Introduction

Data provided by the Australian Institute of Criminology reveals fraud costs the Australian community more than any other crime type, more than $8 billion annually by some estimates. A review of the literature suggests that fraud is underreported in the media, where preference is given to reports of violence and conflict. This lack of media coverage in turn obscures the true extent of the problem from the public. This study set out to establish whether fraud is being underreported in New South Wales metropolitan media. The study compares the coverage of six major crime types by Australia’s two dominant newspaper publishers, Fairfax and News Limited by focussing on a case study of their Sydney-based publications, The Sydney Morning Herald and The Daily Telegraph. The findings of the study provide evidence that fraud coverage is more prominent than other crime types. This is particularly the case when the reported fraud involves high profile people or organisations, or where the value defrauded is high. The findings also reveal differences in the way the two newspapers report on fraud.

The cost of crime in Australia

Accurately measuring the financial impact crime has on society is extremely difficult. What constitutes harm? What can and should be measured? How can anxiety, loss, fear and pain be measured in a meaningful manner [1, 2]? These questions pose a significant challenge for researchers interested in understanding the scope and scale of crime and its impact on society.

The Australian Institute of Criminology [3] measures the financial impact of crime by applying a multiplier, which is calculated by determining the difference between self-reported crime victimisation and official crime statistics, to a given year’s official crime statistics in order to provide an inflated figure thought to account for underreporting of criminal incidents [2, 4]. Researchers then factor in medical costs, including treatment both in and out of hospitals [5]; lost output from the loss of paid and unpaid work [5]; intangible costs such as pain, suffering and loss of quality of life drawn from research into road trauma [6]; and the transfer of resources which are the unwanted transfer of property from victim to offender, but excluding wanted transfers, for instance victims receiving insurance claims [2, 7].

While the elements provided by Mayhew [7] for calculating an average unit of cost by crime work for some crimes, not all crimes use all factors. For instance, fraud, burglary, arson, and criminal damage do not include medical costs due to a lack of data, and an example of this can be found when comparing assault and fraud costings. Assault costings vary due to the differing levels of injuries, however Mayhew [7] included medical treatment at $1000 per crime which made up only one-fifth of total assaults, lost output at $3400, and intangible costs at an average of $800 per incident, when averaged these provided an average unit of cost of $1800 per assault, or $1.4 billion [2]. Whereas fraud is complicated by definitional vagaries and fewer available data sources, which in 2003 saw the costing of each officially reported fraud incident at $9900, and unreported fraud incidents at $1600 per potential incident. This was then combined with $1.83 billion worth of frauds against the Commonwealth to provide a combined total of $5.88 billion in 2003 [2].

In 2013, utilising this methodology, the Australian Institute of Criminology [3] estimated crime cost Australia in the vicinity of $36 billion Australian dollars annually, conservatively estimating this as being 4.1% of the nation’s total gross domestic product. Fraud accounted for 40% or $8.5 billion of that total, theft accounted for 11%, burglary offences accounted for 10%, drugs related offences 9%, arson 8%, and assault 7%.

Media and crime

The manner in which the media reports on crime has been the subject of extensive research across a range of discipline areas. A key theme emerging from this research is that the media acts as a gatekeeper, choosing which crimes to report and which ones to discard according to their own beliefs and perceptions of newsworthiness [8, 9]. A consequence of this is the public becomes more aware of certain, often violent crimes, and more interested in the political and policing response to such crimes [9,10,11,12,13]. The more the media reports on the types of crime that evoke the most fear in people, the more the public believes that the crime problem is growing. This is particularly problematic for law makers as public perceptions of crime and criminal justice often influences decisions on law making, operational policing and judicial sentencing [11].

