1. business and industrial

Implicit Discount Rates and the Purchase
of Untried, Energy-Saving Durable Goods
DOUGLAS A. HOUSTON*
Many consumers appear to rationally calculate the net worth of a household investment, but a substantial minority may lack the skills or alertness to perceive
an investment opportunity and initiate analysis. Intentions to engage in complementary activities play an important role in the investment decision.
W
are considered to be related to particular aspects of expected
costs. Other explanatory variables considered are: household i n c o m e , n u m b e r of individuals in the family, housing
type, h o u s i n g a g e , and h o u s e h o l d square footage.
At another level, persistent errors in j u d g m e n t may be
a significant factor in the b e h a v i o r of m a n y potential investors, as H a u s m a n s u g g e s t s — t h a t is, a clear perception of
investment opportunity m a y b e lacking in s o m e respondents
(Kirzner 1973). T h e s e h o u s e h o l d s m a y , m o r e frequently
than others, refuse to c o n s i d e r i n v e s t m e n t s , in spite of the
size of t h e r e t u r n . A s e c o n d p u r p o s e of the study is to
a n a l y z e t h e d i f f e r e n c e s in c h a r a c t e r i s t i c s b e t w e e n unresponsive h o u s e h o l d s and t h o s e that perceived more clearly
an opportunity to invest.
T h e e m p i r i c a l i n v e s t i g a t i o n is b a s e d o n data acquired
from a m a i l s u r v e y of a p p r o x i m a t e l y 3 , 0 0 0 households.
Data w e r e obtained on various aspects of energy use and
a n u m b e r of d e m o g r a p h i c characteristics of the households.
Households also r e s p o n d e d to a hypothetical question of
the necessary annual cost savings required to cause them
to purchase an e n e r g y - s a v i n g durable good with unspecified
characteristics. Initial p u r c h a s e and installation price for the
long-lived d e v i c e w e r e given to the r e s p o n d e n t s . By use of
qualitative c h o i c e m o d e l s , w e examined (1) the relationship
b e t w e e n a set of e x p l a n a t o r y v a r i a b l e s a n d the minimal
a c c e p t a b l e r e t u r n o n t h e i n v e s t m e n t , a n d (2) the factors
which influence a r e s p o n d e n t to avoid considering an investment p r o p o s a l , regardless of potential return.
A n u m b e r of caveats are in order. First, for the purpose
of this p a p e r , w e f o l l o w B e c k e r ' s ( 1 9 6 5 ) model of the
household as a single e c o n o m i c unit with shared consumption and p r o d u c t i o n . W h i l e criticisms of the " n e w home
e c o n o m i c s " a b o u n d (Ferber and Birnbaum 1977; Lancaster
1975; Willis 1973), this p a p e r focuses on the issue of the
investment decision of the h o u s e h o l d as a singular decisionm a k i n g unit. T o the extent that a household utility function
is inappropriate, the results of this paper should be tempered
by j u d g m e n t . S e c o n d , the study does not contain all economic and d e m o g r a p h i c variables likely t o affect the investment decision, and does not contain psychological information about r e s p o n d e n t s . F o r these reasons, as well as
hy d o s o m e consumers require 50 percent or greater
returns on h o u s e h o l d i n v e s t m e n t s ? W h y d o other
consumers reject investments yielding 50 percent or greater
returns? Based on empirical analyses of consumer investm e n t s in e n e r g y - u s i n g d u r a b l e g o o d s , s o m e e c o n o m i s t s
have argued that consumers b e c o m e confused and hesitant
when monetary paybacks stretch over many years; they m a y
b e unable to choose rationally among investment options.
Gately (1980) found that extremely high discount rates were
applied by consumers in the purchase of an energy-using
durable. Hausman (1979) estimated that very high implicit
d i s c o u n t rates w e r e u s e d by l o w e r i n c o m e c o n s u m e r s .
H a u s m a n argued that lower i n c o m e consumers were less
educated and may have suffered (relative to m o r e educated
individuals) from a defective 'telescopic f a c i l i t y " — a n inability to measure correctly the net worth of investments in
durables.
The purpose of this paper is to examine the factors which
influence a household's investment in an untried, energysaving d u r a b l e g o o d . At o n e l e v e l , w e s e e k t o test the
h y p o t h e s i s that a h o u s e h o l d i n v e s t m e n t d e c i s i o n is rational—that is, based upon an analysis of expected costs
and returns for an entire investment p r o g r a m . It would be
proper in a rational investment analysis to include all expected, user-borne costs of initial planning, p u r c h a s e , u s e ,
and resale or disposal. These costs (hereafter termed prog r a m costs) are frequently ignored in empirical analyses,
but if program costs vary substantially across the investing
household population, we should observe systematic, p o s itive variations between individual discount rates and these
additional costs. Higher discount rates would reduce the
likelihood of investment. B e c a u s e program costs are not
directly observable, proxies are used in this study. For exa m p l e , an i n d e x of prior e x p e r i e n c e with e n e r g y - s a v i n g
activities and an index of intentions to engage in energysaving activities are employed in regression analysis; both
4
*Douglas A. Houston is Assistant Professor of Business Economics,
School of Business, University of Kansas, Lawrence, KS 66045. He
thanks Bruce Jaffee for helpful comments on previous drafts of this paper.
236
© JOURNAL OF CONSUMER RESEARCH • V o l . 10 • September 1983
PURCHASE OF ENERGY-SAVING GOODS
the possibility of s o m e n o n r a t i o n a l b e h a v i o r , there is c o n s i d e r a b l e unexplained v a r i a n c e in the e s t i m a t i o n s . T h i r d ,
the investment p r o p o s a l is h y p o t h e t i c a l , so w e cannot be
confident that r e s p o n s e s w o u l d b e identical for an actual
i n v e s t m e n t . Last, t h e p a p e r d o e s not attempt to explain the
c o m p l e x decision p r o c e s s b y w h i c h the i n v e s t m e n t is anal y z e d . Therefore, w e c a n n o t d e t e r m i n e h o w t h e investment
d e c i s i o n is r e a c h e d , but c a n o n l y p r o v i d e s o m e e v i d e n c e
a b o u t the e c o n o m i c rationality of that decision p r o c e s s .
T H E INVESTMENT DECISION MODEL
Framework of the Problem
T h e respondents in o u r s t u d y c o n s i d e r e d the following
question:
1
I f an energy-saving device, with a very long useful life, could
b e purchased and installed in your home for $ 1 0 0 , approximately how much would you have to save in energy costs
e a c h year before you would install the device?
R e s p o n d e n t s w e r e g i v e n s e v e n r e s p o n s e options: (1) less
t h a n $ 1 0 per year; (2) $ 1 0 - $ 2 0 p e r year; (3) $ 2 1 - $ 3 0 per
y e a r ; ( 4 ) $ 3 1 - $ 4 0 p e r year; (5) $ 4 1 - $ 5 0 per year; (6) m o r e
t h a n $ 5 0 per year; a n d (7) d o n ' t k n o w or u n c e r t a i n .
