Tobacco Documents Online

RESTRICTED • 18th May ~977. To P.L. Short, Esq. CHANGES IN CIGARETTE DELIVERY LEVELS (SPECIFICALLY IN CARBON MONOXIDE) TO MEET THE THREAT OF LEAGUE TABLE PUBLICATIONS IN HOLLAND You asked me to respond to the Minute of the M.P.D.C. Meeting of 28th January 1977 which reads as follows:-'- "~. Sheehy asked for a study to be prepared on the.recent league table developments in Holland • with particular reference to the problem facing management under the threat of impending league tabl~ publications and the conflicts which arise where a successful brand might be threatened unless its formulation is changed." With the help of some of those who participated in these developments (Bob Lowe, Pat Crispe, Fred Haslam and members of BAT Nederland) I have set out a timetable of the important events and followed it with comments which are intended to be in the nature of "lessons to be learned". Also at Mr. Sheehy's suggestion, I have attached in an appendix a very brief account of Niemeyer's successful "cornel'ink"of the low delivery/mildest se~,nent of the Dutch market. Its relevance is that i% demonstrates the advantages of seizing the initiative and han~in~ on to it tenaciously even if the competitive marketing advantage is at the cost of breakinE industry agreements. 18 December ~975 Minutes of the P.D.G. record that arisin~ from Pro~ect CHOPIN (carbon monoxide-~eneral) and an M.P.D.C. request, a new pro~ect, BRAH~S, would be initiated. The objective was to develop low carbon monoxide versions of several European brands, includin~ GLADSTONE MILD in Holland. During the following months development work continued on CIIOPIN (with particular re].eval, ce to the needs ~f )~Y LONG in Switzerland) which was to have technical .value for the BRAH~IS Pro~ect. In Holland at %his time it was not known when the next league table would be published although ±he Aut~nn of 1976. was regarded as a possible time. However, it was considered unl~kely that carbon monoxide values would be included in that particular occasion although it was certainly recognised that they would appear sometime in the not too distant future. A prime objective was that when carbon monoxide levels were published GLADSTONE ~IILD should not be sho~a~ to have hi~her levels than PETER STUYVESANT. ./. BAT Industries document for Province of British Columbia 13 November 2000 BAT INDUSTRIES 00290417

~ March 1976 The P.D.G. decided that, because of the complexities of the various published league tables, it would be necessary to draw up a detailed action plan for each Company. However, for GLADSTONE this could not be perfect because at that time the carbon monoxide levels of Dutch brands were not known accurately. Subsequently, all theimportant brands on the Dutch market were sent to P.D.L. Southampton and also to Switzerland for analysis. MeanWhile Benelux ordered equipment to enable the Company to carry out- analysis of CO in Brussels. 8 April 1976 Visits were mahe to the Companies and a provisional target for GLADSTONE MILD agreed as "a reduction of carbon monoxide by 25% with no changes in DPM, nicotine or taste". End April 1976 The first "one-off" results of the analyses were received. These showed that GLADSTONE FIILD and STUYVESANT had identical levels of carbon monoxide: ~.6% (16.6 mg.). Immediately ~|essrs. Fordyce, Lowe, Crispe and Baretta prepared an action plan ~n which the delivery tar~e±s w~re specified as shown below: DPM Nic/Ci~ co % GLADSTONE M. GLADSTONE M. P. STUYVESANT Current Proposed Current 18-~9 18-~9 ~8 ~.3 1.3 ~.~ %.6 (16.6 rag) 3-q (IZ-6 rag) ~.6 (16.6 rag) Experiments to reduce the carbon monoxide level were started. IO August 1976 The results of experiments with varyinK paper porosities indicated that reduction of carbon monoxide could not be achieved without reduction of DPM, nicotine and taste to unacceptable levels. It was necessary, therefore,~to develop new blends and samples were laid do~. Clearly, it had been hoped to reach the target without creatinE new problems in leaf stock holdings which blend chances to a hiEh volume brand would entail. ~e~tember 1976 The official TN0 league ~able was published in the press. This did not include carbon monoxide but it became k,~o~n that carbon monoxide would p~obably be included in the December publication of the Consumer Guide. ./. BAT Industries document for Province of British Columbia 13 November 2000 BAT INDUSTRIES 00290418