Media framing of fraud and other related crimes defines and shapes public debate [14,15,16]. The idea that the media would intentionally impose their own frame or spin on a news story has been thoroughly discussed by researchers involved in the study of communication, sociology and political science [17]. Entman [14] said that to frame is to ‘select some aspects of a perceived reality and make them more salient in a [communication] context.’ Frames that appear in the mass media have the potential to exert powerful influences on public policy definitions, choices and outcomes [18, 19]. Often, journalists will choose a conflict frame for their news story, placing the protagonists of a situation against each other in order to add interest or controversy [20, 21]. Conflict frames are particularly apparent in political news journalism [20], and the often unrealistic and misleading depiction of crime in the media has led to accusations that editors and journalists are out of touch with the public [8, 22].

Newspapers have a significant agenda-setting effect on other media, and are influential in shaping coverage across all mediums [23]. Australia has one of the most concentrated newspaper ownerships in the world, with Fairfax and News Limited accounting for 86% of newspaper sales in 2011 [24]. Advances in technology have significantly impacted on the media landscape leading to a steady decline in the circulation of Australia’s major newspapers over the past decade [25]. While newspaper circulation and crime rates are both declining, the public perceptions of crime prevalence and the fear of crime continues to increase [11]. Fairfax and News Limited publications have a growing online presence, ensuring that newspapers editors and journalists continue to play a significant role in setting the agenda for public debate and discussion.

Print media and its portrayal of crime has a direct impact upon the development of crime images held within the collective social consciousness ([26], p. 38). These conceptual images often portray the criminal as different from the general public in some manner and suggest a weakening of societal moral fabric. This influence, by the media, on public perceptions directly impacts upon crime control. Evidence of this in Australia can be seen in high profile cases such as the death of Skye Sassine, a 19-month-old child killed following a police pursuit, leading to the introduction of Skye’s Law. Similarly the ‘one-punch law’ was recently introduced in New South Wales following a spate of so-called ‘king-hit’ deaths, or deaths linked to alcohol fuelled serious assaults that played out across states media outlets; in each case public out-cry led to the eventual toughening of legislation [27]. While the ‘Skye’s Law’ and the ‘One-Punch Law’ drew intense media coverage and eventual toughening up of legislative responses, only a few financial crimes can claim to have had such an immediate impact upon legislative responses, and those few that did have taken many years to eventually lead to changes.

Fraud in the media

Fraud, according to Edelhertz [28] is ‘an illegal act or series of illegal acts committed by nonphysical means and concealment or guile, to obtain property, to avoid the payment or loss of money or property, or to obtain business or personal advantage’. While it might be contended such a definition may incorporate a minor purchasing (illegal act) alcohol (property) by using a false identity document (illegal act/guile), its interpretation within the scope of this study is that such behaviours seek to achieve a financial advantage.

With up to 40% of Australia’s total cost of crime attributed to fraud and financial crime [3] it is necessary to examine the active role Australia’s media currently takes in reporting these crime types. Levi [29] asserts that the coverage of white-collar crimes by the media is vital as it influences public perceptions of the harmfulness of the problem. Similarly Sullivan and Chermak [30] contend that ‘[t]he construction of these crimes [by] the media is likely to have a tremendous influence on public perception’. Fraud and financial crimes are however, by their very nature, often hidden and complex and, as such, the media have a vital role to play as a social control mechanism [30]. With media shaping crime discourse, as it did in the Skye’s Law’ and the ‘One-Punch Law’ matters, its influence on public perceptions dictates law and order discourse thus affecting the allocation of enforcement resources, decision making by juries and justice policy ([31], p. 1038).

