The Value of Smoking Prevention
An Estimate of the Dollar Value of Preventing One
Otherwise-Smoker from Starting

Prepared by
Patricia Herman, ND, PhD & Michele Walsh, PhD
Evaluation, Research and Development Unit
The University of Arizona
Funded by the
Arizona Department of Health Services

Evaluation, Research and Development Unit
The University of Arizona
PO Box 210462
Tucson, AZ 85721-0462
Phone: (520) 318-7259
Fax: (520) 318-7252
www.evalrdu.org

This Report Was Prepared By
Patricia Herman, ND, PhD & Michele Walsh, PhD
The Evaluation, Research and Development Unit
University of Arizona

Special Thanks and Appreciation are Extended to
Frank A. Sloan, Jan Ostermann, Gabriel Picone, Christopher Conover, and Donald
H. Taylor, Jr. the authors of The Price of Smoking, the source of our baseline cost
estimates

Table of Contents
Executive Summary ........................................................................................................................ 1
Introduction..................................................................................................................................... 2
Baseline Estimate of the Cost of Smoking ..................................................................................... 3
Modifications to the Baseline Estimate for Arizona Youth............................................................ 6
State and federal excise taxes ..................................................................................................... 6
Adjustment to Arizona average wage rates ................................................................................ 7
More recent worker productivity estimates ................................................................................ 7
Adjustment to life insurance cross subsidization estimates........................................................ 7
Spouse reduction in wages and medical costs due to disability from secondhand smoke
exposure ...................................................................................................................................... 8
A relook at the literature regarding mortality and medical costs for children due to secondhand
smoke exposure........................................................................................................................... 8
Discount the costs to average age for prevention services and add consumption costs from start
of smoking to 24 years of age ..................................................................................................... 8
Portion of these costs that directly impact the state budget ........................................................ 9
Best available estimate of the cost of smoking (and value of prevention) for Arizona youth.. 10
How These Estimates Can Be Used and Their Limitations.......................................................... 12
The perspective of the analysis determines which estimates to use ......................................... 12
Limitations to these estimates................................................................................................... 12
Policy Implications ................................................................................................................... 13

List of Tables
Table 1.
Lifetime cost of smoking for a 24-year-old smoker (male and female weighted
average)……………....................................................................................................................... 4
Table 2.
Lifetime cost of smoking for a 10-year-old future smoker (male and female
weighted average) ......................................................................................................................... 11
List of Figures
Figure 1. Estimates of the probability of a 24-year-old smoker smoking in younger years and
average amounts smoked ................................................................................................................ 9

The Value of Smoking Prevention

Executive Summary
Programs to prevent youth smoking are considered beneficial. However, little has been
documented on the extent of these benefits, who benefits, and what these programs are worth.
This report takes advantage of an excellent and detailed national accounting of the cost of
smoking to develop a set of estimates of the net benefits of preventing one Arizona adolescent
from smoking. These figures can then be used to help justify and better target youth smoking
prevention programs.
The baseline national estimates of the cost of smoking are modified to be appropriate for use
valuing prevention efforts targeting youth of about 10 years of age. They are also modified by
adjusting the estimate to reflect 2009 dollars, to include existing federal and state tobacco excise
taxes, and to include better and more recent data on the effect of smoking on worker productivity
and on the health and mortality effects of secondhand smoke exposure.
Because what is considered a cost differs depending on who is asking the question, when talking
about value, costs, and cost savings, the perspective (or point of view) required for each situation
must be acknowledged. Below are the estimates appropriate for use from several possible
perspectives:
•

The perspective of society as a whole. Preventing one 10-year-old who would have otherwise
become a typical smoker from becoming that smoker is worth about $149,000 to Arizonans
as a whole, ignoring for now who is benefiting.

•

The perspective of the portion of society other than the smoker and his or her family. The
vast majority of the costs to society as a whole accrue to the smoker his- or herself (about
$128,000), and the next largest portion of costs accrue to the smoker’s immediate family
($23,000), mostly due to secondhand smoke exposure. Since it is more difficult to justify
dollars to prevent someone from harming him- or herself, the smoker’s costs may not be
appropriate to include as a value to prevention. Whether smokers’ families should be
included within the realm of personal choice or be considered victims is up for debate.
Depending on whether costs to smokers’ families are included or not, preventing one 10year-old who would have otherwise become a typical smoker from becoming that smoker is a
cost of over $2,000 to all Arizonans except the smoker and his or her family, or a gain of
about $21,200 to all Arizonans except the prevented smoker him-or herself.

•

The perspective of Arizona’s state budget. This perspective only includes dollars that would
flow into and out of state coffers. From this perspective, preventing one 10-year-old who
would have otherwise become a typical smoker from becoming that smoker costs the
Arizona state budget about $6,000 over that smoker’s lifetime. Note that this amount, like all
other estimates presented in this report, is in addition to the cost of the prevention program
itself.

