In 1984, the tobacco workers’ union and the Tobacco Institute, which represents US tobacco companies, formed a labor management committee (LMC). The institute relied on LMC unions to resist smoke-free worksite rules.

In a review of the internal tobacco industry documents now publicly available, we found that the LMC succeeded for 2 primary reasons. First, the LMC furthered members’ interests, allowing them to overcome institutional barriers to policy success. Second, the LMC used an “institutions, ideas, and interests” strategy to encourage non-LMC unions to oppose smoke-free worksite rules.

While public health advocates missed an opportunity to partner with unions on the issue of smoke-free worksites during the era studied, they can use a similar strategy to form coalitions with unions.

In 1984, the Bakery, Confectionary and Tobacco Workers International Union (BC&T) and the Tobacco Institute, a trade association representing US tobacco companies, formed a labor management committee (LMC) to work on issues of joint concern. Eventually, 4 other unions contracting with US tobacco companies—the Machinists, Carpenters and Joiners, Sheet Metal Workers, and Firemen and Oilers—also joined the LMC. In 1987, Samuel Chilcote, president of the Tobacco Institute, summarized the LMC’s functions: (1) lobbying and briefing elected officials at all levels of government (primarily at the federal level), (2) discouraging liberal and labor coalitions from taking antitobacco positions, (3) building support for industry positions throughout the labor movement, and (4) communicating with the public.1 The LMC primarily worked to prevent passage of 2 types of legislation—excise taxes and workplace smoking restrictions—and remained in existence for nearly 20 years.

Political coalitions persist when they help member organizations overcome institutional challenges in political arenas.2 In the case of the tobacco industry, the LMC used its union connections with liberal policymakers to further the industry’s policy goals, including opposing smoke-free worksites. For the trade unions, the industry was a rich ally with conservative political connections. The LMC, as is true of many business–labor committees, was perceived as a mutually beneficial alliance by its members.

As of mid-2004,3 7 states (California, Delaware, Massachusetts, Connecticut, Maine, Rhode Island, and New York) and more than 74 municipalities had passed complete smoke-free worksite laws, including smoke-free bars and restaurants, and in many of these cases trade unions supported those laws.4,5 Conversely, as recently as 1995, only 8% of union locals included in a nationally representative survey sample actively opposed indoor smoking restrictions.6

But throughout the 1980s and 1990s, the LMC—the voice of 5 national trade unions—helped create labor opposition to smoke-free worksites, and this effort was particularly effective at the federal level. To date, the US Congress has passed only 1 smoke-free worksite bill, which mandates smoke-free airplanes.7 The persistence of the LMC makes its history an excellent case study in how political coalitions form and endure. Here we analyze how the tobacco industry used a potentially volatile and no-win issue such as workplace smoking to craft alliances with significant elements of organized labor.

While the LMC dealt with both worksite smoking rules and excise taxes, we focus here on smoke-free worksites to allow a more thorough exploration of coalition issues. The next 2 sections of this article lay out the theoretical framework and methods used to analyze the LMC. These sections are followed by a discussion of the LMC’s history and an analysis of how the Tobacco Institute maintained the committee for nearly 20 years. We conclude with a discussion of what public health and tobacco control advocates might learn about coalition building by studying how the Tobacco Institute, ironically, used a health issue to create a coalition that helped to block restrictions on worksite smoking.

Within public health, coalitions are considered “hot strategies” for dealing with health issues.8 This is true because, in addition to addressing an issue, coalitions build a community’s capacity to solve problems. In the case of tobacco control, for example, ASSIST, COMMIT, and SmokeLess States have required use of coalitions relying on the shared goal commitment of members.912

This model of coalition formation, however, does not necessarily describe coalitions organized to influence the political process. When organizations coalesce in an attempt to influence policy, they do so to pursue their own goals and look to form coalitions with other organizations with which they can have at least a symbiotic relationship. Thus, political coalitions do not always rely on shared goal commitment—although they may share some beliefs—as much as they rely on the existence of mutual advantage. While an initial goal brings the group together, this “goal” can be both broad and vague, representing a founding principle perhaps more than an achievable outcome. The resultant coalition then exists to help organizations overcome institutional barriers to policy success.2

By combining these 2 coalition models, we can understand that political coalitions persist as a result of the strategic use of 3 forces: institutions, interests, and ideas.13 “Institutions” are the enduring rules, procedures, and organizations under and through which the relevant policymakers and other parties interact and conduct their business.13 “Interests” refer to the constellation of economic, social, political, or other groupings of people whose preferences are likely to be taken into account by policymakers. “Ideas” include citizens’, stakeholders’, and policymakers’ understandings of a problem and the solutions to that problem.13 Understood in these terms, interests and ideas are, of course, related. Specific ideas help define, or evoke, particular understandings of a group’s interest in a particular policy. We argue that, given the length of time the LMC persisted as a coalition, the tobacco industry successfully managed the confluence of institutions, ideas, and interests in a way that allowed it to block or delay smoke-free worksite rules.

As a result of the Master Settlement Agreement, more than 40 million pages of tobacco industry internal documents produced in the course of litigation are available online through the Legacy Tobacco Documents Library at the University of California at San Francisco ( and through Tobacco Documents Online ( By following established techniques,14 we searched the Legacy library using the keywords “LMC,” “TILMC,” “labor management committee,” “Savarese” (James Savarese, LMC executive director), and “Stuntz” and “Sparber” (Susan Stuntz and Peter Sparber, key Tobacco Institute employees). In June 2004, we searched several key terms (LMC, labor management committee, Stuntz) for documents added recently.

In total, we retrieved more than 5000 documents. Four study team members (E. D. B., V. M., J. P., and a graduate student) assessed documents to determine whether they were “major,” “minor,” or “trivial” in importance given our research questions; a document’s importance could change in status in light of other documents that emerged during searches. This method has been shown to be a reliable way in which to sort documents.15

In July 2003, we searched Tobacco Documents Online to be sure we had reached saturation in terms of finding the documents available online. We searched for the term “LMC” using Tobacco Documents Online’s full text searching capacity. Finally, in 2003 and 2004, we visited the Roswell Park Cancer Institute in Buffalo, NY, where the files of the Tobacco Institute are currently housed. While not all of the institute’s files have yet arrived, we searched the files of key executives (Stuntz, Sparber, Chilcote) and boxes relating to the LMC.