Sullivan and Chermak [30] suggest however that the media are ‘notoriously inept at covering highly complex financial crimes’ and more likely to highlight and sensationalise street crimes. As with all generalisations it is important to keep an open mind as to the effectiveness of media reporting, therefore it is important to consider the extent of research into the areas of media representation of fraud and financial crimes. Stephenson-Burton [32] suggests that researchers have historically paid scant attention to financial crime in the media, and research which does exist tends to focus on aspects of these crimes other than media representation. Sullivan and Chermak [30] suggest that most of the research tends to be anecdotal commentary and not empirical ([30], p. 307). Citing Stephenson-Burton [32], Levi [29] noted that in 1992 articles on fraud contributed only 1.5% of all press coverage. Cavender and Mulcahy [10] suggest at the time that with ‘[c]rime news, a significant portion of news coverage, focuses on crime in the streets, not crime in the suites’. Cavender and Mulcahy [10] however suggest that ‘…corporate crime receives more coverage now than in the past, especially if physical harm is involved’. While violence is not a component of fraud offences, there is often an association between white-collar crime and other crime activity involving drugs and personal violence, and this association makes fraud crime more appealing to newspaper editors and journalists [33]. Indeed, Michael [34] found that while people are undoubtedly sensitive about crimes of the powerful, white collar crimes still do not elicit the same kind of societal response as violent crime. Levi [31] suggested that ‘[i]t is conventional wisdom among those few criminologists and media sociologists who have paid attention to fraud (and corporate crimes generally) that they are neglected by the print and electronic media’; while White and Perrone [26] argue that ‘there is a skewed focus on ‘street crime’ and bizarre events. Meanwhile white-collar crimes tend not to receive the same kind of coverage or treatment by the mainstream media outlets’. While the media generally eschews meaningful reporting of fraud, there are exceptions in some countries. For example, Lundstrom [35] compared the reporting of welfare fraud in the United Kingdom and Sweden by analysing both corporate news media and citizen generated media finding the focus of crime reporting varied significantly between Sweden and the United Kingdom. While British newspapers, some of which are owned by News Limited, were dominated by ‘crime news narratives’, Swedish newspaper articles devoted much more space to the issue of fraud, including information about the cost of fraud and the implementation of anti-fraud policies ([35], p. 638). This supports the position of Gekoski et al. [8] and others that the media acts as a gatekeeper, choosing which crimes to report and which ones to discard according to their own beliefs and perceptions of newsworthiness. However, the prevailing theme in the academic literature is that fraud, as a distinct crime type, is proportionally underreported in the media [31].

  • Hypothesis 1: fraud is less likely to be reported in the media than other crime types.

Sullivan and Chermak [30] suggest that while the reporting of fraud and financial crime may be lower than other crime types, those that do make it into the media are often highly publicized and sensationalized scandals such as HIH Insurance in Australia, Enron in the United States and GlaxoSmithKline in China. The large scale, high profile cases are often there because they, as Levi [31] suggests, are ‘in your face’ reporting of corporate immorality which is highly newsworthy and provides the media with an opportunity to expose the gap between the collective ‘us’ and ‘them’. The challenge associated with reporting on fraud can be attributed to the lack of a clearly definable innocent victim. Providing the example of Nick Leeson, the Hong Kong based trader who brought down Barings Bank, Levi [29] argued that it was difficult to distinguish between the victim and the accused, however in other cases, like the Robert Maxwell Pensions Fraud in the United Kingdom in the 1990s [36], the identification of victims was much more apparent.

Media outlets, it is suggested in the general media literature, tend to draw on stories of violent, unusual and sensational crimes, while eschewing reports of white collar crimes as these investigations are often long and complicated and the victims unseen [15, 20, 37,38,39,40]. If, as Levi [31] suggests, financial crimes are treated as ‘infotainment’ then this in itself still places financial crimes in the media spotlight, albeit for a shorter duration, and perhaps without the requisite focus on its blatant criminality. Recognizable and celebrated cases, according to Sullivan and Chermak [30], receive the majority of attention at the expense of an analysis of societal problems. Levi [29] points out that ‘[a]s with violent crimes, the media emphasise the polar extremes of cases with traumatised victims and offenders that can be portrayed as extravagant and uncaring’. As such ‘fall from grace’ lifestyle stories get almost as much attention. For example, the ‘Chase for Skase’, as it was referred to in the Australian media, led to extensive media coverage of accusations of fraud against media mogul Christopher Skase, who had fled to Spain to avoid prosecution in Australia; a pervasive news story that was fuelled by the availability of images of Skase at his Majorca hideout, and the ‘us and them’ news frame that existed [41].

