North American Philips Corporation

Text (OCR of the scan at left; may contain errors)

IN THE MATTER OF NORTH AMERICAN PHILIPS CORPORATION FINAL ORDER, OPINION , ETC. , IN REGARD TO ALLEGED VIOLATION OF SEC. 5 OF THE FEDERAL TRADE COMMISSION ACT Docket 9209. Gomplaint, Aug. 1987-Hnat Order, Oct. 24, 1988 This Final Order prohibits, among other things, the North Americar(Phijips Corp. Norelco s parent company. from misrepresenting the performance of the Clean Water Machine or any other product that treats water and also from misrepresenting any test or study of its products. The order requires respondent to have substantiation for any performance claims it makes for any electric-powered consumer appliance, including hair dryers, makeup mirrors, coffee makers, and razors.

Appearances For the Commission: Joel C. Winston. For the respondent: Forrest Hainline, III, Swidler Berlin Washington, D.

COMPLAINT The Federal Trade Commission, having reason to believe that North American Philips Corporation, a corporation ("respondent"), has violated the provisions of the Federal Trade Commission Act, and it appearing to the Commission that a proceeding by it in respect thereof would be in the public interest, alleges: PARAGRAPH 1. North American Philips Corporation is a Delaware corporation, with its offices and principal place of business located at 100 East 42nd Street, New York, New York. PAR. 2. Respondent has advertised, offered for sale, sold and distributed the Norelco Clean Water Machine ("Clean Water Machine ) and replaceable filter cartridges ("filters ) for the Clean Water Machine. The Clean Water Machine is a self-contained, tabletop product designed to remove contaminants from tap water by forcing the water through an activated charcoal filter. PAR. 3. The acts or practices of respondent alleged in this complaint have been in or affecting commerce.

(g) ( .. .. . . . . . Complaint 111 F.

PAR. 4. Respondent has disseminated or caused to be disseminated advertisements and promotional materials for the Machine. Typical of respondent' s advertisements and promotional materials for the Machine, but not necessarily all-inclusive thereof, are the advertisements and promotional materials attached hereto as Exhibits A, B, and C. The aforesaid advertisements and promotional materials contain the following statements and depictions: (a) "Does your tap water contain synthetic detergents, chlorine, trihalomethanes and organic wastes7 .. The Norelco Clean Water Machine can remove up to 90% of these impurities from your family s tap water. (Exhibit A) (b) "CONTINUOUS CLEAN-Only the unique Norelco Continuous Clean Setting repeatedly recycles the tap water through the filter to help give you dcaner waterwater that tastes "bottled water" clean. And independent laboratory tests prove it." (Exhibit A) (c) " Do you worry about the taste. the smell. or the clarity of your tap water? The Norelco Clean Water Machine helps remove chlorine, sediment, sulfur, detergent odors, organic chemicals, and other pollutants you may not even be aware of that arc in your tap water. The exclusive fitration system keeps water in contact with the activated charcoal longer for cleaner, clearer water. " (Exhibit B) (d) "Helps Make Tap Water 'Bottled Water' Clean. " (Exhibit C) (e) "Helps remove chlorine, sediment, sulfur, detergent, odors, organic chemicals and other pollutants, " (Exhibit C) (f) "Multilayered activated charcoal filter helps remove chlorine, sediment, sulfur detergent, odors, organic chemicals and other pollutants. Special design keeps tap water in contact with the charcoal filtering surface longer for better, more efficient cleaning. " (Exhibit C) Tap Water Enters (shown passing through "Prefilter Activated Charcoal " and "Postfiter Cleaner Water Exits. " (Exhibit C) (h) "ONE STEP CLEAN Tap water passes through the system once, trapping impurities to create crystal-clear water. Tap Water" shown passing through "Exclusive Charcoal Filter System " to become "Cleaner Water ). (Exhibit C) (i) "Norelco Clean Water Machine" (Exhibits A, B , C) PAl.. 5. Through the use of the statements and depictions referred to in paragraph four above and others in advertisements and promotional materials not specifically set forth herein, respondent has represented, directly or by implication, that: (a) The Clean Water Machine wil effectively help remove organic chemicals from the tap water treated by it, under typical water conditions;

(b) The Clean Water Machine will make the tap water treated by clean or cleaner, under typical water conditions; and NORTH AMERICAN PHIUPS CORPORATION 141 139 Complaint (c) Independent laboratory tests prove that the Clean Water Machine wil make the tap water treated by it clean or cleaner, under typical water conditions.

PAR. 6. In truth and in fact, under typical water conditions (a) The Clean Water Machine will not effectively help remove organic chemicals from the tap water treated by it; (b) The Clean Water Machine wil not make the tap water treated by it clean or cleaner; and .

(c) Independent laboratory tests do not prove that the Clean Water Machine wil make the tap water treated by it clean or cleaner; because, while the Clean Water Machine may help remove organic chemicals, pollutants, and impurities, if any, from tap water, many original and replacement Clean Water Machine filters were assembled by means of a glue which added a substantial amount of methylene chloride, an organic chemical that is potentially hazardous to consumers' health, to the tap water. Therefore, the representations as set forth in paragraph five were and are false and misleading. P AI!. 7. Through the use of the statements and representations set forth in paragraphs four and five and others not specifically set forth herein, respondent has represented, directly or by implication, that, at the time it made the representations, respondent possessed and relied upon a reasonable basis for such representations. PAR. 8. In truth and in fact, at the time respondent made said representations, respondent did not possess and rely upon a reasonable basis for such representations. Therefore, respondent's representation as set forth in paragraph seven was and is false and misleading. PAR. 9. In the advertising and sale of the Clean Water Machine respondent failed to disclose to consumers with typical water conditions that many Clean Water Machines and original and replacement fiters add methylene chloride, an organic chemical that is potentially hazardous to consumers' health, to the tap water processed by the Clean Water Machine. This fact would be material to consumers in deciding whether to purchase the Clean Water Machine and filters. The failure to disclose this fact, in light of the representations made as alleged in paragraph five, was and is a deceptive practice.

PAR. 10. The acts and practices of respondent as alleged in this complaint constitute unfair or deceptive acts or practices in or affecting commerce in violation of Section 5(a) of the Federal Trade Commission Act.

.,.. ...... :;: Complaint 111 F.

EXHIBIT A WILTON RESIDENTS:

DOES YCUR TAP WATER CONTAIN SYNTHETIC DETERGENTS, CHLORINE TRIHAOMETHAES, AND ORGANIC WASTES? Queson It-Residents Ask It llpers THE NORELCO CleaW can remove up Ia 90 of impuries th frm your family' s lap wat. -ClTluaU5PE AVEEQJU-SIICREAL CHLORINE TRIHAL0'ETH4NES 20;1131) SYNTHETIC DETERGENTS ORG.-NIC WASTES -1i11)-cN!1 !Lime :s1t .. YI",lTmdl OM!TCU- .iI" !h5'IlOOwT-Ia!hl;dent #Ore/co mMNUOCt-OnIh\lQ1H! CUInC\Sm ea18 i NtWf Fi1O O8WD Ma\W00 :3: IDntTl=iJ1l .IICClMI1M 'tTKKAIIIEETD 11tt1a IIJIIOOed..IIIMleula Of rRum FOil Item CtmtlIU-_-.s.....-SH II If us. .'50r 5Mdi AVAILABLE AT CALDOR, MACY'S, SASCOA . AND OTHER FINE STORES.

. ...., . . ,..,... ..,...,.,. , , .. ,....: ... , ,,,,,.., . ....,. .: , ! 139 Complaint EXHIBIT B ""'olv u,",nc'M'd'.' .,oun.-" '"",.",e""",,,, "cm- ;J"E S A" 0' CONT'NUOUS Cl '\N The Norelco Clean Water Machine Helps make tap water taste and ;1' smet" Bottled Water" clean in the convenience of your own home for only pennies a gallon.

On. SI." CI.. Conlln", . c:.... Tac ,,,'.r 0."', Incn Un'Quocap"nuou"'''.."on nc., DC"u'. " '0 ""'oclnal. cry"a1 'y"om"o ,Ou' qc......'" 'h.,'.0""0. , '0 tol"O;".,., c'..n No ",.ra/l.rion "fair!K-!/"j compaa In SIZ -.4 Ot. c3paclly- ovyr1utyplaSliccardl" w.r" "a"';ty .rorag8firJ Wafer that HAS NOT BEEN fESfED OR rREAfED FOR HUMAN CONSUMPTION SHOULD NOT BE USED. G,...';tYnqc:,"'''I&3=) rho 0''''"":' , "'"0;"'' o'.'oo o.n "...' ".o,noGJ.H oo..,nQ C...," 01 GO'"'P7 '.. OC' OI.men' ry .,,,,, ,C" 10 , 1m'"Q ,.bl. 0' ., ..."- W".' q Qr.. 4t 0':" "m. ''''"o' a'O.. . "O '0 '''. 'h- 0.""' .o,omc.",mom ' c, ;..n N.,., "'.o 'n., ': ," o.. '0001 "0"9' "d ' 0 ,. '.'" 'to Indepengent tests prove it out cleans I"c,--J ""'"UI the leading fauceHype filter. 'NFI- CI..n W.... 'on'". ;IT.. ' 'h.' Th. "" "O'''"' mo,. :h'n 2 "me, Doyouworrabcutthlitut.. lhls,""U Qrth. 'he .ollv" "n..o "'o ""o clarilvol your tapwallr" ,,,,.m,, II '''.'''.C. .M ''' ""'.'CO c:u cn'cnno 0 J mmi "'.", ""C","" no'"' "'''C' ., 0' ., "c'" ""d . ,"'. '0 . 'u"'Jr " .'.,-.n. .ne h.'.O".O MM"n.nycun"", ""ul.n,. vo" ,.".. n",. :K"",....n". O',w." O."Cc,,Nme"'. '0.' dr" 'Cu. "'" N".r . "01,,,,,. t""",e" ,.".m '.'0' w,'", In coo,.,,, "n '" ,0'''''0. cn.'caaJ IO" m ""0"'. . . :O "n. "oro'".n, ' m..'o...nCU"'o'K""',"cn.,cc..... .em."n. 'c.. ''''8m. n..\)11" . , Complaint 111 F.T.C.

EXHIBIT C BP AES BOTTLED5 1 5 1 2 2 1658 1235 135 101 79.391647 WATER’2 1 6 0 0 0 1541 1325 145 40 -1 3 1 6 1 0 0 1541 1325 145 40 -1 4 1 6 1 1 0 1541 1325 145 40 -1 5 1 6 1 1 1 1541 1325 135 40 25.089493 CLEARS 1 6 1 1 2 1675 1326 11 38 0.000000 ,2 1 7 0 0 0 1441 1564 312 35 -1 3 1 7 1 0 0 1441 1564 312 35 -1 4 1 7 1 1 0 1441 1564 312 35 -1 5 1 7 1 1 1 1441 1564 62 26 20.926674 eThe!5 1 7 1 1 2 1500 1569 47 28 5.248093 tose5 1 7 1 1 3 1555 1566 19 15 87.903984 ef5 1 7 1 1 4 1581 1566 73 33 22.091492 Ectlec5 1 7 1 1 5 1662 1569 58 14 78.604736 waters 1 7 1 1 6 1728 1571 25 12 50.776390 “cr2 1 8 0 0 0 772 2044 361 23 -1 3 1 8 1 0 0 772 2044 361 23 -1 4 1 8 1 1 0 772 2044 361 23 -1 5 1 8 1 1 1 772 2044 361 23 95.000000 , , , . . . ruuJ..LUU.nu .L .L.LB..-'-' C) U'J-'U'JHrL.L-''J'" 139 Complaint EXHIBIT C t.:". "30 . 10";0; . JUORElCQ' MACHINEClea\iter WATER' CLEAN HELPS MA lA WAT 130ffid HO rT'NRIC Wrt a preciselv conliolled fl tte eledc pumping syem UP end Itrouh tie heo.c odaled chcrcal later No yor d.'M top can nave me tce of bcTted waer. The Clen Wct Mccnine hejps remQe imDJries from yor tap ..er, The Narek Cecn Water r-Chine gNe5 you fW seJ1 ONE STEP CLEAR CONlNUOUS CLEAN IC" LO LG lian "'''-,.r&;Ce. 1e..0( rone'D9rv\Oc:ec. or.rTon!oc.ec c:-daawO Tt,CT''T''=''ec''''c'ec :f, LS'f 'i' ,(f(J-c",;OM I' I i '- ;1' \l1J : , ,).! .../#/,;;:.::$...;\ .,.,) !;, . Complaint 111 F.

EXHIBIT C NDRELCD MACHINEterCleaW HELPS MAE lA WATER 'BOmED WATR' CLEAR DUAL -SElNG D"l PQSTFILTR CLEAER Wife ACTIED EXITS CHAi(COA EXCLUSIVE ADVANCED ' 4\ TECHNOLOGY 'i'" ::, FILTRTION SYSTEM PREFILTR . Mdtiloyred acted ChOfcool Mer helps remove chlorine.

seiment. sulfur, detergent. odors organic chemicals and other pollutants .. Special design keeps tap ..er in canted wi the charcool fih"ering surfcce longer for better. more effcient cleaning W' Wife ENTRS . Long-lasng. ea-fa-replace fih-er WATER THAT HAS NOT BEEN TESTED OR TREATED FOR HUMAN CONSUMPTION TAKE THE PLACE SHOULD NOT BE USED THIS APPLIANCE DOES NOT OF ANY \M100 STERILIZATION OR Distillation P:.OCES.'ES YOU WOULD NORMALLY FOLLOW ,(: : ( ,. , :,.: . , .. , \ Complaint EXHIBIT C IVORElCO Clea ter HElps MA MACHINE W' BOT7 \A ' II7ER' ClfAN BETTER WATER FOR BETTER- TASTING FOODS AND DRINKS CLEARER ICE CUBES DRINKING CRYYv' ClfNWATR BffER Baked SPECIAL RECIPESGoODS ANa c-'- DELICIOUS FROZEN JUICES AND PcERfO !MP DRINKS COFFEE AND GREAT- HOT DISHESTANG OJ , . Iia:.

IN 100 '" :g UJ u 0.

II;

NORTH /Vre/co itThis is AMERICAN COPYRIGHT listedsymbol HIGH MAE. byon PHILIPS NORTH the RIDGE PRINTED AND PARK. AMERICAN Underwriters'producfs CORPORAION.MANUFACTURED BY: PHILIPSSERVICED STAFORD. Laboratories. CT IN nameplate Inc. CONSUMER MODEL USA 06904 means \N100 48-CORPORAION PRODUCTS 19B2 Ol DIVISIONS CAPACIl .