At the completion of our search, we filed the documents in chronological order and constructed time lines of key events. E. D. B. and J. P. read all major documents to look for instances of lobbying strategies. All relevant quotations were extracted into preliminary findings memorandums. In spring 2004, we searched for documents related to the Business Council on Indoor Air (BCIA), using “BCIA” or “Business Council on Indoor Air,” repeating the process we use to search for documents related to the LMC. We searched nearly 1000 documents.

We verified our findings by reviewing union publications such as newsletters and newspapers. We were unable to find NEMI News (the National Energy Management Institute [NEMI] is a consortium of the sheet metal workers and contractors), but we did hand search BC&T News and AFL-CIO News for the years 1975 to 2000. Finally, we used Lexis-Nexis to search newspapers in key states in which statewide legislation was being considered to determine labor positions on tobacco issues.

Policy success requires a favorable confluence of institutions, interests, and ideas. The Tobacco Institute used the LMC to overcome 3 major institutional barriers the institute believed it faced in its fight against smoke-free worksites. First, the institute was less influential among Democrats than Republicans, the latter of whom the institute could count on to support its positions in Congress. Second, the institute could most easily control legislation at the federal level, encountering difficulty at the local level. Third, the institute could readily count on unions employed by its member companies for support, but these unions represented a small fraction of the overall unionized workforce.

To more closely align trade union interests with those of the tobacco industry, the Tobacco Institute attempted to broaden the policy focus from environmental tobacco smoke (ETS) to the larger issue of indoor air quality (IAQ) and to use employee objections to unilateral decisionmaking by management in regard to employee worksite concerns. This focus on ideas was critical to attracting trade union support beyond the 5 LMC unions, and this support created a confluence of interests that could be politically powerful. Before we explore these findings in detail, it is useful to understand the LMC’s history.

History of the LMC

Forming a partnership with trade unions was a potentially challenging endeavor for the Tobacco Institute. As James Savarese, of James Savarese and Associates (Savarese would eventually become the executive director of the LMC), explained in 1982:

The Tobacco Institute must reach out to these labor-oriented groups and engage in meaningful vote-trading. At the present time, the Tobacco Institute is viewed as unreliable among these groups and as notorious “takers” without being “givers.” … To counter this perception, it is important for the [institute] to assist some of these potential allies on votes that are of no concern to the [institute], but are of vital importance to these other groups.16

Savarese was formerly a staff member of the American Federation of State, County and Municipal Employees union. His firm worked closely with Ogilvy & Mather, a public relations firm16 often employed by the tobacco industry.

The tobacco industry and the LMC unions had economic incentives for opposing smoke-free worksites. In a February 1985 speech presented at a conference held for these unions, Ernest Pepples, senior vice president and general counsel of the tobacco company Brown & Williamson, “did the math” for the unions:

A poll of smokers who worked under relatively lenient smoking restrictions in their workplaces said that they smoke about one and one quarter fewer cigarettes each day as compared with smokers who were not faced with workplace restrictions. [These] restrictions … may have cost our industry sales of 350 million packs of cigarettes per year, or $233 million per year in revenue. In more personal terms, such restrictions translate into a loss of about 5,000 jobs in our industry, 750 of them in the manufacturing sector alone. If we look at the indirect as well as the direct contribution of the tobacco industry to employment, we are talking about a loss of approximately 23,500 jobs. … As the restrictions go from relatively lenient to a San Francisco-style measure, the costs—in revenue and jobs—can be expected to increase commensurately.17

Pepples was correct. In fact, totally smoke-free worksites have been shown to decrease smoking prevalence by 3.8% and to result in continuing smokers smoking 3.1 fewer cigarettes per day.18 And, as reported by Warner et al.,19 reductions in US smoking prevalence rates might reduce jobs in regions where cigarettes are produced. The employment picture is complex, however, in that overseas cigarette sales could have compensated for dips in domestic sales, saving US manufacturing jobs. At the same time, however, the industry was becoming more automated (requiring fewer workers) and was exporting manufacturing operations and jobs overseas.20 The industry presented the jobs issue to organized labor as being directly tied to smoking restrictions, when, in fact, several of the industry’s own policies were also driving job losses.

By April 1983, Ogilvy & Mather was planning a training session for the Tobacco Institute on working with unions. According to Patricia Milita of Ogilvy & Mather:

A portion of the program is devoted to coalition building and other activities surrounding organized labor. … The presentation should provide [the Tobacco Institute] with a greater awareness of diversity of concerns among unions. That is, an understanding that the labor movement is not a monolith but is comprised of unions and individuals representing a broad spectrum of political, economic, and social views.21

The Tobacco Institute also planned to run an advertisement in several magazines read by liberals (New Republic, Nation, and In These Times). BC&T had apparently suggested that such an advertisement could frame anti-tobacco efforts as antiworker efforts. Following this suggestion, the institute paid for development of the advertisement, which featured a photograph of 3 Philip Morris workers and the caption “We’re the tobacco industry, too.” According to Peter Sparber, the Tobacco Institute’s vice president:

A great number of our critics are liberals who see nothing beneficial about tobacco. … They are utterly convinced that tobacco kills. … As far as anti-smoker liberals are concerned, there is no reason not to attack tobacco. …As a practical matter, BC&T does not have the resources to carry out this project. Our support of their idea would be a positive gesture and could have an interesting effect on the anti-smoker movement. We do very little to deal directly with the anti-smokers. This appears to be a low risk, low expense approach with a good deal of merit.22

In June 1984, plans for the conference and the advertisement were temporarily disrupted when the Tobacco Institute’s law firm informed the institute that, under federal law, it could not pursue either activity unless it formed an LMC. LMCs were required if a company wanted to provide resources for a union under, or potentially under, contract to it. According to Sparber, “We view the committee as more of an opportunity than an obstacle … the Institute’s relations with organized labor have not been particularly good in recent years. … A committee of this sort will provide us with the flexibility we need to continue this important work.”23

The charter was finalized in September 1984,24 and in October, John DeConcini, president of BC&T, wrote to the presidents of the other 4 unions that eventually joined the LMC, inviting them to join:

Our union has never taken a position on whether or not people should smoke. Nor do we intend to. That is, and should be, an individual choice. But as trade unionists we have an interest in defending the jobs of our members from the misguided efforts of prohibitionists. In so doing on this question we have often found ourselves side by side with tobacco manufacturers, instead of across the table. And we have found that working together we can more effectively present to the public the story regarding our members’ jobs.25

