Yesterday morning I received this tweet:
@SincDavidson Hi Sinclair, I have tried to reach you for a story I’m doing on plain packaging. Wondering if you can please respond?
— Carly Weeks (@carlyweeks) October 18, 2016
Seemed very strange. I then got this tweet:
— Carly Weeks (@carlyweeks) October 18, 2016
@carlyweeks have you tried emailing me?
— Sinclair Davidson (@SincDavidson) October 18, 2016
It does seem seem a bit strange that someone would approach me for an interview via twitter as opposed to emailing me. But there you have it. Anyway I then received an email with the questions the journalist wanted to ask.
For your reading pleasure I’m posting the entire exchange below:
J: I have reached out because I’m doing a story on the plain packaging debate and the relationship between tobacco companies and think tanks.
SD: I can’t provide much input here – my involvement with think tanks is some-what limited. I am a full-time employee at an Australian University. In 2008 I participated in a work-experience program where I was seconded to a think tank (the Institute of Public Affairs) to observe how industry operates and bring back that knowledge to the university to better prepare students for a world of work. My role at the IPA consists of me providing a mentoring role for their in-house researchers – some of whom are undertaking Ph.D studies under my supervision. I derive no income from this role. In fact, I pay a membership fee to the IPA. To be clear, I am not an employee of the IPA and derive no income from them.
J: Part of the research I’ve done led me to some of the work you’ve done in this area.
SD: To be clear – a paper in the peer-reviewed journal Agenda and a working paper on the Social Science Research Network.
J: I understand that you were in Canada for a few events earlier this year on plain packaging.
J: I’ve also spoken to a few experts who have been critical of your work showing that contraband tobacco rose after Australia’s plain packaging law was introduced.
SD: Might be a case of mistaken identity. I have done no research into contraband tobacco. I have cited work done by KPMG in the area (this report is also quoted by Australian government agencies), newspaper reports, and submissions made to an Australian Parliamentary inquiry into contraband tobacco. They all suggest contraband has increased since 2012.
J: So I was hoping to speak to you or at least get a few comments for my piece in response.
I’m wondering if you can provide me some of the details of your recent visit to Canada.
SD: I was invited to Canada by the Canadian Convenience Stores Association to talk about my university research into plain packaging. My aunt married a Canadian and lives in Kelowna (with my grandmother) and so this was an opportunity to visit relatives. I sent four days in Kelowna and then gave a series of lunch time talks relating to my research (joint with Ashton de Silva) into plain packaging. The downside of being the lunchtime entertainment was that I got to eat a cold meal each lunch time. I did not receive any payment for the talks I gave. I was happy enough to visit my aunt and grandmother whom I had not seen in over 20 years and 10 years respectively.
J: Was the tobacco industry involved in the visit in any way?
SD: Not to my knowledge.
J: The Atlantic Institute for Market Studies said that their event was held in partnership with Crestview Strategy, a lobbying firm that represents one of Canada’s biggest tobacco companies, so I would like to have some clarity around the involvement of the tobacco industry.
SD: I can’t help you there – I hadn’t heard of the Atlantic Institute for Market Studies before I spoke there, nor have I heard of them since. I also spoke at the Economic Club of Canada meeting in Toronto and Convenience Store meetings in Montreal, Winnipeg, and Vancouver. I have no knowledge as to how the meetings were organised. Beyond ensuring that each venue had a powerpoint projector I had no interest in the organisation of the meetings.
J: I’m also wondering if you can provide me with some detail about your relationship with the tobacco industry.
SD: For several years my relationship with the industry was as a consumer of their product. I smoked as a student but stopped cold-turkey many years ago. Both my parents smoked until my mother contracted a very aggressive smokers’ lung cancer.
I have had contact with people in Canada (obviously – at the talks I gave), the UK, and parts of Europe opposed to plain packaging. These people work in media, think tanks, and consumer rights organisations.
J: Do you receive direct funds from the tobacco industry or research funding?
SD: No. As a condition of my employment I cannot receive any funding from the tobacco industry. This is a University wide policy and I think all Australian universities have this policy as part of the WHO convention on tobacco control.
J: I understand that the Institute of Public Affairs has and/or does receive tobacco industry funding.
SD: I have no knowledge of the IPA funding beyond what is in the annual reports – my understanding is that the bulk of funding comes from personal donations.
J: Can you confirm whether the institute currently receives any funding?
SD: I don’t know if the IPA currently receives funding from the tobacco industry – I have never been told that it does.
J: In addition, I’ve spoken to some experts and read several reports that are critical of the results of your research, saying the findings are flawed and misleading.
SD: Could you please send through those reports? I have only seen the Victorian Cancer Council press release and an op-ed in the Canadian Medical Association Journal. As soon as the end of academic year rush is over I will be contacting the CMAJ ombudsman regarding the false and misleading statements in that op-ed.
Perhaps these are the same experts who claimed I did research into contraband – if so they might be referring to someone else’s work?
