|
Items in cart [4] |
Questions? Call 888-624-8373 |
HARDBACK + PDF HARDBACK PDF BOOK PDF CHAPTERS Free Resources |
||||||||||||||||
|
||||||||||||||||
|
|
|||||||||||||||
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 112
G
The Smog and Levee of
Joint and Service-SpecKic
Vising
n this chapter me discuss research regarding two questions. First is
Mere a minimum level of advert/sing necessary for a cost-e~ective
recru1Ung program' even ~ that advert/sing ~ not necessary to achieve
conte~or~eous enUs~ent conbact goad? Pastoral Men Me recruits
big climate is good and recruits are plenties' m1Utary planers tend to cut
advert/sing budgets Derby contributing to a reducUon in "awareness"
capital and propensity Even. ~~ may possibly set up a boom or bust
cycle' in which propensity faUs' recru1Ung becomes more Dracula and
Men advert/sing funds have to be restored or even increased beyond
what they would have been gout Me 1nhial cut to stimulate propen-
~ty and enUstmenL lo our knowledge' this Sue has not been Budged
extensively' and therefore the best research designs may not be 1mmed1-
ately clear) Is chapter discusses Me poss~11hy of both econometric and
experimental designs to address the quest/on.
Me second quest/on concerns the proper levels of joint and Service-
spec1Ac advert/sing. Certain types of advert/sing themes' such as generic
themes designed to increase overaU propensHy' may be best done as a
joint program' Shoe advert/sing themes Maturing specific benefits of m1~-
1Col~erg and ~~ko (2003) studied aggregate enbstments into the four Sedges over
the and esh~ate larger business cycle gag., unemployment) effects for the Any and the
Navy than are generally estimated in me studies summarized in Chapter 5. hey recom-
mend an early warning system that wiU help the Department of De~nse (DoD) predict
turning points in recruiting and undertake policy actions that wig reduce the
cyclically of ~cmU~g.
112
OCR for page 113
fOINT AND SERVICE-SPECIFIC ADVERTISING
113
tary service are best done in the Service programs. We do not currently
know much about what levels of Service and joint advertising will most
efficiently sustain different levels of propensity and enlistment. Answer-
ing this question is most likely to require a combination of research designs,
both econometric and experimental.
MINIMUM LEVEL OF ADVERTISING
In the discussion in Chapter 5 of the optimal levels of recruiting
resources to achieve a given set of recruiting goals, the optimal amount of
advertising, like the optimal amount of any other recruiting resource, was
the amount that permits goal achievement at lowest cost. Is there some
minimum level of advertising that is above the level of advertising that
minimizes the total cost of achieving current recruiting goals, but may be
optimal in a longer term perspective? For example, consider a case in
which, because of a poor economy and high unemployment rates, or a
significant reduction in the current demand for new recruits, or both,
current recruiting goals can be achieved even if recruiting resources,
including advertising, are drastically reduced. Is there a reason to main-
tain resources, particularly advertising, above the minimum level to achieve
current goals?
If the answer is yes, it must be because greater advertising expendi-
tures in one period will make recruiting goals in a future period less
costly to achieve. That is, there must be dynamic aspects to some types of
recruiting resources such that current expenditures on those resources
affect both current and future recruiting. As we discussed in Chapter 5
and as reported in the literature (Hogan, Dali, Mackin, and Mackie, 1996;
Dertouzos and Garber, 2003), current advertising expenditures contribute
to a stock of information and awareness capital. This stock changes over
time, as the existing stock of information decays but is replenished by
new advertising expenditures.
Hence, under this basic concept of a flow of advertising contributing
to a stock of awareness or information capital, one can anticipate that the
effect of reducing the flow of new advertising on recruiting will, at first,
be small. But as the stock continues to decay without replenishment, the
adverse effect on recruiting can grow. This is illustrated in Figure 6-1,
which assumes a depreciation rate of 10 percent. The stock of information
and awareness capital has been built up to a notional value of 100 units in
month 1. However, from that month forward, new advertising, which
contributes to the stock, is cut to zero. Some level of new advertising is
necessary simply to compensate for the depreciation of the existing stock.
The effect on the ability to recruit is, at first, modest, because the stock of
capital by month 3 is about 80 percent of the original stock. However,
OCR for page 114
114
120
100
80
o
Oh 60
40
20
EVALUATING MILITARY ADVERTISING AND RECRUITING
O-
+ Stock
~ New Adv
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Months
FIGURE 6-1 Stock of information and awareness capital at 10 percent monthly
depreciation (notional).
with no new advertising and continued depreciation, the stock would
quickly be depleted to less than 30 percent of its original value.
