Can internal tobacco industry documents be useful for studying the UK alcohol industry?

Internal tobacco company documents offer researchers unprecedented insights into the ‘black box’ of tobacco industry strategies to oppose evidence based health policies [1]. For those working in the field of alcohol policy, the absence of similar repositories requires researchers to identify alternative data sources for examining industry activities. For example, documentary analyses and semi-structured interviews have been used to study alcohol industry actors within the United Kingdom (UK) [2, 3].

Given the sheer volume of information available in the tobacco documents library, and various connections identified to exist between the industries (e.g. co-ownership), scholars have sought to use these documents to gain insights into the alcohol industry practices [4‐7]. However, this literature is limited, focusing principally on the co-ownership of Phillip Morris [PM] and the Miller Brewing Company [MBC]), and their activities in the US market. Bond et al. [5] found evidence of common regulatory concerns leading to collaborations between PM and MBC in developing political strategies. MBC feared increased taxation and additional regulation of their promotional activities, as well as negative public perceptions of their products, especially in relation to drink driving and ‘binge’ drinking [4]. To counter this, they promoted ideas of individual responsibility, which closely mirrored tobacco industry narratives [4]. As well as studying co-owned companies, Jiang and Ling [6] document how tobacco industry actors faced with a hostile policy environment have sought to build alliances with other industries, particularly alcohol, to oppose regulatory challenges in US tobacco policy that potentially also traverse sectors such as taxation and advertising restrictions. Dearlove et al. [7], meanwhile, catalogue tobacco industry attempts to recruit alcohol companies and the hospitality sector in the US to resist indoor smoking bans.

The UK alcohol industry experienced a process of significant consolidation in the second half of the twentieth century, especially up to 1990 [8‐10]; a period for which there are substantial holdings of tobacco industry documents. This means company operating names changed and may be unknown to researchers unfamiliar with the UK context. In the same time period, tobacco companies undertook a process of diversification into (and divestment out of) different industries, including alcohol [11]. Given the 70 million pages of documents on the tobacco industry released, it appears likely that existing studies have not fully exhausted the resources available to alcohol industry researchers. Moreover, no previous studies have sought specifically to examine the UK context, or indeed other national contexts beyond the US.

The current paper seeks to address this evidence gap by identifying what information may exist within the Truth Tobacco Documents library in relation to the UK beer industry. A scoping study offers an opportunity to investigate proof of concept and consider the potential for further studies of this dataset. This research may be particularly informative about the nature of relationships between the two industries and, by extension, how far further research on the alcohol industry may benefit from what is already known about the tobacco industry.

This study seeks to ascertain the scope of information available in the Truth Tobacco Documents Library on the UK beer industry, and thereby to consider the potential value of this data source for developing understanding of the alcohol industry. More specifically, we identified a priori a series of targets for data collection as being particularly informative about the nature of relationships between the two industries. We sought evidence on:

The methods for conducting tobacco documents research are well established (see [12]). We began with an initial keyword search of the Truth Tobacco Documents Library using the ‘Guided Search’ facility to search for brewing company names using the ‘Organization’ search tab. As our focus was the UK, a predetermined list of company names was developed from historical studies of the UK beer industry. The initial search terms were company names found in overviews of the UK beer industry contained in Gutzke [9] and Gourvish and Wilson [8]. All available libraries and collections present in the Truth Tobacco Library database were searched. No time parameters were imposed on the searches. Multiple variations of names were used (e.g. ‘Scottish and Newcastle’ and ‘Scottish & Newcastle’). Initial searches generated a large volume of documents, which were reviewed for relevance in an exploratory stage. In keeping with established methodologies [12], where larger numbers of documents were found, the returns were sorted by relevance and the first 60–100 documents reviewed, followed by a random selection of subsequent documents. Tobacco documents work requires a trade-off between the exhaustiveness and breadth of analysis, given the sheer volumes of material available in the database.

A decision was taken to focus our analysis in this preliminary investigation on the ‘Big-Six’ brewers which were dominant in the UK by the 1980s – Allied-Lyons, Courage, Scottish and Newcastle (S&N), Bass-Charrington, Grand Metropolitan (Grand Met), and Whitbread – and the predecessor and constituent companies from which these were formed (e.g. Newcastle Breweries, Watney’s). We also examined Guinness, which merged with Grand Met in 1997 to form Diageo, the world’s largest alcohol company at the time. In 1994, Allied-Lyons merged with Pedro Domecq S.A. for form Allied-Domecq, which was subsequently acquired by Pernod-Ricard in 2005. These mergers signal the breakdown of the previously clear separation between brewers and spirits producers, and the emergence of multi-category alcohol corporations, which now dominate the global alcohol market.

