Sandy Slater

Senior Research Specialist

sslater@uic.edu

www.impacteen.org

Why is Tobacco Price Manipulation a

Problem? Point-of-Sale Marketing

and Cigarette Smoking

The National Conference on Tobacco or Health

Phoenix, AZ June 12, 2009

Overview

• Brief review of recent trends in cigarette company marketing expenditures and impact on price

• Brief review of literature on impact of cigarette marketing on youth and adult cigarette smoking

• Effects of POS cigarette marketing on youth smoking uptake – Focus on recent Archives of Pediatrics & Adolescent

Medicine paper by Slater et al. (2007)

Source: Federal Trade Commission, 2005, and author’s calculations

Cigarette Marketing Expenditures

Inflation Adjusted, 1975-2005

$0.00

$2.00

$4.00

$6.00

$8.00

$10.00

$12.00

$14.00

$16.00

$18.00

1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 2005

Year

Billio

n D

ollars (

8/0

7 $

$s)

Marketing Activities

• Image oriented – Newspaper, magazine, outdoor, transit, and point-of-sale

advertising; company websites and other Internet

• Price related – Price discounts, coupons, free samples, multi-pack deals

• most at POS

• Promotional Allowances – Retail and wholesale level; other

• Merchandise – Specialty item distribution (branded and unbranded); retail

value added (gifts with purchase)

• Some at POS

• Other – Sponsorships, direct mail and telephone

Source: Federal Trade Commission, 2005 and author’s groupings

Source: Federal Trade Commission, 2007, and author’s calculations

2005 Cigarette Marketing Expenditures by Category

2%

87%

2%

6%3%

image price merchandise promotional allowances other

Shift in Marketing Activities

• Rise in marketing spending in unconstrained venues following the MSA

– Significant increases in point-of-sale marketing following the implementation of the 1999 ban on billboard advertising (Wakefield, et al., 2002)

– Consistent with empirical evidence on impact of marketing restrictions on smoking behavior (Saffer and Chaloupka, 2000)

• Shift over time to price-related marketing activities

– Appears to follow release of the first economic studies on the impact of price on youth and adult smoking behavior in early 1980s (Chaloupka, 2005)

– Accelerates following the MSA

Source: Federal Trade Commission, 2007, and author’s calculations

Cigarette Marketing Expenditures per Pack

Inflation Adjusted, 1975-2005

$0.00

$0.10

$0.20

$0.30

$0.40

$0.50

$0.60

$0.70

$0.80

$0.90

1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 2005

Year

Do

llars (

Au

gu

st

Non-Price Other Price

Price-Related Cigarette Marketing and

Tobacco Control

• Evidence from internal documents that price-related

marketing used to soften impact of tax increases

(Chaloupka et al., 2002; Chaloupka et al., 1998)

•Greater price-related marketing since the MSA (Ruel,

et al., 2004; Loomis et al., 2006; FTC, 2007)

•More price-related marketing in states with greater

spending on comprehensive tobacco control programs

(Loomis, et al., 2006; Slater et al., 2001)

Tobacco Industry Efforts to Offset Impact of Tax Increases

• Federal Tax Increase: 61.66 cents per pack,

April 1, 2009

• Philip Morris Price Increase: 71 cents per

pack on “growth and support brands” and 78

cents per pack on “non-support brands”

• weeks before FET increase went into effect

• followed by Reynolds and Lorillard

Tobacco Industry Efforts to Offset Impact of Tax Increases

Philip Morris, 2009

On February 4th, 2009, the Federal Government enacted legislation to fund

the expansion of the State Children's Health Insurance Program (SCHIP)

that increases excise taxes on cigarettes by 158%.

As a result, you will see the price of all cigarettes, including ours, increase

in retail stores.