Another theme is the tendency for fraud and financial crimes to be reported in a simplified manner, removing the overt complexity of the crimes. Sullivan and Chermak [30] contend that such levels of complexity may be beyond the scope of the general public. Lampe [41] suggests that there are far more complex fraud offences occurring than are being reported in newspapers because of the difficulty of obtaining good pictures to support such stories, citing the Skase case as a notable exception. Minkes and Minkes [42] assert the media are likely to emphasize a general perception of traditional crimes while ignoring financial crimes, and when covered the language is often ‘sanitised’ thus removing any overt links to criminality; instead such language links the financial crimes to other more newsworthy events such as drugs scandals, celebrities and failing economies ([42], p. 23).

  • Hypothesis 2: high profile and high value frauds dominate media coverage of this crime category.

Current study

In order to explore the comparative reporting of fraud, the researchers identified the six most costly crime types in Australia as reported by the Australian Institute of Criminology [3]. Fraud, arson, drugs, burglary, assault and theft were identified as contributing up to 85% of the total cost of crime across Australia. Using major New South Wales print media outlets as a case study, researchers collected all print media articles with reference to any of these six crime areas. This approach enabled the comparison of levels of reporting in order to establish whether fraud continues to be underrepresented in Sydney print media, as suggested in the literature.

Method

Sample selection and data collection

To test the hypothesis researchers conducted a detailed content analysis of 12 months of newspaper articles featuring fraud, arson, drugs, burglary, assault and theft. The methods used, including the sampling scheme, coding and analysis approaches will now be explained.

The first step was to conduct a literature review to identify the dominant historical positions on fraud reporting in the media. This produced hypotheses which in turn guided the data collection, and analysis.

  1. 1.

    Fraud is less likely to be reported in the media than other crime types.

  2. 2.

    High profile and high value frauds dominate media coverage of this crime category.

The second step was to design an appropriate sampling framework with which to gather data. Data from the Australian Institute of Criminology [3] Cost of Crime Report was used to identify the six most costly crime types for the 2013 calendar year. These crime types - fraud, assault, burglary, arson, theft and drugs - were then compared against the offence categories as listed by the New South Wales Bureau of Crime Statistics and Research [43] which identify a range of reportable crime categories. These categories were used to create a list of relevant synonyms with which to search for related media articles as seen in Table 1.

Table 1 Search strategies
Full size table

The third step was to identify a suitable database from which to draw related articles. While it was possible to conduct a visual search of the printed publications over a 12 month period, the researchers felt an electronic approach to the collection would a) minimise human error associated with missing relevant articles, b) ensure the data collection was done in timely manner, and finally c) provide a replicable method for future researchers to validate the methodology of this study. The researchers were also aware that the text-only nature of electronic news content removed some of the visual impact of the print medium, including images, font size and placement on the page. However, this consideration needed to be measured against the strengths described above. The Dow Jones Factiva database was identified for this purpose. The Factiva database aggregates business and media content from a wide range of Australian and international providers, including full coverage of newspaper content for all Australian metropolitan titles. The sampling strategy using this database was centred around three elements: 1) the topics, consisting of the top six crime categories provided by the Australian Institute of Criminology [3] Cost of Crime Report; 2) the newspaper titles, The Sydney Morning Herald and The Daily Telegraph; and 3) the time frame (January 1 to December 31, 2013) [44]. Factiva provides a subject search option, but this approach to data collection was discounted because although it was possible to identify crimes types that notionally fell under a second level subject heading Crime/Courts, which is a sub-heading under the broader subject Political/General News, the narrower subject headings under the Crime/Courts category were inconsistent with the crime types being researched. The crime types of assault and fraud had matching subjects, however burglary and theft, which are treated as separate crime types by the Bureau of Crime Statistics and Research and the Australian Institute of Criminology were combined in Factiva under the one subject heading. There were also no subject headings for either arson or drugs. As well as the lack of specific subject headings suitable for all six of the crime types under investigation, it was determined that the subject indexing of articles was not applied consistently. This was especially the case for news items reporting on multiple incidents of crime. It was therefore necessary to conduct keyword searches in order to apply the level of specificity that was not available using the subject search. This strategy required the use of the synonyms contained in Table 1 for each of the crime type searches in order to retrieve relevant articles that may not have included specific mention of the crime type term.

