Tobacco Documents Online

October 13, 1982 TO: Mr. J. W. Johnston Mr. M. L. Orlowsky FROM: L.W. Hall, Jr. SUBJECT: PRICE ELASTICITIES Attached are three documents relatlng to price elasticitles for your review. I have asked Jerry Moore to arrange a meeting for the four of us to discuss the implications of this information sometim~ next week. LWH:Is Attachment cc: ~T/. J. R. Moore

C,,O N F I D E N T I A L October 6, 1982 TO~ Mr. L, N. Hall, Jr.~/ FROH~ H~. D, S. Burro~a RE: SBEE HODELS OF P~ICB 8~NSlTIVITY BY ACS~SEX BACi~G1LOUND .... In 1981, the National Bureau of Economic Research (NBSR) published C~o medals based on cons,-~ers' responses to government smokin~ surVeys. Because the NBER models ~ere cross-sectional (taatin~ differences between people rather then over time) they were able to relate price separately to incidence and rata per day, by age and sex. The NBER elasticities may not reflec~ the exact effects of price increases, since their models didn't deal with chan~as over t4me. Howvar, the effects they found for other variables (race, income, ~orking women, ate.) are highly consistent with our understanding of market dynamics. Thus, it is likely that the NBER models have correctly identified relativ..__~_e price sensitivity amon~ age/sex groups. '. detailed critique of the NBSR models is Attac~ent B. SU~¥ OF FINDINGS According to the NBEE models: • Teenagers and younger adult males are highly price sensitive. s Males over 35 have above average price sensitivity. • Women and 26-34 year old men era/relatively immune to price. • a Price effects incidence; r~te per day is virtually unchanged. TEENS 12-17 NBSR PRICE ELASTICITY AHONG: AGES 20-25 AGES 35"4- TOTAL 14ALES TOTAL NALES TOTAL 2O4- EFFECT ON: Incidence -1.19 - .74 -1,28 NA l~ - ,26 Cons,mption -1.44 - ,89 . -1o40 - ,45 - ,66 - ,42 t~ DSS/ch - 10/6/82 Code: 5.21

CO~CLUSIO~,S important, llalf ($OZ) o! the~otel drop £e £nduJtry voltme 18 atLrlbuLtble to (~lculatad In Attac~ent A) long ter~, drying up ~ ~upply st new ~r= ¢o replace ~ old.• ~t= 18 not a fixed 1o88 co the lnd~try: 1t8 ~rtJnce Increases ~th time. In group'8 cons~pClon by more t~n 5OZ. Diane S, Burrows HAIt~TING DEVELOPMENT D~P~RTHE~T DSB/ch *As discussed in ny 9121/82 aerie re '"Estimated Change in Industry Trend Following F.E.T. Increase". DSB/ch - 10/6/82 Code: $.Z~

,IMP.OR, TANCB OF PRICE I)~ACT B.Y ACE/SEX TO TOTAL INDUSTRY IMPORTANCE - X oF aATE P~B ~ 0F TOTAL SMOKER._._~S DAY CONSUMPTION TeenaEere 12-17 4,0 B 17o4 B 2,2 g Melee 18-24 8.0 29.7 7.5 Males 35+ 2807 36.9 33.6 Total 18+ 96.0 31.7 97.8 TOTAL 100.0 31.1 E 100.0 Source: Tracker (Z a~okera first half 1982, rate in year 1981) adjusted for estimated teenage maoking. 1I. IMPOItTANCg TO P]tlCB IMPACT: Sample calculation aeneas IOZ price increase and flat 623.0 b~llion £ndustry. Importance of group to total change is lnde~end, ent of total voluae or price chanEe. IMPORTANCE NBER LOSS FOLLOWING TO JNDU,STRY S aS- lOX e ICB ISCU S BILLIONS TICITY BILLIONS IMPORTA~CB Teenagers 12-17 2.2Z 13.7 -1.44 2.0 7.21 ~le8 18-24 7.5 46.7 -1.40 6.5 23.6 ~1e8 3~ 53.6 209.3 "/ - .66 13.8 50.0 Total ~8+ 97,8 609.3 - .42 25.6 92.8 ~L 100.0 623.0 - .44 g 27.6 100.0 DSB/eh - 10/6/82 Code: 5.21 -3-

";" ATTACHMENT B TECHNICAL SUPPLEMENT AND C[tlTIqUE OF THE NBER MODELS Lewit and Coats of the National Bureau of Economic Research developed two aeries of models relating ctsaratte price• (and other factor•) to reported incidence, rate per day, and "total demand" (incidence times rate). Although reported rat•s understate true consm|ption, the model• would •till be valid if the under-reportinS was constant across the other variables -- geography, age. sex, etc. We ordinarily use this assu~ption our•elves. Both studies ware weakened by essu~ln~ that prices in ell locales/outlets in a state were similar to the star•wide average reported by TNA. plus local tax. This may be why both models had low adjusted R-squared value• (.11 or less, but still significant). Other aspects of the HEER studies ere crlt!qund separately below, since each used a different data source, Clue perlod, end nethodology. THE OVER-20 MODELS SOURC_.__.~E: "The Potential for Usin~ Excise Taxes to Reduce S~okin~," Working Paper No. 764 of the National Bureau of Economic Research, by E. M. Lewit and and D. Coats, September. 1981. This ordinary least, squares model used date topos of Individual re•pences to Lhe Health Intere£ew Survey of 1976. It expre•and ,incidence, rate, and "total demand" of respondents over a~e 20 ae a linear function of: • Retail price, defined as the T~A- reported average in the respondent'8 state, adjusted for any 1oc81 taxes. • Family lnco~e end siae • Sex • Marital Status / • Health status (perceived) • Region and city size • • Race • Working woman or not Using the full sample, price was not found to be a statistically significant factor in incidence, rate, or "total ds~and'. Coefficients for east of the other variables ware significant (at the 5Z level) in s direction consistent with our consumer research findings. DSB/ch- 10/6/82 Code: .5.21

