The atlas network: a "strategic ally" of the tobacco industry.

Amid growing academic and policy interest in the influence of think tanks in public policy processes, this article demonstrates the extent of tobacco industry partnerships with think tanks in the USA, and analyzes how collaborating with a network of think tanks facilitated tobacco industry influence in public health policy. Through analysis of documents from tobacco companies and think tanks, we demonstrate that the Atlas Economic Research Foundation, a network of 449 free market think tanks, acted as a strategic ally to the tobacco industry throughout the 1990s. Atlas headquarters, while receiving donations from the industry, also channeled funding from tobacco corporations to think tank actors to produce publications supportive of industry positions. Thirty-seven per cent of Atlas partner think tanks in the USA received funding from the tobacco industry; the majority of which were also listed as collaborators on public relations strategies or as allies in countering tobacco control efforts. By funding multiple think tanks, within a shared network, the industry was able to generate a conversation among independent policy experts, which reflected its position in tobacco control debates. This demonstrates a coherent strategy by the tobacco industry to work with Atlas to influence public health policies from multiple directions. There is a need for critical analysis of the influence of think tanks in tobacco control and other health policy sectors, as well as greater transparency of their funding and other links to vested interests.

Introduction

The term think tank refers to “organizations that produce research products with the aim of informing policy debates” (Kimenyi and Datta, 2011). There has been much dispute over this definition because of the variety of organizations that adopt the think tank label (Stone, 2007). However, as Thunert (2003) writes, “the desire to influence public policy—either directly through consultation or indirectly through the shaping of the climate of opinion—is common to all the think tanks of the world.” The role of think tanks in public policy processes has attracted increasing attention, particularly in relation to their lack of transparency, connections to corporate and political entities, and considerable influence. Books and documentaries, such as Merchants of Doubt (Oreskes and Conway, 2011), have illustrated how corporate interests use think thanks to shape public opinion and influence policy processes. Investigative journalism has exposed instances of foreign governments funding American think tanks to influence trade and related policies (Lipton et al., 2014). Such exposes have led to calls to “think harder” about the role of think tanks in policy processes (Lozovsky, 2015).

There is little research on the influence of think tanks within health policy processes. Shaw et al. (2014, 448) write, “think tanks remain poorly conceptualized, under‐researched and inadequately understood in the field of health policy and planning.” Smith et al. (2013) found only 14 articles on think tanks and health policy in the UK. Bennett et al. (2012, 195) find that “There has been virtually nothing previously written about health policy analysis institutes, or indeed any form of specialist think tank.”

One of the few areas of health research that has begun to document the role of think tanks is tobacco control. Concerns about the use of think tanks by the tobacco industry, as third parties to undermine tobacco control policies, have been raised (Non‐Smokers' Rights Association, 2008). Both the Institute for Economic Affairs and Adam Smith Institute admit to receiving donations from tobacco corporations while producing policy recommendations regarding plain packaging (Doward, 2013). Muggli et al. (2004) document how Philip Morris (PM) collaborated with the Heritage Foundation to recruit journalists to write articles that questioned the Environmental Protection Agency (EPA) report on secondhand smoke. Tesler and Malone (2010) argue that PM and RJ Reynolds (RJR) worked with the Alexis de Tocqueville Institute and Manhattan Institute to defeat the Clinton Health Plan in the USA. There is evidence that British American Tobacco (BAT) used publications by the Institute for Economic Affairs to influence states' perceptions of trademark laws and international treaties to limit cigarette health warning labels (Crosbie and Glantz, 2014). Smith et al. (2009) describe how BAT liaised with the European Policy Centre to support arguments that consultation with corporations was “good governance,” and commissioned the Adam Smith Institute to brand reports by the industry as its own publications (Smith et al., 2013). However, research to date has been limited to single case studies, does not consider partnerships with think tanks as an ongoing strategy of the tobacco industry, or consider how funding multiple think tanks generates particular advantages for the industry.