According to the BC&T News,26 the LMC’s goals were to improve job security and economic development through education and research on issues affecting the tobacco industry. Although the Tobacco Institute paid all of the LMC bills and determined its budget, throughout the LMC’s life, the BC&T was represented as the organization’s leader. For example, in 1988, the BC&T News referred to the LMC as “the BC&T-led Tobacco Industry Labor-Management Committee.”27(p8)

The Tobacco Institute had to commit substantial time and money to maintaining the LMC and other contacts with trade unions; the institute also claimed to lobby on some union issues. For example, in a 1985 progress report, Sparber noted that “[a]t labor’s request, we were able to generate Congressional support for rehiring of air traffic controllers fired during the PATCO strike several years ago; we also were able to generate support for antipolygraph legislation among southern House members.”28

The partnership was apparently not nearly as important to the trade unions, at least not important enough for them to dedicate many resources to it. According to Susan Stuntz, the Tobacco Institute’s director of issues management:

The Institute and its consultants support the Committee in a variety of ways. In kind services such as drafting testimony and materials … public relations support … remember since these issues are not at the top of labor’s agenda … we often have to give them a leg up if we want something to happen.29

Throughout the 1980s and 1990s, the LMC was an active lobbying presence at the federal level and in some states. Table 1 (assembled from a variety of documents and probably representing the minimum amounts spent in the areas described) provides information on most of the LMC’s key expenditures; the Tobacco Institute was the sole source of LMC money.

In the 1998 Master Settlement Agreement, the Tobacco Institute was disbanded. The institute’s Walter Woodson urged the tobacco companies to continue LMC support, commenting that “[e]ven with all the headaches, the Committee provides a unique resource for the tobacco industry and its workers.”30 Philip Morris took the lead in organizing the committee, setting it up in Virginia.31,32 The new LMC was much smaller. In addition to Philip Morris, its members were BC&T, the machinists, Lorillard, and Brown & Williamson. In 2000, it was intended to have a budget of $1.75 million.33 This abbreviated version of the LMC, however, existed only until February 8, 2000, according to Virginia’s State Corporation Commission, and does not appear to be in existence now.

In the remainder of the article, we address the question of how this coalition remained in existence as long as it did. First, we discuss the institutional difficulties that the LMC helped the tobacco industry confront. This section is followed by a discussion of how ideas helped align the political interests of labor and tobacco.

Institutional Barriers

In May 1984, William Kloepfer Jr, director of public relations at the Tobacco Institute, wrote, “Last month Sam Chilcote told me he believed that the ETS issue, with all of its political, social and economic ramifications, posed a greater threat than any other single tobacco issue. I told him that was inarguable.”34 How did the institute believe the LMC would help it overcome institutional constraints in terms of its ability to block smoke-free worksite rules?

Working both sides of the aisle.

The LMC gave the unions and the tobacco industry opportunities to expand their spheres of influence. With unions typically aligned with Democrats and business with Republicans, the LMC was a bridging organization between the political parties. Such a form of united front among seemingly divergent groups increases the chance that Congress will follow a coalition’s lead.2,35 Before the founding of the LMC, the Tobacco Institute had calculated how many votes it could influence for labor’s benefit in both the US House and Senate. The institute thought it had the ability to “swing” 30 Senate votes and 120 House votes.36

If trade unions were able to obtain support on some of their key issues in exchange for opposing smoke-free worksites, then this was an easy trade for them. In 1989, looking back over the history of the LMC, Stuntz commented that smoke-free worksites were not on unions’ “top 100,”29 which made it easy for unions to trade their opposition to smoke-free worksites for other things they cared more about.

Working all levels of government.

Another institutional challenge facing the tobacco industry was the large number of entities with jurisdiction over ETS issues. In addition to the federal government, states, counties, cities, and school boards, among others, could implement smoke-free worksite rules. In addition, businesses and other private entities could voluntarily implement their own smoke-free worksite rules. Both of these types of rules, legislated and voluntary, were problems for the tobacco industry.36,37

While the Tobacco Institute had worked with BC&T in local political arenas, such as Maryland38 or New York,16 the institute hoped the formation of the LMC would boost this effort. The institute believed trade unions were powerful allies, because they were decentralized and had members in most political jurisdictions. As early as February 1984, Savarese told the institute that “[i]t’s fine to know a Lane Kirkland … but local issues require local union assistance. A local union official can be of great assistance if approached correctly.”39

In addition to legislation, the Tobacco Institute was concerned about voluntary restrictions. In fighting these restrictions, the institute realized that it was in a difficult position. In legislative testimony, the institute had endorsed voluntary action and opposed legislative solutions, even though it did not want voluntary smoke-free rules either. Its concern was that opposing voluntary action now “would seriously undermine more than a decade of legislative testimony and other statements endorsing the voluntary approach over legislative alternatives. … However, we must discourage workplace restrictions privately.”36 The institute decided it “should look for opportunities to help organized labor challenge smoking restrictions imposed outside of collective bargaining.”36 Thus, trade unions were a key element in fighting the implementation of voluntary worksite policies.

In summarizing the benefits the Tobacco Institute believed it had secured in working with trade unions, Stuntz claimed that the institute had “[f]oot soldiers. 14 million of them. Willing and able to carry the regressivity and indoor air quality messages to their elected officials and to their employers.”29

Working with other unions.

The institute could not plan to attract these “foot soldiers,” however, solely by recruiting trade unionists whose jobs depended on the tobacco industry. The LMC unions thus represented a bridge into the larger trade union arena. In 1984, the institute noted that “the national scope of the unions offer the possibility of interventions ‘in solidarity’ by brother and sister unionists in the far corners of the U.S.”40 And, as Stuntz commented 5 years later, “It’s this solidarity … that makes our labor allies unique among the Institute’s coalitions.”29

In addition, a variety of union groups, including the Council of Labor Union Women and the A. Philip Randolph Institute, representing the interests of Black trade unionists, were able to obtain funds or other kinds of support from the Tobacco Institute in exchange for supporting the institute’s worksite smoking position. These 2 unions were seen as “coalitions themselves and [as providing] unique access to virtually every segment of the labor movement. As such, they are important vehicles to carry the IAQ and excise tax messages to organized labor.”41

The fact that smoking issues would not “make the labor movement’s list of the top 100 issues,”29 however, did present the Tobacco Institute with some challenges, including making sure that trade union support of the institute’s positions was real. In a 1988 memorandum, Stuntz, seeking clarification from the institute’s public relations firms, asked “How will you work to identify and provide additional labor representatives for testimony in opposition to smoking restrictions? Will these be LMC consultants, or real labor people?”42

Because the institute had 3 strong reasons for wanting trade unions as allies, it was willing to devote substantial resources to its labor strategy. However, to attract unions other than those whose members were employed by the tobacco industry, the institute needed to move beyond shared interests to an “ideas” strategy.