respondents within 20 m1108 of 8 lo~r priced etnto, on the premise the~ the price they actuaZly paid might be lower than the assigned prior ("bards: effect"). For the restricted sample, pr£oe was found to ~ 8 8La~ls~lc~y • Incidence end "torn1 demand" mong 20-25 year olde. • "Total demand" among those over 35, When the rngresstone ~ere done by age/sex, stlnlflcnnt coef~£clents vere found only for males, Thongh none were found in the ~enalo reKreestons, the higher elnstlctt~nales/fenales combined than for melee, alone uy iaply some effect among females. Selected elasticities are tabulated on the next pass. Only boxed values have 8tat£st/cal etgn£glcance at normal levels. DSB/ch - 1016182 ~de: 5.21

(S~mple ltesLrLcLed to EILu£neLe Border REPO~T~D llEPO~TZD "TO~L DEMAND" ¢ot~X 2o+ ~ - .0~7 I~ .~6.* I 20-25 26-35 - .4~ ,0~ - 3.,~- - Melee 20+ 26-35 - .292 + .029 - .320 35+ - .246 - .20~ I..658. I Females 20+ 20-25 - .136 - .026 - .302 26-35 - .388 - .134 - .577 35+ + .066 - .077 - .118 IHCOHF, Total 20+ + .o3 ~ I- .08-* I *OLS coeff$clent statistically Significant at 5~ level (2-tailed **OLS coeff£c£ent statistically elgn£flcant at 1~ level (2-exiled test), DSB/ch - 10/6/82 Code: 5.21 -6-

Since the NBER models ere besed on Keolrephic differences in retell price st one point in time (1976), they do ~ot trenslste directly to national changes over time: The positive income elasticity eu~Keste that, over time, price elasticity should be epplled to retail price deflated by some measure of coneomer income. However, the price elasticities ere stro~ly Since consumer prlces may contlnue to rise fester than Income (especially amo~ youn~ adults), using the CPI •s • deflator probably yields conservative estimates of the price impact. Restricting the sanple to eliminate "border effects" improved the model's applicability to the nation. Ho~ever, It crested some regional bias -- Westerners rose from 1KZ to 29X of the sample while Hortheasterners fell from 24Z to 11~. This may have made the price elasticity :ore negative than the "true" uattoual value, since positive coefficients were associated tr~th residence in any region except the West. Thus, the adult NBER elesttclties ere /mpreclse for national price chsn~es. However, they probably Klvea reasonable 1des of the groups most affected by price and the orders of magnitude of the effects. TEENAGE HODEL SOURCE: "The Effects of Government Regulation on Teenage Seeking," by K. M. Lewit, Do Coats, and He Groesaau° ~ournal of Law and Economics, ~cumber, 1981. These mod~ls used personal Interview data from Cycle III of the Health Examination Survey~, conducted 1966-70. Incidence and rate findings froaH~S III seem reasonable, perhaps even 8 cut above most teensse studies. Sut it is not recent. A major goal of this study yes pre/post assessment of the Fairness Doctrine, which loaded television with anti-smoking ~ommerctals from 7/67 through 12/70. Other variables Included: / *In Self-Reported Health Behavior end AttXtudes of Youths 12-17 SLates. HEW, 1975. DaB/oh - 1016152 Code: 5.21 Years, United - 7 -

• T14~ retell prices, adjusted to state coet-of-llvLng end changes in the national CPI during the 4-year survey (an improvement over no adjustment, but •till • Fairness Doctrine intensity versus cigarette udvertteing and ~ount of Lime 8~nC ~8~ch~ng ~ • F~y ~ncone def~a~ud by • N~ber of other children • Eap~o~en~/~o~anco of ~uch res~en~ (detailed) • SLuden~ status ~e Sex ~ce • ~giou ~d city •iae Large, ~ga~Xve prX~e eXas~£e£t£e, a~ s~a~is~Xcally 8£gn£flcan~ £n all a~*l s~c~lca~£ons repor~ed. The authors suggest that these lar ~tims me. ~nc_o.r orate o~her effects which the model did not include ~eLy. T~oy suggest tnet the Income related variables may have failed to represent the youth's actual • diacretione:y income. NBER PRIC~ ELASTICITIES ANONG YOUTH 12-17 REPORTgD R~PORTED INCIDgNCZ '~aTE "TUT~L DEMAND" (l C. x Total 12-17 - 1.19 - 1,44 The Fairness Doctrine we8 also found to have an important negetive effect during its first year (a~counting for • 3 percentage point drop in 12-17 incidence), but showed diminishing returns during Ice second and third year. Cigarette advertising on TV was found to partially offset the'Doctrine effects. Thus, the absence of TV advertising after 1970 was implicitly • negative effect, but the absence of the Fairness we8 e larger implicit positive. DSB/ch - 10/6/82 Code= $,2!

Results £o~ othe: variables ~'e~e not repo:Led. • Loose es~1uaclon of price, loosely deflated. • "Xupure" price e~ss~c~y (probsbly somevh8~ ~nfl~Led). • Ds~a ~r~ ~he la~e 1960'8, which ~y no~ re~lec~ ~oday's ~rke~. pr*ce c~nge 10 even ~rse. Code: 5.21