* Scepticism in CSR Advertisements

The aim of the present paper is to make a contribution to CSR communication theory by investigating the level of substantial information provided in CSR print advertisements in the UK and Brazil. The paper evaluates CSR advertisements using textual analysis. The empirical evidence demonstrates that companies provide a low level of substantial information when advertising CSR. The paper provides empirical evidence as to the extent that corporations use substantial information in their advertisements.

theory by investigating the level of substantial information provided in CSR advertisements in the context of two countries: Brazil and the UK.
The rationale behind the choice of these two countries was to ascertain whether the level of development of CSR awareness and visibility in a country has an impact on CSR communication. Therefore, a developed country recognised as one of the leaders in CSR was chosen, namely the UK. Similarly, Brazil, a developing country which has a recent history of CSR was selected. Developing countries face various obstacles to achieving corporate responsibility, mainly because the institutions, standards and systems -which are the foundation of CSR in Europe and the USA -are comparatively weak (Kemp 2001). On the other hand, civil society can stimulate CSR by generating greater societal demands and expectations regarding business responsibility (Chapple/Moon 2005).
The paper is organized as follows: Section 2 examines CSR communication. Section 3 describes the methods for collecting, selecting and analysing the advertisements. The analysis of the advertisements is in Section 4. Finally, the paper concludes with implications for practice and further research.

CSR communication
CSR communication may not capture the complete reality of CSR; it represents the image the company wants to portray (Alon et al. 2010). Thus, communication can be used as a signal of corporate intentions. CSR communication provides an effective tool for promoting the company's engagement in CSR to various stakeholders as well as enhancing the corporate image (Bondy et al. 2004;Logsdon/Wood 2005). Companies communicate their CSR programmes and actions in order to achieve a positive public image, and to gain legitimacy and support from different stakeholder groups (Adams et al. 1998).
Although companies use a wide range of channels for CSR communication (social reports, thematic reports, codes of conduct, websites, stakeholder consultations, internal channels, awards and events, cause-related marketing, product packaging, interventions in the press and on TV and sales points), three channels in particular seem to play a prominent role: social reports, web sites and advertising (Birth et al. 2008). Advertising is the least researched among the three of them (Farache/Perks 2010).
However, for consumers and the general public, external media coverage has greater credibility than communication that the corporations themselves produce (Goodman 1998). Using a MORI (Marketing and Opinion Research International) survey of the British public, Dawkins (2004) found that people attach more credibility to information coming from non-governmental organisations (NGOs) than from company sources. Du et al. (2010) argue that a challenge to an effective CSR strategy is how to reduce stakeholder scepticism and generate a favourable CSR image. To enhance the credibility of CSR communication, and thus reduce scepticism, communication must be informative, educational, avoid the use of an emotional tone and provide a website or contact information that the consumers can use in order to gather further information on CSR actions (Azevedo 2004).
Further, Schlegelmilch and Pollach (2005) and Jahdi and Acikdilli (2009) identify factors that can lead to success in CSR communications. They are: source credibility of the communicator, reliability and honesty of the statements, and the involvement of the audience with the topics being communicated. Aspects that can enhance source credibility are ethics awards, evidence of contributing to NGOs, and news coverage (Hirschland 2006). On the one hand, it is important to communicate CSR, but on the other, excessive CSR communication may have a negative effect, as corporations that communicate intensively tend to be criticised by pressure groups and the media for exploiting their own social and environmental actions (Tixier 2003;Morsing 2005;Schlegelmilch/Pollach 2005).
In order to mitigate scepticism, Morsing et al. (2008) suggest two processes when communicating CSR to a variety of stakeholders who want to know about a corporation's initiatives but are cynical with regard to CSR statements: the expert and the endorsed communication process. The expert process communicates CSR through facts and fi gures. It is directed at politicians, local authorities, NGOs and journalists, who are already familiar with CSR issues. It is expected then, that this expert group will disseminate a favourable image of the company. On the other hand, the endorsed CSR communication process recognises that communicating to an exclusive group of stakeholders is not enough for legitimacy, image or reputation purposes, and attempts to communicate CSR through third party experts. Endorsed CSR communication is perceived to be fundamental if a company is to avoid being labelled as self-serving or self-complacent. Therefore the challenge of communicating CSR is managed by balancing both the expert and endorsed CSR communication processes (Morsing et al. 2008).
CSR communication remains an emerging fi eld in academia, with research focusing on corporate social disclosure mainly through websites and corporate reports, while little is known about CSR print advertising. Even the term 'CSR advertisement' is not widely used, but Schroder defi nes such adverts as those that proclaim: "...a social ethos as they inform about a company's commitment to environmental concerns, community relations, or the future of mankind, without any overt attempt to promote a specifi c product" (Schroder 1997: 277).
