Abstract
The financial crisis that began in autumn 2008 has attracted considerable attention in regard to the role of the media. This article examines both the audience and the content of the coverage of the crisis on the BBC News website, the largest online news provider in the UK. It demonstrates that online news was a significant part of the overall media coverage of the crisis. Online consumption patterns are very different from those of other media, but the claim that online audiences are ‘dumbed down’ or that they were not provided with a sophisticated range of information and analysis is critically examined. The study also questions whether the content of news coverage was as negative as has been suggested. The research is based on unique access to the BBC News web server-logs, which allow researchers to track audiences not only for the online site as a whole, but also for individual stories, and to match that to content analysis. It makes an important contribution to providing evidence-based research to examine the competing claims that have been made about the role of the business media in the financial crisis.
Introduction
This article seeks to examine evidence relating to the role of the media in the critical phase of the financial crisis by focusing on the BBC news website, the largest provider of online news in the UK, and a source of data both on audience and on content. The article examines the size of the online audience for business news, the types of stories that the audience viewed, and how the tone and content of the coverage changed over time.
During the crisis, the financial media was accused of failing to warn the public about the potential risks, and therefore abdicating its public responsibilities to investigate wrong-doing. Other commentators worried that the media coverage of the crisis had caused panic among the public and investors alike, worsening the crisis.
As the second most viewed source of news, online news is at the heart of this debate. The fastest growing form of news consumption, it can provide instantaneous news and live data that bypass official sources, and may get immediate reports from ground level, making it a potent weapon for spreading rumours and fear.
The research covers the period between September and December 2008, and benefits from unique access to the two BBC databases that track the log-server files which have audience details for every one of the hundreds of BBC News stories that appear on the website each day.
The findings suggest that the online audience was not just interested in news, but also sought out further information about the crisis, and in particular turned to the investigative insights of the BBC’s Business Editor, Robert Peston, whose daily blog became popular and influential.
Content analysis is the other tool used to look at how the website covered the crisis. This suggests that the framing of the debate did not always accept established views, although the BBC coverage did tend to report the views of policymakers and politicians more than bankers.
The article discusses the claim of negative media effects sceptically, finding that the BBC output on balance showed an even-handed approach, and used relatively restrained language to describe the crisis.
The BBC coverage is not typical of how all the UK media covered the crisis. However, the size and weight of BBC News, and its position as the most trusted news source in the UK, does mean that its coverage is an important benchmark in judging the performance of the UK media as a whole. The study also throws light on the difference between online news consumption and other traditional media outlets (Schifferes, 2006, 2011b).
Some early internet theorists suggested that online news would be produced directly by users (citizen journalists) and there would be no need for professional journalists or mainstream media organizations (Gillmor, 2006). Others worried that consumers would only select the news that interested them, leading to a fragmentation of news services (Negroponte, 1996; Pariser, 2011; Sunstein, 2000).
This study suggests that in times of crisis, online news consumers return to trusted sources and professional journalists, with little evidence of increased interactivity.
There has been much criticism of the role of these financial journalists during the crisis. This article suggests that some of this criticism may be misplaced, with journalists making critical decisions under extreme time pressure which broadly reflected the information available at the time.
Literature review
Covering financial crises
There is a considerable literature on the role of the financial media in covering financial crises, but its application to the current crisis is still in the early stages.
Much recent discussion has examined the question of whether the media failed to identify the financial crisis in advance. Detailed research by Starkman (2009) suggests that the majority of coverage by major US newspapers from 2003–7 was positive rather than sceptical about the boom which preceded the crisis. Roush (2011), however, argues that the media did produce a significant number of articles warning about the dangers, but they were ignored by policymakers and the public. Other research (Marron et al., 2010) suggested that the UK and Irish media mainly reported on the crisis episodically, and failed to exercise scepticism about what was done by those in political and economic power, a view echoed by Schechter (2009).
Research on earlier financial crises also focused on the role of the professional journalists. Writing about the 1929 Wall Street crash, Carswell (1938) suggests that reporters were inadequately trained and too dependent on experts, a theme echoed by Loeb (1966) and Welles (1973), as well as by Stein (1975), who writes from the point of view of a policymaker. For more recent crises, a sceptical view is espoused by Jensen (1987) in relation to television news, by McKean (1992) for the US savings and loan scandal, by Durham (2000) for the Asian financial crisis, by Goozner (2000) and Henriques (2000) for the dotcom bubble, and by Sherman (2002) for the Enron scandal.
A more nuanced analysis of the role of reporters by Doyle (2006) emphasizes the dynamics of the newsroom and restrictions on available sources as constraining business journalists’ ability to report freely, while Goddard (1998) suggests that the problem is limited public understanding of the issues. Parsons (1989) suggests that the media was more successful in addressing policy elites during the paradigm shift in economic policy from Keynesianism to monetarism in the l970s.
Tumber (1993) suggests that another frame dominates much business news coverage of financial crises, that of scandal, because this type of story would capture the interest of the general public and sell more newspapers.