F a c e d with s u c h a c o s t - s a v i n g i n v e s t m e n t c h o i c e , m a n y
c o n s u m e r s were u n c l e a r a b o u t h o w to p r o c e e d . Q u i t e likely,
s i m p l e rules of t h u m b w e r e a p p l i e d to the p r o b l e m . In sel e c t i n g a particular return (or c h o o s i n g not to r e s p o n d ) , the
d e c i s i o n , to be r a t i o n a l , m u s t i n c o r p o r a t e information about
e x p e c t e d costs a n d r e t u r n s . In t e x t b o o k e x a m p l e s of inv e s t m e n t analysis, t h e c o s t s , r e t u r n s , and d e g r e e of uncert a i n t y a r e generally treated as a m a n a g e a b l e , m a t h e m a t i c a l
e x e r c i s e . On the s u r f a c e , the q u e s t i o n p o s e d about the ene r g y - s a v i n g device m a y a p p e a r to b e easily a p p r o a c h e d b y
a m a t h e m a t i c a l , cost-benefit, p r e s e n t - v a l u e a n a l y s i s . G i v e n
that f r a m e w o r k , t h e c h o i c e v a r i a b l e then is the net implicit
d i s c o u n t rate.
W e h y p o t h e s i z e that the i n f o r m a t i o n p r o v i d e d to the c o n s u m e r i n the i n v e s t m e n t q u e s t i o n is insufficient to allow a
c o m p l e t e analysis, e v e n t h o u g h t h e question clearly states
a n n u a l i z e d cost s a v i n g s . F i r s t , e a c h c o n s u m e r , p r i o r t o
c o m m i t t i n g to an i n v e s t m e n t , h a s an u n d e r s t a n d i n g of his
or h e r skillfulness a n d k n o w l e d g e in u n d e r t a k i n g the inv e s t m e n t p r o g r a m . T h e u s e r - b o r n e p r o g r a m c o s t s — c o s t of
a n a l y s i s , cost of h o u s e h o l d l a b o r in i m p l e m e n t a t i o n , and
c o s t s a s s o c i a t e d with r e s a l e o r d i s p o s a l — s h o u l d be conside r e d b y t h e rational i n v e s t o r ; t h e s e p r o g r a m costs will, in
g e n e r a l , vary a m o n g h o u s e h o l d s . S e c o n d , the d e v i c e ' s p o t e n t i a l c o m p l e m e n t a r i t y (or substitutability) with existing
h o u s e h o l d energy s t o c k s m a y alter the value of the device
and t h e c h o s e n d i s c o u n t r a t e . T h e s e factors w o u l d b e estim a t e d b y the r a t i o n a l i n v e s t o r a n d i n c l u d e d in t h e c o s t benefit analysis.
F o r t h e e n e r g y - s a v i n g i n v e s t m e n t proposal investigated
' R e s p o n d e n t s are individual households in southern and central Indiana.
The d e t a i l s of the data collection method, selection of respondents, and
c h a r a c t e r i s t i c s of the sample are presented later in the paper.
237
in this p a p e r , the various expected costs and values through
t i m e (both m o n e t a r y and nonpecuniary) can be summarized
in the c h o i c e by the c o n s u m e r of a minimally acceptable
a n n u a l i z e d m o n e t a r y cost savings. That choice implies the
e x i s t e n c e of a gross discount rate (GDR) used in the decision. T h e t e r m " g r o s s " is employed to emphasize that the
return selected b y respondents m a y be based on a n u m b e r
of additional p r o g r a m costs and savings beyond those stated
o n the q u e s t i o n n a i r e .
Interpretation of the Response
to the Investment Question
B e c a u s e the hypothetical investment question resulted in
n o a c t u a l c o s t s o r r e w a r d to the r e s p o n d e n t , a p o t e n t i a l
p r o b l e m arises in interpretation of the gross discount rate
r e s p o n s e s . A r e w e e x a m i n i n g the decision that w o u l d be
m a d e in real life? A j u d g m e n t is necessarily speculative.
W h i l e t h e study could have been performed for actual rather
than hypothetical investment behavior, there are two major
p r o b l e m s with the former approach. First, the m o d e l dev e l o p e d in this p a p e r requires detailed information about
h o u s e h o l d s that rarely can b e obtained for households that
h a v e all c o n s i d e r e d the s a m e investment. Second, it is the
e x p e c t e d cost savings that are important to the investment
decision rather than realized savings. T h e study uses the
hypothetical question to achieve the notion of expected cond i t i o n s , b u t it m a y n o t b e fully r e p r e s e n t a t i v e of a c t u a l
i n v e s t m e n t behavior, since n o monetary penalty or r e w a r d
results from the decision.
T h e r e is e v i d e n c e that m a n y respondents answered carefully. At the end of the completed survey, in space provided
for c o m m e n t s , m a n y individuals indicated their concerns
a b o u t t h e s u r v e y i n v e s t m e n t q u e s t i o n . S o m e a d d e d ass u m p t i o n s , such as the high probability that their household
c o u l d n o t a c h i e v e the stated s a v i n g s d u e to current low
energy u s a g e . M a n y expressed skepticism that any energy
device truly achieves its stated cost savings after all cost
factors are c o n s i d e r e d . Such comments suggest that m a n y
r e s p o n d e n t s did not take the explicit cost-saving statement
as the c o m p l e t e picture. Instead, they m a y have modified
their i n v e s t m e n t analysis according to individual p e r c e p tions of o t h e r expected p r o g r a m costs and savings.
A few c o n s u m e r s seemed to believe that the investment
w o u l d in s o m e w a y lead to loss of a currently useful househ o l d d u r a b l e — i . e . , that t h e e n e r g y - s a v i n g d e v i c e w a s a
r e p l a c e m e n t rather than an add-on item. Such an interpretation w a s not intended by the survey team. Under replacem e n t c o n d i t i o n s , a rational investor would reduce the net
present value obtained from the new investment by the rem a i n i n g present value of returns from the existing durable
to b e r e p l a c e d , less any s c r a p v a l u e . T o the extent that
2
Skepticism about payoffs from household energy investments may be
a reasonable response, if prior information tended to be overly optimistic
about expected savings. For example, the effect of installing ceiling insulation on energy usage has tended to be overestimated by both marketers
and government agencies. See Consumer Reports (1978).
THE JOURNAL OF CONSUMER RESEARCH
238
respondents interpreted the investment as a r e p l a c e m e n t
rather than as a new durable, the required return on investment and the observed implicit discount rate were increased.