We know times are tough, so we'd like to help. We invite you to register

at Marlboro.com to become eligible for cigarette coupons and special offers

using this code: MAR1558

Thank You, Philip Morris USA

Price-Related Cigarette Marketing and

Tobacco Control

•More marketing in stores that are more frequented by

youth than in other stores (Henriksen, et al., 2004)

•Growing use of point-of-sale ads to highlight sales

promotions (e.g. special price, special offer, cents off,

reduced price, multi-pack special) (Feighery et al.,

2008)

•Percentage of stores in California with at least one

ad featuring sales promotion rose from 68% in

2002 to 79% in 2005

•More rapid increases in stores located in

neighborhoods with higher than average

proportion of African-Americans

Source: Tax Burden on Tobacco, 2009, and author’s calculations

Cigarette Taxes and Prices, 1976-2008Inflation Adjusted (Feb. 2009 dollars)

$0.00

$0.50

$1.00

$1.50

$2.00

$2.50

$3.00

$3.50

$4.00

$4.50

1955 1957 1959 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007

Year

Do

llars p

er P

ack

State Tax Federal Tax MSA Costs Net Price

Sources: Tax Burden on Tobacco, 2007, FTC, 2007, and author’s calculations

Average Cigarette Prices, 1975-2005

Inflation Adjusted

$0.00

$0.50

$1.00

$1.50

$2.00

$2.50

$3.00

$3.50

$4.00

1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 2005

Year

Do

llars

per

Pack (

Au

gu

st

2007)

Price Price-Price Related Marketing

Cigarette Prices and Smoking

• Higher cigarette prices induce quitting, prevent

relapse, reduce consumption and prevent initiation

• marketing efforts that lower price will offset impact of taxes

and other factors that raise cigarette prices

• 10 percent price rise reduces overall cigarette

consumption by 4 percent • Half of impact on smoking prevalence

•10% price rise leads 10-12% to try and quit; about 2%

successful

• Half of impact on consumption among continuing smokers

• Smoking in low-income populations more sensitive to price

Source: Chaloupka et al., 2000; Chaloupka, in press

Source: Tax Burden on Tobacco, 2007, and author’s calculations

Cigarette Prices and Cigarette Sales

United States, 1970-2008

16500

18500

20500

22500

24500

26500

28500

30500

1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006

Year

Sale

s (

mil

lio

n

packs)

$1.25

$1.75

$2.25

$2.75

$3.25

$3.75

$4.25

Pric

e (

2/0

9 D

oll

ars)

Sales Price

Source: NHIS, Tax Burden on Tobacco, 2009, and author’s calculations

Note: green data points for prevalence are interpolated assuming linear trend

Cigarette Prices and Adult Smoking Prevalence,

United States, 1970-2008

19

23

27

31

35

1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006

Year

Prevale

nce

$1.25

$1.75

$2.25

$2.75

$3.25

$3.75

$4.25

Pric

e (

2/0

9 d

oll

ars)

Prevalence Price

Source: BRFSS, Tax Burden on Tobacco, 2009, and author’s calculations

Cigarette Prices and Adult Prevalence,

50 States & DC, 2007

y = -1.7038x + 27.473

R2 = 0.1756

11

13

15

17

19

21

23

25

27

29

$3.25 $3.75 $4.25 $4.75 $5.25 $5.75 $6.25

Price per Pack

Ad

ult

Pre

vale

nce

Source: BRFSS, Tax Burden on Tobacco, 2006, and author’s calculations

Cigarette Prices and Percentage of Ever Smokers Who Have Quit

Smoking

y = 0.0167x + 0.2478

R2 = 0.1276

18.0%

22.0%

26.0%

30.0%

34.0%

38.0%

42.0%

$3.00 $3.50 $4.00 $4.50 $5.00 $5.50

Price (dollars per pack)

Pe

rc

en

t Q

uit

ters

Young Peopl e More Responsi ve t o Pr i ce

Increases

•Proportion of disposable income youth spends on cigarettes likely

to exceed that for adults

•Peer influences much more important for young smokers than for

adult smokers •recent estimates indicate about 1/3 of overall impact of price on youth

accounted for by indirect impact through peers

•Young smokers less addicted than adult smokers

•Young people tend to discount the future more heavily than adults

•Other spillover effects •for example, through parental smoking

Source: Liang, et al., 2003; Chaloupka 2003

Cigarette Prices And Youth

• A 10% increase in price reduces smoking prevalence

among youth by nearly 7%

• A 10% increase in price reduces average cigarette

consumption among young smokers by over 6%

• Higher cigarette prices significantly reduce teens’

probability of becoming daily, addicted smokers; prevent

moving to later stages of uptake.