There are a number of studies which try to explain why the business press is particularly vulnerable to ‘capture’ by vested interests, both by business and governments. Aeron Davis (2002, 2011) suggests that powerful businesses increasingly employ public relations agencies to shape coverage of their firms, and are increasingly successful in their ability to set the business agenda. It is a view echoed by Miller and Dinan (2000) and Tambini (2010). However, Schultz and Raupp (2010) and McCrystal (2008) dispute this view and point to the variation in the success rate of PR professionals in insulating their companies from criticism.
Other reasons for the alleged subservience of the business press that have been suggested are its increasing control by media empires (Featherstone, 2009); its dependence on corporate advertising and subsidy (particularly important for the specialist press) (Davies, 2009; Tambini, 2010); the difficulty it has in reporting stories without the cooperation of businesses and their sources (Davis, 2011; Tambini, 2010); and the lack of training and the development of a shared belief system reinforced by the close contact of the beat system (Barber, 2009; Davis, 2011; Lloyd, 2011).
Outside the academy, there was much questioning of the role of the media in exacerbating the current crisis by spreading unnecessary panic (Bauder, 2008; Caplen, 2009; Davies, 2009; Lambert, 2008; Lloyd, 2011; Seamark, 2008). But an inquiry by the Treasury Committee of the House of Commons (2009) largely exonerated the media from blame, a view echoed by Greenslade (2008), Barber (2009) and Goodhart (2009).
There is relatively little recent academic research on this issue, although the role of the press in panics has been examined all the way back to the Dutch Tulipmania of the 1630s (Galbraith, 1990; Goldgar, 2004; Kindleberger, 2000; Thompson, 2006). Recent work, drawing on the literature of behavioural psychology, has emphasized the vulnerability of investors, market makers, journalists and the public alike to unconscious psychological biases – such as the herd instinct, group thinking, and the optimism bias (Iyengar, 2010; Kahneman, 2011; Shefrin, 2002; Shiller, 1996; Thaler and Sunstein, 2008).
There is, however, an extensive literature on media effects in politics which looks at how much the media influences elections and voting (see inter alia, Franklin, 1994; Iyengar et al., 2004; Jamieson, 1984; Lazarsfeld et al., 1948; Lloyd, 2004; McGinniss, 1969; Norris, 2000; Prior, 2005, 2007). Its implications for financial media effects are yet to be explored.
The nature of online news
There is considerable research on the way people use the internet to seek and understand the news, and how that differs from traditional media such as print and broadcast.
In the first place, there is evidence that the internet increases audience concentration, with the audience share of the top news sites consistently higher than their share of news through circulation or broadcast (PEJ, 2009b). This can be partly explained by network effects, the tendency of any distributed system on the web to form nodes which attract more viewers. Hindman reports that, for any given topic, 99 per cent of page views go to the top 10 sites, with an exponential drop off in page views for both news websites and blogs after that (2008). This is reinforced by the use of search engines like Google, whose page-rank algorithms also direct viewers to the most popular sites. In general, the information abundance of the internet forces users to rely on short-cuts and depend more on trusted and familiar sites (Graber, 2004; Schudson, 1978). However, news is accessed in different ways, as many more viewers rely on accumulator sites such as Yahoo or Google which select news from many sources, and local sites which bring together local or regional news providers. Only one of the top 25 sites in the USA, the Huffington Post, is based on blogs rather than mainstream news sources (PEJ, 2009a).
In the UK, concentration is even greater, with the BBC traditionally dominating the news space online, although that dominance has recently been challenged by some newspaper websites (ABC, 2008; Schifferes, 2006). As well as network effects, the BBC also had a first mover advantage as it made a major investment in its website well before most newspapers (Schifferes, 2006).
There is also research on the viewing habits of online news consumers. The result of these trends is that, in ordinary times, most internet users only go to news sites occasionally and stay a short time, skimming headlines and perhaps reading one or two news stories. Although the reach of online news (the number of unique viewers per week or month) is large, the number actually viewing particular stories is much smaller (PEJ, 2011; Tewksbury, 2003). For example, the Project on Excellence in Journalism reports that for the top 25 US news websites, the average viewer spends four minutes per month viewing a particular website (PEJ, 2011).
However, retention rates of web news users (how much they can recall of a particular story 24 hours or one week after reading the news) suggest there is little difference between web news users and television news or newspaper consumers.
Other research by Iyengar et al. (2004) suggests that, particularly in relation to political news, internet viewers prefer to read about the ‘horse race’ rather than more sophisticated analysis and opinion – a theme much discussed by those, such as Davies (2009), who believe that the internet is ‘dumbing down’ the news agenda.
However, online news viewing patterns can change dramatically during a major news event. The evidence so far suggests that internet news viewing is episodic – at the time of major news stories (elections, disasters, and deaths of celebrities) there is a huge spike in interest and a major increase in the number of viewers, and an even greater increase in the number of individual stories they read. This is the result of the greater freedom of choice that can be exercised for ‘on-demand’ news. But most of these episodes only last one or two days, after which internet viewing habits return to their previous patterns (Schifferes, 2006; Schifferes et al., 2010).
