«13TH INTERNATIONAL PUBLIC RELATIONS RESEARCH CONFERENCE “Ethical Issues for Public Relations Practice in a Multicultural World” Holiday Inn ...»
In fact, we find that the coverage of a given company is positively and significantly correlated with past day absolute return. There is also some evidence that this relationship is systematically steeper the larger the amount of ads being purchased by that company on that newspaper. However, when distinguishing between positive and negative returns, only the interaction of ads expenditure with positive returns is mildly significant. This latter result is quite robust to the more demanding specification described above, namely controlling for (company × newspaper) fixed effects. In other terms, newspapers appear to be reacting more strongly to (positive) company-specific newsworthy events, the larger the purchases of ads by that company.
Our results are also related to some recent literature on the links between media coverage and the behavior of financial markets. For example Barber and Odean (2008) show that individual investors appear to be net buyers of stocks being featured in the news, irrespective of the positive or negative tone of the news itself. If this is the case, our results point to an interesting synergy between the marketing and the investor relation department within listed companies: according to our findings, when a company buys ads on a newspaper, it also buys some additional attention of the newspaper to newsworthy events, which can induce readers as investors to be more willing to buy shares of that company. This in turn could translate in higher stock prices in equilibrium, i.e. a lower cost of capital for the firm (Fang and Peress 2008). 5 On top of the contributions on advertising & media coverage and on media & finance, our paper is also related to the comparatively larger literature on political media bias (Groseclose and Milyo 2005, Gentzkow and Shapiro 2009, Puglisi 2006, Larcinese et al. 2007, Puglisi and Snyder 2008, Ho and Quinn 2008, Knight and Chiang 2008, Durante and Knight 2009).
The paper is organised as follows. In section 2 we provide some background on the activities performed by firms’ PR departments and on the interactions of those with media organizations. In section 3 we describe the dataset, while in section 4 we present our main results, and in section 5 we perform some robustness checks. Section 5 concludes.
The internal organization of newspapers
The fact that a company and/or its products are covered by the media creates informative externalities. Those externalities (of a positive or a negative sign) can arise both in the product and in the financial market. If this is the case, the company in question can find it optimal to engage in public relation activities, in order to enhance or to influence editorial coverage.
To investigate this information dissemination activity, and its interactions with advertising and news selection, we have carried out 20 personal interviews with the different stakeholders involved (PR agencies, PR executives, newspaper editors in chief, publishers, newspaper sales executives).
Companies value editorial space because (i) it is cheaper than advertising space and (ii) the news selection process carried out by the editorial team can attribute to a piece of news a degree of relevance and credibility that are difficult to obtain only through the means of advertising.
5 Peress (2008) investigates the link between media coverage and the extent of the earnings announcement drift, i.e. the predictability of stock returns after earnings announcements.
335 Even if they usually claim a more elaborate communication support, PR agencies routinely evaluate their output by collecting the articles they have obtained for a particular client, and multiplying the obtained position-weighted space by the appropriate price that one would pay for advertising that covers the same spaces and positions. This sum must be lower than the agency fee as a bottom line for efficiency.
Since the publication of a press release is free –conditional on the fact that the newspaper staff selects it-, companies compete fiercely on the intermediate information “market” where they can obtain valuable editorial space. The journalists we have interviewed, both in newspaper and magazines, declare to receive on average between 20 and 40 press releases on a daily basis, from which they pick up 1-2 news/articles per day. Since the same press release is sent to several journalists working for the same outlet, a conservative estimate of the publication rate on newspapers is around one article for every ten press releases. In specialized magazines, which are typically more focused on a particular topic, and where the matching between company disclosure and editorial interest is easier, the publication rate is higher, usually around one news every 3 or 4 press releases.
But this dissemination activity is important for media companies as well, since it reduces the costs to gather and verify the information to be published. In a typical newspaper, around 60% of published stories originate from some sort of public relation activity performed by firms and organizations.
Moreover, following the enlargement of topics covered and the growth in the number of pages, this share has been steadily growing over the last twenty years (Gambaro 2007, Boyd Barret 1992).
According to Upa, in 2007 Italian companies spent 2013 million euros on public relation activities, and the growth rate in the previous 6 years has been more than double than the one for advertising on newspaper and magazines. This amount represents about 0.15% of GDP and around 10% of the total communication expenditure. Obviously public relations include other activities than media or investor relation, which anyway represent around half of total expenditure.
In Italy Public Relation officers, both internal and outsourced, are estimated in the range between 20,000 and 50,000 and they confront, or try to influence, 10068 journalists that are employed in the Italian press (2007 figure).
The newsgathering activity performed by a newspaper can be suitably described within a principalagent setting, whereas the newspaper acts as an agent that selects news on behalf of its (collective) principal, i.e. the readers. The principal cannot perfectly observe the actions of the agent nor the intermediate market, which is fed by news agencies and press releases. This is the market where the newspaper on a daily basis would pick up the pieces of news to be published.
In a nutshell, the editor-in-chief selects the more relevant news and chooses the level of effort needed to study, check and write them down. When doing so, he would follow some sort of editorial line that is welcome to the readers.
A certain amount of effort and costs would on average translate in a given amount of precision, appropriateness and completeness for each published piece of news.