Some respondents indicated confusion about h o w to proceed in determining the " a p p r o p r i a t e " response. Others
indicated that insufficient information was provided to allow an answer they could support. These and other respondents who offered no explanation of how they answered
may have been perplexed by the question. Confusing conditions are generally present (often in much greater degree)
in a consumer's purchase decision about any untried durable
good. Clearly, more precise and e x t e n s i v e information
would have made the task easier, but it also would have
altered an important, interesting aspect of the investment
problem. Many consumers, facing an unusual investment
proposal, may lack the alertness to conceptualize clearly
the costs and benefits. These confused respondents affect
our analysis in two ways. First, many individuals (approximately one-third of the sample) responded to the investment proposal with " d o n ' t know or u n c e r t a i n . " Second,
the nonrational selection of positive returns by confused
individuals is reflected in the weakening of the economic
model of implicit discount rate, reducing its explanatory
power.
A flow chart of the decision process is s h o w n in the
Figure. We expect respondents who have the alertness and
analytic ability to select a positive savings n e c e s s a r y to
induce investment. In general, these individuals rationally
augment the questionnaire's statement of costs and savings
with other user-borne program costs prior to deciding on
a necessary return. Although it is possible for a rational
individual to select a " d o n ' t know or u n c e r t a i n " response,
the ability to select the highest return (open-ended upward)
should provide an appropriate response for individuals w h o
are highly uncertain about costs or expect to bear high program costs. The response of " d o n ' t k n o w or u n c e r t a i n "
is generally nonrational. Many individuals selecting it lack
a decision rule (rule of thumb) to guide t h e m in comprehending the data. Unfortunately, some of these individuals
may simply randomly select a positive return, thus weakening the linkage of explanatory factors to the discount rate
in estimation.
W e treated the two groups (positive response or " d o n ' t
k n o w " response) separately in empirical analyses. T h o s e
answering with a positive savings r e q u i r e m e n t w e r e e x amined in a tobit regression model of t h e d i s c o u n t r a t e
level. " D o n ' t know or uncertain" answers were examined
in a probit regression of the choice b e t w e e n the " d o n ' t
k n o w " versus a positive return response.
3
Some "don't know or uncertain*' responses, however, can be economically rational. If a respondent expects that the cost of trying to understand
the question sufficiently well to answer thoughtfully is greater than the
value of that effort, he may rationally choose a "don't know or uncertain"
answer. Yet many who would fit into this category may have applied a
similar logic to the entire questionnaire and are therefore nonrespondents.
3
DATA SOURCE
All information used in the estimations was taken from
a m a i l s u r v e y of i n d i v i d u a l h o u s e h o l d s . Approximately
3 , 0 0 0 residential households serviced by 18 rural electric
m e m b e r s h i p cooperatives ( R E M C s ) in southern and central
I n d i a n a were surveyed. T h e s a m p l e was selected from the
r e c o r d s of t h e c o o p e r a t i v e s by a r a n d o m sampling techn i q u e . T o focus upon residential customers having similar
o p p o r t u n i t i e s to a p p l y an e n e r g y - s a v i n g device to their
h o u s e h o l d operations, those households that were engaged
in farming, that did not o w n their dwelling, or that lived
in the unit for only part of the y e a r were eliminated from
t h e analysis. C e n s u s data for h o u s i n g , population, and inc o m e were used to c o m p a r e the respondents with the regional population. T h e respondents were found to be representative.
T h e questionnaire included 35 items related to various
aspects of energy c o n s u m p t i o n in the home, appliance o w n ership and u s e , measures of past a n d intended energy conservation m e a s u r e s , selected demographic characteristics of
t h e household, and an energy-saving investment proposal.
T h e questionnaire was mailed in the summer of 1979 and
w a s returned, fully c o m p l e t e d , b y 8 3 percent of the households.
T h e high response rate was obtained without followup
contact after the initial mailing. Individual users own the
cooperatives and therefore may h a v e felt a commitment to
r e s p o n d to the survey. Prior to t h e mailing of the survey,
t h e g e n e r a l m a n a g e r of e a c h c o o p e r a t i v e sent a letter to
e a c h selected household in his or h e r cooperative requesting
assistance and stressing the importance of the survey to the
cooperative. Enclosed with each mailed survey was a dollar
bill that the recipient was told to k e e p , regardless of whether
the 35-question survey w a s c o m p l e t e d . At a minimum, the
dollar bill m a y h a v e p r o m p t e d a fairly quick response. Interestingly, m a n y people w h o did not complete the questionnaire returned the dollar bill.
Restricting the study to h o u s e h o l d s occupied for the full
y e a r and h o u s e h o l d s without significant farm-related opera t i o n s r e d u c e d t h e n u m b e r of o b s e r v a t i o n s to 1,081. Of
t h e s e , 7 2 5 r e s p o n d e d to t h e i n v e s t m e n t question with a
positive return and 356 a n s w e r e d " d o n ' t know or uncertain."
T h e r e are clearly s o m e p r o b l e m s of measurement with
respect to h o u s e h o l d square footage, annual income, and
the listing of past and intended conservation efforts. For
e x a m p l e , individuals usually h a v e an imprecise notion of
the heated square footage in their homes. These problems
are largely a shortcoming of the survey form of information
gathering.
THE EMPIRICAL STUDY
Tobit Regression Analysis
B a s e d on p o s i t i v e s a v i n g r e s p o n s e s to the investment
question, a gross implicit d i s c o u n t rate (GDR) was calcu-
PURCHASE OF ENERGY-SAVING GOODS
239
FIGURE
THE INVESTMENT PROBLEM
THE JOURNAL OF CONSUMER RESEARCH
240
lated for e a c h respondent. T h e discount rates are directly
calculated o n the assumption that the returns per year are
p e r p e t u a l . T h e perpetual rates are 0 , 10, 2 0 , 30, 4 0 , and
5 0 percent for m e a n annual cost saving responses of $ 0 ,
$ 1 0 , $ 2 0 , $ 3 0 , $ 4 0 , and $ 5 0 , respectively.
T h e multivariate regression m o d e l , GDR = XB, where
X is a matrix of explanatory variable observations and B is
the vector of coefficients, can b e estimated by O L S . H o w ever, O L S will yield biased estimators if there are m a n y
observations clustered at the limiting value of the dependent
variable (a response of 50 percent). This information can
b e applied t o the estimation of the discount r a t e . A tobit
m o d e l is useful in dealing with such a truncated, dependent
variable in multivariate regression analysis (Tobin 1958).
T o b i t is a h y b r i d of probit and m u l t i p l e r e g r e s s i o n . It
yields answers to t w o questions in this study: (1) the probability of a participant responding that the stated payoffs
were sufficient to warrant investment, and (2) the necessary
discount rate to entice investment. In using this procedure,
an i n d e x , /., is estimated which is linearly dependent upon
a set of i n d e p e n d e n t v a r i a b l e s . I = XA, w h e r e / is u n b o u n d e d and A is a vector of normalized coefficients. Each
observation of the dependent variable, GDR , is a function
of the same explanatory variable set, X, through the / variable and a critical value of the index, / * . If the upper b o u n d ,
/ * , is greater than /,., then GDR is the difference between
/, and the critical / * .
