BSA Decisions Ngā Whakatau a te Mana Whanonga Kaipāho

Summary

[This summary does not form part of the decision.]

The Authority has upheld a complaint that comments made by Mike Hosking during his ‘Mike’s Minute’ segment were misleading in breach of the accuracy standard. Mr Hosking made statements referring to death-rate statistics in Italy related to COVID-19, including that ‘99.2% percent died with underlying health issues. In other words, the very things that were killing them anyway, at over 1,600 per day’. The Authority found the comments were misleading as the broadcaster conflated its own conclusions, drawn from a study into Italy’s COVID-19 figures, with the figure of 1,600 deaths per day, which was based on 2018 population data and ignored both cause of death and the notion of ‘excess mortality’. In this respect, the Authority emphasised the importance of data literacy among broadcasters and journalists, to ensure statistics are not misinterpreted or misrepresented.  Finally, the Authority found the comments about people with ‘underlying health conditions’ did not reach the high threshold for finding a breach of the discrimination and denigration standard.

Upheld: Accuracy. Not Upheld: Discrimination and Denigration.

No Order

The broadcast

[1]  In his ‘Mike’s Minute’ segment during Mike Hosking Breakfast, host Mike Hosking considered and gave his views on a range of topics related to COVID-19, for example: whether the Government’s COVID-19 lockdown was ‘working’; whether it should be relaxed to resemble something more like the Australian situation to reduce the impact on the New Zealand economy; why New Zealand wasn’t testing for COVID-19 to its full daily capacity; the importance of remembering the total number of deaths in New Zealand, Spain and Italy per year (as opposed to deaths from COVID-19 specifically); the suggestion that most deaths in those overseas countries could also be attributed to underlying health conditions; and finally the Government’s rationale for imposing lockdown to manage impact on health services (noting at that time they were only at 50% capacity).

[2]  The specific comments that have been complained about were (italicised):

• ‘The alarm you're seeing offshore in places like Italy is false coverage to the extent that many of those who die were dying anyway. It’s a hard thing to say, but it's also true. …There are very few – a very, very few – who you could argue die specifically of the virus.’
• ‘In Italy, [the annual death rate is] over 600,000 a year, over 1,600 a day. …In Italy, 99.2 percent [of people who have died of COVID-19] died with underlying health issues. In other words the very things that were killing them anyway, at over 1,600 per day.’

[3]  The item was broadcast on 6 April 2020 on Newstalk ZB. In considering this complaint, the members of the Authority have listened to a recording of the broadcast, and we have read the correspondence listed in the Appendix.

Summary of the complaint

[4]  Shelley Burne-Field complained that Mr Hosking’s comments about people with underlying health conditions and Italy’s death-rate statistics breached the accuracy and discrimination and denigration standards of the Radio Code of Broadcasting Practice. In summary, this was because:

• The statement, ‘There are very few who you could argue die specifically of the virus’, and the reference to ‘1,600 [deaths] a day’ in Italy, were inaccurate.
• Mr Hosking’s comments were malicious and encouraged discrimination against, and the denigration of, people with underlying health conditions and/or the aged, by suggesting most people who died from COVID ‘were dying anyway’ (by comparing the grave situations in Spain and Italy), and so not worthy of ‘the protection given to ALL citizens in our country through a lockdown’, ‘based on a Hosking-perceived resultant economic down-turn’.

[5]  The complaint and the broadcaster’s response are outlined in more detail in our consideration of the accuracy and discrimination and denigration standards below.

The relevant standards

[6]  The accuracy standard (Standard 9) states that broadcasters should make reasonable efforts to ensure that news, current affairs and factual programming is accurate in relation to all material points of fact and does not mislead. The objective of this standard is to protect audiences from being significantly misinformed.¹

[7]  The discrimination and denigration standard (Standard 6) protects against broadcasts which encourage the denigration of, or discrimination against, any section of the community on account of sex, sexual orientation, race, age, disability, occupational status or as a consequence of legitimate expression of religion, culture or political belief.