, ,, th USA above In ..rontycord.Serve NOTE:Philipsalso allow ma the so VOnot Served do obtainenclosNoeIco warranty.15 OM the tAmerican ' and ma slates 01ne.:rest damages. byNorh '1 right the by Prnted listed So toccwered legal Mode. D"ES, not slate you. Is 10 to Centersprepaid consequential aulhorizedspeclHc or Isyoslate apply Serve Iranslt oral pastage during01gives'rom not CONSEQUENTIA unit ocldenlolma Norek:o vary at OR Ihe occurring which WM100 oftheship writtenwarranty limitations This 01 any rlghhlimitation and warranty INCIDENW or 01 takeDamage lon other other Mochlne clu WARRATY SPECIALserviceSimplCenterNoCorporationhae exclusion Water YE Walerorto at J1IS AN 0: lackproduct NON-AFOR ogreesworranly '00 material voltageBY l.b220:,IU Clean ONE Clean inandTheabuse.the orwllh ASSUMED IS 0:0:0: FUll Noelco specified defectspurchase. PfRfOf1MW misuseprovide the Norelco atIMPQRlAT: new ME against notbut date eah tller) charge.occident, any f1ESPNSIBllIJY Ihe toREPAIRS NO t9B2 wmran"andtramwithoutfromattachment OR cordyearunit anyproduct CENTER 11 atthe SEIMCE ceptane resulting 0 af Catpamtlonre aHixlng COfpoOlion defecHve SEIMCE INEN the period subjecting anydamae Philp' a or \.10 tor corB. IS ModellN1O NOf1HCO Philips Ijmr paris replacecover of American ornot IO'J5 NorhMochlneworkmanshiprepairdosleo50nab!eorWARATYAUTHORIZED eNorihAmerlcan / Initial Decision 111 F.T.C.

INITIAL DECISION By JAMES P. TIMONY, ADMINISTRATIVE LAW JUDGE AucusT 29, 1988 PRELIMINARY STATEMENT On August 3, 1987, the Federal Trade Commission (“Commission’”’) issued an administrative complaint charging North American Philips Company (“NAPC” or “respondent”) with violations of Section 5 of the Federal Trade Commission Act, as amended (15 U.S.C. 45), in connection with claims made by NAPC in the advertising and sale of its Noreleo Clean Water Machine, a home water filtration device. On September 1, 1987, NAPC filed an answer denying that it had violated the Federal Trade Commission Act as charged. Preliminary matters were addressed on April 4, 1988. Evidentiary hearings for the presentation of complaint counsel’s case-in-chief began April 5, 1988, and ended on April 22, 1988. Respondent presented its case from April 25, 1988, to April 28, 1988, and on May 16, 1988. Complaint counsel presented rebuttal testimony on May 27, 1988.

Twenty-four witnesses testified and the transcript of hearings consists of some 2500 pages. More than 250 exhibits were admitted into evidence. [2] SUMMARY OF COMPLAINT ALLEGATIONS This proceeding concerns the Norelco Clean Water Machine and its replaceable filters. The Clean Water machine is a self-contained, table-top product designed to remove contaminants from tap water by forcing the water through an activated charcoal filter. The complaint alleges that respondent’s advertisements and promotional materials for the Clean Water Machine represented that the Machine can effectively help remove organic chemicals from the water it treated and can make the water “clean” or ‘cleaner.’ The advertisements further claimed that the ability of the Machine to clean the water was proven by independent laboratory tests. The complaint alleges that these representations were false and misleading because many original and replacement Clean Water Machine filters were assembled by means of a solvent which added to , n.L'- .L L.LL.UU. U uV.LH vnr111Vl'j iiii 139 Initial Decision the filtered water a substantial amount of methylene chloride, an organic chemical that is potentially hazardous to consumers' health. These filters, known as the first generation (" G 1" ) filters, were manufactured in 1982 , but continued to be sold by respondent into 1986.

The complaint also alleges that respondent's advertising falsely represented that it had a reasonable basis for its performance claims. Finally, the complaint charges that respondent' s failure to disclose to consumers the methylene chloride contamination problem: in the G filters, in light of the representations made, constitutes a deceptive practice. According to the complaint, this fact would have been material to consumers' decisions whether to purchase the Clean Water Machine and filters. (3) FINDINGS OF FACT 1 I. JURISDICTION 1. Respondent North American Philips Corporation ("NAPC") is a Delaware corporation, with its offices and principal place of business at 100 East 42nd Street, New York, New York. (Answer 1.) 2. Respondent, through its Norelco division, sells the Norelco Clean Water Machine and filters for the Clean Water Machine. The Clean Water Machine is designed to remove contaminants from tap water through an activated charcoal filter. (Answer 2; RX 157- 3. NAPC has nationally advertised on network television and in magazines promoting the sale of the Clean Water Machine and its filters. (Answer 4. The acts and practices of respondent alleged in the complaint have been in or affecting commerce. (Answer II. BUSINESS OF RESPONDENT A. North American Philips Corporation 5. NAPC is wholly-owned by N.V. Philips of The Netherlands. (Gains, Tr. 2191.) 6. NAPC manufactures and sells products in three major markets: consumer, industrial and defense. (CX 142-39. 1 The following- abbreviations are used;

Findings of fact F.

'fr. - Transcript of hearings, prer.eded by name of witness ex - Complaint counsel's exhibit RX - Respondent' s exhibit Initial Decision 111 F.T.C.

7. NAPC’s net sales in 1987 were almost $5 billion. (CX 142-40; CX 147-2.) B. Norelco Consumer Products Company 8. The Norelco Consumer Products Company (‘‘Noreleo”), one of NAPC’s 21 divisions and subsidiaries, is headquartered in Stamford, Connecticut. (Dinley, Tr. 2096.) Before January 1988, [4] Norelco was known as the Consumer Products Division (‘‘CPD’’). (RX 157-G; CX 147-8.) 9. Norelco manufactures and sells in the United States small consumer electrical appliances such as electric razors, drip coffee makers, irons, travel appliances, clean air machines, electric knives, hand mixers, can openers, hair dryers, curling irons, make-up mirrors, digital blood pressure monitors, digital scales and digital thermometers. (Dinley, Tr. 2097-98, 2128-29; Gaines, Tr. 2158-59.) 10. Norelco’s sales ranged between $221 million and $242 million in the 1982-1986 period, comprising approximately 5% of total NAPC sales. (CX 142-7; CX 146-1.) III. THE NORELCO CLEAN WATER MACHINE A. Background 11. The Noreleo Clean Water Machine was advertised and sold during 1982-1986, and is still available from retail sources. As of 1985, Norelco had sold 248,000 Clean Water Machines and 435,365 filters. (CX 142-9 to 142-12, 142-20.) 12. Norelco purchased rights to a water filtration device from the Dynek Company. (Campbell, Tr. 1997-1998; RX 49.) Norelco manufactured the device in its Philips Park factory beginning in June 1982. (Campbell, Tr. 1998; RX 157-M; CX 196-1.) 13. Norelco started marketing the Clean Water Machine at the July 1982 Chicago Housewares Show. (Campbell, Tr. 1995, 2001, 2023.) 14. Prior to placing the Machine on the market, Norelco contracted with United States Testing Co., a product test laboratory, to determine the ability of the Machine to remove certain chemicals under specified conditions. (Rider, Tr. 2456-58; RX 55.) U.S. Testing conducted such tests and reported the results to Norelco in June 1982. (RX 56.) These tests were not designed or able, however, to detect methylene chloride, the chemical leached by the device into the filtered water. (Rider, Tr. 2459.) NORTH AMERICAN PHILIPS CORPORATION 153 139 Initial Decision B. Design and Manufacture 15. The Clean Water Machine has three basic parts: the housing, thc fiter cartridge and the plastic water carafe. The housing is a white plastic cylinder. Its interior is divided into two compartments with a pump at the bottom. (RX 55- (5) 16. The blue plastic filter cartridge is designed to fit into one half of the housing cylinder. Tap water poured into the other half of the housing cylinder is pumped upward into the filter, exits the top of the filter, and flows into the carafc adjacent to the housing. (CX 1- 17. The filter itself is a three-stage device: Inside the filter, the top and bottom stages are microporous, polymer discs designed to remove larger particulate matter from the water. The middle stage is a 60gram bed of loose activated carbon (coconut charcoal). (RX 55-D; TX 56- , 56-K.) 18. Activated carbon is the primary filtering ingredient in the Machine. (Coyle, Tr. 1163; CX 2; CX 56- 13; CX 56-20; CX 90-34. 19. Activated carbon has a limited period of usefulness. When a carbon filter becomes saturated, or "loaded " it will release into the filtered water chemicals previously absorbed. (CX 56-17; CX 90-34. 20. The filters have a useful life of 3-4 months. (RX 55-B; CX 2- Norelco sold replaceable filter cartridges. (CX 32-3; CX 142-10 to 142- 11.) C. Distribution System 21. Norelco sold the Clean Water Machine nationwide through wholesale distributors and retailers. (Lenahan, Tr. 186; CX 142-11.) 22. The Clean Water Machine was available at hundreds of retail outlets, such as department and discount stores, catalogue showrooms, drug chains, and hardware co-op organizations. (Oinley, Tr. 2100.

23. Norelco Service, Inc. ("NSI") stores also sold the Machine. NSI operating 16 stores nationwide, was formerly a subsidiary and is now a division of Norelco. (Dinley, Tr. 2139; RX 157-0 to 157- 24. Consumers bought filters directly from Norelco after the products were no longer available from local retailers. (Haag, Tr. 225- 227; Dugan, Tr. 1098-99; Coyle, Tr. 1177. O. Sales 25. Norelco commenced sale of the Clean Water Machine in 1982 and sold a total of 248 401 Clean Water Machines, as follows: (6) Initial Decision 111 F.T.C.

Year CWMs Sold 1982 159,043 1983 28,388 1984 60,970 The average wholesale sales price per Machine was about $30. The suggested retail price was $50 - $55. (CX 142-8 to 142-10.) 26. Norelco’s gross sales receipts from the Clean Water Machine were $5,347,715, as follows:

Year Gross_ Receipts 1982 $4,465,680 1983 740,492 1984 68,430 1985 73,118 (CX 142-10.) 27. Clean Water Machines continue to be available for purchase by consumers. (CX 188, Stip. 49; Physical Ex. 1.) 28. Norelco manufactured three different versions of filters, which were sold with the Clean Water Machine or as replacements. The first generation, or “G1,” filters were made using methylene chloride solvent in three places on each filter; approximately 382,000 were made from June 1982, to mid-November 1982. The second generation, or “G2,” filters, were made using methylene chloride solvent in one place; approximately 108,600 were manufactured, beginning mid- November 1982, through September 1983. The third generation, or “G8,” filters were made with no methylene chloride; production began in March 1986 and as of June 1986, 50,000 had been purchased. (CX 38-1, 33-2.) 29. Only the G1 filters are at issue in this proceeding. Norelco sold 354,708 Gl filters. (CX 142-11, 142-12, 142-20.) The unit sales price to distributors was $2.99 and the suggested retail price was $4.95. Norelco’s gross sales receipts from G1 filters were $1,060,576. (CX 142-11.) [7] E. Consumer Use 30. Many of the consumers used the Machine to filter five to seven glasses of plain drinking water a day, in addition to other uses. (Bergins, Tr. 261-62; Roche, Tr. 1048-50; Dugan, Tr. 1097.) 31. The Machine had two different cycle settings, one-step clean and continuous clean. On the one-step clean cycle, tap water passed through the system once only. On the continuous clean cycle, tap l"LJrUl1 A1Vlr,.tULAN t'tllLlt':; LUKl UKA'lVN 139 Initial Decision water was recycled through the filter for two minutes, or more. According to the package directions, one-step clean makes the water crystal-clear" but the continuous clean cycle makes the water cleaner" or "extra clean. " (CX 1-3; CX 2- 32. Most consumers changed the filters every three months, or more often. (Haag, Tr. 225; Bergins, Tr. 263-64; Campbell, Tr. 2054. IV. ADVERTISING FOR THE CLEAN WATER MACHINE A. Advertising Dissemination 1. Television 33. From September 1982 to January 1984 , Norelco spent 244 600 on nationally broadcast network television commercials for the Clean Water Machine. (CX 16-2 to 16- 2. Magazines.

34. From December 1982 to August 1983, NAPC spent $129 300 on full-page advertisements for the Clean Water Machine in nationally-circulated health-and-fitness related magazines. (CX 16- , 16- CX 188 , Stips. 16-24.

3. Cooperative advertising 35. Norelco s executives believed that localities in which the media had raised suspicions about hazardous chemicals in the drinking water supply would be particularly good markets for the Clean Water Machine. (Dinley, Tr. 2133-34.

36. Norelco spent $25 000 with K-Mart on a cooperative advertising campaign targeting Tampa and St. Petersburg, Florida. (Lenahan, Tr. 165-66; Dinley, Tr. 2133-34; CX 19- 1.) (8) 4. Promotional materials 37. Norelco issued news releases entitled "What' s in Your Tap Water? New Norelco Clean Water Machine Makes Tap Water Taste as Good as Bottled Water " and "Norelco Clean Water Machine Fact Sheet." (CX 151 to 154. ) The news releases focused on cancercausing industrial and agricultural chemicals that have seeped into the water supply" and touted the Clean Water Machine as an effective means of removing known or "suspected" carcinogens from tap water. (CX 152-3; CX 154- , 154- 38. Norelco gave retailers displays advertising the Clean Water Machine. (CX 26; CX 27; CX 31; CX 188, Stips. 43-48.) These items Initial Decision 111 F.T.C.

were designed to attract the attention of browsing shoppers to the hazards of unfiltered tap water and the benefits of using the Clean Water Machine.

5. Packaging 39. The boxes in which the Clean Water Machine and the Clean Water Machine Replacement Filter were packaged included text outlining features of the device. (CX 1; CX 1838; CX 188, Stips. 1-2.) Each product box advertised that consumers could purchase additional replacement filters by mail directly from Noreleo. (CX 7; CX 8; CX 188, Stips. 8-11.) B. Advertising Representations 40. A factor in Norelco’s decision to enter the water filtration business was public concern about the existence of toxic chemicals in tap water. (Lenahan, Tr. 166-67; Campbell, Tr. 2024.) 41. Noreleo advertised the presence of potentially cancer-causing compounds such as “organic wastes,” “industrial chemicals,” and trihalomethanes 2 in drinking water. (Campbell, Tr. 2025; CX 152; CX 154.) 1. Television 42. The first nationally broadcast television commercial emphasized the Clean Water Machine’s ability to make tap water “taste bottledwater clean” and its effectiveness at removing chemicals from drinking water. (CX 10; CX 188, Stip. 14.) [9] 43. The commercial depicted the following: [Open on picturesque mountain stream. Sound effects: babbling stream.] ANNR: “Bottled spring water: clean . . . delicious . . . expensive.” [Freeze picture of mountain stream. Pull back to reveal that it is the label on a container of bottled water.] ANNR: “Now tap water can taste bottled-water clean for pennies a gallon... [Dissolve to Clean Water Machine sitting on a kitchen counter.] ANNR: “. . . with the new Norelco Clean Water Machine.” [Dissolve to animation of Clean Water Machine in operation. Darker water is poured in and swirled through the filter. Lighter water exits.] ANNR: “In minutes, the Norelco system circulates water over and over through its charcoal filter to help trap chlorine, sulphur, odors, rust and other pollutants to give you better water...”

”

? Trihalomethanes are a class of organic chemicals commonly found in water supplies. They are formed when chlorine added to disinfect the water interacts with certain organic matter. Several trihalomethanes are suspected carcinogens. (F. 70, 143.) . . .. . . . .L'lun.ln ft.L.rn.ll.J1UIj rnlLlr;: LU!trU!tAl1Ull .will 139 Initial Decision (Cut to live action of a woman pouring water for a young boy, which he then drinks. BOY: "Thanks, mom.