Framing Ideas to Align Interests

To attract non-LMC unions, the Tobacco Institute needed to create a framing of ideas that would have broader appeal to trade unionists. By creating connections and images, frames inform people as to how they should interpret a message, including what counts as a message and what can be ignored.43 In policy debates, the way in which people think about an issue can influence policy outcomes, if for no other reason than that the frame creates an interest for people in the legislation.44,45 For example, both the industry and tobacco control advocates have worked to frame the smoking issue as a “right”: the right to smoke versus the right to breathe clean air. The language of rights is attractive because laws cannot eliminate a right, which is fundamental. Thus, in the case of both sides, successful invocation of the “rights” frame increases their chances of policy success.

The LMC unions had vested interests in the tobacco industry’s fiscal health and in improvements in ventilation as a substitute for smoke-free worksite rules. But the possibility of shaping, or framing, the issue in such a way that other, less self-interested unions would be supportive relied on the LMC’s ability to make opposition to smoke-free worksites a credible union issue. To attract union opposition to smoke-free worksite rules, the institute, according to Stuntz, “had to redefine the issue. In this case … we emphasized first collective bargaining and … later … indoor air quality in the workplace.”29

Unilateralism: Imposing a change in work rules.

The industry encouraged opposition to smoke-free worksite rules in cases in which management implemented these rules unilaterally. As Sparber noted, “organized labor already sees these sorts of issues as matters for collective bargaining. They can and will be of help in blocking restrictions.”37 Trade unions did not necessarily oppose smoke-free rules but wanted to be involved in decisions regarding such rules.46

The unilateralism framing, however, was not the industry’s strongest position, because it still did not block smoke-free worksites. Even the BC&T, by 1988, was willing to support the creation of designated smoking or nonsmoking areas in cases “where the employers are responding to the legitimate concerns of non-smoking employees.”47 The Tobacco Institute thus made a conscious choice to begin pushing IAQ.

Indoor air quality.

By using the IAQ language, the Tobacco Institute attempted to ensure that smoke became just 1 of a variety of indoor contaminants to be controlled, one that need not be privileged over others. In addition, the institute could then dispute the scientific basis of smoke-free worksites.48 Although the institute could thwart federal law using raw political power, political power coupled with expert analysis would place it in an even stronger position.49

There are 2 ways to improve IAQ: control pollution sources and improve ventilation. Ventilation can improve overall air quality by helping to remove pollutants, and source controls stop the production of recognized toxins, thus limiting the need for ventilation to remove them. Trade unions wanted regulations governing overall air quality, including improved ventilation systems and source control, while the institute wanted a legislative focus on overall IAQ but one that avoided controlling source pollutants such as smoke. The broader IAQ framing allowed the institute and unions to find common ground.

The institute needed trade unions to carry the broader IAQ message. In 1987, in a speech reviewing tobacco industry strategies, Chilcote outlined the 4-part strategy that the institute had been pursuing: (1) “continue positioning ETS as a small part of the overall indoor air quality issue,” (2) “encourage third parties to seek indoor air quality and ventilation standards rather than smoking restrictions,” (3) “continue efforts to encourage scientific scrutiny of anti-smokers’ claims about ETS,” and (4) “attempt to reverse the most visible and damaging restrictions by demonstrating the failure of smoking restrictions to improve indoor air quality; and by showing the economic damage created by anti-smoker extremism.” Under the second point, he emphasized that “organized labor must be politically influential.”50

However, the LMC did not have to manufacture resentment against smoke-free air laws or policies as the only pollutant employers were willing to fix. The AFL-CIO, at the urging of BC&T, passed a resolution against smoke-free worksites in 1980, 4 years before the LMC was formed. According to the resolution:

Some employers have … concluded that it would be unnecessary to control exposure of [sic] these [other toxic] substances, if workers stopped smoking. … We … oppose misuse of scientific data concerning smoking and exposure to toxic substances to serve as a rationale for failure to take necessary steps to prevent worker exposure to toxic substances in the workplace, which are shown to adversely affect their health.51

In 1986, after publication of the surgeon general’s report on ETS,52 the AFL-CIO acted again. Their 1986 statement expressed concern that employers would use the report to “shirk their responsibility to clean up the workplace,” opting to blame smokers for occupational disease, a claim well founded given a grim history of blaming black lung disease among miners on smoking rather than coal dust, for example.53 While this statement included the recognition that smoking was a major public health problem of concern to smokers and nonsmokers alike, it urged that smoking needed to be considered as part of the larger question of IAQ.54 The statement issued by the LMC unions, written by Savarese,55,56 was more direct:

[T]he report’s conclusion that no engineering controls can effectively deal with environmental tobacco smoke is one which concerns us greatly as trade unionists. … This conclusion violates a fundamental tenet of federal work-place health and safety regulations which the labor movement has supported for decades. Under that regulatory principle employers have always had the responsibility for cleaning up the workplace.57

Occupational Safety and Health Administration regulations typically call for reducing rather than eliminating workplace carcinogens such as asbestos and vinyl chloride. From a union’s perspective, therefore, the notion that ETS should be eliminated rather than reduced to acceptable levels did not seem to be in accord with other regulatory battles involving reducing rather than eliminating hazards altogether.