When using these estimates it is important to remember that they are just that: estimates. They
are the best available estimates of the value of smoking prevention, but they are not without
limitations. These limitations should be noted whenever these estimates are used.
These estimates have policy implications for how tobacco tax revenues are used. To avoid
disincentivising tobacco interventions, and in the interest of equity, there is a strong rational for
using these dollars towards tobacco control, which, if done effectively, may phase out the
income stream over time, but also, unlike other uses, reduces the need for the income stream.
Evaluation, Research and Development Unit, The University of Arizona

1

The Value of Smoking Prevention

Introduction
Most people would agree that programs to prevent youth smoking are beneficial. However, it is
much more difficult to say the extent of the benefits, who benefits, and what these programs are
worth. This report presents the development of a set of estimates of the net benefits of preventing
one Arizona adolescent from smoking. These figures can then be used to help justify and better
target youth smoking prevention programs.
This report is organized as follows. The next section presents a set of estimates of the cost of
smoking that will be used as the baseline from which we will then generate estimates more
appropriate to use in valuing smoking prevention in Arizona youth. The third section presents the
modifications to the baseline estimates needed to generate the best available estimate of the value
of smoking prevention programs in Arizona today. These modifications are limited to those
possible using existing data. The fourth section presents information on how the final estimates
could be used, and their limitations.

Evaluation, Research and Development Unit, The University of Arizona

2

The Value of Smoking Prevention

Baseline Estimate of the Cost of Smoking
This analysis takes advantage of an excellent and detailed accounting of the cost of smoking by
Frank Sloan and colleagues in their 2004 book entitled The Price of Smoking.1 We will use these
estimates of the cost of a 24-year-old smoker as a baseline for the estimate of the value of
smoking prevention in Arizona youth.
This work was chosen as the source for the cost estimates used in this report for several reasons.
First, dozens of estimates of the cost of smoking have been published. Although we believe that
the Sloan et al. estimates are some of the most rigorous, any estimate of these costs is just that,
an estimate. Given the imprecise nature of the estimation process, advantages of using the Sloan
et al. estimates include that they are based on nationally representative samples and that their
derivation is well-documented in a 300+ page book. Second, Sloan and colleagues compare
smokers’ costs to that of “non-smoking smokers” rather than to never or former smokers. This
allows a focus on the costs of smoking itself, and separates out any costs associated with lifestyle
risk factors which are often correlated with smoking. Third, Sloan and colleagues capture more
categories of costs than other estimates. That is, they go beyond an estimate of healthcare costs
to include the cost of tobacco use to worker productivity, the smoker’s family’s health (via
secondhand smoke), life insurance premiums, pensions, and tax revenues. Fourth, Sloan and
colleague’s estimate of healthcare costs include these costs for all causes and not just for those
diseases specifically deemed “tobacco-related”—e.g., lung cancer. Fifth, Sloan and colleagues
use a lifecycle approach in estimating their costs, capturing the impacts of smokers’ increased
resource utilization during their lives and of their decreased survival. Lifecycle estimates are
more difficult to derive, but they have the advantage of creating a direct conceptual link between
smoking initiation and its long term costs (see for example, Miller et al., 19972). This link is
especially important when looking at the benefits of programs focused on reducing initiation.
Table 1 presents the Sloan et al. estimates for the lifetime cost of smoking for a 24-year-old
smoker adjusted from year-2000 dollars (as published in their book) to year-2009 dollars using
the consumer price index (CPI). i The CPI for medical care was used for all healthcare costs, and
the CPI for all items less medical care was used for all other cost categories (implicitly assuming
wage increases kept up with inflation during those years). The medical care multiplier to take
year-2000 dollars to year-2009 dollars is 1.440 and the multiplier for all items other than medical
care is 1.235. Negative numbers are cost savings (benefits) from smoking and positive numbers
are increased costs. Equivalently, positive numbers are the benefits from preventing one potential
smoker from smoking (or continuing to smoke) and negative numbers are the increased cost
outcome of this prevention.
As can be seen, smoking has a number of impacts other than increasing healthcare costs, and the
costs of smoking differ dramatically depending on the perspective or point of view of the
analysis. The fourth column is the sum of the first three and contains the costs of smoking from
the perspective of society as a whole—i.e., the cost of smoking when all costs are considered no
matter who incurs the cost. Based on the Sloan et al. estimates, the lifetime cost of smoking to
society for a typical 24-year-old smoker is $211,622 in year-2009 dollars. That is, if this typical

i

The numbers in Table 1 correspond to Tables 11.1, 11.2, and 11.3 on pages 252, 254, and 255, respectively, in
Sloan FA, Ostermann J, Taylor DH. The Price of Smoking. Cambridge, MA: The MIT Press. 2004.