Now, assume that new advertising, equivalent to a contribution to
the stock of 15 units per month, is provided. In this case, the stock of
information, initially at 100, rises eventually to a new equilibrium of 150,
where depreciation (10 percent of 150) equals new advertising of 15 units.
At this point, new advertising exactly offsets the depreciation, maintain-
ing a stock of information and awareness capital of about 150. When new
advertising exceeds depreciation, the stock grows. This is shown in Fig-
ure 6-2.
This simple example suggests how one may reduce advertising
resources and in the short term experience no significant reduction in
recruits. Moreover, assume that, initially, the optimal level (i.e., cost-
minimizing level) of the stock of advertising is 100. Then, because the
economy sours and the demand for recruits declines, the new optimal
level of the stock is 50. That is, significantly less awareness is necessary to
recruit the desired numbers. In this case, it would make sense to reduce
new advertising until the new equilibrium stock is approached, then
. . .
Increase it again.
However, if there is reason to believe that external economic condi-
tions are likely to improve significantly, or that the demand for new
recruits is to increase, it may be imprudent to permit the stock of aware-
ness capital decline to this level. This would be of greater concern if the
marginal cost of increasing awareness capital in a particular period was
OCR for page 115
fOINT AND SERVICE-SPECIFIC ADVERTISING
150
115
100 —
~ New Adv
50~....................................................................................................................................................................................................................................................
O —
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 3 5 7 9 11 13 15
Months
FIGURE 6-2 Stock of information and awareness capital (notional).
increasing. That is, let A be the augmentation to the stock of awareness
capital in a given time period. Then, let C(A) be the cost of increasing
awareness capital by A in a given period (month). We have:
DC(A) / DA, O
C(A) / DA , O
If the marginal cost of increasing the stock of awareness capital
increases with the magnitude of the desired increase in capital, then it
may be less costly to begin increasing the stock of information capital in
periods earlier than is needed for current recruiting.
To abstract from uncertainty over the demand for recruits, let us
assume that the time path of recruit quotas for both highly qualified
recruits (QH) and less-well-qualified recruits (QL) is known and that it
fluctuates over time. One way to express the issue is whether there is a
case for which
T T
~Cmin(QH t,QL t,Pt,Et) / (1+ r) > ~COpt(QH t,QLt,pt,Et) / (1+ r)
t=1 t=1
OCR for page 116
116
EVALUATING MILITARY ADVERTISING AND RECRUITING
where C(....) is the cost function for producing highly qualified and less
qualified recruits, respectively, in period t, and r is the discount rate. The
variable P is the vector of recruiting resource prices in the period and E is
a vector representing the state of the economy. The notation min repre-
sents the cost-minimizing levels of inputs to exactly produce the supply
of recruits to meet goals in period t, while opt is the level of resources that
minimizes, not the costs in a specific period, but the present value of costs
over periods 1 through T. when discounted at interest rate r. In the second
case, potential supply may be greater than necessary to meet current
goals. If so, it is optimal because it lowers the costs of meeting subsequent
goals. If the above inequality holds as a general case, then there is a level
of resources above the level necessary to meet current goals that mini-
mizes the present value of costs over the longer term.
How can we test the hypothesis that the inequality, defined above,
may hold? We are likely to be able to do so only indirectly. If advertising
is the reason the inequality holds, it is necessary that: (1) advertising
expenditures have effects over time and (2) the marginal cost of increas-
ing the stock of information or awareness capital in a given time period is
. .
ncreasmg.
If advertising affects the stock of information capital only in the period
in which it is produced, then the stock of capital and the flow of advertis-
ing in a given period are equal, except for any external factors affecting
the stock. In a sense, the depreciation rate of the inherited stock is 100
percent. The only reason to increase advertising in a given period would
be to affect current enlistment supply, and there would be no reason not
to reduce advertising if not needed in the current period. Furthermore, if
advertising adds to the effective stock of information at a constant mar-
ginal cost, it would be less costly, in a present value sense, to wait until
the advertising is needed before increasing it, and to do it all in one
period. Only if marginal costs are rising, and advertising effects are dynamic,
would it be optimal (i.e., cost minimizing) to incur advertising costs cur-
rently, not because they are needed currently but because they affect
future recruits.
