joined a health commitment, competitive positioning, forecast revenue, and expected profit offsets. Strategic advantage and a purpose-constrained choice can coexist. The selected records do not reveal how each executive or board member privately weighted those considerations, nor do they measure the long-run counterfactual revenue or reputation effect.

Outcome studies support narrower claims

I01-N04, outcome evidence, begins with a CVS-authored study rather than a company success statement. Jennifer Polinski and colleagues compared household purchase panels and state-level point-of-sale data before and after the removal. CVS-exclusive cigarette purchasers were 38% more likely than purchasers who used only other retailers to stop recorded cigarette purchasing for at least six months. In 13 states where CVS had at least 15% retail market share, the modeled change relative to three states without CVS stores was 0.14 fewer packs per smoker per month.⁵

That study measured recorded purchasing, not verified abstinence. All six authors worked for CVS Health; CVS funded the study and participated throughout its design, analysis, interpretation, and publication. Its observational household and state comparisons reduce some alternative explanations but do not randomize exposure, and three no-CVS states form a narrow comparison group.⁵

A later study by CDC researchers linked CVS store density to Behavioral Risk Factor Surveillance System responses. In the two years after removal, smokers in high-density counties had a modeled 2.21 percentage-point increase in quit attempts relative to counties without CVS stores. The estimate was larger in urban areas; there was no statistically significant rural estimate and no detected change between daily and nondaily smoking.⁶ The telephone survey and difference-in-differences design supply evidence independent of CVS authorship, but still do not randomize store exposure or establish successful cessation.

I01-E03 remains Grade B because the two observational designs and data types point to some change while bounding it. The evidence supports changes in recorded purchases and reported quit attempts associated with CVS presence. It does not support a uniform national effect, a rural effect, a change in smoking frequency, or the claim that CVS alone caused every observed change.⁵⁶

I01-E01 through I01-E03 retain working research status because no internal CVS source-record entry is encoded for their structured source fields. Their external evidence is carried by the adjacent notes, and their grades remain bounded by the stated qualifications rather than implying completed peer review.

Approval varied across the people surveyed

A 2014 probability-panel Internet survey of 4,269 U.S. adults asked about prohibiting all tobacco sales in retail pharmacies. Overall support was 66.1%, but it was 46.5% among current cigarette smokers and 48.0% among respondents with household income below $15,000. The authors reported Internet-survey, self-report, and sample-size limitations.⁷ The question concerned pharmacy policy generally, not CVS specifically, and a survey response does not confer or withdraw institutional authority. It does show why aggregate praise cannot stand in for every affected audience.

Independent qualitative research before the CVS decision studied seven California retailers—four pharmacies and three grocers—that had voluntarily stopped tobacco sales. Researchers interviewed owners, managers, and employees, held customer focus groups, and observed the stores. Pharmacy employees interviewed welcomed the change; many customers were initially unaware of it, though focus-group participants largely supported it.⁸ Those small, local, California cases show the kinds of participant evidence a legitimacy inquiry can seek. They are not testimony from CVS workers or customers and cannot be generalized to a national chain.

The adjacent authority, legitimacy, and acceptance question is therefore distinct from mission fidelity. A refusal can show that a declared purpose constrained management. Legitimacy asks how relevant audiences judge the action and the authority behind it, under norms that the organization does not control alone.

Selznick and Suchman provide different analytical lenses

I01-N05, Selznick's institutional embodiment, draws on Philip Selznick's distinction between routine and critical decisions and his chapters on defining mission and embodying purpose. His argument treats leadership as shaping an institution's character and commitments, not merely solving recurring administrative problems.⁹ The book is a primary theoretical work. It does not discuss CVS or demonstrate that embodying a value makes the value just.

I01-E04 is consequently a Grade C interpretive comparison. Tobacco removal can be read as the sort of character-defining operating choice Selznick asks readers to examine.⁹ No cited evidence shows that CVS leaders used his work or that his framework independently evaluated the decision.

I01-N06, Suchman's legitimacy, draws on Mark Suchman's synthesis of organizational-legitimacy research. He defines legitimacy in relation to socially constructed norms and distinguishes pragmatic, moral, and cognitive forms while comparing strategic and institutional approaches.¹⁰ The framework helps separate benefit to an audience, normative approval, and taken- for-granted comprehensibility. It does not say that popularity equals moral legitimacy or provide a neutral score for CVS.

I01-E05 is also Grade C. Suchman's categories clarify why strategic benefit does not automatically negate moral approval and why favorable reputation does not automatically establish it.¹⁰ The comparison is analytical; it is not a documented influence on the 2014 decision.

Purpose remains answerable to effects and contest

I01-N07, purpose as a contestable decision rule, is an editorial research proposition. A useful test asks whom or what the purpose serves, which option it rules out, what effects can disappoint its authors, who bears transition costs, who may challenge interpretation, and how successors may revalidate or change the commitment. This test is not a conclusion of CVS, Selznick, Suchman, or the outcome studies.

The U.N. Guiding Principles on Business and Human Rights offer a narrower normative comparator. For human-rights risk assessment, Principle 18 calls for meaningful consultation with potentially affected groups; later principles address tracking responses, remedy, and legitimate grievance mechanisms.¹¹ The principles do not certify a company's general purpose or evaluate CVS's tobacco decision. They support the limited point that management intention and aggregate outcome evidence cannot replace attention to specific affected people.

I01-E06 is Grade D and marked research needed. CVS motivates a broader beneficiary-refusal-evidence-contest test, but one corporate decision cannot validate that test across public agencies, communities, religious bodies, cooperatives, or firms. No separate counterevidence records are encoded for I01-E01 through I01-E06; their qualifications instead record the forecast, method, motive, lineage, and generalization limits that could otherwise be lost.