Evaluation, Research and Development Unit, The University of Arizona

3

The Value of Smoking Prevention

smoker had not taken up smoking (or had quit smoking before his or her 24th birthday) society as
a whole would have incurred $211,622 less in costs over his or her lifetime.
Table 1. Lifetime cost of smoking for a 24-year-old smoker (male and female weighted average)

Cost component
Cost of cigarettes themselves
Federal excise taxes on tobacco
State excise taxes on tobacco
Smoker’s mortality
Smoker’s disability
Smoker’s medical care
Loss in smoker’s earnings
Lost income taxes on earnings
Work loss (sick leave/absenteeism)
Other productivity losses
SSI outlays and benefits
Private pension outlays and benefits
Life insurance outlays and benefits
Spouse mortality (SHS)
Spouse disability (SHS)
Infant deaths (SHS)
Medical care (SHS)
Totals

Smoker
$12,466
$1,880
$2,121
$107, 880
$18,052
$1,499
$27,411
$0
$0
$0
$5,406
$7,312
($9,509)
$0
$0
$0
$0
$174,519

Smoker’s
family
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
($907)
($640)
$0
$27,655
$1,290
$754
$870
$29,022

Rest of
society
$0
($1,880)
($2,121)
$0
$0
$2,973
$0
$5,482
$4,046
$1,243
($4,499)
($6,672)
$9,509
$0
$0
$0
$0
$8,080

Society as
a whole
$12,466
$0
$0
$107,880
$18,052
$4,472
$27,411
$5,482
$4,046
$1,243
$0
$0
$0
$27,655
$1,290
$754
$870
$211,622

SHS = these costs are due to secondhand smoke exposure

Three points must be made about this number to aid in clarity for the rest of this report. First, this
$211,622 represents what is actually a larger total dollar amount of costs and benefits over this
smoker’s lifetime. In fact, just over half of this total dollar amount represents the substantial cost
to the smoker of future years of lost life due to smoking. The stream of the costs and benefits of
smoking has been “present-valued” to the smoker’s 24th birthday. Because future costs and
benefits are not worth the same as costs and benefits that exist in the present, they are discounted
(using an appropriate discount rate, in this case 3%) to represent their value when he or she is 24
years of age. A second, related point is that these future costs and benefits are discounted to their
value on the smoker’s 24th birthday. Therefore they are appropriate for use regarding the value of
changing whether this individual smokes or not at that point in time. These costs will require
adjustment to be used to value smoking prevention in an individual younger than 24 years of age,
or to estimate the value of smoking cessation in an older smoker. The value to society of
preventing a younger individual (who would have otherwise taken up smoking) from smoking
will be lower due to additional discounting, and will be higher due to any additional costs
incurred by this younger smoker. The value of smoking cessation after 24 years of age will
increase due to the smoker being closer in time to large mortality costs, but will also decrease
somewhat due to the costs already incurred during a longer smoking life. Finally, this estimate of
$211,622 is based on the costs of a “typical smoker” compared to those of a “non-smoking
smoker.” The definition of a typical smoker in the Sloan et al. estimates includes a typical
smoking cessation pattern. Therefore, $211,622 can be considered to be an estimate of the

Evaluation, Research and Development Unit, The University of Arizona

4

The Value of Smoking Prevention

additional present-valued cost to society of a 24-year-old smoker living a typical smoker’s life
(and quitting at some point on his or her own) versus stopping (or never starting) smoking now.
The first column in Table 1 contains the costs of smoking to the smoker him- or herself. As can
be seen, the cost to the smoker makes up the largest component of the total cost of smoking to
society ($174,519 of the $211,622 cost to society). The biggest cost to the smoker is the loss of
years of life (mortality costs) which in this analysis was estimated as $107,880. The only cost
savings (benefits) to the smoker (other than any pleasure derived from smoking, which is
difficult to monetize and not included in the Sloan et al. analyses) are life insurance benefits
(both in terms of less paid in and more paid out). Note that this benefit has been subsidized by
everyone else paying into the life insurance policy. As noted by Sloan et al, these benefits should
decrease as more life insurance policies add surcharges for smokers.
The second column of Table 1 contains the costs of smoking to the smoker’s family. The biggest
costs here are the spouse’s increased mortality and disability costs because of long-term
exposure to secondhand smoke (SHS). The cost savings here are the increased survivor benefits
from private pensions and social security.
The third column presents the costs of smoking to everyone else—i.e., the “external” costs of
smoking. As can be seen, there are both substantial external costs and cost savings (benefits)
from a typical 24-year-old smoker, and that in this analysis these costs exceed the cost savings
by $8,080 over the lifetime of this smoker. The costs include medical costs not paid by the
smoker (and covered by everyone else), reductions in state and federal income tax revenues
because of the smoker’s lower earnings, losses to employers from smokers’ increased sick leave
and other productivity losses (e.g., smoking breaks), and the subsidization of smokers’ life
insurance by other policy holders who live longer. Cost savings or benefits include the federal
and state excise taxes collected on each pack of cigarettes sold in the US, and the monies that
social security and private pension plans save in reduced payments to smokers because of their
shorter lives.