The growth of on-demand news has been documented in other settings (Newman, 2011), and since the time of this study in 2008, there have been further changes in the way people consume online news, with increased use of mobile phones and tablets, as well as social media such as Twitter, all of which increase the speed at which online audiences take up news.
Research hypotheses
A number of hypotheses emerge from the literature review, both in relation to the expected patterns of online news coverage and the content of news stories about the financial crisis.
Patterns of online news use during the financial crisis
H1. News about the financial crisis will be viewed episodically online, with large peaks on certain days and a return to normal viewing levels in-between these peaks. H2. The intensity of viewing, both in terms of page views, number of visits to the site and time spent on the site, will increase during those peaks. H3. There will be some longer lasting effects which raise page views in the period after the sharp increase in episodic viewing. H4. The types of online stories viewed will become more diverse with the greater intensity of viewing and there will be less focus on the immediate ‘horse-race’ type news stories. H5. Interactivity and the use of multi-media (e.g. video) material will increase as intensity of use increases.
Nature and content of stories about the financial crisis
H1. News stories about the crisis will give particularly negative coverage at the initial acute phase of the crisis. H2. News stories will largely be focused on the domestic rather than the international elements of the financial crisis story. H3. News stories online will depend on a limited number of sources. H4. News stories online will generally follow rather than lead the news agenda, with simplified versions of material already reported elsewhere. H5. News stories online will provide less depth or original analysis.
Research findings
Traffic to the BBC news website
Comparative data: BBC vs other business news websites
Comparing traffic to different news websites can be problematic because of the different metrics used by different organizations, with particular differences between the way newspapers and the BBC measure their online audiences. However, there is comparative data for the first month of the crisis from the digital media research company ComScore (2008), which uses the same metric for all business news websites. Table 1 shows very large – and probably unprecedented – increases in a short period of time, especially bearing in mind that the financial crisis started in the middle of September, so essentially the effects are only present for half a month.
Top 10 fastest growing UK business news websites: August to September 2008.
Notes: 1000s monthly unique visitors, age 15+, home and work, excluding mobiles and PDAs. Source: ComScore.
These figures show that the dominance of the BBC News website increased after the crisis. The BBC’s share increased from 20.7 per cent to 29.7 per cent of all online business news site visitors. Indeed, it had 60 per cent more unique visitors than the next four content creators combined. It is also noticeable that the increase in the number of visitors to the next largest site, Yahoo, was considerably less than to the BBC (Yahoo takes content from other providers rather than creating its own, and may have been less trusted in the crisis). The increased concentration of the audience on a few other key sites was also evident from the fact that their viewing figures went up by 40–60 per cent, compared to the overall increase across all sites of just 10 per cent. Two of the smallest (and the Guardian, whose readers traditionally shun business news) actually doubled their audience.
In the next section, we look at BBC visitor numbers and page views using the BBC’s own measurements. Key details of the methodology used, and definitions of terms, are contained in the Methodological Appendix at the end of this report.
Trends in traffic to the BBC news website during the crisis
The increase in traffic to the BBC Business News website seen in the ComScore figures accelerated in the following month, and remained high for the entire research study period, from mid-September to mid-December 2008. The increased traffic from the global financial crisis was important enough to raise the overall traffic levels to the whole BBC News website substantially.
In the five months prior to the study (from the week beginning 20 April 2008 to the week beginning 14 September 2008), unique visitors to the BBC News site averaged 16m per month, recording an average of 270m page views per month. This rose to a weekly average of 20m unique visitors to the BBC News site during the 13 weeks of the study period, a 25 per cent increase. These users viewed an average of 301m pages – an 11 per cent increase in page views. In the six months following the study period (from the week beginning 14 December 2008 to the week beginning 28 June 2009), average weekly unique visitors remained roughly steady at 20m, but page views dropped very slightly to 299m. Unique visitors to the business pages averaged 29 per cent of the total number of visitors to the BBC News website during the crisis – a substantial increase compared to the previous period.
One way of gauging the prominence of financial crisis stories is to count the number of these which featured in the top five most viewed stories on each day. Over the study period, 22 per cent of the top five most viewed stories on the BBC News website as a whole concerned the financial crisis. However, there were large variations: in the week beginning 6 October financial crisis stories accounted for 16 of the top 25 most viewed news stories.
Another measure is to look at the peak news days for the site as a whole. In 2008, two out of the five biggest days ever for traffic to the BBC News website were related to the financial crisis (8 October and 8 November), and they were only surpassed by the two days around the election of President Obama and the Mumbai terrorist attacks in November.
The business section of the BBC News website saw a much sharper increase in visitors, with unique visits and page views almost doubling from the previous average during the crisis survey period.
In the five months prior to the study period, unique visitors to the business pages averaged 3.8m a week, which increased by 55 per cent to 5.9m during the study period. Although the weekly average of unique users fell 13 per cent to 5.1m in the six months after the study, this was still up by a third on the period prior to the crisis. Average weekly page views rose 82 per cent during the study period. They fell by 19 per cent to 34m a week on average in the following six months, but this was still 48 per cent above the pre-crisis period.