ANNR: for your family s food and drinks. (Cut to picture of Clean Water Machine on a kitchen counter. Dissolve to a shot of the Clean Water Machine resting on a rock in front of a bubbling mountain stream. ANNR: "The new Norelco Clean Water Machine. makes tap water taste bottledwater clean.

(CX 10.

44. Norelco executives were dissatisfied with the poor sales that the first commercial generated. N orelco abandoned the first commercial and initiated a second advertising campaign to alert potential customers that there might be toxic wastes and (10) harmful chemicals in their water and that the Clean Water Machine would help solve that problem. (Campbell, Tr. 2025; CX 143- 1.) 45. The second commercial depicted the following: (Open on the image of a small boy drinking water. ANNR: " Is your tap water as clean as it seems? It could contain impurities you don t want your family to drink. Now Norelco can remove up to 90% of organic wastes (The tenn "ORGANIC WASTES" is superimposed over the face of the boy drinking water.

ANNR: "Chlorine. . .

CHLORINE" is superimposed.

ANNR: "Synthetic detergents.

SYNTHETIC DETERGENTS" is superimposed.

ANNR: "and trihalomethancs.

TRIHALOMETHANES" is superimposed. Cut to a shot of the Clean Water Machine operating on a kitchen counter.

ANNR: "Introducing the Norelco Clean Water Machine. Unlike other filter systems it has a unique continuous clean feature to help make your tap water taste bottled water clean. or your money back. Help fight impurities in your tap water. With the Clean Water Machine. New from Norelco.

(CX 11.) 2. Magazines 46. Norelco ran two versions of the same advertisement in six health and fitness related magazines Health, Organic Gardening, Prevention, Runner s World, Fifty Plus and Weight Watchers. (CX 16- 5 to 16-9; CX 188, Stips. 16-24.

47. The magazine advertisement warned consumers that their drinking water might contain potentially hazardous chemicals not Initial Decision 111 F.T.C.

readily detectable by taste or smell. It stated that the Clean Water Machine could help remove these chemicals effectively and presented the device’s ‘“‘Continuous Clean” feature as a particularly desirable means of getting drinking water that is ‘‘ extra clean’ (emphasis in original). (CX 16-6 to 16-9.) [11] 48. In bold capital letters, the headline warned, “YOUR TAP WATER COULD CONTAIN SYNTHETIC DETERGENTS, CHLORINE, TRIHALOMETHANES AND ORGANIC WASTES.” A smaller bold subheading said, ‘The new Noreleo Clean Water Machine can remove up to 90% of these impurities from your family’s tap water.” (CX 16-6 to 16-9.) 49. The text of the ad read as follows:

Is your tap water as clean as it seems? It could contain impurities you may not want your family to drink.

Now, with the new Norelco Clean Water Machine, you can help filter out detergents, chlorine, trihalomethanes, organic wastes plus rust, sediment, sulfur and algae. To give your family water that looks and tastes “bottled-water’’ clean. And independent laboratory tests prove it.

The Norelco system is better than faucet-type filters. For one thing, there’s almost twice as much charcoal in the Norelco multi-layered filter than in the leading faucet filter systems. And there’s no installation required. For One-Step Clean, the electric pumping system draws water up, swirling it through all of the charcoal once.

For Continuous Clean, water is filtered through the charcoal over and over. Your tap water comes in contact with more charcoal longer, making it extra clean. Now your tap water can taste “‘bottled-water” clean for only 6¢ a gallon. The Norelco Clean Water Machine makes your tap water look and taste like expensive bottled water for only pennies a gallon. And it’s so convenient. Now you can have bottled water quality when you want, for less. And better water means better tasting coffee, juices, ice cubes, soups and baked goods for your family. Be completely satisfied or Norelco will refund your money. We’re so sure that you and your family will appreciate the Norelco quality and benefits of the Clean Water Machine that we’re offering you a money-back guarantee. If you’re not completely happy with the machine, just return it prepaid with your sales [12] slip to Noreleo within 15 days of purchase and we’ll send you a full refund. (CX 16-6 to 16-9.) (Emphasis in original.) 50. The advertisement also emphasized the desirability of the Clean Water Machine’s continuous clean feature by including a bar graph depicting the Machine’s purported effectiveness at filtering chemicals from tap water. At the continuous clean setting, the ad claimed that the ‘‘maximum removal” rate was slightly over 90% for chlorine, , NORTH AMERICAN PHILIPS CORPORATION 159 139 Initial Decision about 85% for trihalomethanes and organic wastes, and about 80% for synthetic detergents. (CX 16-6 to 16- 51. The ad also included a drawing of the Clean Water Machine in operation, with arrows pointing to the phrase "cleaner water exits" at the point of exit. (CX 16-6 to 16- 3. Cooperative advertising 52. An advertisement for the Clean Water Machine run in the Tampa Tribune stated Independcnt tests prove The Norelco Clean Water Machine can remove up to 90% of synthetic detergents chlorine, trihalomethanes and organic wastes from your tap water. (CX 199.

53. A radio advertisement for the Clean Water Machine broadcast on Florida stations warned listeners:

Just Jisten to these news items! " Florida town told: Don t drink the water. Pollution contaminates Bjscayne Aquifer." With aU the uncertainties about out tap water, its good to know K-Mart has the new Norelco Clean Water Machine. Laboratory tests prove the Norclco can help remove from tap water up to 90% of synthetic detergents, organic wastes, chlorine, and trihalomethanes. (CX 19- , CX 19- Promotional materials a. News releases 54. In its news release "What's in Your Tap Water'!" Norelco alerted readers to thc "harmful contaminants" present in drinking water and emphasized that many of these pollutants were either known or suspected carcinogcns in humans. (CX 152. (13) 55. The release focused on trihalomethanes such as chloroform which can occur as a by-product of the chlorination process. The Environmental Protection Agency ("EP A") believes chloroform to be carcinogenic. (CX 152- 56. EP A classifies chloroform as a probable human carcinogen (Group "B2"). EP A also classifies methylene chloride as a B2 probable human carcinogen. (Farland, Tr. 787. 57. The release also warned readers that " industrial chemicals such as benzene, vinyl chloride, trichloroethylene (TCE), trichloromethane, tetrachloroethylene and bromodichlorobenzcne have infiltrated many of the waterbeds, aquifers and underground wells that supply the nation s tap water. According to the release, many of these Initial Decision 111 F.T.C.

compounds are either known carcinogens in humans or “‘carcinogen[s] in mice and suspected carcinogen[s] in people.” (CX 152-2.) 58. Like these six industrial chemicals, methylene chloride is also an organic compound widely used in industry. (F. 80-81.) EPA classifies vinyl chloride and benzene as known human carcinogens (Group “A’’). Like methylene chloride, TCE is a B2 probable carcinogen in humans. (Farland, Tr. 787.) 59. ‘““What’s in Your Tap Water” also stated that the nation’s rivers and streams were polluted with ‘‘more than 700 organic chemicals, including 49 known or suspected carcinogenics.” (CX 152-3.) 60. Drinking water that meets government pollution standards may not be good enough, the news release cautioned: Even when these chemicals occur in amounts that fall within U.S. Environmental Protection Agency guidelines, they can affect the taste, appearance and/or odor of water coming out cf your tap. Add to that any dirt or sediment occurring within the water distribution system and you can get water the government considers acceptable but people don’t like to drink.

The release therefore suggested that consumers would be prudent to want tap water of higher quality than what ‘the government considers acceptable.” (CX 152-1.) 61. Another news release, ‘‘Norelco Clean Water Machine Fact Sheet,” claimed that the Clean Water Machine could remove 86% of organic wastes, 98% of chlorine, 81% of synthetic detergents, and 84% of trihalomethanes, It classified organic wastes and trihalomethanes as “carcinogen[s].” (CX 154-1.) [14] .

62. The ‘Fact Sheet” claimed that the Clean Water Machine had a “high removal rate” for pesticides, PCBs, vinyl chloride, and DDT, compounds it labelled “organic contaminants.” PCBs and DDT are considered to be B2 probable carcinogens in humans. (Farland, Tr. 789-90.) According to the news release, the device ‘‘greatly reduces harmful [chemical] contaminants.” (CX 154-2.) b. Point-of-purchase advertising 68. The point-of-purchase display Norelco disseminated to retailers warned consumers, ‘“‘You’re possibly drinking water full of impurities.” Materials sent to retailers claimed that the Clean Water Machine could remove up to 90% of chlorine, trihalomethanes, synthetic detergents and organic wastes. (CX 27; CX 31-2; CX 188, Stips. 45 to 48.) , NORTH AMERICAN PHILIPS CORPORATION 161 139 Initial Decision c. Ad mats 64. Ad mats Norelco disseminated to retailers carried the headline Your tap water could contain synthetic detergents, chlorine, trihalomethanes, and organic wastes. " (CX 22to CX 25; CX 188, Stips. 25- 32.

65. Norelco also disseminated ad mats for retailers to place in their local newspapers. One ad mat for the Clean Water Machine read Fairfield County: Does Your Tap Water Contain SyntJ:etic Detergents, Chlorine, Trihalomethanes, and Organic wastes? Newspapers Question It-Residents Ask It. " (CX 28; CX 188, Stips. 33-34. 66. Norelco disseminated identical ad mats addressed to "Queens and Nassau County" and "Wilton Residents. " (CX 165; CX 166; CX 188, Stips. 37-40.) Norelco also disseminated the same ad mat with a generic "Your County" headline. (CX 167; CX 188 , Stips. 41-42. 67. One of the ad mats Norelco disseminated to its corporate-owned Norelco Service Stores asked Does Your Tap Water Smell Bad Taste Bad or Look Bad? Norelco Service, Inc. wants to see if you can tell the difference our Clean Water Machine can make." The text was accompanied by a picture of a woman looking quizzically at an upraised glass of tap water. The ad mat claimed that the device "helps remove chlorine, sediment, sulfur, detergents, odors, organic chemicals and other pollutants from your tap water. " (CX 29; CX 188 Stips. 35-36.

5. Packaging 68. Norelco s packages for the Clean Water Machine states that it makes tap water ' bottled water' clean " (CX 1- 1 to 1-5; CX 183and "helps remove chlorine, sediments, sulfur (15) detergents, odors organic chemicals and other pollutants." (CX 1- , 1- V. RESPONDENT S ADVERTISING MADE THE REPRESENTATIONS ALLEGED IN PARAGRAPH FIVE OF THE COMPLAINT A. The Organic Chemical Removal Claim 69. Thc television commercial, the national magazine ads, and ad mats sent to retailers all included the express claim that the device helps remove up to 90%" of "organic wastes. " (CX 11; CX 16-6 to 16-9; CX 22 to CX 25.) Other promotional materials, including the NSI ad mats, the product box, the mail and one of theordernews releasesbrochure,claimed that the Machine will "help remove. . . organic chemicals. " (CX 1; CX 9- 7; CX 29; CX 153-1.) Initial Decision 111 F.T.C.

70. Trihalomethanes are a class of organic chemicals which includes many carcinogens found in drinking water. (Farland, Tr. 785-87; CX 50-2.) Noreleo’s advertising and promotional materials alleged the Clean Water Machine’s effectiveness at filtering out up to 90% of these compounds. (CX 11-2; CX 16-9; CX 28.) 71. Magazine advertisements, ad mats, and the packaging depicted the device’s “‘maximum removal rate” on the continuous clean setting as about 85% for these compounds. (CX 16-6 to 16-9.) 72. Norelco’s news releases emphasized that the Clean Water Machine had a “high removal rate” for “organic contaminants” such as pesticides, PCBs, vinyl chloride and DDT, and an 86% removal rate for ‘organic wastes” as a class and for phenol, an organic by-product of petroleum. (RX 56-G; CX 154.) 73. By advertising that the Clean Water Machine could remove high percentages of organic wastes, trihalomethanes, and organic chemicals, respondent represented that the Machine would effectively help remove organic chemicals from the water it treated. B. The “Clean or Cleaner” Water Claim 74. Norelco’s television commercials, promotional materials, and the package itself claimed that the device could ‘‘makes tap water ‘bottled water’ clean.”’ (CX 11-2; CX 25; CX 188.) A magazine advertisement claimed the Machine’s “Continuous Clean” feature makes water “extra clean.” (CX 16-6 to 16-9). (Emphasis in original.) [16] 75. Norelco asserted that water is not “clean” if it contains “impurities,” ‘contaminants’ or “pollutants” such as ‘organic wastes, chlorine, synthetic detergents and trihalomethanes.” (CX 11- 2; CX 27.) The advertisements stressed that the device could make tap water “clean” or “cleaner” by removing “up to 90%” of these chemical compounds. (CX 11-2; CX 16-9.) 76. The product package emphasized that water filtered by the Clean Water Machine would be ‘‘cleaner’” than water from the tap. The back of the Clean Water Machine box depicted tap water entering the device and water labelled ‘cleaner water’’ exiting into the carafe. (CX 1-8; CX 188-3.) C. The Independent Laboratory Tests Claim 77. The advertisements and promotional materials stressed that Norelco’s assertions about the Clean Water Machine’s effectiveness were supported by “independent laboratory tests.’ (CX 16-8, 16-9; CX 28.) l'\VI\ln J\lVl I'lvjU'" t't11Lli ;: tMttt'U.KA'llUN loti 139 Initial Decision D. The Reasonable Basis Claim 78. A television commercial, thc magazine advertisement, and the box itself all listed specific chemicals that the Clean Water Machine could purportedly remove ("organic wastes, chlorine, synthetic detergents, and trihalomethanes ) and, in most cases, what percentage could be removed. (CX 1-1; CX 11- , 11-2; CX 16- , 16- ) The magazine ad, ad mats and point of purchase shelf card included a bar graph that depicted to the precise percentage point the Clean Water Machine s effectiveness at removing certain chemicals. (CX 16-8; CX 23; CX 26. ) Ads referred to independent tests as supporting the claims. (F. 114.

79. Because of the objective and specific nature of these claims N orclco represented to thc consumer that it had a reasonable basis for making them.

VI. Tile METHYLENE CHLORIDE PROBLEM A. The Uses oj Methylene Chloride in the Clean Water Machine 80. Methylcne chloride (CH ), also known as dichloromethane, is a colorless, liquid volatile synthetic organic chemical with no known natural sources. (Zeise, Tr. 1290; CX 85- 81. The major sources of methylene chloride in the environment are from industrial uses. (CX 85- ) Methylene chloride is used in certain paint strippers, aerosol paints (17) adhesives and automotive products. Methylene chloride is also found in metal degreasers and various solvents and cleaners. (Cohn, Tr. 388.

82. Although most water supplies do not contain detectable methylene chloride, industrial activity can cause methylene chloride contamination of water, generally in the few parts per billion or less range. (F. 145-48.) Low levels of the chemical also may be found in certain brands of decaffeinated coffee, spices, and hops. (CX 124- 124-3; CX 156-82.