At the end of 1987, Stuntz was able to claim success, commenting that “[t]his year we were successful in broadening the issue to overall indoor air quality before certain audiences, most notably labor.”58 By this time, the Tobacco Institute had worked with unions to help defeat a 1985 federal law requiring smoke-free federal worksites, and, as of 1989, it had placed labor consultants in 2 regions and 11 states.59

To ensure that labor had “science” on its side, the institute hired Air Conditioning and Ventilation Analysis (ACVA)/Healthy Buildings International to prove that indoor smoking was not a problem. Unions retained ACVA to study building air quality, but ACVA would then invoice the institute for the cost of the study.60,61 Chilcote noted that “[l]abor unions are making heavy use of our ventilation consultant, Gray Robertson.”1

In addition to ACVA, the institute, through the LMC, hired the National Energy Management Institute to help promote ventilation as a solution to IAQ problems.1 The National Energy Management Institute was formed by the sheet metal workers and sheet metal contractors to ensure that sheet metal work continued to be available. Between January 1, 1988, and May 12, 1989, LMC representatives made 47 IAQ presentations in 18 states and Washington, DC, to labor audiences.62

In 1989, however, the Tobacco Institute was confronted with a problem. Senator George Mitchell and Representative Joseph Kennedy introduced bills giving authority to the Environmental Protection Agency (EPA) to regulate indoor air and to research specific indoor pollutants. John Lyons, of the Tobacco Institute, noted that “EPA has proved to be an aggressive and hostile agency.”63 However, he also noted that

[w]e can not [sic] publicly oppose the bills. For the past five years we have been urging legislators to address indoor air quality problems rather than pass smoking restrictions. It would be extremely hypocritical for the industry to oppose legislation that, at least superficially, responds to our indoor air quality arguments.63

The LMC took the lead in pushing changes in the bill, most notably trying to shift jurisdiction from EPA to the Occupational Safety and Health Administration in the belief that labor issues should be dealt with by the Department of Labor.

Business Council on Indoor Air.

Union support was not the only type of support the institute sought; the institute also was working to obtain management support. In pursuing this dual strategy, the institute was helping to ensure its political success in smoke-free worksite issues but was not necessarily working to secure healthier environments for workers, the stance it had taken with trade unions. In 1987, Chilcote realized that “[b]usiness generally resists tougher ventilation standards but tends to prefer ventilation as an alternative to the control of the many substances found in the workplace.”50 Businesses thus emerged as a possible ally, as long as these businesses were ones that preferred a ventilation solution.

Because there was no business coalition to lobby this point of view, the institute helped to form one by becoming a major underwriter of the BCIA. The institute selected Paul Cammer, president of Cammer and Associates, to manage the BCIA. Cammer estimated the start-up costs at no more than $40 000 and anticipated that member dues would eventually support the BCIA.64

Stuntz knew that Tobacco Institute support would continue past the start-up phase, however, commenting that “[w]e expect … The Institute will underwrite dues for several charter members … we have agreed to guarantee him [Cammer] the full amount in the event the coalition fails for lack of support. … Cammer, coordinating closely with Peter Sparber, who reports progress to me regularly, will proceed throughout the summer to recruit members.”64 Between 1989 and 1995, the institute spent more than $1 million supporting the BCIA, including paying ACVA’s dues (Table 1).

However, the institute’s support of the BCIA had to be hidden, because other businesses were reluctant to ally themselves with the institute. According to Lyons,

We are not members of BCIA—the chemical companies said they would not participate if tobacco was involved—but we are able to wield substantial influence over the direction of the group. We pay several of our indoor air consultants to be members of the organization seats on the board of directors [sic]. In addition, we make public relations and legal and legislative counsel available to the BCIA.63

As Lyons put it, “our views will be represented but without our fingerprints.”63 Fearing more negative publicity than they had earned on their own, the chemical companies kept their distance from the tobacco companies.65

The BCIA was prepared to resist legislative and regulatory efforts to clean up indoor pollution sources and to resist tough ventilation regulations, positions not necessarily sympathetic to those of unions. For example, in 1991, “BCIA’s official position on indoor air quality legislation will be that such legislation is not necessary, however if it must be then it should focus on research.”66 This general position placed the BCIA in the tobacco industry’s corner, opposing smoke-free worksites, the only issue the industry cared about.

The persistence of the LMC demonstrates the Tobacco Institute’s ability to manage the confluence of ideas, interests, and institutions that lead to policy success. The institute convinced major elements of organized labor that it had political influence where organized labor did not; it was able to partner with unions for which smoking was a jobs issue to reach those for which it was not; it spent liberally to build interest in institute issues among some unions; and it understood how to use ideas to frame issues to secure the support of large segments of organized labor. A key marker of its success at the federal level is that the only worksites Congress has ever required to be smoke free are airplanes, when flight attendant unions forced the AFL-CIO not to oppose the legislation.7

Public health advocates have interpreted organized labor positions opposing smoke-free worksites as reflecting tobacco industry influence within unions.67 While this influence was certainly present through the LMC, especially its member unions, organized labor is not a monolith. To attract support for smoke-free worksites, public health advocates also can use the “institutions, ideas, and interests” framework in an attempt to understand how to form their own coalitions with organized labor.

The industry was able to “sell” its framing of the worksite smoking issue to most segments of organized labor because this framing resonated with labor. Successful efforts to frame an issue for any audience must grow from an element of truth. Unions were legitimately concerned about larger air quality issues and were opposed to issue framings that singled out a worker behavior (smoking) as the sole cause of poor IAQ, thus relieving management of responsibility.

But the irony for those working in public health is that while they really did care about worker health and IAQ, the Tobacco Institute did not. According to a 1986 strategy document, the institute’s goal was to “[p]osition the industry as concerned about the broad issue of indoor air quality.” Note that although the institute did not have as a goal actually to do anything about IAQ, the institute did take time to understand the larger labor concern and leverage this concern to the institute’s advantage. The industry’s simultaneous pursuit of the business community through the BCIA demonstrates how it used a different, and sometimes contradictory, framing technique to attract business support of the tobacco industry’s agenda. In terms of political strategy, the LMC and the BCIA together served the tobacco industry’s interests effectively, allowing it to hide the extent of its involvement in policymaking. However, as the science surrounding the negative effects of passive smoke became increasingly strong, an increasing number of worksites went smoke free, often with the support of organized labor.