These are notable gaps given that the majority of think tanks linked to the tobacco industry in the existing literature share a common feature—they are all members of the Atlas Network. Founded in 1981, as the Atlas Economic Research Foundation (hereafter Atlas) by British businessman and philanthropist Sir Antony Fisher, Atlas aims “to strengthen the worldwide freedom movement by cultivating a highly effective and expansive network that inspires and incentivizes all committed individuals and organizations to achieve lasting impact” (2014). It acts both as a think tank itself, by producing research and policy commentaries, and as a link between its 449 partner organizations worldwide, making it one of the largest formal networks of policy institutes. Partner organizations include well‐established entities, such as the Cato Institute and Institute for Economic Affairs, as well as newer think tanks around the globe, such as Imani Ghana—an influential think tank in West Africa. All of the partner organizations (the majority of which are think tanks) share a free market, libertarian ideology. Atlas headquarters, based in Washington D.C., provides partners with grants and other resources, as well as networking and training opportunities. In 2014, over 1000 individuals (mostly think tank employees or founders) received training from Atlas. The network awarded US$4 340 000 in grants to 177 partners in 68 countries. In 2014, Atlas' budget was US$9 586 213, having risen each year over the last decade (Atlas 2014). Atlas is governed by a board of 11 directors and is a registered educational charity in the USA.

This article documents participation by Atlas and its affiliated think tanks in policy debates related to tobacco control in the 1990s in order to advance health policy and tobacco control research beyond a limited focus on individual think tanks; to demonstrate the extent of tobacco industry partnerships with think tanks in the USA; and to analyze how collaborating with a network of think tanks facilitated tobacco industry influence in public health policy. Through analysis of documents from tobacco companies and think tanks, we demonstrate that Atlas, and many of its partner think tanks in the USA, acted as a strategic ally to the tobacco industry. We also show how the industry used the Atlas think tank network to covertly seek to influence public health policy.

It is important to analyze past tobacco industry attempts to influence public health policy in order to raise questions about current practices. Despite efforts to mitigate industry influence through legislation, as recommended under Article 5.3 of the WHO Framework Convention on Tobacco Control, which prohibits direct tobacco industry relationships with public health policymakers, evidence suggests that the industry continues to exert influence on policy processes through third parties (Campbell and Balbach, 2009; ASH, 2008). Numerous front groups of the tobacco industry have been revealed (see for example Muggli et al., 2004; Dearlove, 2002), although little has been written about the indirect influence exerted through policy organizations with less explicit ties to the industry. This article, though dependent on historical evidence, seeks to raise questions about these additional and continued avenues of industry influence in public health policy.

Think Tanks and Policy Influence

McGann and Weaver (2002) conceive of think tanks as independent civil society organizations that mediate between the government and the public. This conception is reflected in the mission statements of many think tanks, which frequently describe their role as bridging knowledge producers and users, providing independent advice on public policy, and holding decision‐makers to account (Stone, 2007; Ohemeng, 2005). In particular, many think tanks claim to be free of political influence, citing their refusal to accept government or political party funding as evidence of impartiality. Supporters of think tank initiatives reiterate these claims of independent expertise. For example, the International Development Research Center, which provides funding to think tanks states, “Independent policy research institutions—or think tanks—are in a unique position to effect positive change in their societies. By generating and analyzing credible local data, they can enhance public policy debates and promote more objective, evidence‐based decision‐making that makes real, sustained improvements in people's lives” (IDRC, 2012). From this perspective, think tanks are seen as exerting an inherently positive influence within public policy processes.

Stone (2007) calls for a more critical analysis of the mission of, and interests embedded in, think tanks. She argues that think tanks are elite institutions, noting the frequent revolving door among think tanks, public relations firms, political parties and the civil service, thus blurring relationships between political and other actors. She further argues that, as think tanks are required to market themselves to donors in order to generate resources, they are likely to adopt ideological positions that reflect donor priorities. As a result, think tanks define the public's interest, policy problems and solutions in ways that reflect these ideological assumptions. From Stone's perspective, think tanks are neither “good” nor “bad” actors in public policy processes, but another type of actor whose power and influence must be interrogated.