83. In manufacturing the Clean Water Machine, Norelco used methylene chloride as a solvent or glue to bond three pieccs of the fiter together. (Emmons, Tr. 1935- , 1967-68; CX 195- 84. In August of 1982, Norelco learned that laboratory tests might link methylene chloride to cancer. Quality Engineer David Harcourt spoke with Tom Brown of the Food and Drug Administration ("FDA" about the safety of mcthylene chloride. Brown told him that the Initial Decision 111 F.T.C.

preliminary results of a recently completed two-year study indicated that “methylene chloride may be considered a weak carcinogenic.” (CX 40.) B. Consumer Reports’ Discovery of Methylene Chloride Leaching 85. On November 10, 1982, Mr. Taub, Division Chief for Chemical Testing for Consumer Reports magazine, telephoned Norelco and informed it that the Clean Water Machine was emitting a high level of methylene chloride into the water it treated. (Lenahan, Tr. 168-69; CX 41-4.) Consumer Reports made the discovery while testing several water purifiers for an upcoming article. (CX 41-4.) 86. Taub told Norelco that the methylene chloride emission continued for the life of the filter, that the amount emitted varied from filter to filter, and that methylene chloride was probably coming from a glue or solvent used to weld the plastic parts of the filter together. Taub also informed Norelco that methylene chloride was a carcinogen. (Dinley, Tr. 2134; CX 41-2, 41-5.) In conversation the next day, Taub told Noreleo’s Vice President Patrick Dinley that the levels of methylene chloride the filters emitted were as high as 8 parts per million (ppm). ® [18] Taub said that levels of 1 ppm fell within the range of a “suspected carcinogen.” (CX 41-2.) C. Norelco’s Testing and Redesign of the Filter 87. Norelco officials William Lenahan and J. Richard Gonzalez contacted U.S. Testing, the firm that performed the original efficacy tests on the Clean Water Machine, to have the filters tested. (Lenahan, Tr. 171-72; CX 41-1, 41-6.) A carton containing twelve filters was delivered to U.S. Testing on November 11th. U.S. Testing selected three of them at random for testing. (Rider, Tr. 2462-64; CX 47.) 88. The filters tested were not production G1 filters. A Norelco engineer brought to U.S. Testing in November 1982 special ‘‘worst case samples’? designed to emit as much methylene chloride as possible. (Emmons, Tr. 1941.) 89. U.S. Testing found that three Norelco filters emitted 15, 10 and 3 Parts per million (“ppm”) are used to signify the concentration of a chemical in water or air. For contaminants in water, the concentration may also be given as the number of milligrams of a chemical per liter of water (‘‘mg/1"). Parts per million generally correspond to the equivalent number of milligrams per liter. (Ohanian, Tr. 1587.) Smaller amounts may be signified as micrograms per liter (“‘ug/1”), which corresponds to parts per billion (“‘ppb’’), or one-thousandth of parts per million. (Kern, Tr. 996; Coyle, Tr. 1148.) Therefore, 8 ppm could also be expressed as 8 mg/1, 8000 ppb, and 8000 ug/l. 139 Initial Decision 16 ppm (10 000- 000 ppb) of methylene chloride into the water treated by the Clean Water Machine. (Rider, Tr. 2465; CX 47. 90. Lenahan directed Norelco s engineering staff to experiment with alternative manufacturing methods to reduce the amount of methylene chloride emission. (Lenahan, Tr. 174; CX 42-1; CX 43- The Norelco engineers experimented with various alternative manufacturing techniques to minimize the amount of methylene chloride. (Emmons, Tr. 1948-51; Rider, Tr. 2466-67.) The Norelco engineers experimented to extend the drying time. (CX 42-1; CX 43- 1; CX 44- 1.) 91. Norelco sent three fiters of this type to U. S. Testing on November 12th. U. S. Testing measured the methylene chloride emission of these fiters to be 8.4 ppm, 8.3 ppm, and 5.3 ppm. (CX 43- 1; CX 48-1.) 92. Norelco engineers also experimented with four other sets of filters with varying saturation levels and drying times. Designated as Samples B through E, these sets were also sent to U.S. Testing for analysis on November 16th. (Emmons, Tr. 1949-51; CX 48-1; CX 49. 93. Soon after, Norelco selected Sample E as an acceptable alternative because it emitted between . 005 and .006 ppm (5-6 ppb) of methylene chloride, or about 1000 times less than the current production Gl filters. (Emmons, Tr. 1951-52; CX 44- 1; (19) CX 49- 1.) The process which Norelco used to manufacture these second-generation, or G2, filters, allowed for 24-hour drying time on the pre-filter disc and used a newly-designed retainer rather than methylene chloride to secure the post-filter disc. (Emmons, Tr. 1951-52; CX 33- 2; CX 43- ) The change to the new manufacturing technique cost Norelco a total of $225, which was used to retool the mold for the new retainer. (Emmons, Tr. 1980-82; CX 206.

94. The mold for the new retainer was modified by November 15th and production of the new G2 filters began shortly thereafter. (Emmons, Tr. 1951-52; CX 33- , 33-2; CX 206.) Norelco manufactured a total of 382 000 G 1 filters between the start of production in June 1982 and the retooling in mid-November 1982. About 108 600 G2 fiters were produced between the retooling in mid-November 1982 and 1986. (F. 28.

D. Norelco s Decision to Continue Selling Gl Filters 95. After Consumer Reports informed Norelco of the methylene chloride problem, and while the company experimented with alterna- Initial Dccision 111 F.

tive filter designs, Norelco continued to sell G 1 filters. (Campbell, Tr. 2005.) At that time, Norelco had approximately 214 000 unsold Gl filters in its warehouse. (CX 188- , Stip. 50. ) 4 96. At the time, Norelco did stop manufacturing new Gl filters in its factory. (Campbell, Tr. 2003.) When they first became aware of the problem, Norelco factory employees consulted a reference book on chemicals and learned that methylene chloride was a suspected carcinogen. The employees isolated all of the 4800 filters then in the factory and covered them with tape. (Emmons, Tr. 1971- , 1982-84; CX 206.

97. Within a few weeks, Norelco s Vice President Patrick Campbell decided to release for distribution the filters isolated in the factory. He also decided to continue sellng the G 1 filters in the company warehouse. (Campbell, Tr. 2049-52.

98. Thousands of G 1 replacement filters and Clean Water Machines with G 1 fiters had already been distributed to wholesalers, retailers NSI stores, and consumers. (CX 142-9 to (20)142- 12.) Norelco did not recall these contaminated filters. (Campbell, Tr. 2049, 2057.) Nor did Norelco change its advertising, which claimed that the Machine would effectively remove organic chemicals and clean the water. (Campbell, Tr. 2058-60.

99. Norelco asked U.S. Testing to provide information on methylene chloride. (Campbell, Tr. 2009; Rider, Tr. 2467- , 2491.) Norelco did not ask U. S. Testing to evaluate methylene chloride s toxic or carcinogenic effects. U.S. Testing is a product testing company and does not evaluate scientific data. (Rider, Tr. 2469-70. 100. Eugene Rider of U. S. Testing told Norelco that methylene chloride was a " suspect carcinogen " defining that term as " compound capable of causing cancer in animals. " (Rider, Tr. 2472 2475.) This confirmed information Norelco had obtained from Consumer Reports and FDA (in both August and November 1982). (Lenahan, Tr. 185; Campbell, Tr. 2002, 2036-38; CX 40; CX 41-4; CX 43- 1.) 101. Rider also informed N orelco that the Food and Drug Administration had a regulation limiting the amount of methylene chloride in decaffeinated coffee grounds after processing to 10 ppm. 4 Clean Water Machines and replacement filters were manufactured at Philip:; Park, Nore1ro s factory in Essex, Connedicut. (Emmons, Tr. 1939- 40; Campbell, Tr. 1998; ex 32- 1.) Finished goods were transported from the factory to Norelco s warehouse in umg bland City, New York, where they remained until shipped out in response tu orders from wholesalp.rs, retailers, NSI stores, Of individual mail order customers. (Dinlcy, Tr. 2098- 100, 2139-40; ex 159- , 159- , 159-43. , , ---------. - _u _n- n.._.

139 Initial Decision (Rider, Tr. 2473-74.) The amount in the decaffeinated coffee is about 1 ppm, a level far smaller than the amount leached by the Machine. (Rozman, Tr. 1918-21; CX 124- ) FDA also told Norelco directly about this rule. (CX 43- , 43-2; CX 44- 1.) 102. Norelco s conclusion that the G 1 filters were safe was not supported by the data U.S. Testing provided. Norelco knew that methylene chloride was a suspected carcinogen based on studies that showed the chemical caused cancer in laboratory animals. (Campbell Tr. 2002, 2027- , 2036-38; CX 40; CX 43- 1; CX 159- , 159-36. 103. Patrick Campbell and Emil Misisco, NAPe's Vice President for Quality and Product Assurance, met the Consumer Reports officials to discuss the methylene chloride problem that it had called to Norelco s attention a few weeks earlier. (Campbell, Tr. 2013- 2016; CX 159-2 to 159- ) Campbell told Consumer Reports that Norelco had corrected the problem by designing a new filter that would leach substantially less methylene chloride than the G 1 filter that the magazine had tested. (Campbell, Tr. 2014-15; CX 159- 104. Campbell also informed Consumer Reports that Norelco had G 1 filters that it intended to sell. Campbell assured Consumer Reports that the remaining GIs would be sold out soon. (Campbell, Tr. 2014- 16; CX 159- , 159-16.) (21) 105. On the basis of what Campbell told Consumer Reports at that meeting, the magazine published a review of the device in its February 1983 issue. The article informed readers Company officials told us that they had already changed the manufacturing process to one that doesn t use methylene chloride. " The article continued By the time this report appears, most of the old cartridges should be off the market. " (CX 50- 106. In fact, Gl filers (both in Clean Water Machines and as replacement filters) remained available on retail shelves for years to come. Norelco itself sold 186 000 G 1 filters between late 1982 and early 1986. These filters were distributed both to large retail accounts as well as to consumers ordering replacement filters through the mail. (F. 108- 10.

107. Many consumers purchased a Clean Water Machine after reading the February 1983 Consumer Reports article, in reliance on the assurance that the contaminated filters would no longer be on the market. (Price, Tr. 112- 13; Haag, Tr. 214-15; Bergins, Tr. 254-55. Consumers who read a follow-up Consumer Reports article in March 1986, learned that they were still purchasing Gl filters. (Haag, Tr. 227; Bergins, Tr. 265.

Initial Decision 111 F.

108. From July 1984 to February 1985, Norelco sold 10 221 Clean Water Machines to Philips Electronics, Ltd., a Canadian subsidiary of V. Philips, NAPC's parent company. (Gaines, Tr. 2191; CX 79- Most of the fiters were GIs. (Gaines, Tr. 2192-93; CX 79- 109. Between 1982 and 1986 , the Builders Emporium chain in California acquired 4700 Clean Water Machines. More than 4200 of them were stocked with Gl filters. (RX 157-Z61 to 157-Z63; 157- Z96.

1l0. Norelco also used the remaining inventory of Gl fiters in its warehouse to fil mail orders from consumers. (Haag, Tr. 226, 231; Roche, Tr. 1052-55; Dugan, Tr. 1099, 1102-03; RX 157-Z10, 157- Z55.

111. Thus, more than three years after Norelco learned that G fiters emitted a suspected carcinogen, Norelco itself continued to distribute them to unsuspecting consumers. 112. During this three-year period, Norelco also sold Gl filters through its own NSI stores. Willam Price bought G 1 filters at the San Francisco NSI store in late December 1985. (Price, Tr. 126- , 133- , 150; CX 150.

113. In addition, G 1 filters were on retail shelves during those years. Annette Bergins bought a G 1 from a local DenviIle, New Jersey store in late 1985. The filters on the shelf were GIs. (Bergins, Tr. 268-69. (22) 114. Mel Boynton bought a Gl filter from a St. Paul, Minnesota outlet of a national catalogue showroom chain in late 1985. (Boynton Tr. 625.) Judith Coyle tested a Gl filter in the summer of 1985 that had been purchased from a Philadelphia department store. (Coyle, Tr. 1175.) On August 3 , 1987, an FTC employee bought a Clean Water Machine containing a Gl filter at a Washington, D.C. area catalogue showroom. (CX 188 , Stip. 48.

E. Later Tests of the Clean Water Machine 1. The New Shelter Magazine Tests 115. In 1983 Rodale Press, the publisher of several health and energy-related magazines, tested the performance of the Clean Water Machine for an article published in New Shelter magazine. (Kern, Tr. 975-77; CX 55- 55- ) The tests were conducted in Rodale s product testing department by Mark Kern.

116. Kern used the "TOX" Lest to evaluate the Clean Water Machine s ability to lower the level of chemicals in the water. The 139 Initial Dccision TOX test measured the level of total halogenated organic chemicals in the water. (Kern, Tr. 980-82.

11 7. A halogenated organic chemical is one which contains, in addition to carbon, an element such as fluorine, chlorine, bromine, or iodine. Most of the compounds which arc regulated or required to bc monitored by federal law, including methylene chloride, are halogenated organic chemicals. (Coyle, Tr. 1124-30; CX 56-20. 118. Kern obtained the Clean Water Machine directly from Norelco in January 1983. (Kern, Tr. 1006.) He tested it and found that it added more halogenated organic chemicals than it removed. (Kern, Tr. 1002-05; CX 58.

119. Following completion of the initial tests, Kern called Norelco about the results. Norelco told Kern that he must have tested an "old filter. " (Kcrn, Tr. 1005-06.

120. Kern performed a follow-uptest ofthe Clean Water Machine in July 1983. Using the one-step cycle, a Machine purchased in June 1983 from a Pennsylvania department store increased TOX levels by a factor of 22, from 77 ppb to 1713 ppb. (Kern, Tr. 1007-10; CX 56-44. 121. The Clean Water Machine compared poorly with the other activated carbon fitration devices on the market in 1983 that Kern tested. Whereas the Norelco Clean Water Machine increased TOX by 10-22 times, the remaining nine devices lowcrcd TOX levels by an average of 39% to 73%. (Kern, Tr. 1005; CX 56-20; CX 58.) (23) 122. The October 1983 issue of New Shelter contained an article regarding Kern s tests. It reported that the Clean Water Machine filters increased the level of contaminants in the water, and that this result was repcatcd by a Machine obtained from a department store in the summer of 1983. (CX 55- 2. The North Penn Water Authority Tests 123. Judith Coyle of North Penn Water Authority ("NPW A") tested three Clean Water Machines, and five different filters in 1983 and 1985. Her tests confirmed that the devices were leaching methylene chloride; that contamination levels are higher when the continuous clcan cycle is used; and that contamination continues even after a filter has been used for three months.

124. Coyle is the Water Quality Manager of NPWA, a municipal water supplier in the Philadelphia area, and is an expert in water quality and water contamination testing. (Coyle, Tr. 1118-23. 125. NPWA conducts some 3400 tests each year to determine Initial Decision 111 F. T. whether its drinking water meets the EP A contaminant standards and to comply with monitoring requirements. (Coyle, Tr. 1120-1123. Each day, NPW A determines the specific organic contamination of its water using an EP A-approved method of gas chromatography ("GC" testing. (Coyle, Tr. 1136-37; CX 90-27.