Overview of outcome: Freedom of expression weighed against harm

[8]  The right to freedom of expression, including the broadcaster’s right to impart ideas and information and the public’s right to receive that information, is the starting point in our consideration of complaints. Our task is to weigh the value and public interest in the broadcast complained about, against the level of actual or potential harm that may be caused by the broadcast, with reference to the objectives of the standards described above. We may only interfere and uphold complaints where the limitation on the right to freedom of expression is reasonable and justified.

[9]  We recognise there is value and public interest in scrutinising, and offering critical commentary on, the Government’s actions with respect to combatting COVID-19. We have also previously acknowledged that hosts such as Mr Hosking are well known to offer strong or provocative opinions, and there is value in this approach for the purpose of generating discussion and public discourse. The free and frank expression of opinions, particularly on a topic of high public interest, is protected by the right to free speech, so long as standards are maintained.

[10]  However, when we considered the potential harm caused by the broadcast, we agreed with the complainant’s submissions in relation to the accuracy standard. The manner in which Mr Hosking used data gleaned from internet articles, to bolster his views on the Government’s actions with respect to COVID-19, risked misleading the audience on an issue of high public importance. We therefore concluded that upholding the complaint under the accuracy standard would place a reasonable and justified limit on freedom of expression.

[11]  We explain below the reasons that have led us to this conclusion.

Accuracy

The complaint

[12]  The complainant identified two statements as being inaccurate, as follows:

• In relation to the statement that ‘There are very few who you could argue die specifically of the virus’, Ms Burne-Field said: ‘this is not accurate. COVID-19 has killed over 70,000 people world-wide as stated by the [World Health Organisation (WHO)].’
• Regarding the reference to ‘1,600 [deaths in Italy] a day’, the complainant said: ‘He has upped the ante (condemnation and malice) against a vulnerable cohort of people “with underlying health issues” by using an inaccurate statistic which tries to show that more people in the Italian community (comparable to more people in the NZ community) possess “the very things that were killing them anyway, at over 1600 per day” – when in fact he is quoting the national statistic for all deaths.’

The broadcaster’s response

[13]  NZME found no breach of the accuracy standard, saying:

• While the BSA has recently determined that ‘Mike Hosking Breakfast amounts to a news and current affairs programme to which the accuracy standard applies’,² Guideline 9a states that the standard does not apply to statements which are clearly distinguishable as analysis, comment or opinion.
• The Authority found in another recent decision considering ‘Mike’s Minute’ that:³

There is an established audience expectation that the ‘Mike’s Minute’ segment is an opinion piece in which Mr Hosking gives his own, often strongly expressed, views on his chosen topic. Audiences also expect this from Mr Hosking and Newstalk ZB generally. Therefore listeners were likely to take this piece as being Mr Hosking’s own opinions, rather than unqualified statements of fact.

• When Mr Hosking stated that ‘there are very few who you could argue die specifically of the virus’, he was making the point that the vast majority of those who succumbed to Covid-19 had underlying health issues and died from a combination of factors. This was analysis and opinion rather than an unqualified statement of fact.
• When Mr Hosking stated that, ‘In Italy, 99.2 percent died with underlying health issues. In other words the very things that were killing them anyway, at over 1600 per day’, the reference to ‘1,600 a day’ was clearly a reference to the total number of daily deaths in Italy. This was not inaccurate or misleading.

Request for further information about sources relied on by Mr Hosking

[14]  To assist us in our consideration of whether the comments complained about were inaccurate or misleading, we requested further information from NZME regarding the sources and dates of the information Mr Hosking relied on in making the comments, as at the time of broadcast. NZME responded that:

• The average death rates quoted for New Zealand, Spain and Italy were taken from a website (in Italy’s case, 10.5 deaths/1,000 population (2018 est.)),⁴ and ‘the approximate total annual deaths was then calculated using the total population for each country. For example, for New Zealand: 4.7m (in 2018)/1000 x 7.6 = 36,480’.
• With respect to the statements made about Italy’s COVID-19 death rates, the broadcaster relied on two articles:
  • ‘More Than 60 Doctors in Italy Have Died in COVID-19 Pandemic’ (Medscape, updated 31 March 2020).⁵ NZME noted the article states that:
    • ‘…In Italy, the case-fatality statistics are based on defining COVID-19-related deaths as those occurring in patients who test positive for [COVID-19], independent from pre-existing diseases that may have caused death. …this may have led to an overestimation of the case-fatality rate, [experts] warn…’
    • ‘An analysis of a subgroup of 355 patients who died from COVID-19 shows that they were elderly and had many comorbidities. The mean age was 79.5 years and half of the patients (48.5%) had three or more underlying diseases, while the remainder had either one or two underlying conditions (25% in each group). The underlying conditions included ischemic heart disease (30%), diabetes (35.5%), active cancer (20%), and atrial fibrillation (24.5%). Only 3 patients (0.8%) had no underlying disease.’
  • ‘99% of Those Who Died From Virus Had Other Illness, Italy Says’ (Bloomberg, 19 March 2020).⁶

[15]  In response to this further information, the complainant argued:

• Everyone who contracts COVID-19 is ‘dying anyway’ in the sense that as human beings we will all die eventually. However, Mr Hosking was adamant the deaths in Italy from COVID-19 were caused by something else and, in the complainant’s words, “their lives were forfeit, so why worry about it?” Mr Hosking did not question when those deaths were going to happen.
• ‘Mr Hosking’s lay-person’s conclusions from information he has found on the internet are not accurate’ and the sources relied on do not back up his claims.
• The statistics quoted relate to a subsample of 355 patients who officially died of COVID-19 (out of 2,003 total deaths at the time). ‘The 99.2% of deaths that Mr Hosking refers to is actually 99.2% of the 355 patients who died, or 352 deaths, NOT “at over 1,600 a day”.’
• The Medscape article is not about evaluating the cause of death but rather considers possible factors contributing to Italy’s high case-fatality rate. It ‘does not make any conclusions regarding whether comorbidities (pre-existing conditions) cause death rather than COVID-19’.
• Many other medical conditions besides those listed in the study (heart disease, active cancer etc) could be considered ‘underlying conditions’. The complainant gave numerous examples of conditions people can live with and successfully manage for many years (eg high blood pressure, diabetes).
• ‘Mr Hosking has drawn a conclusion that people with these conditions are ready to drop dead any moment. He has no background to this information. … It is VERY clear that the ISS study and the JAMA commentary, heavily used and cited in the Medscape and Bloomberg articles [relied on by the broadcaster], are about the increased risks for various groups of people in contracting COVID-19, and consequently dying from it. They are NOT about the fact that these people would have died anyway!’

Our analysis and the outcome

[16]  Determination of a complaint under the accuracy standard occurs in two steps. The first step is to consider whether the programme was inaccurate on a material point of fact, or misleading. The second step is to consider whether reasonable efforts were made by the broadcaster to ensure that the programme was accurate and did not mislead.⁷

[17]  In this case, we have focused our deliberations on the latter requirement to ensure the programme was not ‘misleading’. The key question in our view was whether the manner in which Mr Hosking used particular figures and statistics to support his views relating to COVID-19 death rates, had the potential to mislead the audience. Being ‘misled’ is defined as being given ‘a wrong idea or impression of the facts’.⁸ Programmes may be misleading by omission.⁹

[18]  We have carefully considered the information and comments from both the complainant and the broadcaster, and reached the view that first, the comments complained about were misleading by omission, and second, the broadcaster did not make reasonable efforts to ensure the programme was not misleading in that respect. The key factors that have contributed to our finding that the comments were misleading are:

• The focus of the first article cited by the broadcaster is variations in countries’ case-fatality rates and the factors that may have contributed to Italy having a high case-fatality rate compared to other countries, namely:
  • Italy’s older age distribution
  • the way Italy defined COVID-19 deaths (anyone who tested positive)
  • limited testing for those who are asymptomatic or only present mild symptoms, resulting in a high proportion of positive results and an increased case-fatality rate overall.