The next step was the selection of newspaper titles where the preliminary searches were limited to the Combined Australian Newspapers Search feature of the Factiva database. The titles in this set consist of the daily and weekend editions of all territory and state capital city titles and the national broadsheets, The Australian and the Australian Financial Review (AFR). The results from this search with the initial limitations produced what was considered an unmanageably large dataset. A smaller, more manageable data set of articles sourced from the daily and weekend editions of the newspapers in circulation in Sydney, the Sydney Morning Herald and the Daily Telegraph was preferred. This data set is to be referred to as the Primary Data Set (Table 2).

Table 2 Search results and sampling
Full size table

The time frame of the research period from 1 January to 31 December 2013 was selected as it was the last completed calendar year prior to the commencement of the research. This choice also limited the number of results to a more manageable set. It was considered that the selection of a full year for the searches was preferable to other sampling options. Limiting results to alternate days, weeks or months, or to a shorter consecutive period such as a quarter, may not have provided representative results.

Coding and analysis

The Primary Data Set consists of 16,547 articles as represented in the second column of Table 2 by crime type. For each crime type a Secondary Data Set consisting of 20%, or one in five, of the Primary Data Set articles were selected to be used in the research project. To achieve this a systemic sampling method was used whereby, once date ordered, every fifth article was selected to be included in the final sample as shown in the third column of Table 2.

Each of the resulting Secondary Data Set’s contained within the third column in Table 2 were coded according to metadata elements present within, or applied to, each article as well as some interpretive considerations as determined by the authors. Initial interpretive coding was used to disqualify articles containing false hits and, for those articles retained, giving some classification based on qualitative measures. False hits included instances in which the context rendered the article irrelevant. Examples of this include reporting on circumstances other than crime, such as legal pharmaceutical drugs and fictional crime, such as in reviews of television programs. False hits were also retrieved from articles in which the search words were used figuratively (‘assault on the senses’, ‘stole my heart’). It was considered preferable to eliminate these false hits from the systematically selected Secondary Data Set rather than conducting this activity with the Primary Data Set.

The metadata elements consisted of descriptive and quantitative elements, the most basic of which were; the publication title and section, article title, date, page number, and word count. Each article was allocated a record number (001–999) which was combined with a crime type identifier (D for drugs, AS for assault, AR for arson, etc). There were also a number of elements that required more interpretive consideration. The first of these related to continuing themes. It was clear many articles stemmed from particular events for which there was media interest over an extended period, for example the Australian Sports Anti-Doping Authority (ASADA) investigation into the Essendon Football Club. For relevant articles on recurring topics suitably titled themes were allocated thereby measuring the impact of these themes on the quantity of records returned from the searches. This factor is explored further in the “Discussion” section.

A number of articles also related to more than one of the crime types. Streetwatch, a regular feature in The Daily Telegraph, generally includes three or more short reports of distinctly separate arrests or criminal incidents that occurred the previous day. A number of other articles with a central focus on one crime type also made reference to other types of crime. Articles including more than one crime type were therefore coded with a primary crime type for the first mentioned crime or the crime of major focus in the article. Subsequent or more briefly mentioned crime types were coded as secondary. Each of the distinct and separate events included in a Streetwatch report and relating to one of the crime types was treated as an individual record with its own word count, but with coding to indicate primary or secondary type according to the order in which it appeared in the article. This better reflected the impact of Streetwatch on search results.