Tobacco control advocates, by isolating tobacco smoke as an issue, may have missed the opportunity to partner with organized labor on air quality issues. Interestingly, the institute knew that health groups could also take advantage of the IAQ issue, and it consciously attempted to prevent a public health and labor alliance from existing. In 1989, according to Stuntz,

We are in the process of building a fire wall around organized labor, a wall that the anti-smokers have not succeeded in penetrating. They have tried to do so. Remember, there are a number of anti-smokers in the labor movement, and some have succeeded in blocking labor activity on some of our issues. The AFL-CIO, for example, took no position on the airlines smoking issue, since the pilots oppose a ban while the flight attendants supported it.29

The institute’s firewall consisted of large amounts of money (Table 1), an institutional mechanism for influencing policy and organized labor (the LMC), and an understanding of how to frame issues in a way that would resonate with organized labor. The institute had lost 1 federal battle, the issue involving smoke-free airlines, because it lacked the support of organized labor. It spent millions each year to avoid other, similar losses.

By 1992, the year of the EPA report, the institute was reflecting on the difficulties of maintaining partnerships with labor, particularly the non-LMC unions. As noted in a document titled “Coalitions,”

Allies’ greatest strength—independence—can limit the effectiveness of many of these coalitions on our issues. Allies may not agree or even have an interest in all industry issues, and may not be willing or able to assist in all way that might be requested. … Release of the EPA’s environmental tobacco smoke risk assessment and workplace smoking guidelines, together with the Occupational Safety and Health Administration’s consideration of a rule-making on environmental tobacco smoke as a possible workplace health hazard, will put strain on existing alliances with organized labor and make coalition efforts on workplace issues more difficult.68

The difficulty was clear within 10 years. In October 1988, Savarese’s summary of Massachusetts was that it had “the most sophisticated labor coalition program of any state already in place. It has all the necessary ingredients for success—a local labor consultant, a committed Institute regional vice-president, and a receptive State Federation. We will continue to work through these established channels on any labor IAQ projects.”41

By 2003, however, several state-level federations of the AFL-CIO, including those in Massachusetts, were supporting local and statewide smoke-free worksite laws (see letters endorsing smoke-free worksites at This shift came as a result of outreach efforts by several tobacco control organizations to labor unions, highlighting a shared concern about ETS as a threat to worker safety and health (see

Thus, public health can work, and has worked, with organized labor. But doing so requires that public health advocates craft their own “institutions, ideas, and interests” strategy. As the Tobacco Institute demonstrated, coalitions can form around health issues. And public health advocates have the advantage of a professional commitment to worker health. While public health advocates lack the money available to the Tobacco Institute, they can take time to learn about the structure and concerns of organized labor and use this knowledge to forge political coalitions around health issues.

TABLE 1— Labor Management Committee Budget ($): Groups and Individuals Supported by the Tobacco Institute in Work Involving Labor-Related Issues
TABLE 1— Labor Management Committee Budget ($): Groups and Individuals Supported by the Tobacco Institute in Work Involving Labor-Related Issues
     Membership Groups    
YearACVA/HBIBCIANEMIJames SavareseCLUWLCLAAAPRILabor ConsultantsOgilvy Adams and RhinehartTaxesJohn Jarvis
19846410 000400 00010 000
19856530 000112 00085 00050 000
198666120 000139 000
198767604 000180 000200 000234 000200 00040 00075 000
198868585 000200 000385 00010 00024 00010 000290 000645 000334 000110 000
198955200 000470 000355 00010 00012 00010 000401 000650 000435 000125 000
199069200 000650 000340 00075 00010 00025 0001 015 000675 000993 000160 000
199170377 000251 000636 000437 00020 0005 00020 0001 010 000775 0001 335 000140 000
199271450 000 B280 000780 000445 00020 00030 00020 000897 000685 000978 000120 000
199372,73760 000 B250 000830 00035 00025 000823 000720 000932 000156 000
199474,75130 000595 096368 62316 50016 00015 000370 000332 000151 806
199573,76150 000 B50 000 B850 000300 00015 00015 00015 000558 000287 000150 000
199677,78801 000560 000300 00015 00015 00015 000635 000302 000
199779400 000 B560 000300 00015 00015 00015 000696 500599 000
199880420 000300 000592 000131 000150 000
199981100 000180 000353 000125 000150 000
     Total4 287 0001 201 0006 831 0964 478 623221 500177 000170 0008 089 5004 750 0007 022 0001 487 806

Note. ACVA/HBI = Air Conditioning and Ventilation Analysis/Healthy Buildings International; BCIA = Business Council on Indoor Air; NEMI = National Energy Management Institute; CLUW = Coalition of Labor Union Women; LCLAA = Labor Council for Latin American Advancement; APRI = A. Philip Randolph Institute. All amounts reflect estimated expenses taken from the previous year’s budget unless indicated by “B” for budgeted expense. ACVA/HBI column refers to indoor air quality program: building studies, testimonies, scientific witness, and media tours. In regard to the BCIA, the Tobacco Institute was represented by various consultants (e.g., Gray Robertson of ACVA/HBI). The NEMI received grants from the institute to conduct building air quality studies and provide ventilation advice. James Savarese was the executive director of the committee. Pre-1993, estimates were added for work as LMC labor counsel and economic consultant. Labor consultants worked in various regions of the country, making contact with local labor leaders (includes General Counsel Michael Forscey). Ogilvy Adams and Rhinehart provided public relations and media support for committee activities. National fair tax organizations such as Citizens for Tax Justice worked with the Tobacco Institute in conducting economic studies and argued against regressive taxes through its state affiliates. John Jarvis is the LMC federal lobbyist with contacts in the AFL-CIO.

This work was supported by National Cancer Institute (grant 1RO1 CA095964).