T h e probability that GDR is less than the limit can be
e x a m i n e d for any observation b y using the standard normal
cumulative distribution to transform /, into probabilities.
T h e normalized coefficients in the m o d e l then can be transf o r m e d to r e g r e s s i o n coefficients b y m u l t i p l y i n g b y the
standard error of estimate. Hypothesis testing can be performed using the asymptotic standard errors or likelihood
ratio tests.
T h o s e households that answered " d o n ' t k n o w or uncert a i n " ( a p p r o x i m a t e l y o n e - t h i r d of t h e s a m p l e ) w e r e e x cluded from the analysis of the discount rate level. Because
the investment question allowed any level of positive savings to be indicated through a r e s p o n s e of " $ 5 0 or m o r e , "
the failure to m a k e a positive response m a y indicate that
the " d o n ' t k n o w or u n c e r t a i n " respondents were unable to
formulate the problem sufficiently well to perceive an opp o r t u n i t y . T h u s c o s t s a v i n g s on t h e i n v e s t m e n t m a y be
meaningless to t h e m . Using a probit m o d e l , we e x a m i n e d
the factors that affect the probability of a " d o n ' t k n o w "
response.
4
The Empirical Model of the Discount Rate
T h e c o n s u m e r ' s choice of a m i n i m u m return on investm e n t is e m p i r i c a l l y e x a m i n e d w i t h a d i s a g g r e g a t e tobit
model:
GDR
Choosing a different investment life will slightly alter the discount
rates, but the estimated coefficients do not appear sensitive. When a 20year life is used, very few differences emerge in regression results.
Regardless of the regression model used, the results are robust, providing similar regression coefficient values and levels of statistical significance for the explanatory variables. Another regression model that could
be used, in lieu of tobit, is an extension of the dichotomous probit model
which allows a categorical, discontinuous variable (see McKelvey and
Zavoina 1975).
4
5
+ B EXP
EXP
INT
INC
HSF
NUM
DTY
DAG
+
X
BJNT
+ B INC
+ B HSF
+ B DTY
+
6
where:
GDR
t
t
0
3
5
t
= B
4
+
+ B NUM
5
(1)
B DAG
n
= the gross, implicit discount rate, implicitly selected by the r e s p o n d e n t t h r o u g h his or h e r
choice of a m i n i m u m annual savings from the
investment
- a variable of major h o u s e h o l d energy conservation measures performed within the last 2 4
m o n t h s ; 11 options w e r e equally weighted in a n
index (see A p p e n d i x for details)
= a variable of stated intentions to invest in e n e r g y c o n s e r v i n g m e a s u r e s in the coming 12
m o n t h s ; 11 options w e r e equally weighted in a n
index (see A p p e n d i x for details)
= annual i n c o m e for the household
= heated square footage of the household
= n u m b e r of i n d i v i d u a l s l i v i n g full-time in t h e
household
= d u m m y variable for h o u s i n g type (1 = mobile
h o m e ; 0 = conventional)
= d u m m y variable for housing construction d a t e
(1 = p r e - 1 9 7 4 ; 0 = 1974 or newer)
In estimating Equation 1, w e seek to explain the relationship between the level of the discount rate for the inv e s t m e n t and a set of e x p l a n a t o r y e c o n o m i c and d e m o g r a p h i c v a r i a b l e s . T h e e x p e c t e d s i g n s of t h e r e g r e s s i o n
coefficients are displayed above the variables in the equation.
Testable Hypotheses
T h e tobit model of the gross discount rate level (Equation
1) provides the basis for a n u m b e r of testable hypotheses.
In general, to be consistent with e c o n o m i c a l l y rational b e havior, the model should support a positive relationship
b e t w e e n higher expected p r o g r a m costs to the investor a n d
the required return on investment (indicated by GDR). A s
d i s c u s s e d , m a n y e x p e c t e d u s e r - b o r n e c o s t s m a y be e x cluded from the explicit statements of costs and savings in
the mail survey. Factors indicating higher p r o g r a m costs to
a c o n s u m e r (compared to others in the sample) should p u s h
u p the individual's discount r a t e . C o n v e r s e l y , factors indicating reduced relative e x p e c t e d p r o g r a m costs should
reduce the discount rate necessary to make the investment
w o r t h w h i l e . T h e v a r i a b l e s i n c l u d e d in E q u a t i o n 1 a r e
briefly discussed in w h a t follows.
PURCHASE OF ENERGY-SAVING GOODS
241
1. Greater experience
(EXP) and more planned
actions
for saving energy (INT) should negatively
affect the discount rate (GDR).
Tversky and K a h n e m a n (1973) and
Kahneman and T v e r s k y (1972) h a v e s h o w n that as events
become harder for an individual to picture, s/he will consider the events less likely to occur. T h u s , past experience
with investment a n a l y s i s for similar types of durables or
the present, o n g o i n g consideration of similar investments
should m a k e a c o n s u m e r m o r e willing to engage in analysis.
Prior e x p e r i e n c e w i t h similar investments m a y alter perception of investing skills. F o r an investment in a technology with m a n y u n c l e a r or unstated attributes, a c o n s u m e r ' s
d e p e n d e n c e on p r i o r e x p e r i e n c e w i t h i n v e s t m e n t g o o d s
sharing similar g e n e r a l characteristics is hypothesized to
r e d u c e the e x p e c t e d c o s t from i m p l e m e n t a t i o n of an investment p r o g r a m . F o r e x a m p l e , the recent purchase and
installation of insulation or a heating unit may increase a
c o n s u m e r ' s confidence that s/he can easily incorporate more
i m p r o v e m e n t s into the h o u s e h o l d .
(1979) o b s e r v e s , as i n c o m e and marginal tax rate rise, s o m e
added s a v i n g s a c c r u e to the investor in c o n s u m e r d u r a b l e s ,
r e d u c i n g t h e n e e d e d return.
T h e first t w o effects (suggesting a negative i n c o m e - t o GDR r e l a t i o n s h i p ) m a y b e countered by the incentive for
higher i n c o m e families t o increase w o r k outside the h o m e ;
this s t a t e m e n t p r e s u m e s that a higher income level is indicative of increased labor earning power. Substitution
a w a y from h o u s e h o l d production activities w h e n real w a g e s
rise has b e e n o b s e r v e d , a m o n g others, by M e n e f e e (1982).