To determine if there is a minimal level of advertising expenditure
that is above that necessary to meet current goals, an econometric model
must be able to capture the effects of advertising over time and allow for
nonlinear effects of advertising on recruiting in a given time period.2
2Earlier we presented the notion that advertising contributed to a stock of information or
awareness capital, and that this stock then affected recruiting. This is a useful concept for
exposition purposes. However, the nonlinear effect, in practice, is between advertising in a
given period and enlistments in that and subsequent periods. There is no necessity for the
concept of the stock.
OCR for page 117
fOINT AND SERVICE-SPECIFIC ADVERTISING
117
Dertouzos and Garber (2003) estimate a supply curve that includes advertis-
ing. As discussed in Chapter 5, this specification permitted both dynamic
effects of advertising that is, it allowed advertising in one period to
affect recruiting not only in the same period but in subsequent periods-
and nonlinear effects within a given time period. Advertising in each
period, both current and lagged, enters as a nonlinear S-shaped or logistic
function. The functional form provides ranges for small marginal effects
of advertising, then increasingly larger marginal effects, followed again
by increasingly small marginal effects, as if there were a saturation level
for advertising. If the unit price of advertising is constant, this would
correspond to areas of high but decreasing marginal costs, followed by an
inflection point and increasing marginal costs. Hence, this specification
has the ability to capture the two features of advertising that would be
required if there is a minimum level of advertising that is above the cost-
minimizing level when only current recruiting goals are considered.
The findings reported in Dertouzos and Garber (2003) and those from
the literature reviewed briefly in Chapter 4 are that although advertising
has dynamic effects, the lag structure suggests that most effects are real-
ized within about two to three months of the advertising expenditure.
These data suggest that current advertising may affect recruiting one,
two, or three months in the future, but it does not affect recruiting six
months or a year in the future. If the data on which these findings are
based are strong and compelling, it would mean that a compelling case
could not be made for maintaining advertising expenditures in a given
period, if they are not needed to achieve current goals, solely because it is
desirable to increase recruit supply six months or a year in the future.
However, there is not a large body of evidence on the issue, and most
authors who do provide some evidence also note that the data from which
the estimates are derived are weak, and that more research is necessary.
Hence, because of the paucity and limitations of evidence available to
date, we conclude that the issue remains open and subject to further
research. In our view it would be a serious mistake to view the available
research as sufficient grounds for drawing conclusions one way or another
about the effects of current advertising on future outcomes.
Furthermore, the functional form used by Dertouzos and Garber
(2003), while allowing for nonlinear effects of advertising a necessary
feature for establishing some minimal level of advertising constrains
the nonlinearity to follow a particular form. Because of this, it imposes the
nature of the nonlinearity, rather than having a very flexible functional
form that would allow the data to determine the nature of the nonlinearity.
Hence, while the S-shaped functional form for advertising is plausible, it
cannot be used by itself to test the hypothesis that advertising has non-
OCR for page 118
118
EVALUATING MILITARY ADVERTISING AND RECRUITING
linear effects on recruiting.3 Instead, a more flexible functional form, one
that permits lagged effects of advertising but also nonlinear effects of
advertising within a period and does not constrain the precise nonlinear
relationship, should be applied.
The data necessary to estimate more flexible functional forms are
typically quite demanding. Because the functional form itself places fewer
constraints on the nature of the relationship between the supply of recruits,
recruiting resources, and other factors affecting recruiting, the variation
in the data must be sufficient to support distinguishing among alternative
relationships. It is probably not the case that the most recent data over the
l990s and early 2000s are sufficient to estimate these more demanding
relationships. Dertouzos and Garber (2003), for example, were not able to
estimate a Service-specific enlistment supply function from data over the
l990s, even though they placed constraints on the nature of the functional
relationships.
A controlled experiment using a quasi-experimental design to gener-
ate the data necessary to estimate supply relationships that incorporate
the potential for both dynamic effects of advertising and nonlinear rela-
tionships is likely to be impractical. The reason is that the effects to be
estimated are not simple impact effects of an intervention, but a more
complicated set of relationships regarding the dynamic, nonlinear struc-
ture of advertising effects. To generate sufficient data to estimate such
relationships would require that the experiment be continued over an
extended period of time at least three or four years. In an operational
environment in which there are changing supply conditions and real
recruiting demands to be met, this is unlikely.
