Article Text

Download PDFPDF

‘We have a rich heritage and, we believe, a bright future’: how transnational tobacco companies are using Twitter to oppose policy and shape their public identity
  1. Christina Watts1,2,
  2. Marita Hefler3,
  3. Becky Freeman1,4
  1. 1 School of Public Health, Sydney Medical School, University of Sydney, Sydney, New South Wales, Australia
  2. 2 Cancer Council NSW, Sydney, New South Wales, Australia
  3. 3 Menzies School of Health Research, Darwin, Northern Territory, Australia
  4. 4 Charles Perkins Centre, University of Sydney, Sydney, New South Wales, Australia
  1. Correspondence to Christina Watts, Cancer Council New South Wales, 153 Dowling Street, Woolloomooloo, New South Wales, 2011, Australia; christinaw{at}nswcc.org.au

Abstract

Background The tobacco industry has a long history of opposing tobacco control policy and promoting socially responsible business practices. With the rise of social media platforms, like Twitter, the tobacco industry is enabled to readily and easily communicate these messages.

Methods All tweets published by the primary corporate Twitter accounts of British American Tobacco (BAT), Imperial Brands PLC (Imperial), Philip Morris International (PMI) and Japan Tobacco International (JTI) were downloaded in May 2017 and manually coded under 30 topic categories.

Results A total of 3301 tweets across the four accounts were analysed. Overall, the most prominent categories of tweets were topics that opposed or critiqued tobacco control policies (36.3% of BAT’s tweets, 35.1% of Imperial’s tweets, 34.0% of JTI’s tweets and 9.6% of PMI’s tweets). All companies consistently tweeted to promote an image of being socially and environmentally responsible. Tweets of this nature comprised 29.1% of PMI’s tweets, 20.9% of JTI’s tweets, 18.4% of Imperial’s tweets and 18.4% of BAT’s tweets. BAT, Imperial, JTI and PMI also frequently used Twitter to advertise career opportunities, highlight employee benefits, promote positive working environments and bring attention to awards and certifications that the company had received (11.6%, 11.1%, 19.3% and 45.7% of the total tweets published by each account, respectively).

Conclusions Transnational tobacco companies are using Twitter to oppose tobacco control policy and shape their public identity by promoting corporate social responsibility initiatives in violation of WHO Framework Convention on Tobacco Control. Regulation of the tobacco industry’s global online activities is required.

  • advertising and promotion
  • public policy
  • tobacco industry
  • media

Statistics from Altmetric.com

Video abstract

Introduction

Despite transnational tobacco companies (TTCs) being the cause of 7 million deaths worldwide annually1 and continually opposing evidence-based tobacco control policies,2–5 the tobacco industry has relentlessly pursued strategies to improve their corporate image. Publicly available internal documents have revealed that TTCs have planned to improve their corporate image for almost two decades6 and that corporate social responsibility (CSR) is a significant component of the industry’s overall marketing strategy.7 This is done in various ways, including efforts to improve their image among women,8 through charitable donations,9 10 by contributing to youth prevention work,11 preventing and raising awareness of child labour11 and promoting environmental sustainability and stewardship.11–15

With the recent and rapid rise of social media, TTCs now have a platform where they can readily and openly publish their CSR agenda. TTCs have already embraced online media and appear to be using it to purposefully oppose, weaken and resist tobacco control policy.16 This is unsurprising given online media is not currently restricted by tobacco regulations. An example of this was seen in British American Tobacco (BAT) New Zealand’s online media campaign ‘Agree–Disagree’, which was launched to undermine the New Zealand government’s plain packaging consultations.3 Given TTCs’ aggressive record of undermining tobacco control policies,2–5 bidding for political support,17 continuing advertising and promotion through covert means18–21 and long-standing history of attempting to improve their image through CSR activities,6 it is important to monitor the content being published by TTCs on social media.