25 November 1976 The P.D.G. reported that among four experimental product.s %we' appeared to meet the criteri'a for carbon monoxide. These were:- % Nicotine Nic/Cig DPM CO BLEND in Tobacco 2~7 2.~7 I.~I 16.~ 12.0 250 2.~7 ~.50 18.7 12.9 TARGET 1.30 18.01.19.0 12 o 6 Blend 250 in particular was regarded as very acceptable. Both DPM and carbon monoxide were within the target ranges and nicotine was even a little higher with a very satisfactory Nic/DPM ratio. Subjectively it smoked with good flavour and satisfaction. In spite of having achieved the required delivery levels and the a~reeable subjective smoking responses, the Marketing Director, Holland, decided %o have a consumer test carried out on this blend versus two others and the current blendl An important consideration was that as GLADSTONE ~IILD is the Company's main volume brand such a ma3or product chan~e, in the absence of consumer research evidence, would invoke a high de~ree of risk. This was also the opinion of all ~io had been concerned with the development. Consumer test results becatne available on 15th ~|arch and Sustified the decision entirely. There was a significant preference for the current blend and re3ection of all new blends by consumers. At the time that the decision was taken to consumer research (mid November) there was no certain knowledge that brand delivery /igt~es, including carbon monoxide, were about to be published by the Consumer Guide. This occurred on ~Oth November and was followed by R T.V. Drogra,,me the same evenin.~ in which it was shown that GLADSTONE NILD and BELINDA KS had the hi,<l,est carbon monoxide :levels .and mnade references to the "poisonous ~ases in tobacco smoke". It is probable that the brands for analysis had been collected in May. It was also unfortunate that the pub- lished figure& of carbon monoxide for GLADST.ONE~|ILD were hi~her (19.0 rag.) than those for STUY~rESANT (16.O rag.) and higher than the levels reported in the P.D.L. analysis carried out at about the same time (16.9 - 17.1 rag.). The effect of the pub].ication of carbon monoxide on the sales of GLADSTONE MILD and BELINDA durin~ December and the first part of January appeared to be nil. During February and March there was a small but significant weakenin~ of the shares of the two brands attributable to the publication of carbon monoxide levels.. Fortunately %he position appeared to have stabilised by April/May. ./. BAT Industries document for Province of British Columbia 13 November 2000 BAT INDUSTRIES 00290419

Finally, there m~st be a continuin~ anxiety about the date of the next league table publication and the need to have a lowered carbon monoxide version of the brands on the market in time to be picked up for analysis. It is con- sidered by BAT (Nederland) that the Consumer Association is most unlikely to carry out another analysis for at least a year and a half. Furthermore, for future official league tables the brands will be picked up from the manufacturers and not from the market. Thus the Dutch Company has at the very least six months to complete the project to be • sure of meetin~ the nex£ publication date ~ith a lowered carbon monoxide product. -. Comments 1. I~ has tO be. said, being wise after the event, that althqu~h the right actions were taken, they were taken too late. As yet it is not possible to be certain how much too late because the project is not complete. It is assumed that the brands were collected for analyis in May 1976 and as the project had not been completed by June 1977 it is at least 13- 14 months too late. 2. Certainly, in theory, the project could have been initiated earlier because the Centre and the companies have been aware, in a general sense, since 196~" that carbon monoxide measurements would be included evcn- tually in publications. Nevertheless, on the best information available at the time it was the Dutch Company's ~udEement that a carbon monoxide league table was not imminent. Another factor which probably acted as a disincentive to initiatin~ development by the Dutch Company itself was the lack of expert knowledge of the techniques for reducing carbon monoxide levels within the Company. 3. ~n such situations it is temptin~ to ask whether the sta~es of development after initiation could have been speeded up. There is no evidence that the project could have been carried through with greater speed at any stage. However, it is a fair point that ti,,e could have been saved if the Company had previously meas,,red and maintained up-to-date records of all the.relevant constituents of the major brands on the market. Conclusions 1. For all companies estimates should be made at least annually of i) the time at which each new elcment will come - under attnck (i.e. will be measured brand b~ brand and publicised) within own and/or ~djacent marRets ii) ihe time required by the Company to take appropriate action, inc].udin~ consumer testing of candidate produc£ formulation. * Surgeon General's Report ./. BAT Industries document for Province of British Columbia 13 November 2000 BAT INDUSTRIES 00290420