On one hand, it could be argued that this supports the point Mr Hosking was making, that coverage of Italy’s death rate from COVID-19 was alarmist without acknowledging there may have been other factors at play both in the deaths and in the statistics.

• However, Mr Hosking in our view overstated the position outlined in the sources relied on, namely that 99.2% of people who died in Italy from COVID-19 were ‘dying anyway’ or had conditions that were ‘killing them anyway’. The Medscape article makes no statement that these people were ‘dying anyway’; it says only that ‘The presence of these comorbidities might have increased the risk of mortality independent of COVID-19 infection’ [emphasis added]. As noted by the complainant, the article is also not focussed on evaluating the cause of death but rather commenting on the factors that possibly contributed to Italy’s high case-fatality rate.
• Mr Hosking did not make it clear that the figure of 99.2% of victims he referred to as having ‘underlying conditions’ and ‘the very things that were killing them anyway’ was actually 99.2% of an early study of a subgroup of 355 out of a total 2,003 deaths in Italy at that time.
• Mr Hosking then conflated his own selective conclusions drawn from the study and the articles relied on, with his own calculation of Italy’s annual death rate, which appears to have been based on 2018 estimated population data and disregards any other causes of death included in the annual death rate (besides people with ‘underlying conditions’).
• As noted by the complainant, many other medical conditions besides those listed in the study could be considered ‘underlying conditions’, which people can live with and successfully manage for many years. There was a risk of listeners being misled as to what Mr Hosking was categorising as an ‘underlying condition’ that was ‘killing them [the people in Italy who died of COVID-19] anyway’. Mr Hosking then used this to bolster his suggestion that the effectiveness of the Government’s (in his view) extreme measures to combat COVID-19 (for example the lockdown) may not outweigh the significant impact on the economy – or, as the complainant articulated, ‘why bother’ with measures to protect people with underlying conditions or comorbidities from the virus?
• Finally, and importantly, in our view, Mr Hosking did not consider or acknowledge the possibility that the 352 deaths (with presence of comorbidities) considered in the Italy study he relied on may have died a lot sooner than they might otherwise have, due to contracting COVID-19. We understand ‘excess mortality’ is a term used in epidemiology and public health that refers to the number of deaths above and beyond what we would have expected to see under ‘normal’ conditions. In the context of a period of crisis, it is described as ‘mortality above what would be expected based on the non-crisis mortality rate in the population of interest’ or ‘mortality that is attributable to the crisis conditions’ (in this case as a result of COVID-19).¹⁰ One online source we considered suggests that, for the period 1-7 April 2020, around the time of the broadcast on 6 April, Italy was experiencing a 58% deviation from its expected death rate.¹¹

[19]  For similar reasons, we also concluded the broadcaster did not make reasonable efforts to ensure the programme was not misleading in these respects. Relevant contextual information was omitted that was easily discernible from the same sources relied on for Mr Hosking’s comments, with the result that the broadcast had the potential to mislead listeners about the facts behind Mr Hosking’s relatively strong views critiquing the Government’s measures to manage the impact of COVID-19.

[20]  As we have mentioned in paragraphs [8]-[10] above, the final step in our analysis is to complete the balancing exercise to determine whether the accuracy complaint ought to be upheld, weighing the important right to freedom of expression and any public interest in Mr Hosking’s views in this broadcast, against the harm alleged to have been caused. Recognising that Mr Hosking is in a privileged and potentially influential position as a well-known national broadcaster, we are satisfied that upholding the accuracy complaint in this case does not unreasonably limit the broadcaster’s free speech or prevent Mr Hosking from expressing his views. Rather, it reasonably requires Mr Hosking to express his views in a way that does not propagate a selective or misleading interpretation of the factual sources relied on, or omit important contextual information that may alter listeners’ understanding of the views presented on a topic of high public interest.