126. Coylc tested her first Clean Water Machine in April 1983. (Coyle, Tr. 1141-43; CX 59- 127. Coyle first tested the Clean Water Machine with distilled water, known to be free of organic chemicals. The carafe of distilled water filtered through the Machine registered a methylene chloride level of 3360 ppb (3.4 ppmJ. (Coyle, Tr. 1145-48; CX 59-6 to 59- 128. Tests of the unfiltered well water showed methylene chloride in the range of 1.3 ppb to 8.7 ppb. When this well water was filtered through the Machine, however, the methylene chloride level was increased to 1990 to 3450 ppb. (Coyle, Tr. 1142- , 1147-49; CX 59- , 59- ) Coyle filtered ordinary tap water, measurcd to contain no mcthylene chloride, through the Machine. After filtering, methylene chloride was found in concentrations up to 3670 ppb. (Coyle, Tr. 1156- 59; CX 59- , 59-10.

129. In 1985, Coyle tested a second Machine containing a Gl filter which was purchased from a department store in mid-July, 1985. This machine added methylene chloride to the water in a concentration 2300 ppb (2.3 ppm), using one-step filtration. Moreover, it increased contamination levels of 4700 ppb (4. (24) ppm) after five minutcs on the continuous clean cycle. (Coyle, Tr. 1167-70; CX 59- 130. Also in 1985, Coyle tested a third Machine using two different fiters. The first filter tested, which had already been in use for three months, added methylenc chloride to the water in concentrations of 1300 ppb. The Machine was then tested using a new filter, purchased directly from Norelco by mail-order in March of 1985; that filter leached methylene chloride in a concentration greater than 2000 ppb. (Coyle, Tr. 1170-72; CX 59- 131. The Clean Water Machine added to the water more chemicals than it removed. (Coylc, Tr. 1199.) For example, when Coyle filtered ordinary tap water through the Machine in her 1983 tests, it increased the level of total specified chemicals (including methylcne chloride) from 10 ppb to between 2079 and 3670 ppb. (Coyle, Tr. 1158, 1199- 1200; CX 59- 10.

139 Initial Decision 3. The 1985 Consumer Reports Tests 132. In late 1985 Consumer Reports conducted follow-up tests of eight Clean Water Machine fiters purchased in four cities. When new five of the eight filters emitted methylene chloride in concentrations between 2 and 5. 5 ppm. (CX 51; CX 54- 133. One filter was subjected to further testing. That filter initially leached 3. 5 ppm. Its methylene chloride levels dropped to 1.5 ppm after the first 20 gallons, and continued to decline slowly, to less than 1 ppm, after 50 gallons. (CX 51; CX 54- 134. Consumer Reports advised Norelco in-house counsel Lynne Bezikos of these test results in January 1986. (CX 51; CX 54-6; F. 602.

4. The 1986 U. S. Testing Tests 135. In January 1986, after being advised by the EPA of Coyle 1985 test results, Norelco submitted three Gl filters to U.S. Testing for analysis. When new, the filters leached methylene chloride in concentrations of 1200 to 1900 ppb on the one-step cycle. Thereafter the amount of the chemical increased to 1700 to 2300 ppb, using the continuous clean cycle. (RX 30.

5. The 1987 Intech Biolabs Tests 136. In September 1987, Norelco retained L. Wendell Haymon Ph. , President of Intech Biolabs, to test a single G 1 filter over a period of time. (RX 156- , RX 156- , RX 156-Z5. (25) 137. Dr. Haymon filtered spring water, containing . 13 ppb methylene chloride, through the Clean Water Machine. Initially, filtering increased the level of methylene chloride in the water to 2400 ppb (2.4 ppm J. The levels of methylene chloride in the filtered water declined over time, to 1180 ppb after 50 carafes. (RX 156- , 156- ZI9.

6. Summary of Test Results 138. The tests of the Clean Water Machine show that G1 filters leach methylene chloride in concentrations of 1.2 to 5. 5 ppm. (F. 120 127-131 , 132- 133 , 135- 137.

139. The tests demonstrate that levels of methylene chloride are higher when the continuous clean cycle is used. Machine users favored the continuous clean cycle, due to Norelco s claim that it would make 31.)the water "cleaner." (F Initial Decision 111 F. T. YII. THE FALSITY OF RESPONDENT S ADVERTISING REPRESENTATIONS A. The Organic Chemical Removal Claim 140. Norelco s reprcscntations that under typical water conditions the Clean Water Machine effectively helpcd remove organic chemicals from the water it was treating were false for the 354 708 G Ffilters sold from 1982 to at least early 1986.

1. The organic chemical content of drinking water 141. The term "organic chemicals" includes all substanccs which contain carbon. (Coylc, Tr. 1124; Ohanian, Tr. 1554-55.) These substances can be divided into natural (primarily decayed vegetation) and synthetic organic chemicals (industrial chemicals). (Coylc, Tr. 1124-25; Ohanian, Tr. 1590.) Natural organics arc not regulated by thc government because they are not hazardous. (Coyle, Tr. 1127. 142. Synthetic chemicals are made up of volatile (or purgeable) organic chemicals ("YOCs ) and nonvolatile (nonpurgeablc) organics. (Coyle, Tr. 1126-27; Zeise, Tr. 1291; Ohanian, Tr. 1555 , 1582.) A chemical is considered volatile if it has a relatively low boiling point and readily escapes into the atmosphere. (Zeise, Tr. 1291; Ohanian Tr. 1554-55.

143. Many of the chcmicals in drinking water of greatest concern from a health standpoint are YOCs. (Coyle, Tr. 1127; Ohanian, Tr. 1555-56; CX 88- ) These include, for example, trihalomethanes, a class of YOCs mentioned specifically in the advertising for the Clean Water Machine. (Coyle, Tr. 1132; Zeise, Tr. 1291-92.) Trihalomethanes include chemicals such as chloroform which are formcd as byproducts of the chlorination (26) process used to disinfect water. (Coyle, Tr. 1126-27; Zeise, Tr. 1292.

144. Methylene chloride is a synthctic organic chemical and a YOC. (Zeise, Tr. 1290; CX 96- 15.) It is not a trihalomethane, but a dihalomethane. (Zeise, Tr. 1294.

a. Methylene chloride 145. Drinking water in this country comes from two sources; groundwater (below-ground) and surface water (rivers, lakes, and other above-ground). (Coyle, Tr. 1134.) Methylene chloride is usually not found in groundwater, surfacc water, or tap water. (Coyle, Tr. 1139 , 1159-60; Zeise, Tr. 1297; CX 156-82. 139 Initial Decision 146. Some water does contain methylenc chloride, primarily as a result of industrial activity. (CX 96- 15.) Ordinary levels of methylene chloride in such water supplies are in the range of 1.5-2 ppb. (Ohanian, Tr. 1649-50.) The Ogalalla Aquifer, an underground water source that supplies drinking water to much of the western United States, has less than 5 ppb of methylene chloride. (Rozman, Tr. 1921- 22.

147. All groundwater drinking water systems and 98% of all surface water systems contain less than 0.5 ppb of methylene chloride. None of the 2% of surface water systems containing more than 0.5 ppb are expected to have levels above 50 ppb. (CX 96- 148. A national survey conducted for EPA, the National Organic Monitoring Survey, found methylene chloride in 15 of 109 water samples, with a median concentration between 1 and 2 ppb. (Ohanian Tr. 1586-88; CX 186- ) A second survey, the National Scrcening Program for Organics in Drinking Water, found detectable methylene chloride in 5 of 118 water systems survcyed. Those five systems had average concentrations of 0. 6 ppb. (CX 96- , 96-25. b. Total organic chemicals 149. A national survey of hundreds of water supplies found average concentrations of total organic carbon to be 1.5- 6 ppm. (Ohanian Tr. 1589-90; CX 95-44.

150. Typical VOC concentrations in public water systems are low. A national survey of nearly 1000 groundwatcr supplies found that only about 0.5% of thc samples had total VOC lcvels above 50 ppb, and none was above 100 ppb. (Ohanian, Tr. 1588-89; CX 184- 151. Surveys have also been conducted on the levels of nonvolatile organic chemicals in water supplies. The National (27) Organics Reconnaissance Survey found a median nonvolatie organics concentration of 1.5 ppm. (Ohanian, Tr. 1582-84; CX 185- ) The National Organic Monitoring survey of 113 community water supplies found average nonvolatile organics levels of 1. 2 ppm. (Ohanian, Tr. 1586-87; CX 186- 152. Drinking water typically contains about 1-2 parts per milion of organic chemicals, with most of that consisting of non-hazardous natural organics. (Coyle, Tr. 1135.

2. The Clean Water Machine s chemical removal capabilities 153. U. S. Testing measured the efficacy of the Clean Water Initial Decision 111 ".

Machine for removing chloroform, a trihalomethane, in mid- 1982. (RX 56-U to 56- ) The test found that after filtering by the Machine, the levels were rcduccd by 78-90%. (RX 56- 154. Consumer Reports as part of its February 1983 article, also tested the Machine s chloroform-removal capabilitics. In these tests thc Machine removed 70-85%. (CX 50- 3. The methylcne chloride emission 155. The total organic content of typical water is 1 or 2 ppm, and the Machine removed about 80% of it. The Machine was adding as much or more organic chemicals (in the form of methylene chloride) as it could have been removing under typical water conditions. (F. 138. 156. Most of the organic content of water is natural. (F. 150-52. Therefore, the amount of synthetic organics that the Machine would be removing under typical conditions would be far less than the amount of methylene chloride it was adding. 157. New Shelter tested the halogenated organic content of tap water before and after filtering by the Machine and found that the levels increased by 10 to 22 times. (F. 119-21.) Judith Coylc s tests found larger increases in the chemical content of the water after filtering. (F. 131.) B. The Clean or Cleaner Water Claim 158. Thc Clean Water Machine also did not make the water treatcd clean or cleancr under typical conditions, as advertised by Norelco. The amount of methylcnc chloride lcached by the Machine G 1 filters was above typical levels of all organic chcmicals combined. 159. The Clean Water Machine also did not make the water cleaner" because, under typical water conditions, it added as (28) much or more mcthylene chloride than all of the organic chemicals it could have removcd.

C. The Independenl Laboratory Test Claim 160. Norelco s representation that independent laboratory tests proved that the Clean Water Machine made typical tap water clean or cleaner was also false. Since the Machinc, in fact, did not make the water clean or cleaner, tests could not have proven that it did. D. The Reasonable Basis Claim 161. Norelco s advertising implied that the company had a .''-H .U. ..aL'-H' LL.LL.'-'" oJ '-L J.L\J" 139 Initial Decision reasonable basis for its performance claims. (F. 78-79.) In fact because of the methylene chloride problem, these claims were false and unsubstantiated. Since November 1982, Norelco had no basis for claiming that the Machine effectively removed organic chemicals and made the water clean or cleaner.

VIII. RESPONDBNT S FAILURE TO DISCLOSE THE MBTI!YLENE CHLORIDE CONTAMINATION 162. Respondent represented that the Clean Water Machine was effectively removing organic chemicals and cleaning the water, while it continued to sell contaminated Gl filters, from 1982 to 1986. When Norelco was informed of the problem in November 1982, it did not change its advertising or disclose the methylene chloride emission. (F. 98.) Thus, respondent failed to disclose a fact necessary to correct the misleading impression it had created that the Machine was effectively reducing organic chemical levels and cleaning the water. IX. THE MATF.RIALITY OF RF.SPONDENT s MISREPRESF.NTATIONS AND FAILURE TO DISCLOSF.

A. Respondent's Deceptive PTactices Wen Material 163. Respondent's misrepresentations about the efficacy of the Clean Water Machine, and its failure to disclose the methylene chloride problem, were material.

164. Respondent's advertising expressly represented that the Machine would effectively help remove organic chemicals, would make tap water clean or cleaner, and that independent tests supported these claims. (F. 69-77.) These representations were made deliberately. (F. 40-41.) 165. The representations and failure to disclose relate to a central feature of the product-its ability to remove (29) chemicals and clean the water. These capabilities are the reasons why consumers purchased it. (F. 168- 179.

166. The representations also implied that the Machine made the water safer by removing potentially hazardous chemicals. The fact that the Machine added to the water a chemical considered by government agencies to be potentially hazardous would have been an important factor in consumers' purchase and use decisions. (Price, Tr. 140-41; Roche, Tr. 1061-62; Louie, Tr. 1217- 18. 167. Other water filtration devices on the market, which did not emit chemicals into the water, removed organic chemicals and made Initial Decision 111 F.

the water cleaner. (CX 50; CX 55.) Absent respondent' s deceptive representations and omission of fact, consumers could have purchased such other, effective devices. (Price, Tr. 140. B. Consumer Testimony 168. The advertising presented the Clean Water Machine as an effective means of removing chemicals from tap water. The primary reason consumers bought the Clean Water Machine was because they were led to believe that it could minimize their cxposure to chemicals and contaminants in tap water. (Price, Tr. 119-20; Haag, Tr. 218-19; Bergins, Tr. 250-51; Boynton, Tr. 615, 617 , 623; Roche, Tr. 1046; Maranki, Tr. 1071- , 1088.

169. Consumers who bought the Clean Water Machine were attracted to the name. They understood it to mean that the device would "clcan the water (of) contaminants ma(k)e it safer to drink." (Bergins, Tr. 256; Rochc, Tr. 1047. 170. Norelco promoted the Clean Water Machine to consumers interested in taking prcvcntive steps to minimize risks to their health. (Haag, Tr. 213; Boynton, Tr. 627; Maranki, Tr. 1089; Campbell, Tr. 2204-26; CX 9-7; CX 19- 171. Even when scientists disagrcc about thc hazards of a certain compound, the consumers want to avoid ingesting chemicals whose potential risks are unknown. (Boynton, Tr. 1218; Roche, Tr. 1062; Maranki, Tr. 1089-90; Dugan, Tr. 1110; Louie, Tr. 1218. 172. Consumers were aware that thc ingcstion of chemicals had been linked to cancer, and wanted to avoid that risk. (Bergins, Tr. 251; Roche, Tr. 1054; Maranki, Tr. 1089; Dugan, Tr. 1094. 173. Some bought the Clean Water Machine to protect the more vulnerable members of their families, such as babies, the elderly and the infirm. (Boynton, Tr. 516; Maranki, Tr. 1070 , 1075; Louie, Tr. 1204-05; CX 76- 3; CX 135-4; CX 138. (30) 174. Some consumers were also aware that chloroform, a suspected carcinogcn in humans, was present in tap water as a by-product of the chlorination process. (Price, Tr. 111; Haag, Tr. 219; CX 152- 175. For some consumers, their purchase of the Clean Water Machine was part of their overall interest in reducing the risks associated with thc ingestion of chemicals. (Bergins, Tr. 251; Boynton, Tr. 627-28; Roche, Tr. 1053.

176. The consumers who testified were unanimous in their opinion that they would not have purchased the Machine or continucd to use NORTH AMERICAN PHILIPS CORPORATION 177 139 Initia! Decision the Machine had they known that the filters they bought injected into their drinking water a chemical that the EP A and other scientific organizations thought to be a possiblc or probable human carcinogen. (Price, Tr. 141; Haag, Tr. 234; Bergins, Tr. 270-71; Boynton, Tr. 628- 29; Roche, Tr. 1061-62; Maranki, Tr. 1088; Dugan, Tr. 1109; Louie Tr. 1217- 18.