Looking at the figures on a monthly basis also illustrates the dramatic rise in visitors and page views which is sustained after the acute phase of the crisis. The total number of visitors to the business news section increased from under 10m to 17m at the peak of the crisis, but was still well above 10m in the later phases. Page views which had increased from around 150m to 250m were still around 200m in the months after the crisis (Figure 1).
Month average page views and unique visitors to News Business – April 2008 to July 2009. Source: Sage.
So, to some extent, the boost in audience interest in business news was sustained even after the acute phase of the crisis ended.
Comparison with television and radio audiences
The increase in traffic to the BBC News website far exceeded the increased viewing figures for the main TV and radio news bulletins. For example, the audience for the Ten O’Clock News increased from an average of 4.9 million to 5.7m when the US investment bank Lehman collapsed on 15 September and to 5.8m when world interest rates were cut on 8 October – an increase of 18.7 per cent. These were not the same as the peak days for viewing on the news website, where news about the difficulties of the UK banks and a further interest rate cut in November were the most-visited days.
With the exception of these two news bulletins, more people read about the financial crisis daily on the BBC News website than watched it on News24 or followed it on radio news channels.
Increased intensity of viewing of business news
The public’s desire for information about the financial crisis was not only manifested in a rise in traffic to the front page of the news site. Viewers did not simply ‘skim’ financial crisis stories featured on the front page of the news site but followed up their interest by visiting the business section to learn more. One measure of this was that, during the study period, the business index link on the front page of the whole website was the most-clicked link on most days (previously it had been the third to fifth most-clicked link).
The rise in page views of business stories was proportionately much greater than the increase in unique users of the site: during the 13 weeks of the study period, while unique users to the BBC News site averaged 20m a week, compared with 16m during the previous five months (a 25% increase), business page views grew from a previous weekly average of 23m to 42m (an increase of 82%). This indicates that the audience was viewing more stories per visit than before the crisis, and also spending more time on the site.
BBC audience research indicated that the minutes per site increased substantially during the crisis, and were higher for the business section than for any other index section of the website.
This quest for ‘depth’ was sustained after the end of the study period: weekly averages of unique visitors to the BBC News website remained 25 per cent up on the period prior to 15 September but business page views were up 48 per cent.
Using a slightly different measure from the BBC (LiveStats, which samples the news audience every few minutes and provides more granular figures), the audience figures are much larger. Over the study period, LiveStats recorded a weekly average (news web total) of approximately 29.7m unique users, viewing an average of 252m pages per week (Monday to Friday). Business pages users averaged 8,868,352 per week over the study period, ranging from a high of 12,476,707 in the week beginning 6 October to a low of 7,173,283 in the week beginning 1 December. This would have included substantial numbers of return viewers and people who visited the site several times a day (the average UK BBC Online user comes to the site several times a week).
There was more daily variation in the data on business story page views over the study period. These averaged 39,332,342 per week (Monday to Friday), with daily totals ranging from a high of 15,573,453 on 10 October to a low of 3,646,694 on 5 December. Business page views averaged 14 per cent of the news site total over the study period (Figures 2 and 3). The smaller number reflects the large number of news stories on the whole BBC site (over 200 on a typical day).
Weekly business unique users as percentage of site. Source: LiveStats.
Weekly business page view as percentage of site. Source: LiveStats.
How many stories related directly to the financial crisis?
Over the study period, financial crisis stories accounted for just over a third of all business story page views, on average. This reached a high of 49 per cent in the week beginning 6 October and a low of 17 per cent the week beginning 1 December. Over the whole BBC News website, financial crisis stories accounted for an average of 6 per cent of all page views, ranging from 13 per cent in the week beginning 6 October to a low of 2 per cent in the week beginning 1 December.
Looking at the top 10 most popular financial crisis stories over the study period as a whole (Table 2) shows just how big the spikes were for days of intense news interest. It is unusual on the BBC News website for stories to get 1m page views, and only infrequently do they get as many as 2m. The top stories reflect the overall pattern on the site, with four relating to UK banks, three relating to stock market jitters, and two to broader policy issues. Only one (the Lehman crash) specifically relates to US banks.
Top 10 BBC News online stories on the crisis.
Note: September to December 2008, by page views. Source: LiveStats.
What type of stories did people view?
By a very wide margin, the pages the audiences most wanted to view related to news stories about the crisis, followed by the market data pages and the blog by the BBC Business Editor Robert Peston.
The stories read most often in the daily top five of most viewed financial crisis stories were: news stories (71,706,274 page views, or 66% of the total); market data pages (27,609,483 page views, 25%); and the Peston blog (8,587,245 page views, 8%). Other types of stories accounted for 1 per cent of total page views. News stories tended to predominate on days when economic or policy issues were important. Market data pages were more in evidence on days when Wall Street or the FTSE saw large gains or falls. For example, data pages occupied two of the top three positions on 10 October 2008 when the FTSE 100 dropped below 4000.