Because the cost of crime is a central factor addressed in the Australian Institute of Criminology [3] report the researchers also coded articles that included statements relating to the cost, whether that was the cost of crime to society more broadly or the value associated with a specific crime event. Similarly, researchers also coded articles that included information on penalties applied for the crimes referred to. This information was less likely to be included in articles reporting immediate crime and was more often seen in court reporting. Location was also coded whereby articles were coded as relating to international or Australian stories, with the relevant country included for international stories and the relevant state for Australian stories.

Findings

Fraud attracts consistent levels of media coverage in comparison to other crime types

The results of the initial search in the Factiva data base seemed to conform to relevant academic research that suggested fraud is underreported in Sydney print media [10, 26, 29,30,31,32]. The initial search identified a Primary Data Set consisting of 16,547 articles (Table 2). Of this 31.66% of all crime media coverage was linked to drug matters, assault accounted for 28.64% of the media coverage, fraud accounted for 19.65%, and theft accounted for 16.17% of the coverage, while arson accounted for only 2.37% of the coverage and burglary accounted for 1.52%. Analysis of the Secondary Data Set containing 3307 articles provided an almost identical picture with 31.66% of the articles being linked with drugs, 28.64% linking to assaults, 19.66% linking to frauds, 16.18% linking to thefts, 2.36% linking to Arsons and 1.51% linking to burglaries.

Analysis revealed much of the news coverage on the topics of assault and drugs had no direct link with the commission or investigation of crime. For example, individuals in news articles were often described as drug-users or under the influence of a drug. Many articles discussed the use of performance enhancing drugs in sport, which, while prohibited by a sport’s governing body, are not a criminal offence in most circumstances. Numerous articles discussed drugs in the context of health treatment and disease prevention. The term assault was also used liberally in news articles and many articles discussed ‘hits’ that took place in sport, which is not a criminal offence in most circumstances. The analysis did not include the AIC topics of homicide, sexual assault and criminal damage, thus any use of the word assault to describe these offences was omitted.

When subjected to closer scrutiny, articles with no connection to crime were removed resulting in the Final Data Set of 935 articles. This final data set provided a clearer picture of fraud reporting in the media that contrasts sharply with commonly held perceptions in the surveyed literature. Fraud was the most common crime type reported featuring 34% of the final data set, assaults and drug related offences both accounted for 24% each, while theft 13%, arson 3% and burglary 2% rounded out the final data set.

The initial hypothesis derived from the literature was fraud is less likely to be reported in the media than other crime types. Much of the literature surveyed asserted that fraud was a bigger problem in society than the media gives it credit for; this was in turn supported by the data from the Australian Institute of Criminology [3] which suggested fraud was the most costly crime in Australia. The data gathered in this research however fails to support the stated hypothesis (Fig. 1). Subsequently it is proposed that within the constraints of the current research, fraud is the most reported crime type by number and as such it can be stated that fraud is not proportionally underrepresented in Sydney print media.

Fig. 1
figure 1

Sydney daily newspaper reporting by crime type in 2013

Full size image

To negate suggestions that 2013 was an anomaly in respect of its representation of fraud in the media, raw data for each offence type was gathered between 2010 and 2014 (Fig. 2). From this raw data, fraud, arson, drugs, burglary, and assault all remain relatively constant while the reporting of theft appears to have fallen year on year.

Fig. 2
figure 2

Crime types raw data 2010–2014

Full size image

High profile and high value frauds dominate media coverage

Coverage of fraud-related crime in 2013 was dominated by reports of government corruption. Of the 312 fraud related articles contained within the final sample, 122 related to a series of investigations and hearings by the NSW Independent Commission Against Corruption relating to former and sitting members of the NSW Parliament. Nineteen articles discussed federal politics and the investigation of MP Craig Thompson over the alleged misuse of a union credit card to procure the services of prostitutes. Eleven articles related to Australian cricket and soccer match fixing, seven related to celebrity chef Nigella Lawson and her husband’s fraud trial, horse race fixing accounted for six, while the activities of Australian’s Matt Joyce and Marcus Lee in Dubai accounted for three articles. Together these potentially ‘high profile’ cases accounted for 53.85% of all fraud coverage within the final sample. When examining the remaining 46.15% of articles published it became apparent that instead of solely concentrating only upon ‘high profile’ cases, the reporting covered a range of cases including insider trading, door-to-door conmen, customs frauds, frauds against the Catholic Church, the Commonwealth Bank, Elders and a range of other businesses and individuals.