Instead, we propose a focused effort to maintain data, especially
advertising data, in a systematic and careful way, for the purposes of
estimating a supply curve that incorporates the potential both for dynamic
and nonlinear advertising effects. This is consistent with our recommen-
dation in Chapter 5 for better and more thorough data collection and
estimation using more flexible functional forms.
Finally, it is important to note that most recruiting advertising has
been targeted directly on recruit-age youth. Advertising designed to influ-
ence parents, others who counsel youth, and youth who are not yet old
enough to enlist may have effects that extend over time. Tests for the
dynamic effects of advertising should attempt to distinguish between
3It could be used in conjunction with other constrained forms to test the hypothesis. In
that case, one would specify several alternative forms, including a linear form, and deter-
mine which form explained the observed data the best, and which could provide the best
out-of-sample prediction.
OCR for page 119
fOINT AND SERVICE-SPECIFIC ADVERTISING
119
advertising targeted at recruit-age cohorts and advertising targeted at
influencers or those who are not yet of recruiting age, if possible.
ADVERTISING TARGETED AT THE LONGER TERM
Historically, advertising for recruiting has focused almost exclusively
on recruit-age cohorts of youth. That is, it has been focused on providing
17-21-year-olds with information about opportunities in the Services and
on convincing recruit-age youth that enlisting is an option that they should
seriously consider. This being the case, it is perhaps not surprising that
the literature finds that current advertising expenditures affect recruiting
only in the near term two or three months in the future and do not
have effects on the market in the longer term six months or a year or
more in the future. An alternative type of advertising would target audi-
ences that have, perhaps, a longer time horizon for affecting the recruiting
market. This would include targeting youth earlier in the career decision
process before they can actually enlist, say at ages 14 through 16 and
targeting parents and other adult authority figures who may influence
the decisions of youth both immediately and in the future.4
This type of advertising has not been systematically employed by the
Services or the Department of Defense. Arguably, it is a possible role for
joint advertising. There is a solid logical argument for this type of adver-
tising, based on analyses of survey and other data (see National Research
Council, 2002, for further discussion). However, because it has not been
tested in the recruiting market, little is known about its effects.
The first step toward understanding its potential effectiveness and
how it may influence the recruiting market over time would be to develop
advertising of this nature and test it in the recruiting market. Because we
anticipate that this type of advertising has the potential for affecting the
market over longer periods of time, the experiment should be designed to
devote advertising expenditures of this nature in the recruiting market
for an extended period of time at least two years.
Ideally, the quasi-experimental design should attempt to include
variation over time and cross-sectionally in the advertising test. However,
in an operational environment, a simple pre-post design may be both
4warner, Simon, and Payne (2002) provide evidence that military advertising affects pro-
pensity to enlist. In an analysis of the Youth Attitude Tracking Study propensity over the
1988-2000 time period, they found a youths propensity to enlist to be positively related to
total military advertising expenditures per youth in the youths state of residence in the 12
months prior to the survey. But data limitations prevented determining whether the impact
of advertising on propensity varies by media or by whether it is joint or Service-specific
advertising.
OCR for page 120
120
EVALUATING MILITARY ADVERTISING AND RECRUITING
more workable and may serve to deliver the advertising in a more effi-
cient way.5 If the experimental period is of sufficient duration, it would
be useful to vary the level of advertising over time, rather than keeping it
at a single level "dose." In addition to econometric analyses of effects, one
would also include more qualitative methods of evaluation, including
surveys and focus groups.
THE LEVELS OF SERVICE-SPECIFIC AND JOINT ADVERTISING
In fiscal year (FY) 1989, the last large recruiting year before the draw-
down, the Services spent $100 million on enlisted advertising compared
with a joint level of $20 million. In that year, almost the entire joint budget
was for TV advertising. By FY 2000, joint advertising had dwindled to less
than $5 million despite Service advertising of $250 million (Office of the
Assistant Secretary of Defense [Force Management Policy], 2000~. The
scaled-back joint program has been either direct mail or magazine adver-
tising. The joint program raises four fundamental questions:
· What should be the content of joint advertising, and should it differ
from the content of Service-specific advertising?
· To whom should it be targeted?
· What media should the joint program use?
· Do the efficient levels of joint and Service-specific advertising depend
on the scale of the recruiting effort?
As discussed below, the answers to these questions are interrelated.