Perhaps surprisingly, given the complexity of the issues, explainers and Q&As featured very rarely in the top five list of most viewed stories. Personal finance stories dealing with the impact of the crisis on household finances featured in the top five on only three occasions (UK edition), although they were more in evidence in lower ranked positions.
Was the international audience different from the domestic one?
The BBC News website has a substantial international audience, whose viewing patterns are quite different from the domestic audience, with fewer frequent visits per month and a lower intensity.
UK readers accounted for roughly two-thirds of page views on the BBC News website over the study period, followed by the USA (c.15%); Canada (c.3%); Germany (c.1.5%) and Australia (c.1%). LiveStats permits a detailed breakdown of page views according to whether they were viewed by domestic or international readers.
In general, financial crisis stories seem to have been viewed less heavily by the BBC News website’s international audience than by the UK. At least five financial crisis stories featured in the list of the BBC News website’s top 50 most viewed stories for the domestic UK audience each day of the study period (a total of 325 stories), with a total of 90 financial crisis stories in the website’s daily top five (28% of the total). For international users, financial crisis stories featured in the daily top 50 on fewer occasions (284) with a smaller proportion of these appearing in the top five (13%).
This may be a reflection of the UK bias of the stories, as shown by content analysis. By November, US viewers were drawn to the extensive coverage of the US Presidential election. In the US-dominated international edition, only one financial crisis story makes it even into the top 50 most viewed on both 4 and 5 November – the date of the election.
Examination of the content of these stories also reveals a marked preference among the international audience for news stories over data pages and other story types. A breakdown of financial crisis stories in the daily top 50 shows 71 per cent of stories viewed by international users were straightforward news stories, compared with 40 per cent for UK users. Data pages were less popular among international users (22% compared with 42% among UK users). The daily top 50 story data also show how readership of the Peston blog was almost entirely composed of UK users, while international users viewed a slightly wider range of stories than UK users.
The balance between the UK, US and international coverage shifted during the period as the crisis intensified. While there was substantial initial interest in the difficulties of the US banking system, which mirrored in many ways those of the UK, as well as in the battle over the bailout plan in Congress, by mid-November the focus shifted back to the deteriorating economic situation in the UK. There was sporadic interest in other areas when these impacted on UK savers (the collapse of the Icelandic banks) or stock markets (the announcement of a stimulus package by the Chinese government on 10 November which boosted shares), but otherwise attention remained fixed close to home.
Robert Peston blog: Dramatic increases in audience
How significant was ‘the Peston phenomenon’? The BBC’s Business Editor emerged as an important commentator and Peston’s blog became a crucial source of analysis for many readers. He was also widely quoted as an authority in other BBC News online stories. For the domestic audience, Peston’s blog was in the top five most viewed daily stories on 37 days out of 65. The Peston blog was probably the principal source of news analysis for most readers and was an important part of the BBC’s coverage. Peston also featured prominently as a named source in straightforward news stories, being quoted in 32 of the 178 stories. Overall, viewership of the Peston blog increased from an average of 150,000 unique weekly visitors in the summer to over 1m during the crisis, with daily traffic reaching 3m when some exclusive stories were revealed. The popularity of his blog played a key role in catapulting Mr Peston into the top-rated position among all business journalists, according to peer-ranking studies (Oldfield, 2009; Ponsford, 2009).
Content analysis
This section moves from analysing the online audience’s revealed preferences to an analysis of the content of those top stories. The advantage of this approach is that we are measuring those stories that the public actually read the most. However, we cannot measure directly what the audience thought of the content of these stories, or how much they understood them.
Did the tone of the BBC online coverage contribute to the scaremongering?
Content analysis of news stories in the survey sample has helped to draw some general conclusions about how the financial crisis was reported. A key issue concerns the ‘tone’ of the coverage and whether journalists contributed to the panic by scaremongering, for example by using alarmist language.
The top three stories by page view on each day of the financial crisis (a total of 178 stories) were read and assigned a value depending on whether they were largely ‘positive’ or ‘negative’ in tone (based on whether the story conveyed the impression that the events described or analysed would make the situation better or worse) (Figure 4). The key words used in the story (‘recession’, ‘downturn’, etc.) to describe the situation were also noted and compared.
‘Positive’ and ‘negative’ stories. Source: LiveStats and author’s calculations.
Several conclusions emerge from this analysis. First, examination of the tone of financial crisis stories on BBC News online reveals dramatic swings in sentiment over the course of the study period. Despite the generally downbeat subject matter of the coverage (major bank insolvencies, massive losses on financial markets, etc.), this was not reflected uniformly in the tone of the stories. In particular, the early weeks of the crisis (spanning the manoeuvrings over the bailout in the US Congress and a number of bank rescues, nationalizations and takeovers) saw a number of stories expressing a more or less optimistic feel (in that they implied that the situation might improve as a result of the actions described in the news story). This was manifested in some of the quotations and sources used, as well as in how the implications of the events described in the story were phrased.