[21]  This case has highlighted, in our view, the importance of data literacy in a broadcasting context and in news and current affairs commentary in particular. We urge broadcasters to take care when sourcing and interpreting statistics, and drawing conclusions from scientific or other studies, given that audiences rely heavily on mainstream media to provide authoritative, reliable information on matters of public importance.

Discrimination and Denigration

The complaint

[22]  The complainant considered Mr Hosking’s comments breached the discrimination and denigration standard for the following reasons:

• ‘Hosking’s article/broadcast absolutely denigrates and encourages discrimination against a signification section of NZ society – those who may be disabled by chronic illness – cited by Hosking in the context of Covid-19 as those with “underlying conditions”, and/or the aged.’
• ‘He actively denigrates NZ known populations living with… underlying health issues – who contribute to the NZ economy on an equal footing with other NZ citizens.’
• ‘Their lives and their health, age, and disability status are denigrated by Hosking who shows them as not worthy of saving if that act adversely affects NZ’s economic status. His own words “How sorry will we be, as thousands face joblessness, we face debt for generations, and an economy in recession for what turned out to be, potentially, not a single death that would not have occurred anyway.” This is outrageous.’
• ‘His condemnation and malice doubles down throughout the article’s tone. … [The] constant ‘economy vs health’ stance seriously discriminates against the cohort mentioned above… encouraging listeners to reach Hosking’s obvious discriminatory belief that “those people [with underlying conditions] die anyway”.’

The broadcaster’s response

[23]  NZME did not find any breach of the standard, saying:

• The BSA’s well-established guidance and approach provides that a high level of condemnation, often including an element of malice or nastiness, is required to find a breach in light of the important right to freedom of expression.
• The BSA has also recognised talkback radio is a robust, opinionated environment, where hosts are known for making provocative statements to stimulate robust debate, and audience expectations reflect this.
• The Authority has previously stated that ‘audiences expect that Newstalk ZB will often feature conservative and controversial opinions.’¹² Mr Hosking is well-known for his conservative and controversial views and the Authority has stated that speech which offends is not determinative of a breach.
• While others may disagree with Mr Hosking’s opinion, it was clearly an opinion and one he was entitled to give – this is recognised in guidelines 6b (freedom of expression) and 6c (‘this standard is not intended to prevent the broadcast of material that is… a genuine expression of serious comment, analysis or opinion’).
• ‘Mr Hosking gives his view that the New Zealand Government should be relaxing the lockdown to something resembling the Australian approach, on the basis that this would be less damaging to the economy. His point regarding Covid-19 fatalities in Italy and Spain is that viewed in isolation the statistics are frightening but that it should be remembered that many of those who died had underlying health conditions. This does not equate to encouraging discrimination or denigration against those with underlying health issues or constitute a high level of condemnation such as to constitute a breach of this standard.’
• ‘Mr Hosking further states his view that “this health stance at the expense of the economy was launched through a fear our health service would be overwhelmed.” We consider that this is a correct assessment of the rationale for the imposition of the Level 4 lockdown, as the Government had seen clearly what was happening in other countries and acted to prevent New Zealand suffering the same fate. His point is that since this had not happened, one of the principal reasons for the lockdown was no longer a relevant consideration.’

Our analysis

[24]  The first question we considered was whether Mr Hosking’s comments about people with ‘underlying health conditions’ referred to a recognised ‘section of the community’ for the purposes of the standard, on the basis of ‘disability’ (the wording of the standard is set out in paragraph [7] above).

[25]  The Human Rights Act 1993 prohibited grounds of discrimination (on which the standard is based) define ‘disability’ broadly, including: physical disability or impairment; physical illness; psychiatric illness; intellectual or psychological disability or impairment; any other loss or abnormality of psychological, physiological, or anatomical structure or function; or presence in the body of organisms capable of causing illness.¹³ On this basis, we think people with underlying health conditions could reasonably be considered a section of the community to which the discrimination and denigration standard applies.