177. Consumers paid $50 for a product that they thought would promote good health, only to find that it might pose an additional potential hazard. (Boynton, Tr. 618; Maranki, Tr. 1088-89; CX 138. 178. The Clean Water Machinc instructions included thc following warning in two scparate places;

WATER THAT HAS NOT BEEN TESTED OR TREATED FOR HUMAN CONSUMPTION SHOULD NOT BE USIW. Only tap water which has been rated as safe for drinking should be used. The Norelco Clean Water Machine wil help remove elements which adversely affect taste, odor, clarity, and color of tap water. THIS APPIJANCE DOES NOT TAKE Tile PLACE OF ANY STERILIZATION OR DISTILLTION PROCESSES YOU WOULD NORMALLY FOLLOW. (Emphasis in original.) (CX 2- 1.) 179. Consumers used the Clcan Water Machine to provide their familics water freer of chemicals than what government agencies deemed to bc allowable. (Price, Tr. 118- 19; Haag, Tr. 222; Bergins Tr. 259; Boynton, Tr. 626-27; Maranki, Tr. 1087-88; Louie, Tr. 1217. X. THE POTENTIAL HAZARDS OF METHYLENE CHLORIDE 180. The consensus of govcrnmental and scientific bodies is that methylene chloride is potentially hazardous. Consumers would not have purchased a device to remove potentially hazardous (31) chemicals when the device added a chemical that many scientists and government agencies consider hazardous. (Price, Tr. 140; Haag, Tr. 233-34; Bergins, Tr. 270-71; Boynton, Tr. 628; Roche, Tr. 1061-62; Maranki, Tr. 1089; Dugan, Tr. 1110; Louie, Tr. 1218. 181. Both parties' experts agreed that it would be prudent to avoid unnecessary ingestion of methylene chloride at the levels lcachcd by the Clean Water Machine. (Cohn, Tr. 468; Farland, Tr. 827; Rozman Tr. 1922- 23; Klaassen, Tr. 2434.

A. The Carcinogenicity of Methylene Chloride 182. Several government agencics and scientific organizations have concluded that methylene chloride causes cancer in laboratory animals Initial Decision 111 F.T.C.

and is a possible or probable human carcinogen. (Farland, Tr. 956-67; Zeise, Tr. 1812-18; CX 105-21.) Respondent’s expert witnesses agreed that methylene chloride causes cancer in laboratory mice and can reasonably be considered a possible human carcinogen. (Rozman, Tr. 1803-04; Klaassen, Tr. 2398-96.) 183. Complaint counsel’s expert witnesses testified that ingestion of the methylene chloride leached by the Clean Water Machine raises the risk that users could contract cancer. (Cohn, Tr. 454; Zeise, Tr. 1398.) The existence of this risk supports the conclusion that respondent’s deceptive practices were material. All of the experts agreed that it would be prudent for consumers to avoid drinking water containing these levels of methylene chloride. (Cohn, Tr. 468, 554; Farland, Tr. 827, 969-70; Zeise, Tr. 1454-55; Ohanian, Tr. 1561-62; Rozman, Tr. 1922-25; Klaassen, Tr. 24384.) 1. Actions by government agencies and scientific organizations a. Environmental Protection Agency 184. Since early 1985, EPA has classified methylene chloride as a “probable human carcinogen” (Category B2) under its cancer assessment guidelines. (Farland, Tr. 717, 946-47.) This category covers substances for which there is sufficient evidence of carcinogenicity in tests of laboratory animals and inadequate human evidence. (Farland, Tr. 718; CX 110-10.) 185. The EPA Guidelines establish five categories of evidence of carcinogenicity: [82] Group A— Human carcinogen Group B— Probable human carcinogen 5 Group C— Possible human carcinogen Group D— Not classifiable as to human carcinogenicity Group E— Evidence of non-carcinogenicity for humans (Farland, Tr. 716-17; CX 110-10.) b. Consumer Product Safety Commission 186. In September 1987, the Consumer Product Safety Commission (“CPSC”) issued a notice of interpretation and enforcement policy, announcing its intention to prosecute manufacturers who fail to affix warning labels. (Cohn, Tr. 319-21; CX 121-1, 121-2.) > Group B is divided into two subclassifications. Group B1 covers chemicals for which the human evidence is limited, while Group B2 covers chemicals for which the human evidence is inadequate, but the animal evidence is sufficient. (Farland, Tr. 717-18; CX 110-10.) 139 Initial Decision 187. In issuing this notice, CPSC concluded: After considering the comments on the proposed rul( and other available evidence the Commission has concluded that there is litte or no uncertainty involved in a determination that household products containing methylene chloride and presenting significant exposures to consumers may pose a carcinogenic risk to humans unless and until persuasive evidence to the contrary is obtained. (CX 121c. Food and Drug Administration 188. Methylene chloride is used by some companies to decaffeinate coffee. (CX 124- ) Since 1967, an FDA regulation has set a maximum level of 10 ppm residue in the ground coffee beans. (CX 122-1.) 189. At the maximum level of 10 ppm in the grounds, brewed liquid coffee would contain, at most, about 0. 1 ppm. (Rozman, Tr. 1918-21; CX 124- 190. The FDA has also proposed to ban methylene chloride in cosmetics, including aerosol hair sprays. (CX 124. ) The agency (33) has concluded that methylene chloride causes cancer in animals and may be carcinogenic to humans. (CX 124d. Occupational Safety and Health Administration 191. In 1971 , the Occupational Safety and Health Administration OSHA") established a maximum permissible level of exposure to methylene chloride in the workplace of 500 ppm for an eight-hour day. In 1986, OSHA issued an advanced notice of proposed rulemaking to lower the limit, because the 500 ppm standard may not adequately protect workers against potential cancer and other risks. (CX 128-1.) OSHA concluded that methylene chloride was a proven carcinogen in laboratory animals. (CX 128e. International Agency for Research on Cancer 192. The International Agency for Research on Cancer ("IARC" ) is an international organization of cancer experts. (Farland, Tr. 716; Klaassen, Tr. 2372-73.

193. IARC's most recent review of methylene chloride was published in 1986 and reached the same conclusion as EP A. (Farland Tr. 723- 26; CX 113-63.

g.

180 FEDERAL TRADE COMMISSION DBCISIONS Initial Decision 111 F. T. f. State of California 194. The State of California has had in effect since 1982 or 1983 a nonregulatory guideline, or "action level " for methylene chloride in drinking water of 40 ppb. (Zeise, Tr. 1402-03; CX 130- 195. California has a law that requires that bottes used for bottled water not leach more than 1 ppb of methylene chloride into the water. (Zeise, Tr. 1397-98; CX 131- 196. The 1987 California Department of Health Services report on the health effects of methylene chloride concluded that it is a probable human carcinogen. (Zeise, Tr. 1312; CX 132-96. Other states 197. Ten states have guidelines for methylene chloride in drinking water. The maximum levels range from 2 ppb to 150 ppb. (Ohanian Tr. 1594; CX 190-117.

h. Private organizations 198. The National Sanitation Foundation is a non-profit organization that develops standards and tests products for water filtration devices to ensure that drinking water is not toxic. (Bell, Tr. 1241; CX 115- , 115-20.) The acceptable (34) level for methylene chloride is 5 ppb. (Bell, Tr. 1244-45; CX 115-23.

199. The Water Quality Association, an international trade association of water treatment device manufacturers, has developed voluntary guidelines for the use of solvents in the manufacture of such devices. (Bell, Tr. 1246-47; CX 114.) The guidelines allow 5 ppb of methylene chloride. (Bell, Tr. 1247; CX 114-14. 200. No government agency or scientific organization has concluded that methylene chloride is not a carcinogenic risk to humans. (Cohn Tr. 466; Zeise, Tr. 1313; Klaassen, Tr. 2399. 2. The scientific evidence 201. Methylene chloride is a probable human carcinogen. (Cohn, Tr. 391; Farland, Tr. 824; Zeise, Tr. 1312, 1393.) The opinion is shared by a significant portion of the scientific community. (Cohn, Tr. 391.) B. The Non-Cancer Toxic Effects of Methylene Chloride 202. Methylene chloride ingestion can also result in adverse health effects other than cancer. Methylene chloride added to water by the Clean Water Machine exceeds safe levels set by the EPA, and raises a notpnti"l risk to the health of its users. NORTH AMERICAN PHlUPS CUKPUJtl\ l!V"

139 Initial Decision 203. Methylene chloride is a toxic chemical that can produce a number of adverse health effects. (Ohanian, Tr. 1501-11; CX 127- 11; CX 128- 204. Methylene chloride poses risks to individuals with heart disease. (Ohanian, Tr. 1502-03; CX 83- , 83- , 83- 205. Some people are more sensitive to carbon monoxide formed by ingestion of methylene chloride, including individuals with ischemic heart disease, pregnant women, and children. (Ohanian, Tr. 1509; CX 83-4 to 83-5; CX 99-104.

206. The National Academy of Sciences Drinking Water and Health issued in 1980 , evaluated the toxicity of methylene chloride. (Ohanian, Tr. 1520; CX 94- 16 'to 94- 20.) The Academy established recommended safe levels of ingestion of methylene chloride by adults for one and seven-day periods. (Ohanian, Tr. 1526; CX 94- 16 to 94- 20.) These levels, termed "SNARLs" (Suggested No Adverse Response Levels), are based only on non-cancer toxic effects. (Ohanian Tr. 1518.

207. The one-day SNARL for methylene chloride was set at 35 mg/l (35 ppm). The seven-day SNARL was 5 mg/l, meaning that it (35) is considered safe to ingest that amount of methylene chloride per day for up to seven days. (Ohanian, Tr. 1526; CX 94- 94-20.) 6 The longer the period of exposure, the less of the chemical is acceptable. (Ohanian, Tr. 1527; CX 94-20.

208. The EPA's Office of Drinking Water established its own SNARLs for methylene chloride on March 14 , 1981. (Ohanian, Tr. 1529; CX 83.) EPA issues Health Advisories for chemicals for which binding maximum contaminant levels have not yet been set. (Ohanian Tr. 1529.

209. EPA' s Health Advisories are not legally enforceable. (Ohanian Tr. 1622; CX 83-1.) Unlike the maximum contaminant levels MCLs ), the Health Advisories are strictly health-based and do not consider costs, technology, or other factors. (CX 83-1.) 210. Most toxic effects other than cancer are believed to have a threshold" dose. Doses below that level are considered safe. (Zeise Tr. 1364-68.

211. The SNARL is based on the highest dose level in the study at which no adverse effects were detected the "NOEL" or no observec effect level. (Zeise, Tr. 1364; Ohanian, Tr. 1518-22. 6 The ca1rulation of these SNARl.,; contained a computational error. The one-day SNARL should be 4 mg-/I and the seven-day SNARL should be 6.4 rng/I. (Ohanian, Tr. 1526; ex 83- , 83- Initial Decision 111 F.

212. In setting the SNARL, the NOEL is reduced by a safety factor. A factor of ten is applied because of the uncertainties of extrapolating from animal studies to humans. (Ohanian, Tr. 1522-24; Klaassen, Tr. 2292; RX 133-Z13.

213. EPA's SNARLs are calculated for children, while the National , Tr.Academy of Sciences' SNARLs are based on adults. (Ohania 1531; CX 83-1) Since children drink more water per unit of body weight and are therefore considered more sensitive, the EP A SNARLs are lower. (Ohanian, Tr. 1531 , 1606-07.

214. In 1981 EPA calculated SNARLs for methylene chloride for the following durations of exposure:

one-day 13. mg/l (13ppm) ten-day 1.50 mg/1 per day longer term mg/l per day (Ohanian, Tr. 1531-32; CX 83-5 to 83- (36) 215. The EPA SNARLs were updated in 1985 based on new animal data. (Ohanian, Tr. 1535-37; CX 84.) The EPA SNARLs were also updated in 1987. (Ohanian, Tr. 1539; CX 85.) EPA declined to set a lifetime SNARL because of methylene chloride s carcinogenicity. (Ohanian, Tr. 1539-42; CX 85-9 to 85- 11.) 216. The levels of methylene chloride added by the Clean Water Machine exceed the safe levels set by EPA in 1981 , 1985 and 1987. Most tests of the Machine detected amounts exceeding the ten-day SNARL of 1.5 mg/1. (F. 138.

217. The fact that the amount of methylene chloride added by the Machine exceeds levels recommended by EP A as safe would have been material information to consumers in purchasing or using the Machine. (Price, Tr. 140; Haag, Tr. 234; Bergins, Tr. 270; Boynton fr. 628; Roche, Tr. 1061; Maranki, Tr. 1088; Dugan, Tr. 1109; Louie Cr. 1217.

C. The Evidence A va,ilable In 1982 218. Norelco was informed in November 1982 of the methylene chloride leaching problem and continued sellng Gl fiters. (F. 95-98. 219. By November 1982, Norelco was aware that methylene chloride was a suspected carcinogen through its discussions with U. testing, Consumer Reports and FDA. (F. 100. 220. Two studies suggesting methylene chloride caused cancer in boratory animals, by Dow Chemical and the National Toxicolop" l'jUltTtl AMt.ltiCAN Philip:: L;UH.PUKATION 139 Initial Decision Program ("NTP"), were available in November 1982. (Farland, Tr. 825 , 944-45; Zeise, Tr. 1385-86.

221. The Dow Chemical study found in salivary gland tumors in male rats and benign mammary gland tumors in rats of both sexes. (Cohn, Tr. 393; Zcise, Tr. 1322; CX 103- ) The NTP gavage study reported cancer responses. (Cohn, Tr. 450-51; Farland, Tr. 825; Zeise Tr. 1386-88.

222. Mcthylene chloride also had been shown to be genotoxic in bacteria tests, and carcinogenic potential. (Cohn, Tr. 450; Farland, Tr. 944; Zeise, Tr. 1388-89.

223. There was sufficient evidence in 1982 that individuals drinking water fitered by thc Clean Water Machine were potentially at risk of developing cancer. (Cohn, Tr. 450- 54; Farland, Tr. 826; Zeise, Tr. 1392.

224. By 1982, both thc National Academy of Sciences and EPA had set recommended safe levels of ingestion of methylene chloride in drinking water for non-cancer effects. The levels of (37) methylene chloride leached by the Clean Water Machine exceeded these recommcnded safe levels. (Ohanian, Tr. 1534- , 1562-63. 225. By 1982, the National Acadcmy of Sciences guides for methylene chloride were published. (Ohanian, Tr. 1529; CX 94.) The EPA Health Advisory was also available. (Ohanian, Tr. 1533-34. 226. Norelco was aware of the FDA' s rule on methylene chloride levels in decaffeinated coffee grounds. (F. 101.) 227. Had Norelco consulted qualified toxicologists in 1982, they would have warned the company that the methylcne chloride leached by the Clean Water Machine created a potential carcinogenic hazard and would have advised it not to expose people to the chemical. (Zeise Tr. 1392; Rider, Tr. 2482-83.

228. Had Norelco contacted EPA's Office of Drinking Water, it would have been told of the existence of the Health Advisory for methylene chloride and could have ascertained that they were exceeding the guideline. (Ohanian, Tr. 1534 , 1562-63. 229. EPA had by that time also published in the Federal Register proposed rulemakings relating to volatile organic chemicals generally and methylene chloride specifically. (Ohanian, Tr. 1566; CX 86; CX 88.