TTCs such as BAT, Imperial Brands PLC (Imperial), Philip Morris International (PMI) and Japan Tobacco International (JTI) each have active accounts on Twitter, which is an online microblogging platform where users publish and read 280 character messages, known as ‘tweets’. Prior to November 2017, tweets were limited to a maximum of 140 characters. As of 2017, Twitter is being used by 342 million active users globally22 and is also widely used by corporations for promotion and advertising purposes. Content published on Twitter is primarily user generated and instantaneously published and other Twitter users are able to engage in the content through likes, retweets and replies. How TTCs are using Twitter and the content of their posts needs to be carefully monitored, to keep atop of the industry’s evolving strategies to undermine policy and to ensure that tobacco regulation is current, relevant and effective.

To date, there have been no published studies on TTCs public messaging on their own corporate social media accounts. This study analyses and describes the content published by the global accounts of BAT, Imperial, PMI and JTI on Twitter. The current study examines the use of Twitter specifically, because unlike other high reach social media platforms such as Facebook, Twitter is being overtly and actively used by BAT, Imperial, PMI and JTI and at least three tweets per week are published by each account. This study aims to uncover common themes in the content published by these companies, to understand how social media marketing and public relations strategies are being used to shape the industry’s public identity and interfere with policy.

Methods

Tweets published by the global corporate Twitter accounts of BAT, Imperial, PMI and JTI were downloaded using the web-browser extension, NCapture. All tweets published since the accounts were created, including retweets from other accounts, but excluding replies, up until and including 7 May 2017 were downloaded.

The global corporate Twitter accounts of BAT, Imperial, PMI and JTI were included as they represent four of the world’s six top tobacco companies by market share.23 China National Tobacco Corporation, which has the largest share of the global market,23 does not have a Twitter account and therefore could not be included in the study. Country-specific and region-specific accounts, employment recruitment accounts and research and development accounts from the four chosen companies were excluded. Altria/Philip Morris USA is the other company in the top six, however given its close alignment with PMI, it could be considered a country-specific/region-specific account.

All tweets downloaded by NCapture were imported into NVivo; a qualitative data analysis programme. One hundred tweets from each account were coded initially, using an iterative process of coding tweets and adding content categories. In the first instance, coding was based on the text in the tweet. For tweets that were published with less than five words and/or had an unclear message, weblinks and/or images included in the tweet were also viewed for clarification. The final set of categories were then agreed by authors CW and BF. All Tweets were manually coded by CW. A random sample of 30 tweets, which were from the majority of the categories, were coded by five independent coders, using a coding rulebook, to test inter-rater reliability. The inter-rater reliability was 96%. Each tweet was coded in 1 of 10 overarching categories, with subcategories within four of these overarching categories. There were 30 categories in total, including subcategories. All coding categories were mutually exclusive, therefore, tweets could only be coded under one category. For tweets that could be coded under multiple categories, a coding hierarchy was formulated. This was primarily based on the relative frequency of the possible categories. For example, a tweet that contained messaging about illicit and counterfeit tobacco and plain packaging would be coded as plain packaging as there were fewer tweets overall about plain packaging, than illicit and counterfeit tobacco. Using this methodology, topics less frequently tweeted about were represented in the analysis. Subcategories also were given higher priority in the hierarchy than the overarching category, to ensure that the specific detail of the tweet was more accurately represented. For example, tweets about research into electronic cigarettes were coded as electronic cigarette research and product developments, rather than within the overarching category of research and development. Online supplementary file 1 shows the coding guidelines and hierarchy used for this study.

Results

A total of 3301 tweets were captured from the four Twitter accounts. This included 2659 original tweets (ie, tweets composed by the company) and 642 retweets (ie, tweets published by other sources and retweeted by the tobacco industry corporate accounts). There were no duplicates among the 3301 tweets collected. BAT had a relatively high proportion of retweets (35.1%, n=464) compared with Imperial (9.9%, n=94), PMI (7.5%, n=49) and JTI (9.4%, n=35). Each account was created between March 2012 and March 2014 and at the time of data collection, followers numbered between 2815 and 18 486 (table 1).