Past research is of little use in answering them. The first two questions
have barely been addressed in previous research (Carroll, 1987~. Several
econometric studies have tried to answer the latter two questions. Here
the research strategy has been to estimate the responsiveness of enlist-
ments to Service-specific and joint advertising in different media and then
infer the cost-effectiveness of the various forms of advertising in generat-
ing enlistments. Evidence from such studies about the relative effective-
ness of joint advertising is, at best, mixed (see Table 5-2~. Hogan et al.
(1996) found that joint TV and joint direct mailing were as effective as
Navy advertising in the same media. But estimates by Dertouzos (1989),
Warner (1991), and Warner, Simon, and Payne (2001) are less kind to the
. .
Font program.
5Excluding geographic areas to induce cross-sectional variation in advertising is difficult
and likely to be cumbersome to recruiting operations. Moreover, it may preclude some
efficient ways of delivering advertising, such as national magazines and network television.
OCR for page 121
fOINT AND SERVICE-SPECIFIC ADVERTISING
121
Dertouzos and Garber (2003) question the validity of past estimates of
the effects of advertising (summarized in Chapter 5~. They argue that
enlistments follow an S-shaped relationship with advertising and that the
form of the relationship varies by media. If their analysis is correct, then
we actually know very little about the effectiveness of joint advertising
(and therefore the answers to the above questions). In fact, the insignifi-
cance of the joint program in past econometric studies may have resulted
from its being below the minimum efficient scale (MES) in Figure 5-1.
Despite criticizing past studies on theoretical grounds, estimates pro-
vided in Dertouzos and Garber (2003) do not help answer the above ques-
tions. Their updated analysis of the Ad Mix Test data does not distinguish
joint advertising from Army advertising, their analysis of the 1993-1997
recruiting experience is conducted at the DoD level, and the joint pro-
gram is added to Service-specific advertising to derive measures of total
DoD advertising. Since past work is of little use in answering the four
fundamental questions above, this section now explores a series of ques-
tions that need to be addressed in order to adequately answer these fun-
damental questions. It is apparent from this discussion that an experiment
is needed.
Advertising Content
If the joint program advertises in the same media as the Service pro-
grams with the same message content, it is just an add-on to Service-
specific advertising. That is not necessarily bad. Joint advertising that
mirrors the Service programs essentially expands the scale of a common
program and thereby just moves enlistments up the S-shaped curve por-
trayed in Figure 5-1. If the unit cost of joint and Service-specific advertis-
ing are the same, joint advertising that mimics the message and content of
Service advertising would be a perfect substitute for Service advertising.
Such advertising would be pointless since an equivalent expansion of the
Service programs would yield the same result. In fact, due to the over-
head costs of ad development and effects of program scale on advertising
rates negotiated with ad agencies, newspapers, and other media, it is
unlikely that the joint program could perfectly replicate the Service pro-
grams at the same cost. In order to be useful, joint advertising must do
something that Service-specific advertising does not.
One way for joint advertising to distinguish itself is in the content of
its messages. The committee's previous report (National Research Coun-
cil, 2003, p. 227-230) discussed the message strategies in current military
advertising. Service-specific advertising appears to be occupationally ori-
ented, career-oriented, or oriented toward appealing to youths' desires
for adventure, challenge, self-development, and self-actualization. The
OCR for page 122
122
EVALUATING MILITARY ADVERTISING AND RECRUITING
report labeled these "package-oriented" advertising and went on to note
that only Marine advertising seems directed toward the more noble virtues
of patriotism, self-sacrifice, and service to country. (This seems to have
changed with recent Service advertising in the wake of September 11,
2001, and Operation Iraqi Freedom.)
Two recent DoD study groups the Defense Science Board Task Force
on Human Resources Strategy (DSB) and the DoD Quality of Life Panel
(also called the Jeremiah panel) expressed a belief that DoD needs to do
more to engage the American public about the importance of public
service.6 More values-oriented advertising that stresses the virtues of
patriotism, self-sacrifice, and service to country would serve that pur-
pose. Of course, the ultimate goal of advertising is to make youth more
inclined to serve (i.e., increase propensity), so that more of them walk into
recruiting offices on their own and more respond positively when they
are contacted by military recruiters. One major research question, there-
fore, is whether the message content of military advertising in fact affects
propensity and whether a move toward more values-oriented advertising
would increase it. The other question is whether a move toward values-
oriented advertising is more efficiently accomplished in a joint program
or whether redirected Service programs would have the same impact at
the same cost.