The second hypothesis suggested that high profile and high value frauds dominate media coverage of this crime category. However instead of concentrating on so-called ‘high profile’ cases a significant percentage of the articles in this sample covered a diverse range of fraudulent activities. What came through however was that the articles, for the most part were linked, not to ‘high profile celebrity’ cases as a common theme, but to structural and well know publicly or privately listed organisations that most Australians interact with on a daily basis. The reporting of fraud in this data set is not viewed as being represented as ‘infotainment’ or ‘sensationalised crime’ as described by Levi [31]; Sullivan and Chermak [30]. Instead it is evidence that the reporting appeared to relate more readily to either the direct or indirect impact fraudulent activities were having on these organisations and the subsequent flow-on effect these crimes have on the end user’s costs of living, security and safety. While potentially high profile cases marginally dominated the data set at 53.85%, it is suggested that a significant number of fraud cases are reported on because of the broader impact they have on society as a whole. This suggests that editors and journalists are focussed to some extent on a more mature level of reporting that deals with real world problems and consequences instead of simply as another form of entertainment.

Fairfax versus News Ltd

The results show that the Fairfax newspaper (Sydney Morning Herald) in NSW is more likely to report on fraud than the News Limited newspaper (The Daily Telegraph), although the opposite is true for all other crime types. Of the 935 crime articles found to be relevant, 540 (58%) were published by News Limited, showing that Fairfax published fewer articles relating to crime. Despite this 46% of the crime-related articles published by Fairfax were fraud related, whereas only 24% of News Limited crime-related articles were fraud-related (Fig. 3). It might be suggested such a finding could be attributed to the Fairfax newspaper having a substantially larger business section compared to the News Limited newspaper. Analysis of this reveals that 7.7% (n = 14) of the Fairfax fraud articles were located in the business section, while only 4.6% (n = 5) were located in the News Limited business or money sections. The analysis reveals that both publishers were more likely to treat fraud as general news content rather than business or money content. Reports of fraud crime were also more prominent on the front pages of Fairfax newspapers, although a report on fraud crime was more likely to make it onto the front page of a News Limited publication than other crime types (Fig. 4). Generally, newspaper articles about fraud crime were longer than articles for all other crime types (Fig. 5). On average, Fairfax published more words per article than News Limited for all crime types.

Fig. 3
figure 3

Comparison of Fairfax and News Ltd reporting of crime

Full size image
Fig. 4
figure 4

Number of page 1 articles by publisher

Full size image
Fig. 5
figure 5

Average number of words per article by crime type

Full size image

Cost of crime and sentencing

The Final Data Set provided 312 articles relating to fraud and financial crime. Of those, 22.4%, or 70 articles contained specific reference to the reported costs associated with the crime. Fifty one of those articles related to high profile cases with a combined crime cost in excess of $500 billion. High profile cases involved those where multiple articles were published across the sampled newspapers on a single event such as the Craig Thompson or ICAC inquiries. A further 13 articles reported on non-high profile cases with a combined crime cost in excess of $600 million.

Sentencing data was also gathered during the analysis of the cases. Of the 32 articles, 10.2% of the total 312 fraud articles contained within the Final Data Set that contained references to sentencing, 19 articles reported custodial sentences, 6 articles reported monetary fines, 5 reported the awarding of unspecified sentences while suspended sentences, community service, home detention and the removal of a practicing license all received one mention in the articles examined as shown in Table 3.