Lack of transparency is a particular characteristic of think tanks. There is little information available about who funds most think tanks. Under current US regulation, as in most countries, there is no requirement for think tanks to report who their donors are. As a result, the majority of think tanks do not voluntarily disclose all of their funding sources (Lozovsky, 2015). One study of 164 think tanks worldwide found only 31 achieved the highest level of transparency about funding sources (Transparify, 2014). While there have been calls for greater transparency, and promises from some think tanks to improve financial disclosure, the current norm is nondisclosure.

This lack of financial transparency, within the context of Stone's critical perspective, is notable considering the influence that think tanks seek to exert in policy processes. Think tanks aim to influence policy making in a variety of ways. Their staff may write editorials and opinion pieces, appear in the media, testify in public hearings (Abelson, 2011), cultivate close relationships with politicians, build coalitions around specific policy issues (Monbiot, 2012) and shape public debates through publications and statements (Silverstein and Williams, 2013). In Do Think Tanks Matter, Abelson (2002) argues that think tanks in the USA and Canada can influence political dialogue, policy preferences and the choices of decision‐makers. There is evidence that US government representatives use think tanks research outputs more often than the Congressional Research Service (Bartlett, 2012). McGann and Sabatini (2011) argue that the growing number of think tanks worldwide demonstrates the sectors' expanding impact.

While there is general consensus that think tanks can be influential actors in public policy, demonstrating this influence over specific policy issues remains challenging (Stone, 2004; Abelson, 2002). A number of quantitative approaches have emerged to measure think tank influence. In 2004, the Think Tanks and Civil Societies Program, at the University of Pennsylvania, initiated a Global Go‐To Think Tank Index that ranks think tanks annually by influence based on surveys of opinion leaders and journalists, media mentions and online presence. In 2013, the Center for Global Development, itself a think tank, devised a methodology to rank US and international development think tanks by their public profile and social media presence. Abelson (2006, 161) notes that, while such measures demonstrate “how active think tanks are in a particular policy debate, such information provides little insight into how much influence they have had in shaping public policy.” Quantitative rankings are useful in comparing think tanks to each other, but say little about actual influence over policy outcomes (Thunert, 2003).

Within democratic systems, given the numerous players involved in influencing government policy action, it is often impossible to credit a single institution or actor with responsibility for any one particular outcome. Because of the non‐linear process of policy development, it is difficult to determine whether think tanks set the agenda, expand policy debates, impact policy decisions or simply follow policy trends (Abelson, 2006). Stone (1996) argues that think tanks influence how policy is “debated and decided,” as opposed to specific policy choices. She writes, “they help to provide the conceptual language, the ruling paradigms, the empirical examples that become the accepted assumptions for those in charge of making policy” (Stone, 1996, 110). Plehwe (2014) suggests that the influence of think tanks is best analyzed by using a network approach rather than by looking at individual agents in isolation. This approach recognizes that think tanks shape processes through ongoing policy conversations with influence distributed among multiple policy‐actors (Ohemeng, 2014).

Noting the challenges of demonstrating think tank influence in health policy, we aim to illustrate the influence of Atlas by assessing the media coverage of outputs related to tobacco control, connections between policy actors and the scope of the network. We also refer to quantitative studies to locate Atlas partner think tanks in influential health policy circles and consider donor perceptions of Atlas' influence and corresponding support. Thus, while it is difficult to identify Atlas' direct influence on tobacco control policy developments, we aim to demonstrate Atlas' substantial contributions to shaping policy conversations around tobacco control.

Methods

To investigate connections between Atlas and the tobacco industry, we searched the Truth Tobacco Industry Documents (TTID) database of over 4 million documents released by tobacco companies as a result of litigation and whistleblowers. While the TTID is a substantial collection of documents, providing insight into corporate activities and strategies, it is also recognized to be incomplete, with some documents withheld, missing, destroyed or redacted (Mackenzie et al., 2003). As a result, the connections that emerge are often more suggestive than conclusive, but still crucial to understanding industry influence in health policy.