The fact of the matter is this – there is an academic dispute as to the impact of the policy. Some academics have written articles saying the policy has worked and I have written articles (joint with an RMIT colleague) that the policy has not worked. This is what academics do and how academia operates.
Now you have probably read the Victorian Cancer Council press release responding to the Social Science Research Network paper. There are many things wrong with it – but indulge me by doing just one quick check. In that press release they say:
Contrary to the statement by Davidson and de Silva (page 15), this analysis [the Post-Implementation Review] did control for several major tax increases that occurred in Australia prior to the implementation of the tobacco plain packaging policy and during one and two years afterwards …
Go and have a look at our page 15. See what it is.
If you’re game do a search though our entire paper for the PIR or Chipty analysis. See what comes up.
Then form your own opinion as to the rest of the press release.
J: The Australian government’s figures do show that smoking has decreased since the plain packaging law was introduced, for instance. It would be great if you could provide a response to those criticisms so I can include it in the piece.
SD: I don’t know which figures precisely you are referring to. But if it is the Chipty study then I’m afraid her analysis shows no such thing. She found that plain packaging had reduced smoking prevalence by 0.55% over three years over and above the pre-existing downward trend. I have found out that the sample error in the underlying data is 0.6%. That is bigger than the effect that she found. In other words the 0.55% difference is not statistically significantly different from zero.
In her modelling she also uses a rather unusual base case: an unmarried, Australian born, 14 – 17 year old, male, with a tertiary qualification, employed full time, but with an income less than $6000, and living in Victoria. I would like to test whether the choice of such an unusual base case distorts Chipty’s results. The government to-date is refusing to release the data for that study.
If you are referring to the Wakefield studies then please note their disclaimer after the Social Science Research Network paper got published:
The NTPPS was quite explicitly not designed to assess quitting success or change in smoking prevalence but rather focussed on the immediate impact of the legislation on perceptions of the pack, effects of health warnings and understanding of product harmfulness.
So not only am I arguing that the results in the Wakefield studies didn’t show a decline in smoking prevalence, the authors (via the Victorian Cancer Council) are now claiming the study wasn’t intended or designed to show that result.
If you are referring to the Australian Bureau of Statistics Household Expenditure data then that is the very data series that I (and Ashton de Silva) analysed in theAgenda paper. If experts think that analysis is wrong, they should write a rejoinder and publish it in the journal, not make complaints to newspaper journalists.
J: I’m working on a tight deadline at this point, so any help you can give me today would be great. I may be able to add some changes in later to the story once it has already been published, but I’m not certain that this will be possible.
SD: The point is this: There is an academic dispute about the effectiveness of a government policy. This is not unusual or even rare. Writing articles, giving presentations, talking about my research to interested parties, including the media, is actually my job. I teach a subject called Public Sector Economics – so critiquing government is what I research and what I teach.
Over the past few years I have published papers that critique government claims relating to the Fuelwatch Scheme (a government mandated price fixing scheme), the claims relating to stimulus spending efficacy, the claims made about the neutrality of the proposed mining tax, government forecasting of corporate tax revenue, claims about subsidies paid to the mining industry, government claims about base erosion and profit shifting (forthcoming), claims about the effectiveness of the new innovation policy. The list goes on … but I hope the pattern is clear.
I critique government policy – plain packaging is just another government policy.
I have learned, unfortunately, that government cannot be trusted when making claims as to the effectiveness of its own policies. You need to check the data yourself. Often when I do check the data I find the government (and its agencies etc.) are over-egging the pudding. (I hope that expression makes sense in Canada).
What is happening here is that the anti-tobacco lobby is shooting the messenger. I am not in the pay of big tobacco; I am an academic economist who tests and critiques government policy. I have a track record of publications doing just that. The Australian government’s claims relating to plain packaging – like so many other claims I have investigated – do not stack up. That is what I have found, that is what I have reported.
As it turns out I had a long discussion with Garfield Mahood in Toronto during the Q&A session of my talk at the Economic Club and also again after the session. He put to me the same questions with the same underlying premise that somehow I am corrupt, or on the take, or that my motives are base, or that I am inadvertently benefiting the tobacco industry, etc. etc. that you have put to me. Mind you, he was very quick to back away from stating that premise when I asked him if that is what he was implying. In the end he seemed happy to accept that I am an academic doing research and publishing results, and my motive to come to Canada was to visit my relatives.
I understand that the public health lobby don’t like the results of my research – but, in that, they are no different from any of the Australian government agencies (including politicians) that haven’t liked me critiquing their particular policies. I don’t resile from that – it would be dishonest and irresponsible to cover up the fact that a government policy had failed to meet its stated objectives.
This represents the moral bankruptcy of the political left – rather than engage in the issues – the boring stuff like sample error and regression diagnostics (that show some regressions demonstrating that the plain packaging policy “worked” have 100% miss rates on a hit-miss table) and the like, they want to go the smear campaign.