T h e r e s p o n d e n t s ' p e r c e p t i o n s of the complementarity of labor input t o the o p e r a t i o n of the device is u n k n o w n . If, in
fact, the capital services from the investment are perceived
as substitutes for l a b o r , then additional implicit cost saving
w o u l d o c c u r , and the gross implicit discount rate w o u l d be
affected n e g a t i v e l y . A l t h o u g h there may b e s o m e varying
directions of influence, the overall effect of i n c o m e on the
discount r a t e is p r e d i c t e d to b e negative.
Where intentions to e n g a g e in p l a n n e d energy-saving activities are s t r o n g , there m a y be expected cost savings ass o c i a t e d w i t h i m p l e m e n t i n g a n d m a n a g i n g a n u m b e r of
complementary i n v e s t m e n t s . With many similar investm e n t s e x p e c t e d , a c o n s u m e r is likely to view the costs of
adding another as l o w ; an investment in learning h o w to
u s e such durables h a s already b e e n " s u n k . " T h e intention
variable also m a y b e positively correlated with c o n s u m e r
attitudes favoring c o n s e r v a t i o n as a desired end in itself. If
s o , a c o n s u m e r r e s p o n d i n g with high n u m b e r s for the INT
v a r i a b l e w o u l d r e q u i r e a s m a l l e r m o n e t a r y p a y b a c k . It
should be e m p h a s i z e d , h o w e v e r , that the intentions variable
is not an attempt to directly m e a s u r e attitudes.
B e c a u s e t h e t w o v a r i a b l e s , EXP a n d INT, are indices
formed as equally w e i g h t e d s u m m a t i o n s of 11 different possible conservation a c t i o n s , the reliability of both indices
should be e x a m i n e d . A n internal consistency p r o c e d u r e , the
C r o n b a c h alpha coefficient, w a s e m p l o y e d to e x a m i n e reliability. T h e coefficients for EXP and INT were 0 . 6 5 and
0 . 6 8 , respectively. W h i l e there is no statistical test for acc e p t a n c e or rejection of the v a l u e , alpha values of this size
g i v e some d e g r e e of confidence that the indices are reliable
(Nunnally 1978, p p . 2 2 9 - 2 3 0 ) .
3. Greater household
size, measured in square
footage
of heated dwelling
space (HSF), should negatively
affect
the discount
rate.
For household square footage to directly r e d u c e GDR, the respondent must expect cost savings
(beyond the direct functioning of the device) from a larger
h o u s e h o l d . H o u s e h o l d size, for e x a m p l e , may b e positively
correlated w i t h h e a t i n g and cooling system stocks and other
e n e r g y - u s i n g c a p i t a l . If a respondent expects scale econo m i e s with respect to the energy-using household capital,
then additional s a v i n g s m a y be anticipated b e y o n d those
stated for t h e d e v i c e .
H o w e v e r , h o u s e h o l d size m a y be positively correlated
with w e a l t h . If greater wealth reduces the overall savings
rate of c o n s u m e r s (see Ferber 1973), the HSF variable m a y
be a c t i n g , in p a r t , t o r e d u c e i n v e s t m e n t s in a h o u s e h o l d
durable. O t h e r lines of causality could be presented. A spurious c a u s e of a p o s i t i v e r e l a t i o n s h i p b e t w e e n HSF and
GDR is d u e to the correlation of HSF with i n c o m e . T h e
s i m p l e c o r r e l a t i o n b e t w e e n t h e t w o v a r i a b l e s is -f-0.46.
Therefore, i n c o m e a n d household square footage effects
m a y best be interpreted jointly.
A d d i t i o n a l d u m m y v a r i a b l e s for m o b i l e h o m e v e r s u s
conventional h o m e (DTY) and for older (pre-1974) h o m e s
(DAG) w e r e i n c l u d e d in the empirical model to control for
variability d u e t o structural differences in dwellings.
2 . Higher household
income (INC) should
negatively
affect the discount rate.
T h e r e are several effects of inc o m e upon the d i s c o u n t rate level. First, i n c o m e m a y be
positively correlated with educational level. Individuals
w i t h more e d u c a t i o n m a y b e m o r e capable of understanding
t h e investment p r o p o s a l than those with less education. If
l o w - i n c o m e families r e c o g n i z e a high probability of their
m a k i n g large e r r o r s , t h e n they rationally choose high rates
o f return to c o m p e n s a t e for the r i s k . S e c o n d , as H a u s m a n
6
Lower income families also may tend to ignore the investment and
respond ''don't know or uncertain." These responses are not included in
the analysis of positive discount rates; they are examined in a following
section.
6
4 . As the number
of household
members
(NUM)
increases, the required discount rate should increase.
There
are two influences t o consider. First, as the n u m b e r of individuals u n d e r o n e roof increases, the i n c o m e available to
each falls. T h e r e is considerable evidence that lower p e r
capita i n c o m e leads to a decrease in the short-run savingsto-consumption ratio (Ferber 1973). A s s u m i n g the average
household m e m b e r ' s well-being is considered in spending
decisions, as i n c o m e falls, an increase in the required discount rate on any g i v e n investment is expected.
T h e s e c o n d influence concerns the complementarity or
substitutability of h o u s e h o l d labor for the capital services
THE JOURNAL OF CONSUMER RESEARCH
242
of t h e p r o p o s e d investment. Labor complementarity suggests a d o w n w a r d revision in the stated cost savings from
the d e v i c e . H o w e v e r , if the new capital is expected to be
a substitute for s o m e household labor, then the cost savings
m a y b e e n h a n c e d . T h e true relationship is not clear. Another p r o b l e m of inferring expected labor cost for the n u m ber of individuals in the household is the u n k n o w n quality
of t h e l a b o r . F o r e x a m p l e , small c h i l d r e n m a y b e u n a c ceptable " w o r k e r s . " On balance, the per capita income
effect is expected to b e larger than any effects of labor in
household production, and the expected sign of NUM is
positive.
T o s u m m a r i z e , the expected effects of explanatory variables upon the h o u s e h o l d ' s selection of a discount rate are:
TABLE 1
RESULTS OF TOBIT ANALYSIS
Normalized
coefficient
Variable name
Housing type (DTY)
Housing age (DAG)
Number (NUM)
Experience (EXP)
Income (INC)
Intentions (INT)
Home size (HSF)
Constant
R
C
2
d
T-ratio
0.20427
1.56
-0.24192
-2.92
0.07601
2.65
-0.00841
-0.42
-0.00001
-0.98
2.66
-0.08133
3.47
-0.00019
1.6242
10.29
.0563
3
Regression
coefficient
3.8824
-4.5981'
1.4447'
-0.1559
-0.0001
-1.5459'
-0.0037'
30.870'
Elasticity
b
e
+ .197
-.059
-.227
Given nonlimit observation; upper limit response = 50 percent. Dependent variable
= gross, implicit discount rate.
Elasticities calculated at mean values (percentage change in size of discount rate for
small percentage change in explanatory variable, given nonlimit observation).