An advantage of loading all advertising into Service advertising was
alluded to above: namely, that advertising unit costs are lower in pro-
grams of larger scale. This advantage probably disappears once programs
reach some minimal scale. In fact, Navy and Army TV advertising seem
to generate about the same impressions per dollar of spending despite the
much larger scale of the Army TV program.7 A potential advantage of
having both joint and Service-specific advertising is that there are more
competition and more innovation in the development and delivery of
advertising the more firms are involved in the program.8
6The Dss report is available in The Defense science Board Task Force on Human Resources
strategy, Office of the Under secretary of Defense for Acquisition, Technology, and Logistics,
Washington DC 20301-3140, February 2000. The Quality of Life panel was commissioned by
secretary of Defense Donald Rumsfeld and was chaired by Admiral David Jeremiah tRety.
It did not publish a formal report, but an informal report and briefing of its findings and
recommendations are available from the Undersecretary of Defense for Personnel and
Research.
Unpublished calculations using the data supplied by PEP Research.
sin fact, there is some evidence from weapons programs that the cost-reducing competi-
tion from a second supplier outweighs the economies of scale that are lost when procure-
ment is split between two firms.
OCR for page 123
fOINT AND SERVICE-SPECIFIC ADVERTISING
The Target Audience
123
Current Service advertising is directed at youth themselves. The
committee's previous report expressed the belief that some advertising
needs to be directed toward civilian influencers of youth: parents, school
teachers and counselors, and others (for the role of family in influencing
behavior, see, for example, Sewell and Hauser, 1972; Otto, 2000; Mortimer
and Finch, 1996~. Such advertising, if effective, would build up the sup-
port base from which youth make early career decisions upon leaving
high school. Changing the attitudes of adult influencers would not happen
overnight and would require a sustained effort over time. One purpose of
targeting adult influencers would be to counteract the decline in the adult
veteran population, which one study (Warner et al., 2001) found to be
related to the decline in enlistment in the 1990s.
An advertising campaign aimed at adult influencers would be consis-
tent with the DSB recommendation to engage the American public about
the value of public service.9 It is an empirical question whether such
advertising, if done properly, would have more impact on youth propen-
sity and the ultimate goal, enlistment than advertising directed at the
youth themselves.
Advertising aimed at adult influencers is likely to be more effectively
accomplished in the context of a generic (joint) program than via Service-
specific advertising whose purpose is to channel recruits to particular
services. But that is an empirically testable hypothesis.
The Media
The Services advertise in a number of media: TV, radio, magazines,
newspapers, direct mail, billboards, etc. In recent years about 60 percent
of Army and Navy advertising expenditures have been for TV. Until
recently, most Marine Corps advertising was TV and most Air Force
advertising was non-TV. According to the Dertouzos-Garber model, an
organization with a low advertising budget should spend the bulk of its
budget on non-TV media. Non-TV advertising generates many more im-
pressions per dollar than TV advertising (according to PEP data) and
(theoretically) has a lower MES. But, theoretically, such advertising
reaches its saturation point at a relatively low level of expenditure. TV
advertising becomes cost-effective only at higher budget levels due to its
higher MES, but it remains cost-effective much longer due to the fact that
the saturation point is not reached until a fairly high level of expenditure.
9 The committee's previous report also expressed the belief that more advertising should
be values-oriented (National Research Council, 2003, p. 233-234~.
OCR for page 124
124
EVALUATING MILITARY ADVERTISING AND RECRUITING
These considerations suggest that the decision to eliminate joint TV adver-
tising during the latter half of the 1990s as the joint program was scaled
back made sense. However, devoting almost the whole 1989 joint budget
of $20 million to TV advertising probably did not make sense.
Factors that would have to be considered upon any rebuilding of the
joint program, or during experiments with advertising, are the message
content and the target audience. Ads that appeal to patriotism, for exam-
ple, hope to evoke a strong emotional response from viewers. TV or radio
ads may be more likely to evoke such responses than magazine or news-
paper ads. Messages that are targeted to youth in one medium may be
more productively targeted to adults via other media. For example, if
adults are more likely to read newspapers and less likely to watch MTV,
ads aimed at adult influencers might be more effective in newspapers
than on MTV.
Overall Scale of Recruiting
For several reasons alluded to above, the optimal levels of joint and
Service-specific advertising are likely to depend on the overall scale of the
advertising effort. In particular, joint advertising is likely to be more cost-
effective in the context of a large recruiting effort than a small recruiting
effort. This section discusses another reason why this might be the case.