230. There was sufficient information available in 1982 that the methylene chloride leached by the Clean Water Machine posed a potential carcinogenic and toxic risk to users. N ore1co knew that Initial Decision . 111 F.T.C.

methylene chloride was a suspected carcinogen based on laboratory animal tests. Additional information confirming methylene chloride’s hazards, and indicating the government’s attempts to limit exposure to it, was available. Norelco’s search for such information was inadequate. (Rider, Tr. 2467-70.) XI. RESPONDENT’S 1986 FILTER REPLACEMENT PROGRAM A. Reappearance of the Methylene Chloride Problem 231. Judith Coyle’s 1985 tests showed that two Clean Water Machines leached methylene chloride. (Coyle, Tr. 1172.) In the late summer of 1985 she telephoned Frank Bell of EPA’s Office of Drinking Water and sent him a copy of her test report. (Coyle, Tr. 1178-74; Bell, Tr. 1229; CX 59-8, 59-5.) 282. In mid-September 1985, Bell contacted Robert Gaines, manager of Norelco’s Health Care division. (Bell, Tr. 1229-30; Gaines, Tr. 2159-60.) Bell told Gaines that, according to Coyle’s tests, the Clean Water Machine added a potential [88] carcinogen to the water in high quantities, and that it was not proper for Norelco to continue selling the product. (Bell, Tr. 1229-30.) 233. Bell sent Gaines a copy of Coyle’s report and asked him to respond. (Bell, Tr. 1280.) Shortly thereafter, Coyle also spoke with Gaines, and told him the filters that she had tested were G1 filters. (Coyle, Tr. 1173-74, 1177; Gaines, Tr. 2163-65, 2177-79.) 234. Over the following week or so, Gaines did some research regarding methylene chloride, and learned that it was a suspected carcinogen. (Gaines, Tr. 2168, 2175-78.) In addition, he learned that Norelco had continued to sell G1 filters in the filter design change in 1982. (Gaines, Tr. 2175-76, 2184.) 235. Norelco’s legal staff instructed Gaines to await an EPA letter before acting. (Gaines, Tr. 2184-85.) 236. Norelco personnel would not return Coyle’s calls when she tried to relay further information to the company. (Coyle, Tr. 1177- 78.) When Consumer Reports called Norelco in January 1986 to alert it of the continuing contamination problem, Norelco insisted that it thought that its filters were “problem-free.” (CX 51; CX 54-6.) 237. Despite its knowledge of the methylene chloride problem Norelco continued to sell G1 filters to consumers until early 1986. (Roche, Tr. 1052-55; Dugan, Tr. 1102-03.) 238. In mid-February 1986, the March 1986 issue of Conswmer Reports contained an article entitled ‘““A good reason to shut off vn.ln l11nl',1\.1\.d1.1 I111IJU '. \Jv u u ,.n V"' 139 Initial De( ision Norelco s water filter." The article reported that Norelco had broken its "promise (of) corrective action " to Consumer Reports and had continued to sell contaminated filters. (CX 51.) 239. Norelco s President Kress then ordered that all Clean Water Machines and filters be removed from the NSI sales shelves. (RX 157- Z15 to 157-Z18 , 157-Z81.) 240. On February 25, 1986, Norelco ordered the destruction of the remaining Gl filters. (Crowley, Tr. 2218; RX 157-Z37 , 157-Z87. B. The F'later Replacement Program 241. In February, 1986 , Norelco decided to initiate a filter replacement program, whereby it would replace consumers ' G 1 filters with the newly designed G3 filters which contained no methylene chloride, beginning in March 1986. (Gaines, Tr. 2190; Crowley, Tr. 2206-07; CX 33- ) This action was taken in (39) response to government pressure. (Dinley, Tr. 2138-39; Gaines 2190. 1. The number of notifications made 242. Norelco sold 248 000 Clean Water Machines, and 354 000 G J filters. (F. 25 , 29.) A total of 19 188 consumer notification letters were sent out. (CX 33-6; CX 144- , 144- 243. Norelco sent 1505 letters to trade customers, notifying them of its replacement program. (CX 144- 2. Norelco s replacement program letter 244. Norelco s letter to consumers announcing the filter replacement program was dated March 14 , 1986:

Norelco first marketed the Clean Water Machine in 1982 after extensively testing its ability to fiter drinking water. In late 1982, subsequent independent testing demonstrated that the Clean Water Machine filters contained traces of methylene chloride in the glue used to seal the filter cartridge. Notwithstanding the lack of conclusive data regarding methylene chloride, Norelco changed its fijter design in late 1982 to remedy any potential problem.

Recently it has come to our attention that a few of the Clean Water Machine filters of the old design could stil be found on retail shelves. There are no established EP A standards for concentration levels of methylene chloride in water. Methylene chloride is found in decaffeinated coffee, spices, hairspray, paint thinner and other consumer products. However, Norelco has undertaken a voluntary replacement of fiters of the old design with fiters of a design which do not produce any methylene chloride. Norelco is offering to replace all old design filters with fiters of a new design. you have a fiter with a date code of 472 or lower or no date code at all, we will , 186 EDERAL TRADE COMMISSION DECISIONS Initial Decision 111 F.

replace it for you free of charge. The date code is molded on the underside of the filter dial.

To obtain replacement filters, send your existing fiters to: Norelco Service Inc.

30- 10 Review Avenue Long Island City, NY 11101 140J Nore1eo will reimburse you for your postage. (CX 69.

245. The consumer letter contained misleading statements that minimized the methylene chloride problem. Norclco knew that EP A listed methylene chloride as a probable human carcinogen and that G filters emitted mcthylene chloride in levels above those deemcd safe for non-cancer effccts by the 1985 EPA Health Advisory. (Gaines, Tr. 2195-96; Crowley, Tr. 2229; CX 65- , 65-4; CX 66. 246. The March 1986 Consumer Reports article generated consumcr complaints to Norclco. In response Norelco stated that there were no known health risks and no established standards which the G 1 filters did not mect. (Crowley, Tr. 2237. 3. Consumer response to the replacement program 247. In response to the replacement program, Norclco received 2460 claims from consumers and trade customers. (CX 142-22. 248. In response to these claims, Norelco distributed 18 998 filters to consumers, and refunded a total of $12 268, some of which was for returned 'Machines. In addition, it distributed 239 filters to trade customers, and replaced filters in 3000 Clcan Water Machines that had yet to bc sold to retail customers. (CX 142-22. 249. Norelco s cost for the replacement program was $132 000. (RX 157-Z96.

DISCUSSION XII. THE FACTS Norelco began production of the Clean Water Machinc in June 1982 using a chemical solvent, methylene chloride, to glue together parts of , 83. ) In thc Machinc s replaceable filter cartridge. (F 12 , 13 November 1982, Norelco launched an advertising campaign that encouraged consumers to question thc quality of their tap water. (F. 44.

These advcrtiscments asked Is your tap water as clean as it seems? It could contain impurities you don t want your family to , ,. ., ,,. - h' -AH 'U' '''A "'...U '-.. 'U' 139 Initial Decision drink." According to the ads independent tests prove " that the Clean Water Machine can "remove up to 90%" of "organic wastes chlorine, synthetic detergents and trihalomethanes" from drinking water. Urging consumers to "help clean up (their) tap (41) water Norelco promised that the device could make tap water "bottled-water clean. " (F. 45-51.) In November of 1982 Consumer Reports informed Norelco that thc Clean Water Machine s replaccable fiter was emitting high lcvels of methylene chloride, a suspected carcinogen, into the filtered water. (F. 85-86.) Norelco had learned from the Food and Drug Administration three months earlier that mcthylene chloride, the solvent used in the filters, was linked to cancer. (F. 84.) Norelco verified this information through tests performed by the United States Testing Company. (F. 87-89.

By that time, Norelco had already distributed thousands of the contaminated "G 1" filters to wholesalers and retailers. 7 (F. 94-95. About 214 000 remained in Norelco s warehouse. (F. 95.) Company officials met with the Consumer Reports staff and said that Norelco had changed its manufacturing method to produce a second-generation ("G2") filter that leached less methylcne chloride, and that the G 1 filters would sell out soon. The magazine reported this information in its February 1983 issue. (F. 103- 107.) Norelco, however, continued to sell GIs for the next three years. (F. 106- 114. In September of 1985 , the Environmental Protection Agency contacted Norelco after it learned that Gl filters were still being sold. (F. 232. Consumer Reports working on a follow-up to the February 1983 article, questioned Norelco in January of 1986 about the continued availability of filters. (F. 236.) Norelco finally undertook a Gl filer replacemcnt program in March 1986. (F. 241.) Norelco contacted fewer than 10% of all Clean Water Machine owners about the availability of an improved filter, and did not mention that the GIs emitted a suspected carcinogen into the water. (F. 242-244.) Of the more than 354 000 Gl filters that Norelco sold, it replaced about 000. (F. 242 , 248.

XII. THE VIOLATIONS Respondent' s television and radio commercials, magazine and newspaper ads, promotional materials and packaging expressly 7 Norelco dislributed GIs as original equipment in Clean Water Machinr.s, which retailed for about $50 each as well as separately-boxed replacement fiters. which retailed for about $5 each. (F. 25- 29. Initial Decision 111 F.

represented that the Clean Water Machine would effectively help remove organic chemicals from water, that it would make water clean" or "cleaner " and that independent tests proved that the device made water "clean " or "cleaner." (F. 45, 48- , 63- , 69-77. (42) An advertisement is deceptive if it contains a material representation or omission that would be likely to mislead reasonable consumers. Cliffdale Associates, Inc. 103 FTC 110, 164- 65 (1984), appeal dismissed sub nom., Koven v. FTC No. 84-5337 (11th Cir. October 1984); Thompson Medical Co. 104 FTC 648, 816- 17 (1984) afi'd 791 F.2d 189 (D. C. Cir. 1986), cert. denied 107 S.Ct. 1289 (1987). The G 1 filters emitted levels of methylene chloride far higher than those found in typical tap water. Methylene chloride is a synthetic organic chemical associated with industrial waste, which thc advertising promised the Clean Water Machine would filter out. A reasonable consumer would have read these ads to mean that water coming out of the Clcan Water Machine would contain lower levels of organic chemicals and would be clean or cleaner than water put into it. Water filtered by the Clean Water Machine, however, contained a higher level of organic chemicals (and a far higher level of industrial contaminants) than typical tap water. Consumers were, therefore misled by Norelco s representations.

The claims are material. Accurate information about the fiter emission of methylene chloride would likely have affected a reasonable consumer s decision to buy the Clean Water Machine. American Home Products Corp. 98 FTC 136 , 368 (1981), afi'd 695 F.2d 681 (3d Cir. 1982).

Respondent expressly promised that the Clean Water Machine would "help remove up to 90% of organic chemicals" and would make tap water "clean" or "cleaner." The claims relate directly to the primary purpose of thc product: to remove potentially hazardous or cleaner. " 8chemicals from tap water and make it clean Consumers bought the Clean Water Machine to remove chemicals from their drinking water. They would not have bought the Machine had they known that the device, in fact, added an organic contaminant to the water. (F. 168, 176.

The consumers who bought the Clean Water Machine hoped to gain an "extra margin of safety" by filtering chemicals out of their tap Fer1rier, Car. 85 ITC :J8, 61 (1975),petition disissed 529 F. 2d 1398 (2d Cir.),cert. denied 429 U. 818(1976). (l. 169.

. . ._ 139 Initial Decision water-including chemicals they "may not even be aware of. " (F. 171- 172.) They wanted to minimize the risks of drinking chemicals in tap water. (F. 169 , 170.) Even when scientists might disagree about the hazards of a certain chemical, consumers wanted to avoid any unnecessary exposure. (F. 171.) (43) Norelco knowingly exposed consumers to the hazard that the company s advertisements promised to correct. Consumers who took advantage of the continuous-clean function and changed the filter received a higher dose of methylene chloride. (F. 139. A. Amount of Methylene Chloride Leached The Gl filters used in the Clean Water Machine add from 1.2 to 5. ppm of methylene chloride to tap water. (F. 138.) Ordinary levels of organic chemicals in water are about 1.5 to 2.6 ppm. Those organic chemicals include mostly harmless decayed vegetation that naturally occurs in the water supply but also potentially hazardous synthetic organic chemicals from industrial pollution and agricultural run-off. (F. 149-152.

Norelco s advertisements for the Clean Water Machine focused not on the presence of decaying leaves, bark or vegetation in tap water but rather on synthetic organics; "chlorine, synthetic detergents organic wastes, and trihalomethanes. " (F. 40- , 69-73.) Volatie organic chemicals in drinking water are well below 100 ppb (. 1 ppm). (F. 150.) Consumers bought the product to remove these industrial contaminants, many of which have been linked to cancer or other adverse health effects. (F. 168-179.) The Clean Water Machine added many times the amount of synthetic organics than would be typically present in consumers' tap water. Consumers did not get what they paid for, and would have been better off drinking water straight from the tap.

B. Alleged Embargo Norelco s Vice President Patrick Campbell testified that after Christmas 1982, he knew that G 1 filters would be on retail shelves far longer than what he had told Consumer Reports. He testified that sometime in 1983 he ordered that the replacement G Is that remained in the warehouse be "embargoed " that is, segregated and not distributed. (Campbell, Tr. 2017-20.

The record does not support Campbell's testimony. At trial Campbell could recall few details about the embargo, such as how he 190 FEDERAL TRADE COMMISSION DBCISIONS Initial Decision 111 F. implemented it or how he communicated it through the distribution chain. Campbell' s testimony was also disputed by every other current or former Norelco official who took the stand. (Lenahan, Tr. 160, 185- 90; Dinley, Tr. 2135-37; Gaines, Tr. 2175-76; Crowley, Tr. 2219-21; CX 159-23 to CX 159-26.

The company sold many GIs from its warehouse after the purported embargo began. Almost 90% of the 214 000 Gl filters in Ndtelco warehouse in November 1982 were sold by 1986 to consumers, trade customers and others. The inventory of Gl and G2 filters was commingled in the warehouse. (Gaines, Tr. 2017- , 2192-93; Crowley, Tr. 2219. ) (44) C. Scientific Evidence Every expert who testified stated that methylene chloride provides no benefit to humans and that a consumet would be prudent to avoid ingesting it. Respondent' s experts would not have bought the Clean Water Machine for their own use. (F. 181 , 183. Every official body that evaluated methylene chloride (Environmental Protection Agency, Consumer Product Safety Commission, International Association for Research on Cancer, the State of California) has found it to be a possible or probable human carcinogen and potentially hazardous to humans. (F. 184-200.) This fact alone, makes the representations material. 9 Simeon Management Corp. 87 FTC 1184 (1976), afI'd 579 F.2d 1137 (9th Cir. 1978). XIV. RESPONDBNT S CONDUCT Norelco learned from Consumer Reports in November 1982 that the Clean Water Machine leached significant amounts of methylene chloride. (F. 85-86.) Norelco officials met with the magazine staff before press time and told them that it had designed a new filter that leached substantially less methylene chloride and that the G 1 filters would be sold out within a short period of time. The magazine reported this in its February 1983 article and consumers relied on the article in purchasing the Machine. (F. 103-107.) Norelco allowed unsuspecting consumers to buy and use the remaining 214 000 contaminated filters that remained in its warehouse and the thousands stil available on the retail shelf and in Norelco s own NSI stores. (F. 106-114.