Dummy variable for housing type (1 = mobile home; 0 = conventional).
Dummy variable for housing age (1 = pre-1974; 0 = 1974 or newer).
Significant for 1 (statistical tests are based on asymptotic properties of normalized
coefficients).
'Significant for 1 .
a
• Greater income or greater household square footage is expected to reduce an individual household's discount rate.
Because of collinearity between these two explanatory
variables, individual effects on GDR may be obscured and
only a joint interpretation of effect may be reasonable.
• Both experience with energy-conserving activities and intentions to engage in such activities are hypothesized to
reduce the discount rate.
c
d
9 0
9 9
• More household members are expected to increase the discount rate.
R e l a t i o n s h i p s b e t w e e n e x p l a n a t o r y v a r i a b l e s and the
level of t h e discount rate (dependent variable) can b e interpreted as follows:
• The higher the discount rate, the less likely that a similar
real investment will be undertaken.
• Thus any variable that positively (negatively) affects the
discount rate will reduce (increase) the likelihood of a
household undertaking energy-conserving investments.
Results of the Discount Rate Estimation
B e c a u s e the cost s a v i n g r e s p o n s e is t r u n c a t e d ( $ 5 0 or
m o r e ) , a tobit model was used in estimating the discount
rate with t h e limiting value of the discount rate at 50 percent. A t t h e mean of continuous variables and modal values
of the d u m m i e s , the probability of a nonlimit observation
(positive return less than 50 percent) is 8 7 . 9 percent; the
observed frequency is 87.7 percent. Table 1 shows the normalized coefficients, r-ratios, estimated regression coefficients, and the elasticities at the m e a n s of continuous variables. T h e regression coefficients and elasticities apply to
the relationship between the explanatory variables and the
level of t h e gross discount rate, given that GDR is a nonlimit observation. S u m m a r y statistics for the explanatory
variables are s h o w n in Table 2 (left c o l u m n s ) , and the simple correlation coefficient matrix is shown in the u p p e r section of T a b l e 3 .
C h a n g e s in the level of the discount rate are most sensitive to t h e h o u s e h o l d square footage (HSF). A l s o important in t e r m s of the percentage change in the discount rate
l e v e l r e s u l t i n g from a p e r c e n t c h a n g e in an e x p l a n a t o r y
variable (elasticity) are n u m b e r of family m e m b e r s (NUM)
and intentions to conserve (INT):
Elasticity
Household square footage
Number in family
Intentions to conserve
— .227
+ .197
— .059
A l t h o u g h t h e s e t h r e e v a r i a b l e s act in a c c o r d with the
h y p o t h e s e s , i n c o m e , surprisingly, does not. T h e coefficient
is negative but insignificant. T h e r e are a n u m b e r of possible
explanations for the result. First, the high correlation bet w e e n i n c o m e a n d h o u s e h o l d s q u a r e f o o t a g e ( + 0.46)
m a k e s it difficult t o i n t e r p r e t t h e t w o v a r i a b l e s independently. T h u s , the testing of independent effects on the discount rate from each of these variables m a y not be meaningful. Second, it m a y b e that t h e various counterbalancing
influences of t h e i n c o m e v a r i a b l e p r e v i o u s l y discussed
weakens the linkage to the discount rate. Last, the overall
magnitude of the effect of i n c o m e on the discount rate may
be slight, m a k i n g findings of statistical significance difficult
to obtain. Interestingly, the role of i n c o m e is greatly expanded in explaining the choice of a " d o n ' t know or unc e r t a i n " r e s p o n s e , suggesting that at this m o r e elementary
level of c h o i c e , i n c o m e plays a different role. This is discussed in the following section.
For energy-saving d u r a b l e s , a d a m p e n i n g influence on
investment m a y result from a trend decline in the size of
housing units. F o r e x a m p l e , the construction of smaller new
h o m e s m a y l e a d to a d e c l i n e in t h e a c c e p t a n c e of new
energy-saving i n v e s t m e n t s . A countering effect may be a
decline in the average family s i z e — a trend which would
act to l o w e r t h e r e q u i r e d d i s c o u n t rate a n d increase the
probability of investment.
Although household size and n u m b e r in the family are
important variables, the intentions to engage in conserving
activities (INT) m a y h a v e a m o r e important short-term influence on energy-saving investments. Household square
PURCHASE OF ENERGY-SAVING GOODS
243
TABLE 2
SUMMARY STATISTICS OF VARIABLES
Positive savings responses only
(725 observations)
Variable name
Housing type (DTY)
Housing age (DAG)
Number (NUM)
Experience (EXP)
Income (INC)
Intentions (INT)
Home size (HSF)
a
b
a
b
"Don't know or uncertain" responses
(356 observations)
Mean
Standard
deviation
Mean
Standard
deviation
0.10897
0.67724
3.2938
1.9117
19,612.
0.92690
1,485.7
0.31181
0.46785
1.3748
1.9032
9,863.
1.2411
801.82
0.17697
0.80337
3.0758
1.5112
12,969.
.5787
1,032.2
0.38218
0.39801
1.5771
1.8877
8,781.
1.0546
752.18
Dummy variable for housing type (1 = mobile home; 0 = conventional).
Dummy variable for housing age (1 = pre-1974;0 = 1974 or newer).
f o o t a g e , i n c o m e , a n d n u m b e r of f a m i l y m e m b e r s are
" s t o c k " v a r i a b l e s and generally c h a n g e quite slowly. On
t h e other h a n d , large p e r c e n t a g e c h a n g e s in the intentions
i n d e x are not i m p r o b a b l e w h e n the current level of intent i o n s is l o w . F o r e x a m p l e , w h e n the current intentions index
is at 1 (activity), an i n c r e a s e by j u s t 1 in intended activities
w o u l d decrease the estimated discount rate from 2 2 . 2 to
2 0 . 9 percent in o u r m o d e l . T o obtain a like c h a n g e in the
discount rate, m e a n family size w o u l d h a v e to rise from
3 . 2 9 to 4 . 2 7 m e m b e r s . B e c a u s e i n c o m e a n d h o u s e h o l d
s q u a r e footage are h i g h l y correlated, an individual variable
interpretation m a y be i m p r o p e r . If considered jointly, inc o m e and h o u s e h o l d square footage still m u s t rise far above
n o r m a l trend to h a v e the impact of a 1-unit increase in the
intentions i n d e x . P r o m o t i n g the c o n c e p t of energy-conservi n g activities m a y be o n e w a y of altering the intentions of
c o n s u m e r s to p e r f o r m these activities. T h u s successful prom o t i o n s that substantially increase intentions should h a v e
a large influence on the actual a c c e p t a n c e rate of untried,
energy-saving d u r a b l e g o o d s .