Table 1

Characteristics of selected Twitter accounts

The top three categories that comprised the highest proportions of tweets across all accounts were: policy opposition (30.4%), CSR (20.8%) and career opportunities, company awards and employee benefits (19.4%). Tweets under these categories comprised over 70% of the total number of tweets. A detailed summary of the findings relating to these top three categories only will therefore follow. Table 2 shows the number and per cent of tweets coded under the 30 categories and subcategories by account.

Table 2

Frequencies and proportions of tweets by coded category

Policy opposition

BAT, Imperial and JTI most frequently tweeted about topics that either outright opposed or critiqued tobacco control policies or aimed to reduce the effectiveness or legitimacy of the measures (36.3%, 35.1% and 34.0%, respectively). Such tweets were coded as policy opposition. PMI tweeted less frequently on this area, with just 9.6% of their total tweets directly resisting or opposing tobacco control measures. One topic mentioned frequently across all accounts was illicit and counterfeit tobacco, which represented 12.6% of all tweets. BAT had a focus on the issue; 18.6% of their tweets were solely about illicit or counterfeit tobacco (n=246). Illicit and counterfeit tobacco was also regularly tweeted about by Imperial (15.7%, n=149), and to a lesser extent by JTI (11%, n=41) and PMI (5.3%, n=35). Examples of tweets about illicit and counterfeit tobacco include:

Portuguese #police seize more than 182 tons of tobacco leaf, the largest seizure ever seen in Europe https://t.co/cmXOxRV3Cc#illegaltobacco

-@JTI_Global 3 Feb 2016

Every fourth cigarette sold in Pakistan is illicit. Illegal cigarette trade expected to reach 27.1bn sticks by 2015 ➜ http://t.co/ngxtlniVE8

-@BATPress 4 Feb 2014

Tweets opposing marketing and promotional policies also had a high frequency across the tweets from all four accounts (10.6%, n=349). This category encompassed plain packaging (9.0%, n=297), the ‘slippery slope’ of tobacco marketing restrictions and how other industries could be impacted (0.8%, n=25) and tweets that generally resisted or opposed tobacco marketing and promotion restrictions (0.8%, n=27). Imperial and BAT had a focus on tweeting about marketing and promotional restrictions. 12.5% of all tweets from Imperial (n=119) and 11.6% of all tweets from BAT (n=153) opposed plain packaging legislation. Examples of tweets opposing plain packaging include:

Given evidence from Australia, the conclusion that #plainpacks is an effective measure defies logic. Our statement: http://t.co/DxILgd1eB4

-@BATPress, 3 April 2014

Myth 5: Tobacco Control said #plainpacks would stop young ppl from taking up smoking. Govt stats show this isn’t true

@ImperialBrands, 5 December 2014

Three of the four companies specifically warned of how policy-makers and regulators would impose tobacco control measures, like plain packaging, on other industries:

Is wine in Australia the next target? https://t.co/nN0Xyzpbs1#thefutureofbrands#slipperyslope

@JTI_Global, 26 April 2017

Regulators are copy-and-pasting tobacco-style regulations into other sectors without any thought as to whether they worked elsewhere

@JTI_Global, 20 March 2017

Of the four accounts, JTI most frequently posted tweets of this nature (5.3%, n=20) and linked to a website established by JTI, www.thefutureofbrands.com, which is designed as a digital abandoned sideshow with games such as the ‘Wheel of misfortune—round and round the regulators go, where they stop, nobody knows’ and the ‘Fortune Teller—Has the future of your brand already been written?’, to warn that the food and drink industries will be the next target of the ‘malicious’ tobacco regulators.24

Tweets opposing tobacco taxes, tobacco-related trade agreements and smoke-free environments also appeared across the accounts, however, frequencies of these topics were low. BAT, however, led the way with tweets opposing the actions of the WHO with 19 of the total 22 published tweets across the 4 accounts portraying the WHO as an unethical organisation that is unfairly targeting tobacco farmers through regulation. For example:

WHO proposals put livelihoods of 2.7 m Filipino farmers at risk without offering economically viable alternatives http://t.co/rxTwQe3b#fctc

- @BATPress, 18 September 2012

Corporate social responsibility

Alongside tweets opposing and resisting tobacco control policy, all four tobacco companies consistently and regularly tweeted about the positive impact that the companies are having on society and the environment. Such tweets were coded under CSR and made up 29.1% of PMI’s tweets, 20.9% of JTI’s tweets, 18.4% of Imperial’s tweets and 18.4% of BAT’s tweets. Tweets about the companies’ philanthropic works featured on all accounts (5.2% of the total tweets, n=171), however, Imperial had a focus on highlighting their philanthropic strategies, with 12.7% of their tweets coded under Philanthropy and staff volunteering. Examples include:

Mobilise For May’ update: We’re proud of our 6000 employees who’ve spent 50 000 hours volunteering to support 150 good causes worldwide.

-@ImperialBrands, 4 July 2016

Giving back to the community matters to PMI. In #Mexico our team donated gifts to local organizations that help vulnerable youth. #InsidePMI https://t.co/HSozVoloSB

-@InsidePMI, 6 April 2017

All four TTCs tweeted about social and environmental issues such as child labour and human rights, environmental sustainability and racial and gender diversity and inclusion. 9.1% of JTI’s tweets were about issues of child labour and human rights, many of which referenced the ‘Achieving Reduction of Child Labour in Support of Education Programme’, which is a joint venture of JTI, Winrock International and the International Labour Organisation to reduce child labour in the countries where JTI operates.25 Tweets to promote perceptions of the companies’ practices as environmentally sustainable were common, however, were most prominent on PMI’s account (6.1%, n=40). PMI also highlighted their values to accept and include all people, and promote gender equality in society and the workplace. These tweets were coded as Racial and gender diversity and inclusion and made up 8.7% of PMI’s total tweets (n=57). Examples of such tweets include:

Biases cloud our view. Let’s aim to be more inclusive. ACT to #UNLABEL – and foster #diversity & #inclusion https://t.co/N8aYTmiwKO

@InsidePMI, 9 December 2016

Fewer than 5% of firms have a female #CEO. How can we inspire future female leaders? https://t.co/ICZF4UId7H

-@InsidePMI, 5 July 2016

Career opportunities, company awards and employee benefits

All companies used their respective Twitter accounts to advertise and promote career opportunities, highlight employee benefits and positive working cultures, and bring attention to awards and nominations and certifications that the company had received. Such tweets featured particularly heavily for PMI, making up 45.7% of their tweets. For PMI, the majority of tweets within this category specifically highlighted workplace culture and employee benefits (34.0%, n=300). Examples include:

It’s your work. Do it your way. That’s the #freedom Michel gets at PMI. And what makes him #passionate about it https://t.co/5x62pnMhbg

-@InsidePMI, 12 February 2016

We empower our 80 000 people to achieve their career goals. Proud to be named a #GlobalTopEmployer! #UnlimitYourself https://t.co/Pithpc6qKN

-@InsidePMI, 1 March 2017

All four companies also frequently tweeted about their status as a ‘top employer’ and/or being awarded as a ‘top employer’ in various countries and regions. Such tweets were included under Company awards and recognition and made up 4.8% (n=158) of the total number of tweets. For example:

We’re proud to be frequently rated as a top employer around the world #wearebat #topemployer https://t.co/zQOyi2Ehwq https://t.co/mLBrr3J7JV

-@BATPress, 12 August 2016

Other themes and topics

The remaining 29.8% of tweets from the four accounts were coded across a further seven categories: research and development (10.4%, n=344), company strategy, investments and results (9.8%, n=324), customer service (6.9%, n=227), media, public relations and events (1.8%, n=58), holiday, birthday and anniversary messages (0.4%, n=13), non-English language (0.3%, n=9) and general facts about tobacco (0.2%, n=7). Although tweets overall were less frequently coded under these categories, it should be noted that 21.3% (n=312) of tweets published by BAT were about research and development, including electronic cigarette research and product development and reduced-risk products and harm reduction. This was the second highest category for BAT after policy opposition.