1. Chilcote S. Memorandum, May 29 1987. Bates Number: TI DN 0002799–2803. Available at: Accessed June 29, 2004. Google Scholar
2. Hula KW. Lobbying Together: Interest Group Coalitions in Legislative Politics. Washington, DC: Georgetown University Press; 1999. Google Scholar
3. American Nonsmokers’ Rights Foundation. Overview List—How Many Local Smokefree Laws? Berkeley, Calif: American Nonsmokers’ Rights Foundation; 2004. Google Scholar
4. Massachusetts AFL-CIO Praises Statewide Smoke-Free Worksites Law. Boston, Mass: Massachusetts AFL-CIO; 2004. Google Scholar
5. Shufro J. Memorandum on New York smokefree worksites. Available at: Accessed June 29, 2004. Google Scholar
6. Sorensen G, Stoddard A, Youngstrom R, et al. Local labor unions’ positions on worksite tobacco control. Am J Public Health.2000;90:18–20. LinkGoogle Scholar
7. Pan J, Barbeau E, Levenstein C, Balbach E. The airplane smoking ban and role of organized labor: a case study. Am J Public Health. In press. Google Scholar
8. Wandersman A, Goodman R, Butterfoss F. Understanding coalitions and how they operate. In: Minkler M, ed. Community Organizing and Community Building for Health. New Brunswick, NJ: Rutgers University Press; 1997:261–277. Google Scholar
9. Stillman FA. Evaluation of the American Stop Smoking Intervention Study (ASSIST): a report of outcomes. J Natl Cancer Inst.2003;95:1681–1691. Crossref, MedlineGoogle Scholar
10. White J, Bero LA. Public health under attack: the American Stop Smoking Intervention Study (ASSIST) and the tobacco industry. Am J Public Health.2004;94:240–250. LinkGoogle Scholar
11. COMMIT Research Group. Community Intervention Trial for Smoking Cessation (COMMIT): I. Cohort results from a four-year community intervention. Am J Public Health.1995;85:183–192. LinkGoogle Scholar
12. Goldsmith MF. AMA office directs SmokeLess States. JAMA.1994;271:643. Crossref, MedlineGoogle Scholar
13. Heclo H. Ideas, interests, and institutions. In: Dodd LC, Jillson C, eds. The Dynamics of American Politics. Boulder, Colo: Westview Press; 1994:366–395. Google Scholar
14. Malone R, Balbach ED. Tobacco industry documents: treasure trove or quagmire? Tob Control.2000; 9:334–338. Crossref, MedlineGoogle Scholar
15. Balbach ED, Gasior R, Barbeau EM. Tobacco industry documents: comparing the Minnesota depository and Internet access. Tob Control.2002;11:68–72. Crossref, MedlineGoogle Scholar
16. Savarese J. Memorandum to Michael J. Kerrigan, November 18, 1982. Bates No. TI04940503–0507. Available at: Accessed February 23, 2004. Google Scholar
17. Pepples E. Speech to legislative conference. Bates No. 521040556–0563. Available at: Accessed July 23, 2003. Google Scholar
18. Fichtenberg CM, Glantz SA. Effect of smoke-free workplaces on smoking behavior: systematic review. BMJ.2002;325:188–194. Crossref, MedlineGoogle Scholar
19. Warner KE, Fulton G, Grimes N. Employment implications of declining tobacco product sales for the regional economies of the United States. JAMA.1996; 275:1241–1246. Crossref, MedlineGoogle Scholar
20. Barbeau E, Levenstein C. Feeling the heat: displaced tobacco workers in North Carolina. New Solutions J Occup Environ Policy.1999;9:65–80. Crossref, MedlineGoogle Scholar
21. Milita P. Memorandum, April 20, 1983. Bates No. TIOK 0028737–8742. Available at: Accessed March 15, 2003. Google Scholar
22. Sparber PG. Memorandum, April 23, 1984. Bates No. 503866571–6573. Available at: Accessed June 6, 2003. Google Scholar
23. Sparber PG. Labor relations. Bates No. TIOK0022813–2814. Available at: Accessed August 3, 2002. Google Scholar
24. Stuntz SM. Letter to John Rupp. Bates No. TIOP0022699. Available at: Accessed June 19, 2003. Google Scholar
25. DeConcini J. Letter to Carlough. Bates No. 88115775–5776. Available at: Accessed October 10, 2003. Google Scholar
26. Tobacco committee formed by BCT, T.I. BC&T News. November/December 1984:8. Google Scholar
27. AFL-CIO passes resolution concerning clean indoor air. BC&T News. June 1988:1. Google Scholar
28. Sparber PG. Public smoking. Bates No. TIMN285670–5679. Available at: Accessed July 16, 2003. Google Scholar
29. Stuntz SM. Remarks to the Executive Committee. Bates No. TI11871026–1076. Available at: Accessed March 25, 2004. Google Scholar
30. Woodson W. Memorandum to Dawson, Milstein, and Scruggs. Bates No. 20648200318. Available at: Accessed July 7, 2004. Google Scholar
31. Liebengood H, Scruggs J, Scully T. Memorandum to Steve Parrish, January 5 1999. Bates No. 2077130682–0683. Available at: Accessed July 22, 2003. Google Scholar
32. Peck JH. Virginia Co. tobacco industry labor/management committee. Bates No. 106012911. Available at: Accessed February 8, 2002. Google Scholar
33. Fry A, Allen F. TILMC 2000 dues revised. Bates No. 2070750614. Available at: Accessed July 13, 2003. Google Scholar
34. Kloepfer WJ. Memorandum, May 24, 1984. Bates No. TI1041–1104. Available at: Accessed June 29, 2004. Google Scholar
35. Birnbaum J. Capitol Hill listens to coalitions. Washington Post. May 3, 2004:E1. Google Scholar
36. Tobacco Institute. Workplace Smoking Restrictions Program. Bates No. 521046375–6537. Available at: Accessed June 25, 2004. Google Scholar
37. Sparber PG. Presentation, March 30, 1984. New York State Archives. Bates No. TI0487–1091/1105. Google Scholar
38. Model ‘no smoking’ law stopped with BC&T help. BC&T News. October 1982:7. Google Scholar
39. Woodson W. Labor briefing at Ogilvy & Mather. Bates No. TCAL0469326–9328. Available at: Accessed April 4, 2001. Google Scholar
40. Proposal for a legislative conference for U.S. unions in the tobacco industry. Bates No. 8815805–5809. Available at: Accessed July 13, 2003. Google Scholar
41. Savarese J. Memorandum to Jeff Ross and Walter Woodson. Bates No. TIDN0010058–0061. Available at: Accessed July 13, 2003. Google Scholar
42. Stuntz SM. 1988 Public Smoking Program. Bates No. TIDN0019193–9195. Available at: Accessed July 13, 2003. Google Scholar
43. Giliam FD, Bales SN. Strategic Frame Analysis: Re-framing America’s Youth. Los Angeles, Calif: University of California at Los Angeles; 2000. Google Scholar