Evaluation, Research and Development Unit, The University of Arizona

5

The Value of Smoking Prevention

Modifications to the Baseline Estimate for Arizona Youth
Several adjustments must be made to the estimates shown in Table 1 in order to generate the best
available estimates of the value of smoking prevention for Arizona youth. The modifications
presented in this section are limited to those possible with existing data. The next section
presents the results of these modifications and a list of future improvements to consider. The
modifications we will make in this report to the estimates shown in Table 1 are the following and
each is discussed in detail below:
•

Adjust the state and federal excise taxes up to present levels

•

Adjust national wage estimates to reflect Arizona’s slightly lower average wage rate

•

Adjust worker productivity estimate to reflect values seen in more recent studies

•

Adjust the life insurance impacts for the fact that current non-employer-based life insurance
policies in Arizona all charge higher premiums to smokers

•

Add medical costs for the spouse and a reduction in the spouse’s wages both due to disability
from secondhand smoke exposure

•

Reevaluate estimates used for the mortality and medical costs for infants and children due to
secondhand smoke exposure

•

Discount costs back to the average age at which the smoking prevention intervention is
targeted and add cigarette consumption from start of smoking to 24 years of age

•

Add a column to show the cost of smoking to Arizona’s state budget

State and federal excise taxes
As can be seen in Table 1, state and federal tobacco excise taxes are a cost to smokers and a
benefit to the rest of society. In these estimates the total cost of these taxes to the smoker over his
or her lifetime is $4,001 ($1,880 plus $2,121), and those dollars flow into state and federal
coffers to provide benefit the rest of society. ii Note that this $4,001 partially offsets the total
costs imposed by a smoker on others. Without the benefit of these tax dollars the total external
costs of smoking would be $12,082 ($8,080 plus $4,001). The estimates in Table 1 use a federal
tobacco tax rate of $0.44 per pack and a state tax rate of $0.50 per pack (in 2009 dollar terms). If
this combined rate of $0.94 is increased, the external cost of smoking is reduced.
Before April 1, 2009, the federal excise tax on cigarettes was 39 cents per pack.3 After that date
the federal excise tax increased to $1.0066 per pack. On December 7, 2006, Arizona’s state
excise tax on cigarettes rose from $1.18 to $2.00 per pack.
[www.taxadmin.org/fta/rate/cig_inc02.html] These excise taxes transfer dollars from smokers to
various state and federal funds which, hopefully, benefit the citizens of Arizona and the nation as
a whole.
ii

Technically, some small portion of these dollars may also flow back to the smoker and his or her family if they are
a beneficiary of any state or federal program that is funded by these taxes. However, for the purposes of this analysis
any benefits returning to one smoker’s family are considered insignificant when compared to the number of people
who make up the rest of society. Note that each smoker is considered alone. When considering a fellow smoker, the
first smoker is considered to be part of the “everyone else” group.

Evaluation, Research and Development Unit, The University of Arizona

6

The Value of Smoking Prevention

Adjustment to Arizona average wage rates
The estimates made by Sloan et al. used year 2000 median daily earnings of $98.20 for females
and $129.20 for males for absenteeism and other productivity losses. The CPI adjustment used to
create the estimates in Table 1 (multiplying by 1.235) means that year 2009 median daily
earnings of $121.24 for females and $159.52 for males were used in that table. US median usual
weekly earnings for full-time wage and salary workers (the measure used in the Sloan study) in
2008 was $638 for females $798 for males, and $722 for females and males combined.
[http://www.bls.gov/cps/cpswom2008.pdf] Translating these figures into median daily earnings
(by dividing by 5) results in $127.60 for females, $159.60 for males, and $144.40 for both
combined—adjusting these to 2009 using the non-medical CPI gives $126.85, $158.66, and
$143.55, respectively. These numbers are not far off from Sloan’s original estimates adjusted to
2009 values. Similar 2009 numbers for Arizona are $130.63 for females, $153.89 for males, and
$141.96 for both combined. The Arizona numbers are 98.8% of the national values, therefore,
the productivity estimates are adjusted down by a factor of 0.988.

More recent worker productivity estimates
The Sloan et al. estimates for losses in worker productivity (other than absenteeism) are based on
a 2000 study that estimated losses due to smoking breaks in workplaces using an estimate of 30
minutes per day.4 More recent studies have used worker’s self-report estimates of their
productivity at work to estimate productivity loss due to smoking.5, 6 Estimates from the more
recent of these studies were used to estimate productivity loss due to smoking in Arizona. This
study used results from about 35,000 workers to estimate an average loss of 53.6 hours per year
due to absenteeism and 76.5 hours due to presenteeism (reduced productivity while at work).5
The absenteeism estimate is within the range used by the Sloan et al. estimates, so we kept the
Sloan et al. estimate for absenteeism. We used the ratio between the presenteeism and
absenteeism losses in this new study to generate a better estimate for presenteeism losses due to
smoking.