Discussion

BAT, Imperial, PMI and JTI are using their respective corporate Twitter accounts to oppose and critique tobacco control policies, to promote an image of being socially responsible corporations and position themselves as favourable workplaces. These social media strategies highlight that the tobacco industry is continuing to heavily promote CSR activities to shape their public identity and oppose tobacco control policies in a public domain.

Online communications, particularly through social media, are currently poorly regulated and TTCs are evidently exploiting these avenues for their own corporate gain. The lack of regulation on social media has opened an opportunity for TTCs to also use Twitter to highlight information that is potentially false or misleading and create and support a narrative that TTCs are valuable corporations to society. Across the four Twitter accounts analysed in the current study, a relatively large proportion of tweets highlighted environmentally sustainable business practices and efforts to have a positive social impact. However, without knowing that tobacco growing and manufacturing causes significant damage to land and agriculture, contributes to poverty and food insecurity, uses children for labour, consumes comparatively large amounts of energy and water, pollutes the air and generates colossal amounts of waste,26 tweets of this nature have the potential to mislead readers that TTC’s are making a positive impact on these issues, rather than significantly causing or contributing to them. For example, tweets such as ‘140 million trees planted between 2007 and 2012 through our afforestation programmes #trees #afforestation http://t.co/WtdnRGHuUY’, published by BAT, are misleading as tobacco growing and curing is a major cause of deforestation and is responsible for the loss of an estimated 1.5 billion hectares of forest since the 1970s.26 BAT, Imperial, PMI and JTI also largely present false or misleading information when opposing tobacco control measures, particularly in regards to illicit tobacco and plain packaging, in order to discredit the effectiveness of these measures. For example, false tweets such as, ‘Myth 5: Tobacco Control said #plainpacks would stop young ppl from taking up smoking. Govt stats show this isn’t true’, were published by Imperial, despite evidence that plain packaging reduces the appeal of cigarette packs to adolescents.27 28 It is unsurprising that tweets of this nature are published by TTCs considering the long history the tobacco industry has of misrepresenting evidence to interfere with tobacco control.5

BAT, Imperial, PMI and JTI also each promoted and celebrated being awarded as a ‘Top Employer’ on several occasions, yet on investigation, the ‘award’ was found to be a simple certification achieved through application to the Top Employers Institute, which is a for-profit company based in the Netherlands.29 Tweets of this nature, along with tweets that aim to highlight TTCs’ philanthropic efforts such as, ‘Giving back to the community matters to PMI. In #Mexico our team donated gifts to local organizations that help vulnerable youth. #InsidePMI’, have the potential to shift public sentiment on TTCs as they imply TTCs have an altruistic benefit to societies and economies across the world.

Illicit and counterfeit tobacco was a major theme in this study. Illicit tobacco is particularly useful to the tobacco industry as a platform for public health collaboration, as it can be presented as an area where its interests overlap with government agencies. An important component of tobacco industry strategy to undermine tobacco control has been to focus on engagement with associations and businesses that represent the interests of the tobacco industry and to partner with agencies with portfolio responsibilities such as finance, trade and law enforcement to weaken the influence of health agencies.30–32 This has been evident in Australia, where leaked internal emails recently revealed that tobacco companies provided Australian authorities with intelligence, helped plan operations, identified targets and paid for surveillance technology to aid government efforts in illicit tobacco law enforcement.33 In statements responding to this finding, BAT, Imperial Brands and PMI, all defended such actions as being within the interests of the industry to assist the government to curb illicit tobacco.33 The prominence of tweets about illicit and counterfeit tobacco across all four accounts could be a reflection of the long-standing agenda of the industry to emphasise the issue of illicit tobacco in order to regain legitimacy among governments and a seat at the public policy table.