Studies find a strong link between propensity to enlist, as measured
from surveys such as the Youth Attitude Tracking Study, and actual enlist-
ment (e.g., Warner et al., 2002~. Given the youth propensity to enlist, the
lower the military's demand for recruits, the greater the fraction of enlist-
ments that will come from youth who are already positively inclined to
enlist without military advertising to attract them. In such an environ-
ment, advertising can be targeted at those with a propensity and can be
largely informational and aimed at informing them about career opportu-
nities in particular Services and the like. It need not attempt to persuade
youth that military service is a good thing; they are already so inclined.
But the larger the demand for recruits, the more effort the military
must make to attract youth with no propensity to enlist. The issue is not
"high" versus "low" demand on some absolute scale, but demand rela-
tive to propensity. Holding constant the recruiting goal, any change in
propensity arising from factors other than military advertising (e.g., a
lower unemployment rate) can influence the degree to which advertising
needs to build propensity in the youth population rather than channel
those who have already decided to join to a specific Service.
How a change in the scale of the overall recruiting effort affects the
levels of joint and Service advertising depends in part on whether the two
sources of advertising differ in the degrees to which they build propen-
OCR for page 125
fOINT AND SERVICE-SPECIFIC ADVERTISING
125
sity. Much Service advertising is aimed at inducing youth already with a
propensity to enlist to join a particular Service and may not affect overall
propensity in the youth population. To the extent that joint advertising
can affect the propensity base, it should play a larger role in the overall
advertising program when the larger the recruiting mission is compared
with the base of youth with a propensity to enlist. As an example, joint
advertising might not have been very productive in 1992, a very easy
recruiting year, but it could have been very productive in the late 1990s,
when propensity was lower and recruiting was much more difficult.
CONCLUSIONS AND RECOMMENDATIONS
We have addressed the question of whether there is a minimum level
of advertising necessary for a cost-effective recruiting program, even if
that advertising is not necessary to achieve current enlistment contract
goals. Historically, when the recruiting climate is good and recruits are
plentiful, military planners tend to cut advertising budgets, thereby con-
tributing to a reduction in awareness capital and propensity levels. This
may possibly set up a boom or bust cycle, in which propensity falls,
recruiting becomes more difficult, and advertising funds have to be
restored. We presented a model that describes the conditions under which
it would be cost-effective to advertise in the interests of future enlistment
supply, and we reviewed research to date that speaks to the issue. While
extant research suggests that advertising may have effects only for a short
period of time, the data available to previous researchers are limited for
several reasons. First, they do not permit examining both lagged effects
and nonlinear effects within a time period. Second, they focus on adver-
tising aimed at youth at the point of the enlistment decision and do not
permit examining possible supplemental advertising approaches, such as
those aimed a youth several years prior to an enlistment decision or those
aimed at adult influencers, such as parents.
As a result, research to date does not permit a definitive answer to the
question of the cost-effectiveness of advertising above and beyond that
which is necessary to achieve current recruiting goals. We recommend a
focused effort to maintain advertising data in a systematic way for pur-
poses of estimating a supply curve that incorporates the potential for both
time-lagged and nonlinear advertising effects. We further recommend a
program of research, incorporating quasi-experimental methods, to
examine advertising effects over an extended period of time.
We then turn to the optimal levels of joint and Service advertising. It
is our opinion that certain types of advertising themes, such as generic
themes designed to increase overall propensity, are best done as a joint
program, while advertising themes featuring specific benefits of military
OCR for page 126
126
EVALUATING MILITARY ADVERTISING AND RECRUITING
service are best done in the Service programs. What we do not know is
whether there is an optimal level of joint and Service advertising or, more
specifically, what advertising fund level should be allocated to joint pro-
grams. We note that issues of scale play a role in addressing this question,
as certain types of advertising (e.g., television) do not appear to have a
constant effect across levels of expenditure. The larger the recruiting effort
and the larger the budget, the greater the potential value of a multifaceted
campaign, with some resources targeted toward providing information
about specific Services to those already with a propensity to enlist and
others targeted toward increasing propensity among those currently with-
out it.
We recommend a program of research aimed at examining the effects
and cost-effectiveness of information-oriented versus values-oriented
advertising in joint and Service advertising programs.
Representative terms from entire chapter:
awareness capital