Paradoxically multinational corporations are the recipients of growing distrust (Verschoor 2008) while at the same time, they are increasingly using CSR advertising campaigns (image advertising campaigns based on CSR initiatives) to publicise their impact on society (Becker-Olsen et al. 2006). These CSR campaigns attempt to disseminate a positive corporate identity that aims to infl uence corporate image (Van Rekon 1997).
The benefi ts of CSR advertising are that it enhances and protects reputation and can help to build morale among employees, attracting the best staff at both entry and senior levels (McDonald et al. 2001). Pomering and Johnson (2009) acknowledge CSR advertising as a tool for promoting a desired corporate image, although they also recognise that CSR advertising may lead to consumer scepticism.
Advertising receives a fair amount of criticism regarding, for example, the way it increases consumerism, how minorities are portrayed, and how messages are delivered in accordance with the sender's objectives. The manipulation and the effects of advertising have also been widely discussed by a number of authors (Dyer 1995;Leiss et al. 2005). Similarly, corporations are under constant surveillance regarding the way they behave in developing countries, the suppliers they work with and the way the workforce is treated (Julholin 2004). Likewise, CSR actions are often perceived as merely marketing or PR tools. Banerjee defi nes CSR actions as "narrow business interests [that]… serve to curtail interests of external stakeholders" (2008: 51). In this scenario, CSR advertisements face a considerable amount of criticism related to the nature of advertising and the behaviour of corporations.
In order to avoid these negative interpretations and increase credibility in their advertisements, companies should provide substantial information on the programmes or actions they are publicising, such as investments, benefi ciaries and possible outcomes. Corporations should communicate CSR together with the necessary support in case questions about the veracity of the statements arises (Polonsky/Jevons 2009). Azevedo (2004) suggests that communication must be informative and provide a website or contact details which consumers can use in order to gather more information about CSR actions.
Truthful information is also expected in an area that already receives an increasing amount of public scrutiny (Schlegelmich/Pollach 2005). As Van de Ven argues "… accuracy should be the basis for every assertion in CSR-related corporate communication" (2008: 348). In order to increase the credibility of CSR communication, the literature suggests that companies should as-sociate themselves with third parties, such as NGOs, the government and associations (Azevedo 2004;Morsing 2005;Schlegelmilch/Pollach 2005). When communicating their association with NGOs, companies intend to transfer credibility from the NGOs to themselves as people have a higher level of trust in NGOs than in corporations (Dawkins 2004). This holds true in both the UK and Brazil (Akatu 2006).
Association with third parties can also be understood as what Morsing et al. (2008) call endorsed CSR communication, avoiding the self-serving label that comes with CSR communication. Similarly, when a company discloses its CSR actions and links itself with an NGO, government or association, the reader perceives the corporation as more socially responsible in its activities. Therefore, it is understood that the third parties associated with the corporation are in agreement with its actions. The present paper investigates the level of substantial and third party association information provided in CSR advertisements in Brazil and the UK.

Method
Over a period of a year, the authors gathered CSR advertisements from four magazines: Veja and Exame in Brazil and The Economist and Time Magazine in the UK. These are a mix of weekly news and business magazines. Although the advertisements were collected from two very different countries, the readerships of these magazines are quite similar: well-educated, well-informed readers from the upper social classes. All the magazines selected were leaders in terms of circulation in each country at the time of the data collection. They are also comparable in terms of type and are recognized as infl uential in each research context. Weekly news and business magazines were chosen because they enable coverage of a wide readership, reaching not only those interested in economic issues, but readers of the news as well.
Veja was chosen because it is the leading weekly news magazine in Brazil, with an average circulation of more than 1 million copies in 2007 (IVC 2007). It is the largest Brazilian publication in terms of circulation, advertising, billing and subscribers. This places Veja among the largest international publications. In fact, it is the fourth largest weekly magazine worldwide in terms of circulation, just below Time US, Newsweek US and US News. The magazine has 58% of the market share among weekly informative periodicals in Brazil (IVC 2007). It covers politics, economics and world events as well as art and culture. Veja reaches on average 6,973,000 readers annually (IVC 2008).
Exame is a leading business magazine in Brazil, with a circulation of 180,000 copies in 2007 (IVC 2007). The magazine is directed at executives and reaches 70% of them in Brazil and is the leader in the business segment with circulation double that of all its competitors combined. The magazine is published every two weeks and has approximately 899,000 readers annually (IVC 2007).
For the UK, The Economist was chosen, which is considered to be Britain's most successful political and business magazine, with a weekly circulation of more than 180,000 copies in the UK and approximately 670,000 readers in 2007 (ABC 2007). Although The Economist was originally produced for the UK, it has become a global publication with more than four-fi fths of its circulation outside of the UK. Every week, over a million copies of the magazine are sold through subscriptions and at news-stands in over 200 countries. This demonstrates the relevance of the magazine to opinion leaders worldwide.