7
T h e h y p o t h e s i z e d n e g a t i v e effect of the experience varia b l e on the d i s c o u n t rate w a s not o b s e r v e d to be statistically
significant. E x p e r i e n c e and intentions are not necessarily
p o s i t i v e l y l i n k e d ; e a c h m a y p l a y a different r o l e in the
m o d e l . The s i m p l e correlation coefficient in this sample is
+ O . 0 0 8 . C o r r e l a t i o n c o u l d be n e g a t i v e if, for e x a m p l e ,
t h e r e were a m a j o r revision in c o n s u m e r p l a n s . T h e role of
e x p e r i e n c e in p r o v i d i n g a s s u r a n c e to an individual that s/he
h a s some basic c o m p e t e n c y in an investment area m a y be
i m p o r t a n t in e x p l a i n i n g the decision to e x a m i n e an investm e n t proposal on e c o n o m i c g r o u n d s . That subject is disc u s s e d in'the n e x t section.
W h e t h e r h o u s i n g is m o b i l e h o m e or conventional (DTY)
is statistically significant at the 90-percent confidence int e r v a l . T h e p o s i t i v e s i g n o f DTY m a y b e influenced by
collinearity with h o u s e h o l d s q u a r e footage (HSF). T h e sim-
Income may be considered as a stock (of wealth) if it is interpreted as
a proxy for permanent income.
7
pie correlation coefficient of HSF to DTY is — 0 . 3 2 5 . Surprisingly, D A G , the d u m m y on h o u s i n g age (1 = p r e - 1 9 7 4
construction), h a d a n e g a t i v e effect on the discount rate.
An explanation of this result is not o b v i o u s . O n e line of
r e a s o n i n g is t h a t o l d e r h o u s i n g u n i t s are o n a v e r a g e in
poorer repair so that t h e probability of an e n e r g y - s a v i n g
device achieving c l a i m e d results m a y be considered h i g h .
T h e mean o b s e r v e d discount rate of 22.5 percent is not
unusually high for an i n v e s t m e n t in household d u r a b l e s .
H a u s m a n ( 1 9 7 9 ) , in a study of c o n s u m e r purchases of energy-saving a p p l i a n c e s , found a 2 6 . 4 percent m e a n level of
the discount r a t e . H e estimated m u c h higher rates for low
income levels and m u c h l o w e r rates for very high i n c o m e
levels. By contrast, o u r results d o not suggest this linkage
of the discount rate level to i n c o m e . T h e discount rates,
calculated by i n c o m e r a n g e s , are:
Annual
income range
Discount rate
Under $5,000
$5,000-$ 10,000
$10,001-$15,000
$15,001-$25,000
Over $25,000
21.6%
25.4%
25.3%
23.0%
19.9%
These results suggest that, o n c e a c o n s u m e r is willing to
consider an i n v e s t m e n t in t e r m s of expected costs and benefits, income m a y p l a y a w e a k role in predicting the discount rate level and t h e c o n s e q u e n t acceptance rate. In the
next section, w e e x a m i n e the role of income in the decision
not to attempt an e c o n o m i c appraisal of the i n v e s t m e n t .
The "Don't Know or Uncertain" Response
H e r e w e look briefly at the factors that m a y p u s h a h o u s e h o l d r e s p o n d e n t i n t o t h e " d o n ' t k n o w or u n c e r t a i n " r e sponse on the i n v e s t m e n t q u e s t i o n . W e h a v e argued p r e v i o u s l y that t h i s r e s p o n s e i s , in g e n e r a l , m a d e w i t h o u t
consideration of either future costs or values surrounding
the durable. A decision m a d e w i t h o u t examining costs o r
benefits cannot b e c o n s i d e r e d as rational investment anal-
THE JOURNAL OF CONSUMER RESEARCH
244
TABLE 3
SIMPLE CORRELATION COEFFICIENT MATRICES
A. Positive return respondents:
Variable
DTY
DAG
NUM
EXP
INC
INT
HSF
VAR
1.00000
-0.00475
-0.05867
-0.07920
-0.23658
-0.04720
-0.32477
1.00000
-0.05208
-0.07548
-0.14029
0.04970
-0.15728
DTY
DAG
1.00000
0.08119
0.19194
0.10084
0.21622
NUM
1.00000
0.00341
0.00837
0.02379
EXP
1.00000
0.07655
0.45788
INC
1.00000
0.01756
INT
1.00000
HSF
1.00000
-0.13324
-0.12828
-0.09785
-0.07043
-0.17657
DAG
1.00000
0.09765
0.27892
0.18017
0.15246
NUM
1.00000
0.09261
0.22879
0.12702
EXP
1.00000
0.16453
0.36634
INC
1.00000
0.11127
INT
1.00000
HSF
"Don't know or uncertain" respondents:
Variable
DTY
DAG
NUM
EXP
INC
INT
HSF
VAR
1.00000
0.00718
-0.07374
-0.15700
-0.16050
-0.08006
-0.21721
DTY
y s i s . This assertion rests upon the option not chosen by the
' ' d o n ' t k n o w " r e s p o n d e n t s — i . e . , to i n d i c a t e the o p e n e n d e d h i g h e s t return on the investment; s o m e positive return
m u s t b e sufficient to i n d u c e i n v e s t m e n t for a n y r a t i o n a l
individual.
T h e s a m e set of explanatory variables from Equation 1
is used in a probit regression, where 1 . = " d o n ' t k n o w or
u n c e r t a i n " r e s p o n s e . T a b l e 2 , in t h e far r i g h t c o l u m n s ,
indicates m e a n s and standard deviations of each explanatory
variable for the " d o n ' t k n o w " r e s p o n d e n t s , and Table 3
s h o w s the simple correlation coefficient m a t r i x . T h e regression coefficient estimates and r-ratios are:
Variable
DTY
DAG
NUM
EXP
INC
INT
HSF
Constant
R =
2
Estimate
T-ratio
-.06194
.20775
- .03736
- .04239
-.00003
-.11330
-.00026
.34647
.1440
-0.499
2.110
1.269
-1.883
-6.733
-3.039
-4.117
2.139
T h e largest difference b e t w e e n the probit model and the
m o d e l of t h e discount rate level is the g r e a t e r m a g n i t u d e of
t h e i n c o m e effect in the probit a n a l y s i s . In addition, exp e r i e n c e is a statistically significant v a r i a b l e in the probit
m o d e l . A l l other variables that were significant in the tobit
analysis of the level of the discount rate r e m a i n s o , and the
size of t h e h o u s e h o l d square footage variable effect is inc r e a s e d . A s discussed previously, c o n s u m e r s with low inc o m e s m a y lack the tools to grasp the i n v e s t m e n t p r o b l e m .