We searched the TTID using keywords for mentions of Atlas and the names of partner think tanks in the USA. We focused on the USA because the country is where the network began and has the most partners. Because the majority of documents in the TTID date from before 2003, we searched documents from the decade between 1990 and 2000. This is when key policy debates emerged regarding the harms of secondhand smoke, public smoking bans and litigation against tobacco companies. Searches were conducted for the terms “Atlas Network” and “Atlas Economic Research Foundation,” as well as for each of the current 155 Atlas partners based in the USA, as listed on the Atlas website. Think tanks' full names, as well as acronyms or abbreviated names, were searched for. Documents retrieved were analyzed for evidence of financial support from the tobacco industry, funding purpose, relationships between think tanks and corporations (such as alliances) and personal links between the industry and Atlas partners (such as board membership).

We searched the websites of Atlas partners mentioned in TTID documents for policy outputs related to tobacco. This enabled us to trace activities mentioned in the TTID documents to think tank activities and outputs relevant to tobacco control. We then used LexisNexis (NY, USA) to search for media mentions of think tank outputs related to tobacco control. Finally, we searched sites such as Sourcewatch.org and PRwatch.org, as well as conducted broad Google searches, to follow up on further potential links. The earlier data was organized chronologically and triangulated, and interpretations of documents were crossed checked by the three authors.

Findings

Tobacco industry connections with atlas

Truth Tobacco Industry Documents suggest that Atlas served as a link between the tobacco industry and those acting on its behalf to influence health policy. In 1994, an economics professor at George Mason University, James Bennett, wrote to the tobacco company RJR describing plans to publicize a book co‐authored with Thomas J. Dilorenzo entitled Unhealthy Charities: Hazardous to Your Health and Wealth (Bennett and DiLorenzo, 1994). The letter describes plans to write another book, Cancer Scam: The Diversion of Federal Cancer Funds to Politics, and for the US$100 000 payment for this work to be sent via Atlas.

Documents suggest that, not only did RJR use Atlas as an intermediary to pay the authors for writing these books, but the authors had multiple connections to the tobacco industry through Atlas partner organizations. Bennett was an adjunct scholar at the Heritage Foundation, which received funding from PM in 1995 and 2000 (PM, 1995, 2001), and was a member of the Mont Pelerine Society, an Atlas partner described by PM as an “ally” in an external affairs plan (PM, 2000). Co‐author Dilorenzo was also a member of the Mont Pelerine Society, as well as a fellow at the Independent Institute which also received tobacco industry funding (PM, 1995, 2001). Peter Sparber, the Vice President of the Tobacco Institute (an industry created and funded trade organisation, which was also an Atlas partner), provided advice on early drafts of Cancer Scam (Sparber, [Link]), and three draft copies of the manuscript were found among PM documents dating 4 years before the book was published (White and Bero, 2004; Morley et al., 2002).

The publication of the books, and related commentary, was timely for the tobacco industry. The US government was providing increasing funding for comprehensive tobacco control projects, such as the American Stop Smoking Intervention Study (ASSIST), which aimed to build capacity of policy‐focused interventions to impact individual behaviors. In response, tobacco companies initiated a sustained campaign to discourage and divert federal funding (White and Bero, 2004). As part of this campaign, the books by Bennett and Dilorenzo, and related publications, propagated arguments that government‐funded organizations, such as the American Cancer Society and US National Cancer Institute, used taxpayer funds for “narrow political advancement” under the guise of public health, and argued that taxpayer dollars would be better spent on medical research (Bennett and DiLorenzo, 1994 & 1997). Cancer Scam, in particular, used American Stop Smoking Intervention Study as an example of what it claimed was government waste, greed and cronyism.

The books appear to have received a high degree of public and media attention. An August 1994 handwritten memo from a PM executive's office describes plans to train Bennett in speaking to the media, publicize his books and ensure he is interviewed by “friendly media” (Anon, 1994). The books were reviewed in the New York Times (Taylor, 1994; Henig, 1997), Wall Street Journal (Boot, 1994) and academic journals (Schughart 2000). None of the reviews mentioned that the authors had received payment from, or had any connections to, the tobacco industry. Bennett (1995b) also promoted the books without disclosing his connections to the tobacco industry. The documents suggest that the role Atlas played in receiving funding from RJR, on the authors' behalf, obscured relationships with the tobacco industry, allowing the books to be presented as independent research by policy experts.