The early period of the crisis also saw wild gyrations in share prices, with record daily gains posted following the launch of the initial US bailout plan on 19 September as well as equally large falls four days afterwards. These fluctuations in market sentiment were obviously reflected in the tone of the coverage. Indeed, in the first three weeks of the study period the number of ‘positive’ and ‘negative’ stories was roughly equal, with a slight preponderance of negative stories. Thereafter, the tone of the stories became almost uniformly negative as the market contagion began to impact the real economy, although the mood picked up towards the end of November as the banking system appeared to be stabilizing.
The second observation to be drawn concerns the alleged use of alarmist language by the BBC. Clearly, the tone of a story cannot be considered independently of the event it is describing, or the sources and quotations used. The keyword analysis reveals a wide range of terms used to describe the unfolding situation. Many of these are clearly alarmist: ‘Armageddon’, ‘an economic Pearl Harbour’ and ‘catastrophe’ crop up, although mostly in quotation. On the other hand, the most widely used words (‘crisis’, ‘recession’, ‘downturn’ and ‘slowdown’) are much more neutral and descriptive. These were the words used most consistently in financial crisis stories and they do not seem out of place, given the context (Figure 5).
Keywords. Source: LiveStats; author’s calculations.
Framing the crisis: sources and actors
From what perspective was the financial crisis covered by BBC News online? Our analysis indicates that much of the coverage was focused on the fortunes and actions of a small group of individuals and institutions, and had a marked UK focus.
Although the crisis was a story of major institutions being laid low and economic prosperity crumbling, it also foregrounded a small group of key individuals regarded as pivotal to managing the situation. Figure 6 indicates the top five individuals who were most often the subject of financial crisis stories in our survey. In each of the four monthly periods there were only a small handful of such individuals – probably no more than a dozen – who featured at all in financial crisis stories.
Key individuals. Source: LiveStats; author’s calculations.
There was a predictable focus on government leaders and central bankers. CEOs of commercial banks, however, rarely featured, despite these institutions being initially the focus of attention. Gordon Brown, the British Prime Minister; Alistair Darling, his Chancellor; Mervyn King, the Governor of the Bank of England; George Bush, the US President; and Hank Paulson, the US Treasury Secretary, dominate the list. The most positive coverage was of Gordon Brown, Alistair Darling and Mervyn King, who were shown as grappling actively with the situation. George Bush was portrayed as heavily involved in seeking a resolution, albeit largely ineffectually, and Hank Paulson was invariably the villain of the piece – a former banker lacking impartiality.
The same pattern held in relation to institutions, with the focus shifting from individual banks and the coordinated action intended to stabilize these in the first weeks of the crisis, to more general reporting of the policy response to the crisis later on, from the end of October and early November. The central banks in the USA and the UK became particularly prominent in this period, although there was also much interest in the nationalization/bailout of Lloyds and RBS.
Analysis of results
Key findings
There was a sharp increase in both traffic (unique users) and usage (page views) for the business pages of the BBC News website over the three-month study period, compared with the previous five months.
This increase was largely sustained over the six months following the end of the study period.
The increase in traffic and usage for the site generally led to more viewers accessing the business pages for specific information about the crisis. Financial crisis stories were regularly among the most popular of all stories during the study period.
News summaries and market data pages were by far the most popular types of financial crisis stories. Q&As, explainers, graphic pages, etc. were considerably less popular, although many news pages included some of these features.
However, the Robert Peston blog was extraordinarily popular, with a 10-fold increase in audience.
Financial crisis stories were more popular with the domestic audience, compared with international users.
BBC News coverage of the financial crisis tended to focus on UK stories, or international stories with a significant impact on the UK, such as the US bank bailout. They also tended to focus on the actions of a relatively small number of key individuals and institutions (mainly British and American leaders).
The tone of the BBC News online coverage was not particularly alarmist, although rapid swings in sentiment between ‘positive’ and ‘negative’ stories were in evidence.
Patterns of online news consumption during the crisis
This research confirms that the news audience consumes online news differently from radio and television news. It also questions whether our current understanding of online news consumption is adequate to explain the pattern found during the financial crisis.
In particular, the first hypothesis that online news is dominated by casual viewers with short attention spans, with sharp but transient peaks when a particular news event becomes salient, is not confirmed by the evidence of this study. There is evidence of episodic incidents, with huge peaks in viewing during some of the key days of the crisis, and large viewing figures for particular stories. However, there is not a single peak, but a series of peaks which grow in size over the first two months of the crisis. The overall increase in the size of the audience was sustained for the whole of the three-month study period, as was the increase in intensity. Therefore, the peaks were less marked because of the big increase in the audience size overall.
So it may be that for a news event of this size, where there are fresh developments almost every week, more and more viewers are making ‘an appointment with the news’ online as well as offline. This is a new and important finding which suggests that the pattern of online news consumption can shift considerably during major ongoing news events.
The second hypothesis, the increased intensity of viewing during the episodic peaks, is given more support. The number of page views rose more quickly than the number of unique visitors, and showed a bigger variation over time, reflecting an increased intensity during the key moments of the crisis. However, it is important to note that there was also an increase in the audience who were ‘skimming’ by just looking at the index page, with its news summary, as well as those who were looking in more depth at stories.