Time Magazine, Europe Edition, was selected because it is a leading weekly news magazine in the UK, with average circulation of 135,000 copies in 2007. The UK has the highest circulation fi gures among all countries, accounting for more than 25% of the European Edition, which is 525,000 copies, including Europe, the Middle East and Africa (ABC 2007). The magazine covers politics, culture, economics,business, lifestyle and sports and reaches a readership of approximately 500,000 people in the UK.
The authors selected the three companies that advertised the most in each country: Banco Real, Banco Bradesco and Unilever in Brazil; Chevron, Shell and Total in the UK. In order to select the advertising campaigns to be analysed, the same rationale applied to the selection of the advertisements was used. The more prominent campaigns placed the greatest number of times were selected. Following this rationale, six CSR advertising campaigns, made up of 26 advertisements in total, were selected.
To determine the amount of substantial information provided in each advertisement, we took an essentially qualitative approach to the analysis of the data. The method of analysing the advertisements was systematic, assessed and coded for low, medium or high levels of substantial information. This relates to information such as investments, number of people benefi ted and the CSR programme. Associations with third parties, such as non-governmental organisations, government, and partnerships with associations and other businesses, were also analysed.

Findings
The empirical evidence demonstrates that four companies in the sample tend to disclose a low amount of substantial information. For instance, a Brazilian bank addresses the issue of using recycled paper in its cheque books, although it does not provide supporting information such as the amount of paper being used, the number of trees being saved, the investment made by the bank or how this action will affect the environment. The three advertisements have the same copy, the content of which is displayed below.
Environmentally friendly cheque books. When you use them, you are saying what you expect from a Bank and from your life. And you are contributing in one of many ways to the preservation of our natural resources. Banco Real uses recycled paper or white paper produced in an environmentally friendly manner for 100% of its materials. Banco Real also gives you the opportunity to choose what type of paper you want in your cheque book. For further details, visit www.bancoreal.com. br/sustentabilidade. (Banco Real,'Environmental cheque book') Shell follows the same pattern and displays a low amount of substantial information. Interestingly, all the advertisements in their Employees campaign provide a numerical fi gure that does nothing to enhance the content of the information. The company is making use of the artifi ce of providing data that is not related to its impact on the environment. A selection of extracts illustrating these fi ndings is provided below.
Hillary Mercer is helping to build a new energy future in the region. She manages a 7500-strong workforce constructing Sakhalin II, one of the world's largest integrated oil and gas projects, on Sakhalin Island in far-eastern Russia. (Shell, 'Hillary Mercer') So, as demand for cleaner energy grows, projects such Sakhalin II, one of the world's largest integrated oil and gas projects, are ever more important. Although the plant is still being built, more than 75% of its future capacity has already been sold. (Shell, 'Oleg Smirnov') In its campaign Our energy is your energy, Total also fails to present any signifi cant substantial information. Information related to the renewable energy being advertised is not provided. The same occurs with the notion of oil exploration while protecting marine life and reducing energy consumption. Excerpts from the advertisements demonstrating this fi nding are given below.
Extending the life of oil also means reducing energy use in industry -something we're already doing in our own refi ning and petrochemicals facilities. (Total, 'Looking after a scarce resource') Tomorrow we're aiming for 6,000 metres down. But even there, we're not alone -respecting and protecting marine life is a constant priority during the lifetime of our projects. (Total, 'You can look for oil at great depths') Because the world is going to need alternative energies and because exploration is our core business, Total is developing next-generation wind turbines. (Total, 'Wind, one of the most natural ways to move forward') Unilever, Bradesco and Chevron provide more substantial information in their advertisements. Chevron, for instance, gives a great deal of substantial information in all three of its examples. The company supplies a large amount of text, discussing in depth the issues proposed. The information addresses the steps the company has taken, the amount of money being invested and its objectives. Extracts are printed below.
Since 1992, we have made our own energy go further by increasing our effi ciency by 24%. (Chevron,'There are 193 countries') Helping the U.S. government save taxpayers $151 million while reducing greenhouse gas emission by an expected 1.5 million tons. (Chevron, 'A 5% reduction') Spending more than $1 billion over the next several years on next generation, ultra clean diesel fuel from natural gas. (Chevron, 'Russia, Iran and Qatar') Although the level of information from Bradesco varies, the company is considered to provide a high amount of substantial information. Three out of four advertisements display high volumes of information regarding investments made by the company, number of people benefi ted by its programmes, the actions performed, donations, its projects and its partners. Examples are provided below.