In addition to channeling industry funding for such publications, Atlas received US$275 000 from PM in 1994 for “Tobacco Issues Management” (Borelli, 1994), and received “public policy grants” ranging from US$75 000 to US$475 000 until at least 2000 (PM, 1995, 2001). While the purposes of these donations are not specified, PM documents describe Atlas as one of its “strategic allies” (PM, 1999) and “third party contacts” (PM, 1998).

One PM document describes the value of partnering with Atlas as derived from the latter's “positive influencing of legislative and regulatory climate, policy debate and public opinion on critical issues” (Marden, 1998). Regulation and public opinion were primary concerns of the tobacco industry during the 1990s. The US federal, as well as state, governments were debating the adoption of public smoking bans and other tobacco control measures. Further revelations about the health harms of tobacco use, and industry efforts to deny or obscure them, created a public relations crisis (Brandt, 2009). The industry needed allies to counter‐growing criticism and the move towards stronger regulation. Free market think tanks, like Atlas, helped to package stronger regulation as unnecessary government interference in individual freedoms, and positioned tobacco companies as important business actors contributing to economic prosperity (Hoek, 2015). As described by PM staff, “the ideals and activities of the (Atlas) foundation enhance an improved operating environment for all PM businesses and encourage opinion leaders and decision‐makers to take a critical look at regulatory or policy initiatives aimed at suppressing free enterprise, free speech or the rule of law” (PM, 2001a). In short, Atlas was seen as an influential ally in public policy debates over tobacco regulation.

Documents suggest PM staff aimed to use Atlas to expand the company's policy influence and build new alliances beyond the USA. Staff nominated Atlas for a US$150 000 grant “to positively impact the regulatory environment, particularly in Latin America, through the production and dissemination of publications and articles, through organizing public speaking events and seminars, as well as through support, guidance and assistance to in‐market think tanks and institutes” (PM, 2001). Like in the USA, tobacco control was being publicly debated in Latin America, a region where the tobacco industry had sought to expand market to offset declining demand in North America. Atlas, with its global network of think tanks and track record of opposing regulation, provided an entry point into the region's policy community. Available TTID do not confirm whether the award was made. Documents do show that Atlas invited PM Manager of Public Policy and Regulatory Affairs, Matthew Winokur, to attend a conference on deregulation in Argentina in 1998, suggesting at least some of the activities listed in the proposal were carried out (Atlas, 1998).

Tobacco industry connections with atlas partner think tanks

The extent of the tobacco industry's connections with Atlas partner think tanks between 1990 and 2000 is presented in Table 1. Fifty‐seven Atlas partner think tanks (37%) in the USA are identified in TTID as receiving funding from the tobacco industry, listed as allies or partners, and/or as implementing partners in policy campaigns. While it is not possible, within the scope of this article, to trace each of these connections, the multiple instances of think tanks receiving industry funding, and being listed as allies or as part of strategies to influence policy (32 of 57 think tanks in Table 1), suggests relationships whereby tobacco companies provided resources in return, or in payment, for think tanks participation in policy debates. One list of PM “third party allies” notes that Atlas partners the Claremont Institute, Hoover Institute, Independent Institute, Pacific Research Institute and Reason Foundation all received funding from PM (PM, 1999). Another PM document lists “national allies” including Atlas partners such as the Cato Institute, Frontiers of Freedom, Mackinac Center for Public Policy, Manhattan Institute for Policy Research and National Center for Public Policy Research (PM, 1998), all of which received PM funding in 1999 and 2000 (PM, 1999a, 2001).