Adjustment to life insurance cross subsidization estimates
In the book the Price of Smoking, the authors note that if smokers were universally charged “the
actuarial value of the loss in death-contingent payments that smoking causes” there would no
longer be any cross subsidization between nonsmokers and smokers in terms of life insurance.1
(pg193)
Informal calls to insurance brokers in Arizona indicate that all life insurance policies sold
to individuals have higher premiums for smokers. However, life insurance policies sold to
businesses to cover their employees do not distinguish between smokers and nonsmokers, or
even between the mix of smokers and nonsmokers in a particular employee mix. Therefore, it is
still likely that cross subsidization is happening. However, it is not likely to be as much as was
seen in the older data used in the Sloan et al. estimates, which were based on the 1992 through
2000 panels of the Health and Retirement Study. We adjusted the life insurance cross
subsidization estimates down by 25 percent to account for the likelihood that less cross
subsidization is ocurring now and likely less will happen in the future due to more awareness of
the mortality costs of smoking.

Evaluation, Research and Development Unit, The University of Arizona

7

The Value of Smoking Prevention

Spouse reduction in wages and medical costs due to disability from
secondhand smoke exposure
The Sloan et al. estimates include estimates for loss of life and quality of life reductions for
disability for both the smoker and his or her spouse. However, no loss of earnings and no
medical costs for the spouse’s disability are included. We used the ratio between smokers’
disability costs and their lost earnings and medical costs to calculate and estimate of lost earnings
and medical costs for smokers’ spouses based on the estimate for spousal disability. The medical
costs for smokers’ spouses will now be included in the medical costs due to secondhand smoke
(SHS) exposure, which in the Sloan et al. estimates only contain medical costs for smokers’
children and infants.

A relook at the literature regarding mortality and medical costs for
children due to secondhand smoke exposure
The Sloan et al. estimates use an estimate for infant mortality based on infant exposure to SHS
and to prenatal maternal smoking that assumes 599 male and 409 female deaths due to smoking
and that these deaths are valued at $100,000 per year of lost life (the same mortality cost as used
for smokers’ deaths). These estimates of male and female infant deaths are similar to those
calculated by the CDC's Smoking-Attributable Mortality, Morbidity, and Economic Costs
(SAMMEC) system for the 2000-2004 period of 445 male and 331 female deaths.7 We allocated
these mortality costs across all smokers and adjusted them for the usual age at which men and
women have children.
For the direct medical costs for infants and children due to illness from SHS exposure the Sloan
et al. estimates used an estimate that was an average across a number of studies. However, these
studies used a variety of methodologies and targeted different childhood diseases. Therefore, we
took the most detailed and comprehensive of these studies and used these estimates. This study
measured the medical costs from excess cases due to SHS exposure and maternal smoking of
low birth weight infants, respiratory syncytial virus bronchiolitis, acute otitis media and otitis
media with effusion, asthma and burns.8

Discount the costs to average age for prevention services and add
consumption costs from start of smoking to 24 years of age
The Sloan et al. estimates show the cost of smoking for a 24-year-old smoker. In order to use
these cost of smoking estimates as estimates of the value of smoking prevention we have to
discount these costs back to the age targeted for smoking prevention services. Smoking
prevention services in Arizona tend to target children and adolescents in grades 4 through 8 with
most of the services going to those in grades 4 and 5.9 Using an average age of 10 years for
children in those grades, we must discount the 24-year-old estimates back another 14 years (to 10
years of age) using a discount rate of 3% (the value used in Sloan et al). This means that the 24year-old estimates will be multiplied by a factor of 0.661 (1/(1.03)14).
In addition to discounting, we also must be sure to capture any additional costs that would occur
during the period between 10 and 24 years of age for a typical smoker. For most of the cost
categories shown in Table 1 it is reasonable to assume zero costs for the 10 to 24 year old period.
The health impacts of smoking are not yet manifest in this younger population, so mortality,

Evaluation, Research and Development Unit, The University of Arizona

8

The Value of Smoking Prevention

disability, medical care, and productivity costs are zero or negligible. Also, for at least most of
these years these smokers during these ages are not living with spouses or their own children—or
at least not enough to generate large SHS health impacts. However, these smokers do purchase
and consume cigarettes. We used data from the 2009 Youth Risk Behavioral Survey (14 to 18
year olds) and the 2005 Arizona Tobacco Survey (18 to 24 year olds) to estimate the number of
packs per year smoked by these younger smokers, and we used data from the 2005 Arizona
Tobacco Survey to estimate the likely start age for a typical 24-year-old smoker. The additional
consumption each year is then estimated as the proportion of future 24-year old smokers that
would have been smoking by that age times the average pack-per-day consumption for that age.
These packs consumed are then valued in terms of the cost of the cigarettes themselves and in
terms of the excise taxes paid.