Table 3 Reported sentencing
Full size table

Discussion

The major purpose of this study was to compare the prevalence of fraud related activities with the extent to which it is reported in Sydney print media. The study found that one third of crime reports in the two major New South Wales newspaper publications are related to fraud. The finding was contrary to the literature review which suggested that fraud would be underreported when compared to other crime types. The level of coverage suggests that editors and journalists consider fraud to be newsworthy [8]. The extent to which such reports influence public perceptions about the pervasiveness and seriousness of fraud crime is not really known, but when media reporting is concentrated on specific crimes public awareness, and concern is raised to the extent that political and policing responses are often required [12, 45].

The AIC revealed that fraud accounts for about 40% of the cost of crime to society, more than any other crime type. In money terms, fraud is the most serious crime in Australia and so it might be expected that law enforcement provide a proportional response. The absence of media coverage of fraud crime could have the effect of lulling the public into believing that fraud is not as significant a problem as the AIC figures set out. But, as this study shows, Sydney print media is giving the issue of fraud the prominence it warrants. The findings of this research negates the perception that fraud is an under reported crime type within the scope of this research project [29,30,31, 41]. Fraud during this sample period was the most commonly reported crime type. Strengthening these results is the work of Goh and Holmes [46] that identifies in New South Wales during that period there were a total of 49,218 frauds reported to relevant authorities, an increase of 13.2% on the previous 12 month period. The evidence provided in this study about the prominence of fraud coverage in the two influential New South Wales newspaper publications raises fraud as a social issue warranting an appropriate political and policing response.

Of all the crime types, fraud received the most front page coverage in 2013, with 21 page one articles in The Sydney Morning Herald and six in The Daily Telegraph. The decision by The Sydney Morning Herald to give more prominence to fraud is likely to be a reflection of the preferences of its readership compared to those of The Daily Telegraph, or perhaps at least what the editors and journalists believe are their readership’s preferences. The fact that either newspaper has given prominence to the reporting of fraud crime is also likely linked to the availability of pictures and subjects with a high media profile, such as former and serving politicians [41].

All but two of the front page articles in The Sydney Morning Herald and The Daily Telegraph related to the dominant fraud story of 2013, the investigation of current and former politicians by the NSW Independent Commission Against Corruption (ICAC). Journalists had access to the ICAC hearings where allegations against several public figures were laid bare, and pictures of the involved parties were easily obtained, meaning all the necessary ingredients of a viable front page story were present [41, 47, 48]. However, the study found that fraud coverage in 2013 was not unusually high despite the prominence of the ICAC matters, with newspapers giving consistently high coverage to fraud between 2010 and 2014 (Fig. 2). This suggests that the media has a regular supply of similarly high-profile fraud cases on which to report each year, but it also indicates that fraud matters are routinely accepted as newsworthy by editors and journalists, particularly as almost half of the fraud coverage in 2013 was unrelated to the high profile cases.

This study’s focus on sampled articles from two newspaper titles from the same city over a recent calendar year presented a number of useful findings as outlined above. Opportunities for further research based on this study may be informed by these limitations of scope, data collection and data sampling. Expanding the titles selected for data collection to include those from other cities or selecting instead those from another city would provide comparable data that may show location-specific variations or point to a more generalised approach to fraud reporting in print media. Similarly, expanding or varying the year of publication for source articles may demonstrate variations or consistencies over time. Additionally, wider sampling of articles in the search datasets may also be considered. This would ensure greater reliability, but it requires considerably more time for analysis. A more significant variation that future researchers may wish to consider relates to the media type. Research into the treatment of fraud reporting more generally would benefit from studies focused on other media, including radio and television news broadcasting.

Conclusion

This study provided a case study analysis of crime reporting by the two largest newspaper publications in New South Wales for 2013, The Daily Telegraph and The Sydney Morning Herald. Two hypotheses relating to the frequency of fraud coverage in the media and the nature of fraud coverage in the media were developed from a review of the topic literature. The findings were surprising in that they conflicted with the prevailing understanding of media coverage of fraud, finding that fraud attracts significant coverage in the media, more than any other crime type, and this held true for both publications. The implications for law enforcement and law makers are considerable given the influence that media coverage has on public perceptions of crime and public safety.