Table 1

Evidence of connections between Atlas partner think tanks and tobacco companies (1990–2000)

Atlas Partner Think Tank	Company	Grant received	Policy partner	Listed as ally
Acton Institute	PM	95, 98	94	98
American Enterprise Institute	PM	95, 98, 99, 00	94, 95	98
American Legislative Exchange Council	PM	95	––	––
Americans for Tax Reform	PM	95, 98, 99, 00	94, 98, 99	99
Atlantic Legal Foundation	PM	93–97	––	––
Atlas Society	PM	95	––	––
Cascade Policy Institute	PM	98, 00	95	––
Cato Institute	PM	95, 98, 99, 00	94, 95, 98	98, 99
Center of the American Experiment	PM	95	––	––
Claremont Institute	PM	––	95	––
Commonwealth Foundation	PM	98	95	––
Competitive Enterprise Institute	PM	95, 99, 00	95	98
Foundation for Economic Education	PM	00	––	98
Fund for American Studies	TI	97	––	––
Georgia Public Policy Forum	PM	98, 00	––	––
Goldwater Institute	PM	98, 00	––	––
Heartland Institute	PM	98, 00	94, 95, 99	98
Heritage Foundation	PM	95, 97, 98	94, 95	99
Hoover Institution	PM	99	94, 95, 99	98
Hudson Institute	PM	95, 98	95	98
Human Rights Foundation	PM	97	97	––
Independence Institute	PM	94, 98, 99, 00	94	––
Independent Institute	PM	95, 98, 00	95	––
Independent Women's Forum	PM	98, 99, 00	––	––
Institute for Human Studies	RJR	92	95	98
Institute for Justice	PM	95, 98, 99, 00	95	98,99
Institute for Policy Innovation	TI	95	95	––
Intercollegiate Studies Institute Inc	PM	––	95	––
James Madison Institute	PM	97, 98	––	––
John Locke Foundation	RJR	00	––	––
Kansas Policy Institute	PM	00	––	––
Law and Economics Centre	PM	00	––	––
Leadership Institute	TI	98	––	––
Mackinac Center for Public Policy	PM	95, 98, 00	94, 99	98
Manhattan Institute for Policy Research	PM	95, 98	94, 95, 00	99
Media Research Center	RJR	96	96	––
Mercatus Center	PM	00	––	––
Mont Pelerin Society	PM	––	––	00
National Center for Policy Analysis	PM	95, 98, 00	94, 95, 99	98, 99
National Center for Public Policy Research	PM	95, 99, 00	99	––
National Journalism Center	PM	95	94, 99	98, 99
National Review Institution	PM	91	––	––
National Taxpayers Union	PM	95, 99, 00	95, 99	––
Nevada Policy Research Institute	PM	95, 00	95	––
New England Legal Foundation	PM	95	95	––
Pacific Legal Foundation	PM	95, 99, 00	95	98, 99
Pacific Research Institute	PM	94, 99	94, 95, 98, 99	98
Philanthropy Roundtable	PM	––	94	98
Pioneer Institute for Public Policy Research	PM	98	94, 95	99
Reason Foundation	PM	95, 98, 99, 00	95, 99	98
South Carolina Policy Council	RJR	––	––	98
State Policy Network	PM	95	95	––
Tax Foundation	PM	00	95	99
Texas Public Policy Foundation	PM	98	00	––
Thomas Jefferson Institute	PM	99, 00	99	99
Yankee Institute for Public Policy	PM	––	95	––

The columns indicate the years (i.e., 98 = 1998 and 00 = 2000) Atlas partner think tanks received grants, were listed in documents describing plans to produce or disseminate policy ideas, or were described as third party allies in industry documents. This list is based on documents in the Truth Tobacco Industry Documents and may not be a complete representation of the connections between the think tanks and tobacco industry.

PM, Philip Morris; RJR, RJ Reynolds; TI, Tobacco Institute

Atlas partner think tanks have claimed that the prevalence of pro‐industry ideas in their activities and outputs merely reflect a shared ideology, based on free market economics and libertarian perspectives, rather than corporate influence. When accused of representing the interests of the tobacco industry, the Adam Smith Institute (an Atlas partner based in the UK) claimed its report on smoking and taxes “reflects our free market, libertarian principles” (as quoted in Smith et al., 2009). However, the pattern of funding from the industry to think tanks documented here, which concurrently produce publications supporting industry interests, as well as internal documents describing Atlas and its partners as “allies” in public relations strategies, strongly suggests that the policy ideas put forward reflect a strategic approach to knowledge creation and dissemination.