9 Norelco s advertising claims are false or misleading and relate tu something consumers consider important. No proof of actual injury is necessary. Material false claims are legally presumed to cause injury. lntmlational Harvester 104 FTC 949 , 1056 (1984). NORTH AMERICAN PHILIPS CORPORATION 191 139 Initial Decision N orclco continued selling G 1 filters after a cursory evaluation of the information on methylene chloride. U. S. Testing informed Norelco that laboratory studies had found methylene chloride to be carcinogenic in animals. N orelco chose to interpret this to mean that thc levels leached by the Clcan Water Machine would be safe in humans. (F. 99-102.

This interpretation was erroneous. (F. 102.) Chemicals that produce cancer in animals are considered to have the same potential in humans, absent compcllng evidence to the contrary. (Klaasscn, Tr. 2357; CX 105- 115. (45) By November 1982, Dow Chemical and the National Toxicology Program of the National Institutes of Health had releascd studies showing that methylene chloride caused cancer in laboratory animals. (F. 220-221.) Methylene chloride had been shown by this time to be genotoxic (able to cause mutations in DNA) in bactcria tests, an indication that the chemical had cancer-causing potential. (F. 222. The levels of methylene chloride leached by the Clean Water Machine exceeded SNARLs set by the EPA and thc National Academy of Sciences for safe levels of ingestion for non-cancer effects. (F. 224- 225.) The levels of methylenc chloride leached by the Clean Water Machine far exceeded thc levels allowed by the FDA in decaffeinatcd coffee. (F. 101.) Other volatic synthetic organic chemicals, compounds in the same class as methylene chloride, were known or suspected by 1982 to be toxic and carcinogenic. EP A had already announced in the Federal Register its intention to regulatc volatile organic chemicals generally and methylene chloride specifically. (F. 229.

As soon as employees in the N orelco factory had heard about the methylene chloride problem, they consulted a chemistry reference book that was on the premises and learned that methylene chloride was a suspected carcinogen. Concerned about the consumers who were using the filter and the employees who were applying methylenc chloride on the asscmbly line, they taped up and isolated in a corner of the building all 4800 finished filters that were still in the factory. (F. 96.) In a few wecks Vice President Patrick Campbcll directed that the 4800 filters, along with the 214 000 GIs in inventory, be distributed to consumers in the ordinary course of business. (F. 97.) Norelco knew by early 1983 that slow sales would likely keep G Is on the retail shelf for up to fifteen years. (Campbell, Tr. 2071-72. Norelco officials again were told in 1985 that Gl filters were stil Initial Decision 111 F.T.C.

being sold. Judith Coyle of North Penn Water Authority contacted Frank Bell of the EPA in September 1985 and told him that a Clean Water Machine filter recently purchased directly from Norelco leached substantial amounts of methylene chloride. Bell then informed Noreleo’s Health Care Division Manager Robert Gaines that G1 filters were still for sale and sought Norelco’s response. According to Bell, Norelco chose instead to ‘“‘stonewall’’ and provided no response. (F. 231-237.) Norelco took no action until it received a formal letter from EPA in January 1986. (CX 59-0.) Norelco decided to undertake the replacement program a few days before a scheduled meeting with representatives of the EPA, CPSC, FDA, and FTC. (Bell, Tr. 1283-84; F. 241.) Although NAPC sold more than 248,000 Clean Water Machines, it sent notices only to its trade customers and to 19,188 consumers. (F. 242.) [46] Rather than inform consumers that methylene chloride is potentially hazardous, the Norelco notice letter emphasized how widespread its use is, implying that ingestion of the chemical is not a cause for concern. Norelco assured owners that ‘“‘methylene chloride is found in decaffeinated coffee, spices, hairspray, paint thinner and other consumer products,”’ but did not add that the amount leached by the Clean Water Machine far exceeded the levels allowed in decaffeinated coffee and that the Food and Drug Administration had already proposed a rule banning its use in all cosmetic products, including hairspray. (F. 189-190, 240.) The letter represented that ‘there are no established EPA standards for concentration levels of methylene chloride in water,” even though the amount of methylene chloride leached by the Machine exceeded the levels that an EPA Health Advisory had recommended as safe and several other agencies considered the chemical to be a probable human carcinogen. (F. 240; F. 2465.) XV. THE ORDER The order requires respondent to cease and desist from the specific misrepresentations alleged in the complaint. The order also prohibits misrepresentations of test results and requires a reasonable basis for any performance claims for any Norelco electric-powered consumer appliance. These provisions are commonly included in deceptive advertising cases, and are well-within the Commission’s authority to ensure that the violations do not recur. FTC v. Colgate-Palmolive Co., NUKni AM lill.AN PtllLlPti L;UKl U1iAnUN l::M 139 Initial Decision 380 U.S. 374, 394- 95 (1965); Sears, Roebuck Co. v. FTC 676 F. 385 , 394-96 & n. 20 (9th Cir. 1982).

The Commission has entered orders covering many of the company's products on the basis of violations as to a single product. Litton Industries, Inc. 97 FTC 1 , 78- 80 (1981), affd 676 F.2d 364 (9th Cir. 1982); Sears, Roebuck 95 FTC 406, 515- 22 (1980), ajJd 676 F. 385 (9th Cir. 1982).

In fashioning the appropriate remedy, three factors arc relevant: thc seriousness and deliberateness of respondent' s violations, the transferability of the violation to other products, and respondent' 10past record of violations. Sears (47) Roebuck 676 F. 2d at 392. However, not all of these factors need bc present to justify a multiproduct order: "The weight to bc given to a particular factor or element will vary. The more egregious thc facts with respect to a particular clcment, the less important it is that another negative factor be present. Id.

Norelco s violations in this case were serious, deliherate and egregious. Norelco knew about the methylene chloride problem in November of 1982, and continued to sell contaminated G 1 filters unti early 1986. The product exposed consumers to a potential health hazard. American Home Products 695 F. 2d at 706-08. Consumers could not have evaluated on their own thc truthfulness or falsity of the representations. Id. The claims are "clear, direct, unqualified, and explicit " and thus more likely to mislead consumers. Sears, Roebuck 676 F.2d at 393. The size and duration of the deccptive advertising campaign were substantial. American Home Products 695 F.2d at 707 , 709; Sears, Roebuck 676 F. 2d at 394. By cxposing consumers to a potential carcinogcn, N orelco increased risks to consumers' health. The company directed the ad campaign to health-conscious consumers who did not want to subject themselves to contaminants in tap water. That Norclco exposed consumers to methylene chloride for three years knowing that the chemical had 10 In the past five years, the Commission has issued t.three complaints against North American Philips Corporation charging false and unsubstantiated performance claims for Norelco appliances. The other two cases, relating to the "Hlack Pro" shaver s claim to cure razor bumps and the Clean Air Machine s claim to remove indoor pollutants, both resulted in consent orders.Norlh. American Philips Corp. 101 FTC 359 , 363 (1983); Nvrth American Philips Corp. 107 FTC 62, 71 (1986). The Commission has taken into account as evidence of prior misconduct the existence.e of multiple consent orders with the FTC and other agencies. Jay Nor. , Inc. 91 FTC 751 , 856 & n. 33 (197R); see alsoThompson Medico. Go. Inc. 104 forc 648 , R32 n. 78 (1984). This seems to be a mislJse of the stipulation that the "agreement is for settement purposes only and does not constitute an admission by respondent that the law has been violated " and appears to be contrary to law. ITT Continental Baking Co. v. FTC 532 F. 2d 207 , 223 n. 23 (2d Cir. 1976). I have therefore disregarded th('se consent orders as evidence of past violations. 194 EDERAL TRADE COMMISSION DECISIONS Initial Decision 111 F.

bccn linked to cancer demonstrates disregard for the welfare of its customers, and a " blatant and utter disregard" for the law. Sears Roebuck 676 F. 2d at 394.

Norelco s 1986 replacement program did little to mitigate the violations. The company undertook the effort only when threatened with government action and bad publicity from Consumer Reports. Norelco notified fewer than 10% of all Clean Water Machine owners about the program. The letter Norelco sent contained inaccurate statements, tending to downplay the seriousness of the problem. The replacemcnt program was not "a good faith attempt to eliminate (deceptive representations) (48) rapidly," and does littc to tcmper the egregiousness of Norelco s conduct. American Home Products, 695 2d at 708-09.

Norelco no longer manufactures the Clean Water Machine. The advertising strategy used by Norelco in this case, misrepresenting the Clean Water Machine s chemical removal capabilities, is, however readily transferable to Norelco s other products. Sears, Roebuck, 95 FTC at 516; and 676 F.2d at 392. Therefore, a multi-product order is warranted. In Sears, Roebuclc the company s misrepresentations regarding the performance of its dishwasher resulted in an order barring false or unsubstantiated performance claims for 14 major appliances, such as trash compactors and microwave ovens. Sears Roebuck 95 FTC at 515 , 524.

The order in this case requires a reasonable basis for performance claims and prohibits misrepresentations of tests or studies for any Norelco electric-powered consumer appliance. At present, Norelco manufactures fewer than 20 such appliances, including digital thermometers, scales and blood pressure monitors; razors; coffee makers; Clean Air Machines; irons; travel kits; steamers; electric knives; hand mixers; can openers; hair dryers, curling irons and brushes; and makeup mirrors. (Dinley, Tr. 2128-29; Gaines, Tr. 2158- 59.) This product line accounts for less than 5% of NAPC' s annual sales of $4.5 billon and less than 12% of its consumer sales. (CX 188lD; Stip. 51; F. 10.

NORTH AMERICAN PHILIPS CORPORATION 139 Initial Decision CONCLUSIONS OF LAw 1. The Federal Trade Commission has jurisdiction over the advertising and sale of Norelco Clean Water Machines and their replaceable filters under Section 5 of the Federal Trade Commission Act.

2. Respondent's use of false, misleading and deceptive statements and representations, and respondent' s failure to disclose material facts, as herein found, were likely to mislead reasonable. consumers into believing that such statements and representations were true and induced them to purchase substantial quantities of Clean Water Machines and filters by reason of those mistaken beliefs. 3. The acts and practices of respondent as herein found were all to the prejudice and injury of the public and constitute unfair and deceptive acts or practices in or affecting commerce in violation of Section 5(a) of the Federal Trade Commission Act. 4. The accompanying order is necessary and appropriate under applicable legal precedent and the facts in this case. (49) ORDER It is ordered That respondent North American Philips Corporation a corporation, its successor and assigns, and its officers, representatives, agents and employees, directly or through any corporation subsidiary, division or other device, in connection with the advertising, offering for sale, sale or distribution of the Norelco Clean Water Machine, or any other appliance, device or product designed or intended for the purpose of treating water ("device ), in or affecting commerce, as "commerce" is defined in the Federal Trade Commission Act, do forthwith cease and desist from: A. Representing, directly or by implication, contrary to fact, by the " oruse of the words "helps clean helps remove helps eliminate any other words or phrases of similar import, that any such device effectively helps remove organic chemicals or any specified organic chemical from consumers' tap water, under typical water conditions; B. Representing, directly or by implication, contrary to fact, that any such device effectively provides clean or cleaner tap water, under typical water conditions;

Initial Decision 111 F.

C. Representing, directly or by implication, contrary to fact, that independent laboratory tests, or any other tests, prove that any such device effectively provides clean or cleaner water, under typical water conditions; and D. Representing, directly or by implication, contrary to fact, by the " oruse of the words "helps clean helps remove helps eliminatc any other words or phrases of similar import, that any such device effectively helps removc impurities or pollutants, or any specified impurity or pollutant, from consumers' tap water, under typical water conditions.

II.

It is further ordered That North American Philips Corporation, a Corporation, its successors and assigns, and its officers, representatives, agcnts and employees, directly or through any corporation subsidiary, division or other device, in connection with the advertising, offering for sale, sale or (50) distribution of any electric-powered consumer appliance sold under thc "Norelco" trademark, in or affecting commerce, as "commerce" is defined in the Federal Trade Commission Act, do forthwith cease and desist from misrepresenting, in any manner, directly or by implication, the contents, validity, results, conclusions, or interpretations of any test or study. It is further ordered That North American Philips Corporation, a corporation, its succcssors and assigns, and its officers, representatives, agcnts and employees, directly or through any corporation subsidiary, division or other device, in connection with the advertising, offering for sale, sale or distribution of any electric-powered consumer appliance sold under the "Norclco " trademark, in or affecting commerce, as "commerce" is defined in the Federal Trade Commission Act, do forthwith cease and desist from making, directly or hy implication, any performance-related representation for or about such product unless, at the time the representation is made, respondent posscsses and relies upon a reasonable basis, consisting of competent and reliable evidence, that substantiates the representation. For purposes of this provision, to the extent evidence consists of scientific or professional tests, analyses, research. studies or anv other NU1UH AMt;ltlCAN PHILIPS CUHPORATION 139 Initial Dceision evidence based on expertise of professionals in the relevant area, such evidence shall be "reliable and competent" only if those tests analyses, research, studies, or other evidence are conducted and evaluated in an objective manner by persons qualified to do so, using procedures generally accepted in the profession or science to yield accurate and reliable results.

IV.

It is further ordered That for three (3) years from the date that the representations to which they pertain arc last disseminated, respondent shall maintain and upon request make available to the Federal Trade Commission or its staff for inspection and copying: A. All materials relied upon to substantiate any claim or representation covered by this order; and B. All test reports, studies, surveys or other materials in its possession or control or of which it has knowledge that contradict qualify or call into question such representation or the basis upon which (51) respondent relied for such representation, including complaints from consumers.

It is further ordered That respondent shall forthwith distribute a copy of this order to each of its operating divisions and to each of its officers, agents, reprcscntatives or employees engagcd in the preparation and placement of advertisements or other such sales materials covered by this order.

VI.

It is further ordered That respondent shall notify the Commission at least thirty (30) days prior to any proposed change in the corporation such as a dissolution, assignment or sale resulting in the emergence of a successor corporation, thc creation or dissolution of subsidiaries or any other change in the corporation which may affect compliance obligations under this order.

Final Order 111 F.

VII.

It is further ordered That respondent shall, within sixty (60) days after service of this order upon it, and at such other times as the Commission may require, file with the Commission a report, in writing, setting forth in detail the manner and form in which, it has complied with this order.

FINAL ORDBR The Administrative Law Judge filed his Initial Decision in this matter on August 29, 1988, finding that the respondent engaged in unfair and deceptive acts or practices in or affecting commerce in violation of Section 5(a) of the Federal Trade Commission Act, 5 C. 45. An appropriate Order to remedy the violations was appended to the Initial Decision. ' Service of the Initial Decision was completed on September 22 1988. Neither respondent nor complaint counsel filed an appeal. The Commission having determined that this matter should not be placed on its docket for review, and that the Initial Decision and the Order therein shall become effective as provided in Section 3. 51(a) of the Commission s Rules of Practice, 16 CFR 3.51(a), It is ordered That the Initial Decision and the Order therein shall become the Final Order and Opinion of the Commission on the date of issuance of this Order.

lUWA GHAyn;/t UF THE AMERICAN PHYSICAL THERAPY ASSOC. 199 199 Complaint