Can these changes in viewing patterns be explained by our third hypothesis – that of a ‘hysteresis effect’, where people who visit a news website for the first time during a crisis often become regular visitors later on, with overall audience levels after the episodic event higher than before, and with new visitors being attracted to the site? This did happen, with a 50 per cent increase compared to before the crisis. However, during the crisis period, something else was also happening: there was an increased intensity among existing viewers, who visited the site more often and read more stories – and who were the major factor in the big increase in viewing figures and page views during those three months. The hypothesis suggests that overall viewing for non-financial crisis stories would rise equally, but during the crisis it was precisely the increase in the reading of financial crisis stories that was most marked.
One of the most striking findings was the unique pattern of story consumption during the financial crisis. Our fourth hypothesis predicted that there would be a greater diversity of types of stories people read during the intense period of the crisis. However, our research shows that one additional type of webpage dominated – market data pages, which allowed viewers to find out live information about their own shares or about market movements. These viewing figures increased by up to 300 per cent on crisis days compared to the pre-crisis average, and made up nearly 40 per cent of page views. This points to the unique characteristic of web-based financial news, the ability of viewers to interrogate a large and instantly available in-depth database to access information relevant to themselves. Although more conventional types of in-depth explainers and graphics pages were also on the site, these proved less popular – except on a few key days – than access to the financial data itself. This suggests that, rather than seeking sophisticated explanations, viewers were more interested in their own version of the ‘horse race’, i.e. whether they themselves were winning or losing from the financial crisis.
Finally, our fifth hypothesis, on the increased use of interactivity, took a very particular form on the BBC News website. There was relatively limited use of personal commentary by those affected by the crisis, and video was not a central part of the crisis story. Instead, there was a massive increase in the popularity of the blog run by the BBC Business Editor, Robert Peston, which increased its audience 10-fold. This was where key stories were often broken, and where the public left their comments (and sometimes provided information which suggested further investigations). Interactivity can evidently vary considerably across different story types, and may not take the same form for celebrity news as for more serious stories. As Twitter and other social media sites were not in widespread use in 2008, the blog was the most effective way of communicating breaking news.
Nature and content of online news stories about the crisis
The research has led us to question a number of the hypotheses about the nature of online news content, while confirming some others.
In particular, our first hypothesis about the negative nature of news coverage, especially in the early stages of the crisis, has to yield to a more nuanced explanation. In fact, in the first few weeks of the crisis, the tone was surprisingly positive or neutral, reflecting the uncertain understanding of the depth and breadth of the crisis. Indeed, it was only when events began to directly touch on UK shores, particularly with the threat to the entire UK banking sector, that the tone changed dramatically. Whether access to live market data at this point would have reassured or concerned viewers, it is hard to see how our analysis of online news content supports the belief that blind scaremongering by the media panicked the public into an unnecessary crisis.
With respect to our second and third hypotheses, we did find that the BBC News mainly had a domestic rather than international orientation, despite the fact that one-third of the audience was drawn from outside the UK. This was less true at the beginning of the crisis, but as the effects on the UK financial sector grew more serious it may have been inevitable that this became the focus. The international audience itself continued to follow these stories on the BBC, although it had a less intense viewing pattern. In fact, the main reason the audience from outside the UK was reading business news on the BBC was to find out more about the effects of the crisis on Britain.
The most controversial hypothesis, which suggests that online news is ‘dumbed down’ – purely derivative of the mainstream news agenda and focused on simple news stories based on conventional sources, receives only limited support in this study. It is true that our third hypothesis about the limited number of sources, individuals and institutions cited in the main BBC news stories, is confirmed. But it should be remembered that we were measuring this in relation to the top three news stories of the day, which were bound to concentrate on the key players and institutions – although perhaps not to the extent we predicted.
Our fourth hypothesis about whether the online news agenda followed rather than led the news is contradicted by the rise of the Robert Peston blog, which became a leading source of breaking news not just for the BBC but also for the rest of the news media. Robert Peston’s influence has been attested by several studies using peer-based rankings (Oldfield, 2009; Ponsford, 2009). This demonstrates the power of the web to break stories faster, and in more detail, than is possible in other types of media.
There is also limited evidence to support the fifth hypothesis that there was a lack of analysis and in-depth material. In fact, the website was able to provide far more detail – both about market movements and about the causes and consequences of the crisis – than was possible on TV or radio. It is true that some of this material was not as popular with viewers as market data, but it may well have been viewed by the smaller group of more intensive viewers, the ‘news junkies’. In fact, the very severity of the crisis, with new and important developments every day, may explain why many viewers were attracted to the lead story.
One question raised by the study of content is whether the website could have provided more information about how the crisis was likely to affect individuals personally. Very few stories about personal finance were written during the crisis, although the great interest in the market data pages, and other audience research, suggests that this was an area in which the public was particularly interested.