There are more than 108 thousands students -children, youngsters and adults -benefi ting in 40 schools around all the Brazilian states and the Federal District, in an investment that totalled more than R$157 million in the last year alone. (Bradesco, 'The biggest investor in education') The 1.6 million actions were performed at 40 schools of the Bradesco Foundation throughout Brazil as well as at 149 other sites, such as public schools and community centres, expanding the number of people reached. Another surprising fact: we collected approximately 30 tonnes of products, ranging from computers to food, which were distributed to 88 registered institutions. (Bradesco, 'The Bradesco Foundation performed') The amount of substantial information provided by Unilever varies across its fi ve advertisements. For instance, three were categorised as offering low, one medium, and one a high volume of substantial information. A few examples are provided below to illustrate this categorisation. Advertisement providing a low volume of substantial information: …it is to bring vitality not only with quality products, but, above all, with a business model that exhibits social and environmental responsibility. (Unilever, 'How Unilever is moving') Advertisement providing a medium amount of substantial information: There are currently 37 sports centres, with 48 thousand children benefi ted. (Unilever, 'Our team wins') Advertisement providing a high volume of substantial information: 400 schools and 2800 teachers a year take part in the project. 10 rooms were also created for recreation in hospitals, benefi ting approximately 1 million children since the beginning of the project. (Unilever, 'With how many sticks…?') To summarise, three companies present a low amount of information, one provides a mix of low, medium and high levels and only two present a high volume of substantial information in their advertisements (Table 1).
Regarding the association with third parties, among the six companies analysed, only three reported their associations with other entities. Chevron and Unilever highlight their partnerships with government agencies and other businesses, as can be seen in the excerpts below.
Helping the U.S. government save taxpayers $151 million while reducing greenhouse gas emissions by an expected 1.5 million tons. (Chevron, 'A 5% reduction') The project is made in partnership with City Hall as well as municipal and guardianship councils that monitor and denounce abuses. In a joint action, we take children from the fi elds and place them in their proper place: school. (Unilever, 'We want to help change') Their efforts and the partnership between the city, Unilever and other companies have brought surprising results, which refl ect improvements in the areas of health, education and the economic development of the city. (Unilever, 'The best proof') Bradesco is the only company to associate itself with NGOs. The advertisements 'The biggest investor in education' and 'Another piece of our socio-environmental work' inform the reader regarding its partnerships with the NGOs Fundação S.O.S. Mata Atlântica (Atlantic Forest Foundation) and the Ayrton Senna Institute. The former works for the protection of the Atlantic Forest and the latter is related to the development of Brazilian children and young people. An extract from Bradesco's advertisement illustrates the association. In summary, only three of the companies surveyed report their associations with other institutions in their advertisements (Table 2). Therefore, half of the companies do not make use of this strategy, which, according to the literature, would give more credibility to what is being advertised (Morsing 2005;Schlegelmilch/Pollach 2005;Azevedo 2004 The evidence demonstrates that half of the companies tend to disclose low volumes of substantial information. In one of the cases, all the advertisements provide a fi gure that does nothing to enhance the content of the information. The company provides data that is not related to its impact on the environment. On the other hand, the other three companies provide more substantial information in their advertisements, although the level of information varies. Among the six companies analysed, only three reported their associations with other entities.

Further research and practical implications
The present paper investigated the levels of substantial information provided by corporations in relation to CSR advertising. While scholars are aware of the possible pitfalls with advertising focusing on CSR, corporations need to be more careful and transparent when making use of it.
Half of the companies in the sample fail to follow the advice of the literature on CSR communication, which recommends the use of high volumes of substantial information when publicising CSR actions (Polonsky/Jevons 2009;Schlegelmilch/Pollach 2005). Similarly, only half of the companies surveyed report their associations with other institutions in their advertisements. According to the empirical evidence, the advertisements published by corporations do not seem to address the concerns raised by some authors (for example, Morsing et al. 2008;Birth et al. 2008;Pomering/Johnson 2009) regarding scepticism. The way corporations develop their advertisements only provide more fuel for the growing concern regarding the lack of transparency and credibility in CSR communications, specifi cally in advertising. Interestingly, the research found that corporations in Brazil tend to provide a higher level of substantial information than corporations advertising in the UK. This fi nding contradicts the literature on advertising content which specifi es that material from developed countries tends to contain more information than that from developing countries (Abernethy/Franke 1996; Kaynak/Ghauri 1986). More research is needed to determine to what extent Brazilian advertisements can be seen as more informative than British ones. Further research could be carried out with a larger sample in order to provide more evidence, and also to investigate whether the amounts of substantial information and third party association are related to industry type. This paper does not claim that companies which provide high volumes of substantial and third party association information are not subject to scepticism about their CSR messages. The suggestion made here is that in order to increase the credibility of CSR advertisements and attempt to inhibit scepticism, corporations could increase the levels of substantial information in their CSR advertisements. Association with third parties, such as NGOs, could also infl uence perception of the communication as trustworthy. The provision of high levels of substantial and third party association information seems to be a good starting point for communication communicate CSR engagement in a credible way.