Four Workloop proposals remain hypotheses

The I01-P records propose design experiments, not demonstrated effects.

• I01-P01 proposes a durable, scoped intent object with end state, constraints, tradeoffs, risk tolerance, and revalidation triggers. The CVS case and Selznick make consequential commitments visible, but neither tests a software object.¹⁹ Evaluation must compare decisions before and after adoption, test whether coworkers can use the intent, and measure maintenance cost and silent drift. A centrally authored object could become ceremony or freeze a purpose that affected people contest.
• I01-P02 proposes linking every delegation to a higher-scope intent. No cited study tests this rule. A representative audit must measure whether the link improves local tradeoffs or merely adds explanation labor, and whether coercive or low-value work can borrow legitimacy from a desirable parent objective.
• I01-P03 proposes that events challenge governing assumptions without automatically rewriting intent. The two outcome studies show why different measures and geographies can revise the strength of a purpose claim; they do not validate an event-trigger design.⁵⁶ Tests must preserve the original intent and evidence, distinguish noise from material disconfirmation, and reserve amendment authority for an accountable decision.
• I01-P04 proposes making abandonment a first-class decision. Tobacco removal supplies one motivating case, not a general performance result. Trials must follow resources released, obligations honored, worker and customer transitions, and later outcomes against comparable commitments that continued by default. Abandonment language could otherwise shift costs to vulnerable workers or erase inconvenient obligations.

Affected parties and unmeasured consequences bound the case

The company releases speak through senior executives. The CVS-authored study uses purchases; the CDC study uses behavioral-survey responses; the national attitudes study asks about pharmacy policy generally; and the local case study predates CVS and examines different retailers.¹⁵⁶⁷⁸ The selected evidence therefore contains no representative, participant- authored account from CVS store workers, pharmacists, smoking customers, patients, suppliers, competitors, investors, or local communities affected by the nationwide change.

The company's revenue forecast does not quantify how implementation affected jobs, hours, wages, store-level conflict, customer travel, household spending, or returns to particular investors. The outcome studies do not measure those distributions. The absence of those measures is an evidence limit, not a finding that the effects were zero or that every group shared one experience.¹⁵⁶

The selected records also do not assess nonhuman, ecological, waste, or future- generation effects of the removal and replacement of products. That omission is not evidence of no effect. It keeps a purpose stated in terms of human health from silently exhausting the institution's possible obligations.

No structured impact observations are encoded. Their absence means benefits and harms have not been assessed in that part of the record; it does not erase the documented operating decision or the bounded purchase and quit-attempt findings. No reading-dependency records are encoded either. The linked works and cases are comparison routes, not prerequisites, documented lineage, or evidence that one institution inherited another's purpose.

The comparison field tests unlike claims to purpose

Long-lived religious, military, administrative, and industrial paths include the Jesus movement and earliest Christian communities, the pre-Constantinian church, the Benedictine monastery, the Roman Catholic Church, IBM under Thomas J. Watson Sr. and Jr..

Corporate, ownership, and commercial comparison routes include Amazon, Apple, Costco, FedMart, the Green Bay Packers, Hershey, Newman's Own, Novo Nordisk, Vanguard, and Zeiss.

Community, public-service, developmental, coercive, and resistance routes include the ACOFOP community forest concessions, the Africa CDC regional coordination system, the Amul dairy federation, Aravind Eye Care, the Aswan High Dam administration, Bali subak irrigation governance, Banco Palmas, the Botswana diamond developmental state, BRAC, the Brazilian Landless Workers' Movement, the Canadian potlatch ban, Charagua Iyambae, Cherán communal government, the Chile individual-account pension system, the Colombian military false-positives system, the Columbia River Inter-Tribal Fish Commission, the Cuban national literacy campaign, the Dawes Act allotment administration, ECOWAS and the Gambian transition, FaSinPat and Argentine recovered factories, the First Intifada popular committees, the Fujimori reproductive-health program, Grameen Bank and group-based microcredit, Karuk cultural fire institutions, Kōhanga Reo, Kudumbashree, the Malaysia Federal Land Development Authority, Menominee Tribal Enterprises, the MINUSTAH cholera response, Moroccan women's argan cooperatives, the Nauru Phosphate Royalties Trust, the Navajo livestock reduction, Navajo Nation peacemaking, the Nigeria Ebola emergency-operations center, the northern Nigeria polio-vaccine boycott, the Osage guardianship system, PDVSA institutional hollowing, Peru rondas campesinas, the Regional Indigenous Council of Cauca, Rwanda gacaca courts, Saemaul Undong, the Senegal River Basin Organization, SEWA cooperatives, Shining Path, the Somaliland peace conferences, the South Africa Treatment Action Campaign, the South African Truth and Reconciliation Commission, Tanzania ujamaa villagization, Te Awa Tupua governance, the Thailand universal-coverage scheme, the Tunisian National Dialogue Quartet, the U.S. federal Indian boarding-school system, the U.S. federal Indian termination policy, and UZACHI community forestry.

The Rule of St. Benedict, The Functions of the Executive, and The Practice of Management are additional conceptual routes. Adjacency does not imply common beneficiaries, moral equivalence, success, direct influence, or endorsement. The comparison question is always whose good was named, which choice changed, which effects were measured, and who could contest the answer.

Additional reciprocal institution comparisons

Newly developed institutional records add these reciprocal comparison paths:

• European Union — institution-comparison
• Haudenosaunee Confederacy — institution-comparison
• Mondragon — institution-comparison
• Mozilla — institution-comparison
• Paris Commune — institution-comparison
• REI — institution-comparison
• W3C — institution-comparison

Each path identifies a sourced case where this idea is a defining emphasis. The relation is editorial comparison, not evidence of direct influence, shared terminology, or equivalent outcomes.