Packs per year smoked

Figure 1 shows the trajectories of the probability of a typical 24-year-old smoker smoking at
each earlier age and the average packs per year smoked for each of those ages. As can be seen
almost 95 percent of typical 24-year-old smokers were also smoking at 19 years of age, and by
20 years of age these smokers are already smoking as much on average as was found in the Sloan
et al. estimates (about 216 packs per year). Therefore, the cost and excise taxes collected on a
total of almost 1,200 additional packs smoked before 24 years of age were added to the Table 1
estimates.
250

Average packs per year
smoked

100%

200

Probability of a 24-yearold smoking that year

80%

150

60%

100

40%

50

20%

0

0%
10 11 12 13 14 15 16 17 18 19 20 21 22 23
Years of age

Figure 1. Estimates of the probability of a 24-year-old smoker smoking in younger years and average
amounts smoked

Portion of these costs that directly impact the state budget
The last adjustment to these dollars is to include a column acknowledging the impact of the cost
of smoking on Arizona’s state budget. This perspective is important since this is the source for
much of the funding for tobacco control programs, including smoking prevention. As might be
expected, the Arizona state budget perspective includes a subset of the costs included for the rest
of society.
The most obvious dollar amount included in for the state budget is the estimate of state tobacco
excise tax revenue. Other direct dollar impacts on the state budget due to smoking come from

Evaluation, Research and Development Unit, The University of Arizona

9

The Value of Smoking Prevention

Medicaid expenditures iii and lost state income tax revenue due to the lower earnings of smokers.
In the Sloan et al. estimates approximately 43 percent of the increase in medical costs due to
smoking are costs paid by Medicaid. The estimate of the impact on Arizona income tax revenues
used a tax rate of 2.2 percent of taxable earnings

Best available estimate of the cost of smoking (and value of
prevention) for Arizona youth
Table 2 shows the baseline cost of smoking estimates from Table 1 after all the adjustments
discussed above have been made. As can be seen the main differences between the results shown
in Table 2 and what was seen in Table 1 are:
•

That state and federal excise tax revenues collected on cigarettes have gone up substantially.

•

The smoker’s mortality, disability, medical, and lost earnings costs have gone down due to
the fact that these costs are now farther in the future.

•

Smokers’ sick leave has gone down because these costs are also farther in the future, but
other productivity losses have tripled with using newer, better estimates.

•

Transfers of dollars between smokers, their families, and the rest of society for Social
Security Insurance and private pensions have been reduced due to discounting, but the life
insurance transfer has been reduced both because of being farther in the future and because
higher premiums for smokers becoming more common.

•

Finally, costs to the smoker’s family because of their secondhand smoke exposure have gone
down in total due to discounting, despite the addition of lost earnings and medical care costs
for the spouse, and better (and higher) estimates for infant mortality and children’s medical
costs.

iii

Note that technically a substantial portion of Medicaid costs are paid by the Federal government.

Evaluation, Research and Development Unit, The University of Arizona

10

The Value of Smoking Prevention

Table 2. Lifetime cost of smoking for a 10-year-old future smoker (male and female weighted average)

Cost component
Cost of cigarettes themselves
Federal excise taxes on tobacco
State excise taxes on tobacco
Smoker’s mortality
Smoker’s disability
Smoker’s medical care
Loss in smoker’s earnings
Lost income taxes on earnings
Work loss (sick leave)
Other productivity losses
SSI outlays and benefits
Private pension outlays and benefits
Life insurance outlays and benefits
Spouse mortality (SHS)
Spouse disability (SHS)
Spouse lost earnings (SHS)
Infant deaths (SHS)
Medical care (SHS)
Totals

Smoker

Smoker’s
family

Rest of
society

$10,699
$4,054
$7,338
$71,322
$11,934
$991
$17,905
$0
$0
$0
$3,531
$4,776
($4,715)
$0
$0
$0
$0
$0
$127,835

$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
($593)
($418)
$0
$18,283
$853
$1,280
$1,744
$2,073
$23,222

$0
($4,054)
($7,338)
$0
$0
$1,965
$0
$3,581
$2,643
$3,762
($2,939)
($4,358)
$4,715
$0
$0
$0
$0
$0
($2,023)

Evaluation, Research and Development Unit, The University of Arizona

Society Arizona
as a
state
whole
budget
$10,699
$0
$0
($7,338)
$71,322
$11,934
$2,956
$853
$17,905
$3,581
$399
$2,643
$3,762
$0
$0
$0
$18,283
$853
$1,280
$1,744
$2,073
$149,034 ($6,087)

11

The Value of Smoking Prevention

How These Estimates Can Be Used and Their Limitations
This report documents the estimation of the value of smoking prevention efforts in Arizona. This
section presents the appropriate use of these estimates, and their limitations.