Searches were run iteratively to refine the search terms and excluding confounding terms. The optimum search terms for each company were decided, taking into account the volume of hits and to identify ‘saturation points’ (after which useful material is unlikely to be present) [12]. The final set of search terms used and the number of documents returned for each company are shown in Table 1 below.

All documents returned were then accessed and read. All types of document (e.g. letters, memoranda, annual reports) which related to policy relevant activities of the UK alcohol industry were included in the third and final stage of study. Where documents focussed on countries other than the UK (principally the US), these were included as potentially offering wider insights into the modus operandi of these companies, which are increasingly global in character [13]. Duplicate documents and ‘restricted’ documents (which are listed in search returns but cannot be opened and read as the contents were deemed privileged) were included in the number of search returns but were not examined. In total 169 documents were included in the data analysis. The pdfs of included documents were downloaded, saved and read in detail. Information relating to each company was extracted and tabulated. Data analysis involved summarising the nature of the information available in relation to each of our a priori targets for data collection. As these documents are all in the public domain, ethical approval for the study was not required.

The study findings are presented in each of the target data collection areas identified a priori for study, with the exception of shared facilities where we did not identify any relevant data. We found information on shareholding across industries, as a well as more formal and extensive co-ownership. As the former implies a different type of relationship to the latter, we present these two separately. Where necessary for clarification, details of third parties (i.e. companies, associations, membership organisations) are given in parenthesis. A limited amount of material contained within the analysed documents on the activities of the alcohol industry (principally to do with marketing strategies) is unrelated to connections with the tobacco industry, and is not presented here.

This study considers the possible value of the internal tobacco company documents during an important period in the evolution of the emerging global alcohol industry. The documents examined show evidence of significant inter-connections between the alcohol and tobacco industries. Most obviously, this occurred in the form of various instances of co-ownership between sectors, which led in turn to integration of company functions and cross fertilizations of corporate strategy, also aided by common directorships across industries. Corporate cultures, practices and strategies developed within tobacco companies, or through close association with tobacco divisions within alcohol companies, may have diffused widely through the alcohol industry (and vice versa) as suggested by previous studies of the alcohol industry [3, 13, 26, 27]. The specific substance of the interconnections between industries both in the UK and beyond warrants further attention in studies of alcohol industry strategies to advance interests and shape policy.

In addition to the key findings on co-ownership, there are three further observations that can be made about these inter-connections on the basis of this study. Firstly, there are strong informal contacts between industries that serve as the foundation for an array of reciprocal help-seeking activities. These function in similar ways to the cultivation of long-term inter-personal relationships by the alcohol industry with policy actors previously identified [3, 21]. Moreover, these forms of collaboration appear likely in areas of common commercial interest. This finding mirrors those of previous studies [6, 7]. Secondly, the relationships between the alcohol and tobacco sectors needs to be considered within the broader context of cross sector corporate collaboration [28‐30], including membership of third party bodies, attendance at the same meetings and events, and the use of common professional service providers. Thirdly, BAT appears to be particularly prominent and important to study carefully in future investigations of cross industry connections.

This study adds to the existing proof of concept. The Truth Tobacco Documents Library represents a valuable and under-utilised data source for scholars seeking to understand the historical evolution of the alcohol industry, and so to locate the current activities and strategies employed by alcohol companies in their historical context. The tobacco documents offer a resource for examining individual alcohol industry actors and organisations with no previously established connections with the tobacco industry.

The insights presented and discussed here may be further developed through triangulation with other data sets. Previous experience with tobacco documents research underlines the importance of ‘snowball’ techniques in the identification of additional search terms to identify further relevant documents. Further searches may also focus on product brands, for example, rather than company names. Similarly, searching consecutive Bates Numbers for relevant documents may uncover further useful texts.

This study focussed largely on the UK beer sector. It is likely, therefore, that additional information on UK spirits companies is contained within the Truth library. Similar studies on alcohol companies in other countries should be fruitful to undertake, though we make this suggestion without implying that further research should take place only at the national level. This study scopes available information rather than analysing it in depth. Our study was also circumscribed in terms of the data we sought, focussing on connections between tobacco and alcohol companies. As with all studies of the tobacco documents, the insights derived here are of a largely historical nature with the most recent documents identified dating from the early 2000s. This tobacco company research resource offers, nonetheless, the capacity to develop new insights both of a historical nature and with clear contemporary relevance into the alcohol industry.

There is highly relevant material in existence in the Truth Tobacco Documents Library on the alcohol industry, which may provide a valuable data source for researchers interested in better understanding the alcohol industry and its relationships to the tobacco industry. Much of the data we have examined pertains to the period leading up to the consolidation of the global alcohol industry, and offers insights into developments in this key period, which have important health implications for understanding the political significance of the industry today. This preliminary study has identified data that can inform decision-making about future studies. It remains to be seen just how far further research on the alcohol industry may benefit from what is already known about the tobacco industry.