This influence is notable in the publications and activities of Atlas partner think tanks in response to lawsuits filed against the tobacco industry. In 1998, the US Department of Justice filed a criminal legal action against the four largest American tobacco companies (PM, RJR, BAT American subsidiary Brown & Williamson and Lorillard) for violating the federal Racketeer Influenced and Corrupt Organizations Act, and defrauding the American people by lying about the health risks of smoking and by marketing to children (Hurt et al., 2009). Documents describe how PM aimed to use Atlas partners' influence to discredit the lawsuit. A February 1999 “Weekly Activity Report for Issues Management” notes that Thomas Borreli (Corporate Scientific Affairs Manager) “discussed international litigation issues with Atlas Foundation” (Vitobello, 1999). A later email from Roy Marden (PM Manager of Industry Affairs) to Frank Gomez (Director of Public Affairs) lists several Atlas partners beside “Key Actions” related to countering the federal lawsuit (Marden, 1999). For example, the Cato Institute is listed next to “op‐eds, media, policy briefs, LTEs.” The same year, Cato received US$100 000 from PM and published an op‐ed in the Legal Times titled, “The Great Tobacco Robbery: Lawyers Grab Billions,” which referred to the federal lawsuit as “no better than extortion” (Levy 1998). The Cato Institute ranks within the top 10 most influential think tanks in the USA, and the top 10 within domestic health policy (McGann, 2015, 100).

The list of actions also notes plans to “blast faxes to state legislators” through the Heartland Institute. The Heartland Institute, which received funding from PM during the 1990s and included PM Manager Roy Marden on its board, had a program called PolicyFax, which faxed monthly outputs from Atlas think tanks to over 8,000 politicians, as well as other policy‐focused organizations (Heartland Institute, 2015). These links provided another avenue to disseminate industry‐funded opposition to the lawsuit, under the guise of think tank publications, to US politicians and policy makers.

Disseminating policy ideas through think tanks obscures industry links, as think tanks claim to be independent. For example, the Heartland Institute states on its website that it “does not promote policy ideas for money.” This is challenged, however, by a 1999 letter from Heartland President Joe Bast to PM Manager Marden, in which he writes, “Because Heartland does many things that benefit PM's bottom line, things that no other organization does, I hope you will consider boosting your general operating support” (Bast, 1999). The letter further notes that tobacco policy issues are well represented in PolicyFax, as well as Heartland's other publications the Heartlander and Intellectual Ammunition. The letter also mentions specific publications, such as one written by Bast (1996) himself, entitled “Joe Camel is Innocent,” ridiculing claims that the cartoon is used to market cigarettes to children, and arguing that the EPA exaggerated the risks of secondhand smoke. Bast (1999) then thanks PM for US$30 000 in donations the previous year and requests a further US$35 000 for 2000.

Bast's letter further notes plans to collect all tobacco resources in one place on Heartland's website entitled “The Smoker's Lounge.” This webpage, still active at the time of writing, demonstrates not only how likeminded think tanks within the Atlas network disseminated and support each other's ideas, but also the prevalence of industry‐sponsored ideas about tobacco control promoted by these think tanks. At the time of writing, the webpage states,

Welcome to the Smoker's Lounge, the place to go for sound science, economics, and legal commentary on tobacco issues. This ‘issue suite’ cuts through the propaganda and exaggeration of anti‐smoking groups by giving you access to the best available research and commentary from scores of independent research organizations, publications, and government sources (Martin & Bast, 2007).