Another element that was conspicuously missing from the BBC coverage was any attempt to frame the crisis as a scandal, attributing blame for the crisis to individual bankers. This was in contrast to the popular press, where much attention was focused on the misdeeds of Sir Fred Goodwin – the former head of the nationalized bank, RBS. The BBC consciously eschewed such an approach – perhaps a reflection of the very different values and norms of the organization compared to the tabloid press, who may have been more in tune with the popular mood.
Conclusions and future research directions
This study, although limited to a single news organization, has challenged some key assumptions about the pattern of online news consumption. It has suggested that in times of major panics, viewers rely on trusted sources and are not particularly keen on interactivity. It also suggests that online viewers may actually grow more rather than less loyal to particular news websites, and suggests that there are very particular reasons (access to live market data) why online news should be especially popular in times of financial crisis. Most strikingly, it suggests that online viewing patterns can change substantially during a major crisis, with an intensity of use that may come close to ‘making an appointment with the news’ online.
It has also attempted to address some of the concerns about the nature of online coverage, suggesting that it is not as superficial as has sometimes been hypothesized. It has also attempted to throw some light on two questions raised about the nature of the overall media coverage of the crisis: whether journalists were ‘captured’ by special interests and therefore unable to see the looming crisis or explain it properly; and whether the public and the markets were panicked by lurid and irresponsible headlines. This study suggests a more nuanced answer to these claims. Journalists were not so much stoking panic as reflecting real market conditions, and they were trying to judge fast-breaking stories with the limited sources that were available to them at the time.
It is not possible to fully answer these questions in a single study. However, the research has also highlighted some key areas where future research could help improve our understanding.
Sophisticated consumers, or panicked public: The multiple audiences for financial news
This study has examined how the UK’s biggest mainstream news provider covered the crisis online. It would be extremely valuable to compare such coverage, in terms of both content and audience, with that of other news providers, both in the UK and internationally. One attempt at preliminary evidence, looking at the content of four major news websites (FT, BBC, New York Times, Reuters) during the first anniversary of the crisis (in September 2009), suggests that there are very substantial differences in their story selection, presentation, and use of graphics, driven by their different audience profiles.
A crucial part of those different audiences for business news is the distinction between a general news audience and the specialized business audience, who are likely to want more detail and use more than one source. In general, the type of information provided in the specialist press, and the tone of coverage, suggested a much greater awareness of the potential crisis in the year leading up to the Lehman collapse. But we know very little about how market professionals and investors receive market information and deal with rumours (Thompson, 2010).
We also know relatively little about what the general audience thinks of online business news coverage, or how much they understand the economic concepts that they read about. Broader survey research on financial capability suggests that a substantial proportion of the population, perhaps as much as 50 per cent, lacks financial capability – which includes the ability to budget, to make long-term plans, to understand financial statements, and to keep up with the business news (FSA, 2006). Nevertheless, the internet is becoming an increasingly important tool to search for financial information generally, which will enhance financial capability (Dutton et al., 2009; Schifferes, 2012).
Finally, it would be useful to find ways of looking in more detail at the variation in public responses to news announcements to understand when those announcements did or did not cause panic. One explanation for the relatively restrained tone of BBC News coverage of the Lehman collapse in September 2008 is that it had learned the lessons of the Northern Rock collapse in September 2007. The story, broken by BBC News, of a government rescue plan for the troubled former building society panicked investors into queuing up to withdraw their savings (BBC, 2007; Brummer, 2008; Peston, 2008).
Journalists as gatekeepers in the financial crisis: Lessons of history
The complex nature of the financial crisis meant that even in the digital age, the role of journalists in framing news coverage and interpreting events to the general public was crucial.
However, the crisis undermined many of the assumptions that had framed business coverage of the economy for the previous two decades, and indeed undermined public confidence in business journalists (Schifferes, 2012). More research needs to be done about the perceptions, background and training of financial journalists themselves, and their interactions with editors, which played a crucial role in how the crisis was covered. It would be useful to know how many general reporters were thrown into the fray in the heat of the crisis, what their level of knowledge was, and how their coverage differed from that of more specialist reporters.
Business journalists were freed to take a more critical stance in regards to the financial system. But in reality they fell back on some of the alternative but long-established frames to explain financial crises: scandal (blaming the bankers); moral panic (blaming the public); and reform (blaming the policymakers) (Schifferes, 2011a). A further historical analysis of the use of such frames could provide a useful perspective on understanding the coverage of the current crisis (on the media and business in earlier historical eras, see Kaplan, 1975; Kynaston, 1985; McCusker, 2005; Nasaw, 2003; Nye, 2011; Read, 1992; Starr, 2004; Urofksy, 2009; Weinberg, 2008; Winseck and Pike, 2007).
Although journalists have retained their gatekeeping and agenda-setting functions in the online news space at the moment, that situation could change (Gans, 1979; Schlesinger, 1978, Schudson, 1978). The nature of business news coverage – with its complexities and obscure language which is often difficult to understand – does give the business journalist more power to frame the story for the general public. However, the rapid development of social media and new forms of technological interaction may challenge this in the future, and new ways to consume, discover and create news will no doubt have a substantial impact on the way the next financial crisis is covered.