- "5 2. The estimates should be made for 'each company by the appropriate Territorial Team in consultation with R&D (Smoking and Health) after considering all information available and taking into account the views of each Company° Responsibility for communication with Companies to remain with the Territorial Team. 3- For these Companies/Harkets which are assessed as beinK at hiKh risk (criteria is to be defined as for e.g. likely to come under attack within three years) appropriate action will be as follows: i) The Company will be informed and in particular those individuals responsible for each function concerned with product development i.e. Marketing, • Lea£/Chemist, Production. ii) All main brands and all brands likely to have a total or partial health role will be analysed and measurements made of those elements identified as likely to be featured in publications. This will be carried out regularly (to be defined). ~i) As soon as the basis for quantification is available plans will be prepared for the development of "reserve" product formulations. The plans will be put into effect according to a time schedule with the object of producing produc~s which meet both delivery requirements and a~reed levels of consumer acceptability. "Reserve" product formulation plans should be included in the SmokinE and Health Key Area in the annual Company Plans. ~. Companies (specifically, the product development functions within them) will be info~,ed on a regular basis of all ~echnical developments in the methodology of reducing, adjusting and balancing the levels of the constituents. This information will be communicated both to the Associated Companies and to the Territorial Teams by the Product Development Adviser, Millbald~. 5. Information on the timinK, frequency and main content of league table and related publications, on the main import of imminent restrictive legislation and on the significant "health" product trends within all markets should be briefly su,u,arized half-yearly in a Bulletin produced by PR and R&D. '6. Companies should be alert to the possibilities of exploitinE future publications of new elements in league tables and turnin~ them to ,harkening advantage. They should not allow themselves %o be hampered by industry a~recments which caus~ them %0 forEo opportunities to be innovative and trulycompetitive in return for the insubstantial protection against outside threats. / . (G.J. SIDDALL) OO BAT Industries document for Province of British Columbia 13 November 2000 BAT INDUSTRIES 00290421

APPENDIX DEVELOPHENT OF TIIE LOW DELIVERY SEC~IENT BY NIEMEYER • In a sense BAT (NederJ.and), in company with near].y all the other Tobacco ~|anufacturers in Holland, has been reacting defensively to the attacks and threats of attack on the Tobacco Industry by Smoking and Health pressures - and specifically by tile publication of League Tables. With a few remarkable exceptions this, indeed, has been the posture adopted by most Companies throughout the world. One partihular exception to the rule is the Dutch Niemeyer Company and the ini%.ia%ive they seized and maintained by positioning their brands (ROXY DUAL FILTER, KELLY HALVARET FILTER, EVEREST FILTER) to be in %he best position in League Tables. Whether ROXY' s favourable position in the firs% League Table, published in 196~. by the Consumer Suide, was by accident or design is not important now, but however it came about, it seemed to point the direction for Niemeyer's future product strategy. In 1971 Niemeyer launched KELLY HALVARET and ROXY DUAL FILTER and in 1972 advertised ROXY DF as low in Tar and Nicotine in viola%ion of the ~ianufacturers a~reement concluded in ~965. Its sales inm~ediately gained market share. Furthermore, they printed the words "Low Nicotine Content" on the labels. In 1973 Niemeyer exploited their advantage and enraKed their competitors further by adopting a "one dot'~ (indicating low delivery levels) system on the ].abels. When in 1976 a ~overrmlent Working Party rcported adversely on the dot system as beJ,,.~ mis- leading Niemeyer started court proceedinKs and obtained a ~udKe,nent in their favour which prevented the publication of the report. Maintaink~their initiative Niemeyer ensured that when the first official League Tables were published by the Ministry of Health in 1975 ROX~ DF was listed near the top in low delivery ranking, in fact second to AUSLESE and on a par "with FLAII[. They also launched an advertisin~ ca.,pai~n to ensure that the smokin~ public ~.as aware of ROXY' s favourab].e position. By Docember 197~ thc brand had reached tl~ Market Shal-e. " Subsequently, other manufacturers launnhed iow delivery brands and in the League Table publication in September 1976 ROXY DF was listed in thirteenth position. The response of Niemeyer to this situation was to launch an advertisin~ campaign ~d~ich was blatantly misleading and was also in contravent~.on of the trade mark laws. An account o~ this and the decision of the Advertising Code Commission against Niemeyer was published in "Marketin~ News" in February 1977. The market share of ROXY DF l-cached ~.~ in ])ece,,,|,er 1976. BAT Industries document for Province of British Columbia 13 November 2000 BAT INDUSTRIES 00290422

To date, however, Niemeyer remains the Company in IIolland which is regarded by smokers as pre-eminent in producing low delivery brands~ This aspect of the Company image has been systematically built up over many years and enhanced by Niemeyer's virtual monopoly of health claim advertising. Whatever the League Table position of their brands is now, or may be in the future, and no matter how the image was developed; whether by being aggressive, misleadin~, immoral, unethical or do~n right dishonest, their re~utation is a fact and like most well established images and reputations it will be very resistant to change. Over and above this, the extra volume of cigarettes already sold because of their refusal to be hampered by threatened restrictions apd gentlemens a~reements is tile most ~lling fact of all. 0 BAT Industries document for Province of British Columbia 13 November 2000 BAT INDUSTRIES 00290423