The perspective of the analysis determines which estimates to use
Because what is considered a cost differs depending on who is asking the question, when talking
about value, costs, and cost savings, we must acknowledge the perspective (or point of view) of
the analysis. This is why Table 2 contains five columns of estimates. As can be seen, the
numbers in each column are related, but the “bottom line” estimate for each column varies
dramatically. Below are listed various possible perspectives and the estimate to be used for each.
•

The perspective of society as a whole. This perspective is appropriate to use when discussing
the benefits of smoking prevention across all individuals in the state, iv no matter who
receives the benefits or pays the costs. From this perspective, preventing one 10-year-old
who would have otherwise become a typical smoker from becoming that smoker is worth
$149,034 to Arizonans as a whole.

•

The perspective of the portion of society other than the smoker and his or her family. As can
be seen in Table 2, the vast majority of the costs to society as a whole accrue to the smoker
his- or herself ($127,835), and the next largest portion of costs accrue to the smoker’s
immediate family ($23,222) mostly in terms of the health and mortality costs of secondhand
smoke exposure. To many, smoking is considered a voluntary, personal decision, and free
will and personal autonomy must be respected. Therefore, it can be easier to justify dollars
for tobacco control programs that prevent costs to everyone else (i.e., reduce the external
costs of smoking) than it is to justify dollars for programs that prevent the costs of smoking
from affecting smokers themselves. Whether smokers’ families should be included within the
realm of personal choice or be considered victims is up for debate. Therefore, depending on
whether costs to smokers’ families are included, preventing one 10-year-old who would have
otherwise become a typical smoker from becoming that smoker is a cost of $2,023 to all
Arizonans except the smoker and his or her family, or a gain of $21,198 to all Arizonans
except the prevented smoker him-or herself ($23,222 gain to the family minus a $2,023 cost
to the rest of the state).

•

The perspective of Arizona’s state budget. This perspective is appropriate to use when
discussing budget issues with the Arizona legislature because it only includes dollars that
would flow into and out of state coffers. From this perspective preventing one 10-year-old
who would have otherwise become a typical smoker from becoming that smoker costs the
Arizona state budget $6,087 over that smoker’s lifetime. Note that this, like all other
estimates presented in this report, is in addition to the cost of the prevention program itself.

Limitations to these estimates
Any estimate of the cost of smoking is just that; an estimate. Therefore, the numbers in Table 2
should be used with caution. First of all there are a number of cost components that were not
iv

This statement assumes that federal dollars paid by Arizonans generally return to Arizonans in the form of various
federal benefits.
Evaluation, Research and Development Unit, The University of Arizona

12

The Value of Smoking Prevention

included in these estimates. It is likely that these costs will be small, but they should be noted for
completeness. Missing costs include lost worker productivity due to caring for children with SHS
caused illnesses, SHS costs to those other than the smokers’ family (these will be especially
small due to the implementation of the comprehensive statewide smoking ban), lost income taxes
on spouse lost earnings, and sales tax revenue on cigarette sales. Sales taxes were not included
because these amounts differ greatly by location and because these revenues mostly benefit local
communities.
Other cost component estimates would benefit from additional, more recent and Arizona-specific
analyses. In particular, it would be beneficial to closely analyze the Arizona Health Care Cost
Containment System’s (AHCCCS’s) costs to determine what portion is due to smoking. The
limitations stated by the authors of The Price of Smoking should also be closely reviewed.1 Each
chapter presents the limitations of the analyses discussed therein, and a summary is presented in
the last chapter of the book, starting on page 267.
Finally, one last limitation for the use of these estimates should be stated. These estimates are
appropriate to use to estimate the value of preventing a 10-year-old, who would have otherwise
become a typical 24-year-old smoker, from becoming a typical 24-year-old smoker. The number
of 10-year-olds prevented from smoking according to this definition is difficult to determine for
at least two reasons. First, because it requires knowing whether a 10-year-old will become a 24year-old typical smoker, and second, because it requires a determination of whether an
intervention prevented this or not.

Policy Implications
The finding that the costs to society as a whole of a 10-year-old taking up smoking is about
$150,000 supports the general notion that preventing smoking is an important and useful
undertaking. The finding that the vast majority of these costs are costs to the smoker him-or
herself and not to the rest of society is important for clarifying the real benefits of tobacco
control programs. However, one of the most striking findings is that preventing a child from
taking up smoking (who otherwise would have) actually costs the State of Arizona about $6,000
over the course of that child’s lifetime.
This situation arises because of the substantial taxes paid on cigarettes in Arizona, which more
than offset the costs of smoking to the rest of society. The State actually makes money from
smokers. Therefore, dollars spent on tobacco use prevention have a double cost—the cost of the
program itself and the cost of the future lost tobacco tax revenues. These programs are still
justifiable as funds spent to help a particular citizen group—otherwise future smokers. However,
if tobacco tax dollars are spent on non-tobacco-related services two issues arise. First, smokers
pay these additional dollars into the State budget, and they are the biggest losers from smoking.
Therefore, in the interest of equity, they should be the beneficiaries of the services funded by
their taxes. This is especially true given that tobacco taxes tend to be regressive—i.e., have a
larger economic impact on lower socioeconomic groups.10-12 Second, tobacco control programs
and any non-tobacco programs dependent upon tobacco tax dollars are put into direct conflict.
The former are actively reducing the future budgets of the latter, which sets up a dis-incentive to
fund tobacco control programs. In contrast, if tobacco tax dollars are used to fund tobacco
control programs, the reduction in future funding caused by the success of the programs will
align with the reduced need over time for those funds.