The type of content published by TTCs on their corporate Twitter accounts gives insight into priority or focus areas for TTCs, which may be reflective of current or future company strategies to regain public support and influence. PMI announced on 13 September 2017 its support for the newly formed non-profit research organisation, foundation for a smoke-free world, with a pledge of US$1 billion over 12 years to fund research to eliminate tobacco smoking globally.34 PMI’s use of its corporate twitter account appears to have aligned with this public identity shift as it has primarily promoted an image of being socially responsible (29.1% of PMI’s tweets) and a favourable workplace (45.7% of PMI’s tweets) and, compared with BAT, Imperial and JTI, rarely opposed or undermined tobacco control policies (9.6% of PMI’s tweets).

The current study reveals that TTCs are capitalising on social media by driving their key messages through Twitter. The tobacco industry is continuing with their long-standing strategy of opposing tobacco control policies and promoting CSR to shape its public identity. However, instead of this information being drawn from multiple channels such as court cases, annual reports, leaked tobacco industry documents and websites, decades of tobacco industry messaging is being published and promoted openly and freely through dedicated social media channels. As such, for the first time, the public is able to easily respond directly to TTCs within a public domain and engage with their content. The regulation of tobacco industry use of social media is urgently needed. Guidelines for implementation of Article 13 of the WHO Framework Convention on Tobacco Control (FCTC) outline that publicly promoting socially responsible business practices constitutes a form of advertising and promotion and contributions from tobacco companies to ‘socially responsible causes’ are a form of sponsorship.35 Parties to the convention, therefore, have a responsibility to legislate against tobacco industry promotion of CSR within their own jurisdictions. It is equally urgent for current gaps in the FCTC to be addressed to ensure tobacco industry use of social media is comprehensively regulated. While Article 13 of the FCTC includes cross-border advertising,36 the implementation guidelines are yet to be detailed in any operational way and must be made a priority.

Strengths and limitations

Likes and replies and retweets by other Twitter accounts of the content published by BAT, JTI, Imperial Brands and PMI were not included in the analysis. Such information could provide greater insight into the effectiveness of types of posts and the topics that gained highest engagement from audiences. The study also did not separately analyse the content of original tweets (ie, tweets published by the TTCs corporate accounts) and retweets (ie, tweets were originally published by other sources and retweeted by the TTCs corporate accounts). An analysis of retweets separately would give insight into the type of accounts that the tobacco industry is retweeting and the type of content that is most frequently retweeted. Another limitation of the study is that each tweet was coded only into one category, using a coding hierarchy, even if two or more topics were featured in the tweet. The proportion of tweets in some categories were therefore conservative. A strength of the study was the manual method of coding, as opposed to using a machine taught auto-coding system. This allowed greater discernment of tweets and it also allowed URLs to be followed and images to be viewed for tweets that did not have a clear central message.

Conclusion

BAT, Imperial Brands, PMI and JTI are actively using Twitter as a new communication platform to oppose tobacco control policy and shape their public identity. For the FCTC to be effective in curbing the influence of the tobacco industry, cross-border advertising guidelines need to be detailed and CSR activities should be more widely legislated against by parties to the convention. This is critical to ensure the FCTC’s application and enforcement is relevant, comprehensive and free of loopholes that the industry can exploit.

What this paper adds

  • The study is the first to analyse public messaging on British American Tobacco, Imperial Brands PLC, Philip Morris International and Japan Tobacco International’s own corporate social media accounts.

  • British American Tobacco, Imperial Brands PLC, Philip Morris International and Japan Tobacco International are all using Twitter with the primary purposes of opposing and critiquing tobacco control policy, promoting an image of being socially and environmentally responsible and portraying themselves as a highly favourable employer.

References

Footnotes

  • Contributors CW was responsible for the collection and analysis of results, interpretation of results and writing the journal paper. MH contributed to interpretation of results, reviewing and revising the paper. BF was responsible for the study design, interpretation of results, and reviewing and revising the paper.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.