An examination of the website's contents, however, shows that the “independent research” cited recently comes largely from fellow Atlas think tanks with industry connections. The site argues that public health advocates have exaggerated the risk of smoking‐related illness, supporting this assertion with reference to a 1998 op‐ed, “Lies, Damned Lies, and 400 000 Smoking‐related Deaths,” by Robert Levy and Rosiland Marimont. Levy is a fellow at the Cato Institute, which as noted earlier received funding from tobacco corporations throughout the 1990s. Similarly, the website notes that “Science writer Michael Fumento and others have documented how the threat of secondhand smoke has been greatly exaggerated.” Fumento is a fellow at the Hudson Institute, a PM‐funded Atlas think tank, and graduated from the PM‐funded National Journalism Center (Young Americans Foundations, 2015). The sites suggests “Three credible and interesting people who write frequently on tobacco issues”: Fermento, Jacob Sullum and Kip Viscusi. Sullum was a fellow at the Reason Foundation, which received grants from PM of US$25 000 in 1995, US$20 000 in 1999 and US$20 000 in 2000 (PM, 1995, 1999a, 2001). He wrote the book, For Your Own Good: The Anti‐Smoking Crusade and Tyrant of Public Health (Sullum, 1998), as well as articles and op‐eds that argued against stronger regulation, questioned the links between smoking and cancer, and decried increasing tobacco taxes. Viscusi has served as a paid expert witness for the tobacco industry on dozens of occasions (Friedman et al., 2005). For example, in 1995, the Tobacco Institute paid Viscusi US$32 810 for comments and testimony he gave on the industry's behalf, in order to help the industry counter Occupational Safety and Health Administration and EPA efforts to address secondhand smoke issues in the workplace (Tobacco Institute, 2003). The knowledge sources, in short, are circular and self‐referential, leading back to Atlas partners funded by the tobacco industry. Figure 1 illustrates the network map of individuals (ovals) and think tanks (squares) mentioned on Smoker's Lounge, and their connections to PM.

Figure 1

Smoker's Lounge Network. [Colour figure can be viewed at wileyonlinelibrary.com]

Conclusion

This article finds that the multiple connections between Atlas, its partner think tanks, and the tobacco industry represent more than a handful of isolated cases. With over one‐third of Atlas partners receiving funding, or working in partnership with tobacco corporations, it is argued that the sum of these relationships generates influence that is greater than apparent from analysis of individual case studies. The evidence reviewed suggests the industry aimed to use Atlas, not only to covertly fund anti‐tobacco control publications, as demonstrated by the channeling of funding for Cancer Scam and Unhealthy Charities, but also to build linkages with and between think tanks to influence health policy. The Atlas think tank network provided means, not only for countering arguments for stronger tobacco control regulation, but also strategic information dissemination such as through PolicyFax. Atlas partners repeatedly quoted and cited each other, as illustrated by the Smoker's Lounge website, to create an illusion of independently supported research. By funding multiple think tanks, within a shared network, the industry was able to generate a conversation among seemingly independent policy experts. This suggests a coherent strategy by the tobacco industry to work with Atlas to influence public health policies from multiple directions.

While it is difficult to demonstrate the direct impact of industry funding on think tank activities and outputs, and think tank influence on public policy processes, the size, connections and resources of Atlas and its partners suggests an influential relationship. A number of Atlas partners are recognized to be among the most influential in the USA suggesting it is highly likely that they shaped public health policy debates around tobacco control by delegitimizing tobacco control advocates and generating contention over federal litigation against tobacco companies. Indeed, it is argued here that it is unlikely that the tobacco industry would have continued funding Atlas and its partners for over a decade if they had little influence.

While it may not be surprising (for tobacco control advocates and academics studying the role of free market‐oriented think tanks) that a large number of Atlas think tanks have shared ideological beliefs with tobacco corporations, the findings of this paper show that it is more than a coincidence that so many Atlas partners received tobacco funding while promoting ideas favorable to the industry. Revealing these instances is critically important to improving transparency of think tank influence within public policy given claims and resultant public perceptions that think tanks provide independent expertise devoid of vested interests.

The continued lack of transparency regarding think tank funding sources and the current limited availability of internal tobacco industry documents in the TTID dating after 2003, means that it is hard to determine if the industry has continued to fund Atlas and its partner think tanks to the present day. Only a few think tanks reveal their donors. For example, the Cato Institute continues to get funding from RJR (Cato, 2014). The findings of this paper support arguments that think tanks engaged in public policy should be required to disclose their funding sources and partnerships. This could be achieved through regulatory change regarding their charitable status and reporting requirements, supported by pressure from civil society and other public interest groups. Until such transparency is achieved, public health policy makers should view the perspectives, activities and outputs of think tanks with critical awareness of their nature, not as independent experts, but as another type of actor whose power and influence must be interrogated.