Evaluation, Research and Development Unit, The University of Arizona

13

The Value of Smoking Prevention

Higher tobacco taxes are seen as part of the arsenal of tobacco control. Indeed, increased taxes
have been shown to decrease cigarette smoking somewhat13-16 particularly among younger
people15, 17—although this effect on youth has been challenged.18, 19 However, it is important to
consider the uses to which this tax revenue is put. To avoid disincentivising tobacco
interventions, and in the interest of equity, there is a strong rational for using these dollars
towards tobacco control, which, if done effectively, may phase out the income stream over time,
but also, unlike other uses, reduces the need for the income stream.

Evaluation, Research and Development Unit, The University of Arizona

14

The Value of Smoking Prevention

Reference List
(1) Sloan FA, Ostermann J, Picone G, Conover C, Taylor DH, Jr. The Price of Smoking.
Cambridge, MA: The MIT Press; 2004.
(2) Miller VP, James CR, Ernst C, Collin F. Smoking Attributable Medical Care Costs:
Models and Results. Berkeley, CA: Berkeley Economic Research Associates; 1997.
(3) Alcohol and Tobacco Tax and Trade Bureau. Tobacco Federal Excise Tax Increase and
Related Provisions. TTB gov 2010 April 20;Available at: URL:
http://www.ttb.gov/main_pages/schip-summary.shtml. AccessedApril 20, 2010.
(4) Parrott S, Godfrey C, Raw M. Costs of smoking in the workplace in Scotland. Tobacco
Control 2000;9:187-92.
(5) Bunn WB, Stave GM, Downs KE, Alvir JMJ, Dirani R. Effect of smoking status on
productivity loss. JOEM 2010;48(10):1099-108.
(6) Halpern MT, Shiklar R, Rentz A.M., Khan ZM. Impact of smoking status on workplace
absenteeism and productivity. Tobacco Control 2001;10:233-8.
(7) Adhikari B, Kahende J, Malarcher A, Pechacek T, Tong V. Smoking-Attributable
Mortality, Years of Potential Life Lost, and Productivity Losses --- United States, 2000-2004. MMWR 2008;57(45):1226-8.
(8) Aligne CA, Stoddard JJ. Tobacco and children. Archives in Pediatric and Adolescent
Medicine 1997;151:648-53.
(9) Malter F. Comprehensive Statewide Evaluation of BTCD's Intensive Schoolbased Tobacco Prevention Education Program in Grades 4 – 8, 2007-08.
Evaluation, Research, and Development Unit; 2009 Sep 1.
(10) Gospodinov N, Irvine I. Tobacco taxes and regressivity. Journal of health economics
2009;28(2):375-84.
(11) Regidor E, Pascual C, Gutiθrrez-Fisac JL. Increasing the price of tobacco: economically
regressive today and probably ineffective tomorrow. European Journal of Cancer
Prevention 2007;16(4):380.
(12) Franks P, Jerant AF, Leigh JP et al. Cigarette prices, smoking, and the poor: implications
of recent trends. American Journal of Public Health 2007;97(10):1873.
(13) van Baal PHM, Brouwer WBF, Hoogenveen RT, Feenstra TL. Increasing tobacco taxes:
a cheap tool to increase public health. Health Policy 2007;82(2):142-52.

Evaluation, Research and Development Unit, The University of Arizona

15

The Value of Smoking Prevention

(14) White VM, Gilpin EA, White MM, Pierce JP. How do smokers control their cigarette
expenditures? Nicotine & tobacco research 2005;7(4):625.
(15) Gallet CA, List JA. Cigarette demand: a meta-analysis of elasticities. Health Economics
2003;12(10):821-35.
(16) Wilson N, Thomson G. Tobacco tax as a health protecting policy: a brief review of the
New Zealand evidence. Journal of the New Zealand Medical Association 2005;118:1213.
(17) Ross H, Chaloupka FJ. The effect of cigarette prices on youth smoking. Health
Economics 2003;12(3):217-30.
(18) Glied S. Is smoking delayed smoking averted? American Journal of Public Health
2003;93(3):412.
(19) DeCicca P, Kenkel D, Mathios A. Cigarette taxes and the transition from youth to adult
smoking: Smoking initiation, cessation, and participation. Journal of health economics
2008;27(4):904-17.

Evaluation, Research and Development Unit, The University of Arizona

16