44. Lakoff G. Moral Politics: How Liberals and Conservatives Think. 2nd ed. Chicago, Ill: University of Chicago Press; 2002. Google Scholar
45. Iyengar S. Is Anyone Responsible? How Television Frames Political Issues. Chicago, Ill: University of Chicago Press; 1991. Google Scholar
46. Sorensen G, Youngstrom R, Maclachan C, Emmons K, Levenstein C. Labor’s position on worksite tobacco control policies: a review of arbitration cases. J Public Health Policy.1997;18:433–452. Crossref, MedlineGoogle Scholar
47. Unions challenge employer’s ability to exert social control over workers. BC&T News. May 1988:2. Google Scholar
48. Drope J, Bialous SA, Glantz SA. Tobacco industry efforts to present ventilation as an alternative to smoke-free environments in North America. Tob Control.2004;13(suppl 11):141–147. CrossrefGoogle Scholar
49. Jenkins-Smith HC, Sabatier PA. The dynamics of policy-orientated learning. In: Sabatier PA, Jenkins-Smith HC, eds. Policy Change and Learning: An Advocacy Coalition Approach. Boulder, Colo: Westview Press; 1993:41–56. Google Scholar
50. Chilcote SD. Memorandum to Executive Committee. Bates No. 86023282–3287. Available at: Accessed August 2, 2002. Google Scholar
51. AFL-CIO policy rejects anti-smoking campaigns. BC&T News. June 1980:1. Google Scholar
52. Health Consequences of Involuntary Smoking. Washington, DC: Office of the Surgeon General; 1986. Google Scholar
53. Smith B. Black lung: the social production of disease. Int J Health Serv.1981;11:343–359. Crossref, MedlineGoogle Scholar
54. AFL-CIO. Smoking and the workplace. Bates No. THK0022692–2693. Available at: Accessed July 7, 2003. Google Scholar
55. Savarese J. Union statement in response to the 1986 surgeon general’s report. Bates No. 2066201880–1881. Available at: Accessed July 13, 2003. Google Scholar
56. Savarese J. Letter to Durst. Bates No. 2066201879. Available at: Accessed July 13, 2003. Google Scholar
57. Surgeon general report attacked by unions. BC&T News. January 1987:1. Google Scholar
58. Stuntz SM. Annual meeting remarks. Bates No. TIDN0009909–9915. Available at: Accessed July 11, 2003. Google Scholar
59. Tobacco Institute. 1990 proposed budget index summary. Bates No. TI38461319–1512. Available at: Accessed February 27, 2004. Google Scholar
60. Robertson G. ACVA quotation. Bates No. TI10160516–0521. Available at: Accessed July 13, 2003. Google Scholar
61. Robertson G. AFGE local 1812. Bates No. TI10160515. Available at: Accessed July 13, 2003. Google Scholar
62. Tobacco Institute. TI scientific consultant activity. Bates No. 92760977–0989. Available at: Accessed December 27, 2002. Google Scholar
63. Lyons J. Mitchell Bill and TI’s allies. Bates No. TI01280616–0618. Available at: Accessed December 11, 2003. Google Scholar
64. Stuntz SM. EPA coalition. Bates No. TI10570304–0305. Available at: Accessed June 23, 2004. Google Scholar
65. Whelan E. A Smoking Gun: How the Tobacco Industry Gets Away With Murder. Philadelphia, Pa: George F. Stickley; 1984. Google Scholar
66. Hamilton J. BCIA board meeting of January 16. Bates No. TIOK0008917–8919. Available at: Accessed April 22, 2004. Google Scholar
67. National Association for Public Health Policy. Finding common ground: how public health can work with organized labor to protect workers from environmental tobacco smoke. J Public Health Policy.1997;18:453–464. Crossref, MedlineGoogle Scholar
68. Tobacco Institute. Proposed budget—1991. Bates No. TIMN0390579–0803. Available at: Accessed September 30, 2002. Google Scholar
69. Tobacco Institute. Tobacco Institute 1985 budget summary. Bates No. 505168230–8250. Available at: Accessed August 6, 2004. Google Scholar
70. Tobacco Institute. Public relations divisions 1986 proposed budget. Bates No. 506644155–4259. Available at: Accessed August 5, 2004. Google Scholar
71. Heger C. Tobacco Institute proposed budget. Bates No. 640518376–8463. Available at: Accessed June 29, 2004. Google Scholar
72. Tobacco Institute. Public affairs division, proposed 1988 operations plan and budget. Bates No. 506644419–4579. Available at: Accessed February 18, 2004. Google Scholar
73. Tobacco Institute. Tobacco Institute public affairs division proposed budget and operating plan 1989. Bates No. TIMN0366674–6895. Available at: Accessed June 29, 2004. Google Scholar
74. Tobacco Institute. Tobacco Institute public affairs division: proposed budget—1991. Bates No. TIMN0377297–7378. Available at: Accessed April 13, 2004. Google Scholar
75. Tobacco Institute. Tobacco Institute 1992 budget, public affairs division. Bates No. I16681420–1499. Available at: Accessed April 7, 2004. Google Scholar
76. Tobacco Institute. Tobacco Institute 1993 budget, public affairs division. Bates No. TI16681190–1262. Available at: Accessed April 13, 2004. Google Scholar
77. Tobacco Institute. Tobacco industry labor management committee. Bates No. TI30849053–9055. Available at: Accessed April 9, 2004. Google Scholar
78. Tobacco Institute. Tobacco Institute labor management committee. Bates No. TI14929409. Available at: Accessed April 9, 2004. Google Scholar
79. Scruggs J. Morris P. Final industry labor management committee budget. Bates No. 2065484578–4579. Available at: Accessed April 9, 2004. Google Scholar
80. Tobacco Institute. 1999 TILMC budget. Bates No. TI14929409. Available at: Accessed April 9, 2004. Google Scholar
81. Tobacco Institute. Final TILMC 1999 budget. Bates No. 2065484579. Available at: Accessed April 9, 2004. Google Scholar


No related items




Edith D. Balbach, PhD, Elizabeth M. Barbeau, MPH, ScD, Viola Manteufel, BA, and Jocelyn Pan, DrPHEdith D. Balbach and Jocelyn Pan are with the Community Health Program, Tufts University, Medford, Mass. Elizabeth M. Barbeau is with the Dana-Farber Cancer Institute and the Harvard School of Public Health, Boston, Mass. Viola Manteufel is with the Community Health Program, Medical College of Wisconsin, Milwaukee. “Political Coalitions for Mutual Advantage: The Case of the Tobacco Institute’s Labor Management Committee”, American Journal of Public Health 95, no. 6 (June 1, 2005): pp. 985-993.

PMID: 15914820