The maximum effort is given to the more relevant stories of the day, but on the margin there are several combinations of effort and relevance that can offer the same output to the reader. From this point of view, the editor can select the stories to be published on the basis of some type of portfolio considerations.
When the piece of information is related with a company or its products, a story-specific investment in disclosure and dissemination (press release, photos, contacts) can lower the cost for the newspaper to produce that news and therefore modify the rank of final relevance.
This is in some sense a grey zone, because it is unclear whether the PR activity performed by firms can be considered as a clear-cut instance of media capture (Besley and Prat 2006). Indeed, as discussed above newsworthiness and precision-enhancing effort must be jointly taken into account and they are both valuable to the reader, so that there is not an ex ante definite ranking of potential stories. On the other hand this is an equilibrium phenomenon, in that a company has no reason to engage in a costly activity if the news regarding it is published anyway.
336 In this setting the focus of the pushing company is on the (potentially) free editorial space, which must be compared with costly adverting, and the informative value given to the published story by the selection process itself.
There are circumstances where the company tries to soften or to erase a bad news. This can only occur in a dynamic setting, i.e. when the PR executive has a stable relationship with the journalist.
Over time she gives the reporter exclusive news or valuable information; once she has credit, she can ask for some deviation from the editorial selection standard. The rational reporter must consider the actual value of the relationship over the future years and trade it off against the intensity of slant which is required on that particular occasion. Private incentives might play a relevant role here: for example, some valuable information can improve the reporter’s position and increase her salary.
A third typical situation is when a company is involved in a rent seeking activity. In this case the publication of a story containing the appropriate information is only a part of a larger process that usually includes (i) the hiring of some academics that would draft a piece of research matching the company goals and (ii) inducing the right person (usually public officials or decision makers) to read and pay attention to the obtained article. The information and the piece of research must be of interest for a large chunk of the readers in order to get published, but the only readers that matter to the company are those that ultimately decide about the rent.
While there are several qualitative and anecdotal papers on the relationship between public relation officials and journalists, and its effects on media output, only recently have more rigorous and quantitative studies emerged, mainly in the financial sector.
Bushee and Miller (2007) study a sample of 184 midsized companies that hired Investor Relation firms, and find that they have significant increase in disclosure, press coverage and trading activity.
In the three setting described above the company can leverage on its advertising expenses over that particular newspaper both with the carrot of spending more, and with the stick of spending less (or completely withdrawing the advertising). While PR practitioners generally claim not to engage in this practice, this is perceived as a problem by journalists. In fact, media organizations adopt several institutional arrangements to deal with this problem.
Other things being equal, this leverage is more powerful when the advertising revenues per copy are larger, the advertising client are more concentrated both in number and by sectors, and when the publication is weaker.
When there is an exchange between advertising and coverage a newspaper can lose copies but the increased advertising can more than compensate this loss (Di Tella Franceschelli 2009).
From the company perspective, private incentives of top managers can play a significant role. Since the information ends up being mixed with entertainment and being personalized in stories and adventures, top managers receive an extra media exposure and can transform the company investment into private benefits such as salaries, stock options and future positions (Nguyen 2005).
Following the seminal work of Reuter and Zitzewitz (2006) some papers have studied the interaction between advertising and coverage.
Gurun and Buttler find that on U.S. newspapers there is on average a more positive slant in articles about local companies (as identified by the distance between the newspaper’s and the company’s headquarters) than about non-local ones, and they present evidence that this slant is linked with local advertising expenditure.
Adverting can influence the editorial line of a newspaper also in more indirect and general way.
Gabszewicz, Laussel and Sonnac (2001) propose a model where in order to get larger advertising revenues the publisher induces the editors to moderate the political message they convey to their readers, so that more readers may buy the newspaper. In this process of convergence the variety of political opinions available to the reader gets reduced.
The same line of reasoning is developed by Strömberg (2004): his political economy model delivers the result that media would find it optimal to give more coverage to topics that interest larger and richer social groups.
337 By the same token, in the case of European public televisions advertising revenues push the programs toward a growing commercial orientation (Gambaro 2005).
The interaction between advertising and editorial coverage depends both on the structure of the advertising market and on the internal organization of newspapers and advertisers.
There are differences between magazines and daily newspapers. In the former there are typically close links between the advertising department and the newsroom, up to the point that sometimes – and of course unofficially- sales department distribute on a weekly or a monthly basis the list of advertisers that should be covered by journalists. On the other hand, those links are typically milder in newspapers, where heavy advertisers rarely happen to be essential sources of newsworthy stories like in specialized magazines. In our interviews, both advertising executives and journalists assert that direct requests of quid pro quo are relatively rare in newspapers, while they are common in magazines.
In 2007 Italian newspapers collected 1702 million euros of advertising, which represents around half of total revenues. This figure is in line with other European countries, while in the U.S. advertising share is generally 80-85% of the total.
Geographic composition is also different. In Italy only 45% of the advertising is local, while in several north European countries the share is around 60%. The U.S. are on the opposite side of the spectrum, since only 20-25% of the advertising is national.
Our paper contributes to analyze the impact of advertising and public relation on media coverage, with a focus on listed companies. Daily returns and trading volume can be considered as the revealed opinion of the market about the value of the news that the subsequent day can be published on the newspaper. The sign of the return give also a rough but measurable indication of whether the news environment on a given day about a given company is positive or negative.