• 10% price increase reduces probability of any initiation by

about 3%, but reduces probability of daily smoking by nearly

9% and reduces probability of heavy daily smoking by over

10%

Sources: Chaloupka and Grossman, 1996; Tauras, et al., 2001; Ross, et al., 2001

Source: NSDUH, Tax Burden on Tobacco, 2009, and author’s calculations

Cigarette Prices and 12-17 Year Old Smoking

Prevalence Rates, 50 States & DC, 2005/06

y = -0.5185x + 13.443

R2 = 0.0452

7

8

9

10

11

12

13

14

15

16

$3.00 $3.50 $4.00 $4.50 $5.00 $5.50 $6.00

Price per Pack

Prevale

nce

Source: MTF, Tax Burden on Tobacco, 2009, and author’s calculations

Cigarette Price and Youth Smoking Prevalence,

United States, 1991-2008

$2.25

$3.00

$3.75

$4.50

129.9 124.1 121 116.9 110.8 99.2 91.3 84.5 78.4

Year

Pric

e p

er p

ack (

2/0

9

do

llars)

6

11

16

21

26

31

36

Sm

okin

g P

revale

nce

Cigarette Price 12th grade prevalence 10th grade prevalence 8th grade prevalence

POS Cigarette Marketing and Youth

Smoking Uptake

• Bridging the Gap •Funded by Robert Wood Johnson Foundation since late-1997

•Focus on adolescent tobacco, alcohol, and illicit drug use; more

recently on physical activity, diet, and obesity

•ImpacTeen project (UIC) •collected observational data on community level cigarette

marketing at the point-of-sale from 1999 through 2003

• detailed state tobacco control policy data

•Much more

•Youth, Education and Society Project (U. of MI, ISR) •Builds upon Monitoring the Future study funded by the National

Institute on Drug Abuse and conducted by Lloyd Johnston and

colleagues

•Focuses on school policies, programs, and other influences on

youth tobacco use

POS Cigarette Marketing and Youth

Smoking Uptake

• Design

• observational data collection at the point-of-sale in communities

around schools participating in the Monitoring the Future survey

• approximately 200 schools per year

• about equally divided between 8th, 10th, and 12th grade

schools

• census of retail outlets selling tobacco in most communities

• random sample of 30 in larger communities

• identified from business lists, verification calls, and on-site

• average of 18.1 stores per community

• 17,476 stores observed 1999-2003

POS Cigarette Marketing and Youth

Smoking Uptake

• Collected data on variety of cigarette marketing at the

point-of-sale • in-store, exterior, and parking lot measures of advertising

• low-height advertising and functional objects

• Promotions: cents-off specials, on-pack coupons, multi-pack

discounts, gifts with purchase (Marlboro and Newport)

•Cigarette prices (Marlboro, Newport, and lowest price)

•Placement (self-service vs. clerk assisted only)

•Measures used in analyses reflect the proportion of

stores in a given community with different types of

marketing • price is average price of premium brands

POS Marketing and Youth Uptake

• Marketing Measures:

Advertising Scale (0-5) 2.56

No self-service placement 83%

Any vs. no promotions 48%

Premium price $3.62

POS Marketing and Youth Uptake

• Outcome:

•Youth smoking uptake – 6 levels based on past and

current smoking behavior and future intentions to smoke

•Validated with longitudinal MTF data •26,301 students

Stage % of youth

Never smoker 53.7%

Puffer 20.7%

Nonrecent Experimenter 4.1%

Former Established Smoker 3.1%

Recent Experimenter 6.9%

Current Established Smoker 11.5%

POS Marketing and Youth Uptake

• Methods:

• Generalized ordered logit model •Allows impact of explanatory variables to have different

impact on movement between levels of smoking uptake

• Account for complex survey design • sampling weights; clustering (school-level) adjustment

•Variety of individual-level control variables • gender, race/ethnicity, grade, student’s income, parents’

education, live with both parents

•State tobacco control policies

• smoke-free air index

• youth access index

• purchase-use-possession index

•Other variables • year, urban/suburban/rural

POS Marketing and Youth Uptake

• Findings:

• Significant impact of advertising on early stages of

uptake (from non-smoker to puffer)

• effect declines as move to later stages of uptake

• statistical significance declines as move to later stages

• Simulations look at impact of different levels of advertising

on stages of uptake

•If all stores had no advertising, estimate that prevalence

of never smoking would rise by nearly 9%

Advertising and Youth Smoking Uptake Simulations

0

0.1

0.2

0.3

0.4

0.5

0.6

Never Smoker Puffer Non-recent Experimenter Former Established Recent Experimenter Current Established

Percen

t o

f Y

ou

th

all ads actual no ads

POS Marketing and Youth Uptake

• Findings:

• Significant impact of promotions on later stages of

uptake

• effect rises as move to later stages of uptake

• statistical significance increases as move to later stages

•Simulations look at impact of different levels of advertising

on stages of uptake

•If all stores had no promotions, estimate that prevalence

of current established smoking would fall by over 13%

Promotions and Youth Smoking Uptake Simulations

0.00%

1.00%

2.00%

3.00%

4.00%

5.00%

6.00%

7.00%

8.00%

9.00%

10.00%

Current Established

Percen

t o

f Y

ou

th

all promos actual no promos

POS Marketing and Youth Uptake

• Findings:

• Youth smoking uptake negatively associated with

higher cigarette prices • small, statistically insignificant impact on transition from

never smoking to puffing

• large, statistically significant and consistent impact on

transitions between other stages of uptake

• $1.00 increase in price would reduce the odds of

moving from one stage to next by 24%

•Consistent with other recent evidence on price and

youth smoking uptake

• Weak association between self-service only

placement and youth smoking uptake • not statistically significant; somewhat stronger for later

stages of uptake

POS Marketing and Youth Uptake

• Limitations:

• Cross-sectional data limits ability to assess causal

impact of POS marketing on youth smoking uptake

• Relatively crude measures of cigarette marketing

•Inability to match store-specific data to youth based

on stores they frequent most

POS Marketing and Youth Smoking

• Recent analysis by Feighery and her colleagues (2006)

• data on POS marketing collected observationally from 53 stores

located near 3 California middle schools

• branded signs

• functional objects

• shelving units & product displays

• shelf-space for specific brands

• Based on observation data and youth self-reports of shopping

behavior, four measures of advertising exposure constructed:

• shopping frequency in stores with more cigarette advertising

• shopping frequency in stores that sell cigarettes

• exposure to brand impressions in stores where students shop

• self-reported exposure to cigarette advertising

POS Marketing and Youth Smoking

• Recent analysis by Feighery and her colleagues (2006)

• youth smoking behavior:

• ever smoking

• susceptibility to smoking

• Key findings:

• ever smoking and susceptibility to smoking positively and

significantly associated with alternative measures of advertising

exposure in all but one of the models estimated

• estimate that youth who are highly exposed to marketing are 2-

3 times more likely to have ever smoked than youth with low

exposure to cigarette marketing

• Similar limitations

Summary

• Cigarette marketing expenditures have increased sharply

since the MSA • some recent declines, but per-pack amount more than double

spending prior to the MSA

• Cigarette marketing increasingly dominated by spending

on price-reducing promotions

• Higher cigarette prices encourage smokers to quit

smoking, prevent former smokers and youth from starting,

and reduce consumption among continuing smokers • increases in price-lowering promotions offsets the impact of higher

cigarette taxes on youth and adult smoking

• Youth smoking uptake associated with point-of-sale

cigarette marketing • advertising has greatest impact on early stages

• price and price-promotions have greater impact on later stages

Sandy Slater

sslater@uic.edu

www.impacteen.org