[26]  However, we found the broadcast did not reach the high threshold for encouraging the denigration of, or discrimination against, these people as a group, taking into account:¹⁴

• Mr Hosking’s views were protected by guidelines 6b and 6c to the standard, as ‘a genuine expression of serious comment, analysis or opinion’ on a topic of high public interest.
• His comments did not carry a high level of condemnation, malice or nastiness towards this group, or amount to hate speech. He was expressing his genuinely held opinion (albeit in a manner that had the potential to mislead the audience, as we have found above) and was relatively measured in his delivery.
• Mr Hosking’s comments made a legitimate contribution to a wider debate and were not calculated to hurt or offend. Mr Hosking explained his views on a range of aspects relating to Government actions to combat COVID-19, its rationale for taking such action, comparisons with the (much graver) situations in Spain and Italy, and the possible impact on New Zealand’s economy.
• Audience expectations that talkback radio, Newstalk ZB as a station, Mike Hosking, and the ‘Mike’s Minute’ segment will offer robust, opinionated discussion and sometimes provocative or controversial views in the interests of generating debate about topics of public interest.

[27]  For these reasons, we have not found actual or potential harm arising from Mr Hosking’s comments of the nature described in the complaint that warrants regulatory intervention with respect to Standard 6.

[28]  We therefore do not uphold this part of the complaint.

For the above reasons, the Authority upholds the complaint that the broadcast by NZME Radio Ltd of the ‘Mike’s Minute’ segment during Mike Hosking Breakfast on 6 April 2020 breached Standard 9 (Accuracy) of the Radio Code of Broadcasting Practice.

[29]  Having upheld part of the complaint, we may make orders under sections 13 and 16 of the Broadcasting Act 1989. We have concluded no order is warranted in this case. This is on the basis that publication of our decision is sufficient to: publicly notify the breach of the accuracy standard and censure the broadcaster; and to provide guidance to NZME and other broadcasters regarding the importance of data literacy and of ensuring statistics are not used to bolster opinions in a way that risks misleading audiences on matters of public importance, through the omission of relevant information.

Signed for and on behalf of the Authority

Judge Bill Hastings

Chair
14 September 2020

Appendix

The correspondence listed below was received and considered by the Authority when it determined this complaint:

1  Shelley Burne-Field’s original complaint – 7 April 2020

2  NZME’s decision on the complaint – 10 May 2020

3  Ms Burne-Field’s referral to the Authority – 21 May 2020

4  NZME’s response to the referral – 10 June 2020

5  NZME’s response to Authority’s request for further information relating to the accuracy standard – 16 July 2020

6  Ms Burne-Field’s comments responding to the further information – 17 July 2020

¹ Commentary: Accuracy, Broadcasting Standards in New Zealand Codebook, page 18
² Wilson and NZME Radio Ltd, Decision No. 2019-067 at [14]
³ Woolrych & Glennie and NZME Radio Ltd, Decision No. 2019-100 at [24]
⁴ <https://www.indexmundi.com/italy/death_rate.html>
⁵ <https://www.medscape.com/viewarticle/927753#vp_2>
⁶ <https://www.bloomberg.com/news/articles/2020-03-18/99-of-those-who-died-from-virus-had-other-illness-italy-says>
⁷ Commentary: Accuracy, Broadcasting Standards in New Zealand Codebook, page 19
⁸ Attorney General of Samoa v TVWorks Ltd, CIV-2011-485-1110
⁹ Commentary: Accuracy, Broadcasting Standards in New Zealand Codebook, page 19
¹⁰ Hannah Ritchie, Max Roser, Esteban Ortiz-Ospina and Joe Hasell, ‘Excess mortality from the Coronavirus pandemic (COVID-19)’: <https://ourworldindata.org/excess-mortality-covid>
¹¹ <https://www.economist.com/graphic-detail/2020/07/15/tracking-covid-19-excess-deaths-across-countries>
¹² Day & Moss and NZME Radio Ltd, Decision No. 2018-090 at [21]
¹³ Section 21(1)(h)
¹⁴ With reference to the guidelines to Standard 6 and the factors outlined in Commentary: Discrimination and Denigration, Broadcasting Standards in New Zealand Codebook, page 16