T h u s i n c o m e m a y act as a " f i l t e r , " p u s h i n g such individuals into a " d o n ' t k n o w " n o n r e s p o n s e . T h e relationship
b e t w e e n i n c o m e level a n d t h e p r o b a b i l i t y of t h e " d o n ' t
k n o w , " nonrational response is:
Income level
Probability of "don't know
or uncertain" response
$5,000
$10,000
$15,000
$20,000
$25,000
$35,000
42.0%
35.3%
29.2%
23.4%
18.5%
10.7%
H i g h positive correlation b e t w e e n i n c o m e and household
s q u a r e f o o t a g e s u g g e s t s t h e s a m e w a r n i n g as before—
n a m e l y , that the t w o variables m a y best b e treated jointly.
T h e m a g n i t u d e of the c o m b i n e d effect of t h e s e t w o varia b l e s , m o v i n g in u n i s o n , should a p p r o x i m a t e the effects
d e m o n s t r a t e d in the tabular display j u s t g i v e n .
G r e a t e r e x p e r i e n c e w i t h e n e r g y - c o n s e r v i n g activities
t e n d s to increase the probability that a positive savings res p o n s e will b e obtained from an individual. T h e fact that
t h e e x p e r i e n c e variable (like i n c o m e ) is significant at this
level of analysis but is not significant in the regression on
t h e d i s c o u n t rate level is consistent with the following arg u m e n t : (1) i n d i v i d u a l s ' conceptual abilities are linked to
prior e x p e r i e n c e as well as to i n c o m e ; (2) o n c e this conc e p t u a l p r o b l e m is o v e r c o m e , t h e i n v e s t m e n t analysis
h i n g e s u p o n expected future benefits and c o s t s , which are
essentially unrelated to i n c o m e or prior e x p e r i e n c e . While
this line of reasoning has plausibility, there m a y be other
a r g u m e n t s that could be offered.
PURCHASE OF E N E R G Y - S A V I N G GOODS
245
Unexplained Variance
The low R values f r o m t h e estimations suggests that
these basic m o d e l s m a y b e w e a k in a n u m b e r of w a y s . T o
obtain a m e a s u r e of p r e d i c t i v e v a l i d i t y , cross-validated correlations were o b t a i n e d o n t h e p r o b i t m o d e l . T h i s w a s d o n e
by splitting the s a m p l e r a n d o m l y into an estimation a n d a
holdout s a m p l e . R e g r e s s i o n p a r a m e t e r s from the e s t i m a t i o n
sample were u s e d to o b t a i n p r e d i c t i o n s of the d e p e n d e n t
variable in t h e h o l d o u t s a m p l e . T h e c o r r e l a t i o n o f t h e s e
predictions with the a c t u a l h o l d o u t values of the d e p e n d e n t
variable was m a d e . T h i s p r o c e d u r e w a s then r e v e r s e d , with
the holdout s a m p l e b e c o m i n g the estimation s a m p l e . F o r
the overall s a m p l e e s t i m a t i o n , t h e unadjusted R is 0 . 1 4 4 .
T h e two cross-validation c o r r e l a t i o n s are 0 . 1 5 1 and 0 . 1 3 4 .
2
2
The low R values r e p o r t e d for the estimated m o d e l s d o
not signal the failure o f t h e m o d e l s . Part of t h e l o w e x plained variance in t h e a n a l y s i s of the discount rate level
m a y result from r a n d o m n e s s of r e s p o n s e by those w h o h a d
no rational basis for c h o i c e . U n d o u b t e d l y , our m o d e l s c o n tain insufficient i n f o r m a t i o n to explain the initial r e s p o n s e s
to a tough i n v e s t m e n t d e c i s i o n . Personality characteristics,
attitudes, and family life s t y l e — a s well as additional e c o nomic c o n s i d e r a t i o n s — m a y b e quite important. Studies
w i t h m o r e specific a n d e x t e n s i v e data on t h e s e f a c t o r s
should throw m o r e light o n t h e influences behind decisions
about investments in u n t r i e d d u r a b l e s . H o w e v e r , it is likely
that low R v a l u e s w o u l d b e o b t a i n e d with a m u c h richer
data set. Q u e s t i o n s t h a t p r o v i d e very little detailed information about a l o n g - t e r m a c t i v i t y and that require substantial conceptual a n d a n a l y t i c ability will invariably p u z z l e
prospective d e c i s i o n - m a k e r s . Our model only provides
c l u e s to s o m e of t h e r e l e v a n t f a c t o r s b e h i n d i n v e s t m e n t
behavior.
2
2
SUMMARY
In a mail s u r v e y , h o u s e h o l d s w e r e asked to choose either
the minimal a c c e p t a b l e a n n u a l cost reduction on a longlived energy-saving d e v i c e ( g i v e n an initial purchase and
installation price) or to c h o o s e not to respond by indicating
" d o n ' t k n o w or u n c e r t a i n . " W e then sought to e x a m i n e
factors influencing the c h o i c e s of h o u s e h o l d s .
Factors related to e x p e c t e d c o s t and value influence the
decisions of i n d i v i d u a l s w h o c a n conceptualize an investm e n t p r o b l e m . G e n e r a l l y , c o n s u m e r s acted as if there were
additional costs b e y o n d t h o s e stated in the survey question,
and they consistently a t t e m p t e d to integrate the incremental
costs into their i n v e s t m e n t d e c i s i o n s . In the study, this w a s
reflected by l o w e r i n d i v i d u a l d i s c o u n t rates for those bearing lower e x p e c t e d c o s t s ( o r greater value) for the investment.
Factors that l o w e r t h e d i s c o u n t rate increase the rate of
acceptance of an untried d u r a b l e g o o d . Variables important
to explaining the d i s c o u n t r a t e are: household square foota g e , n u m b e r of family m e m b e r s , age of housing, and intentions to u n d e r t a k e e n e r g y - c o n s e r v i n g activities. A major
short-term d e c r e a s e in t h e d i s c o u n t rate and increase in the
a c c e p t a n c e rate for an investment is unlikely unless intentions are substantially revised upward. Unlike other variables in the m o d e l (such as i n c o m e , family s i z e , and h o u s e h o l d square footage), intentions are subject t o larger and
q u i c k e r adjustments that can rapidly alter the prospects for
investment.
F o r individuals w h o do not have the conceptual tools to
analyze the i n v e s t m e n t p r o b l e m , a response to the investm e n t q u e s t i o n i s , in g e n e r a l , n o n r a t i o n a l ( m a d e w i t h o u t
c o n s i d e r a t i o n of costs or benefits). T h e bulk of these individuals r e s p o n d e d to the question with a " d o n ' t k n o w o r
u n c e r t a i n " r e s p o n s e . C o m p a r e d to those w h o went ahead
and a n s w e r e d the investment question, the " d o n ' t k n o w "
r e s p o n d e n t s w e r e characterized by lower i n c o m e , smaller
h o m e s , larger families, less experience with energy-conserving activities, and lower intentions to e n g a g e in future
e n e r g y - c o n s e r v i n g activities.
T h e study did not find that c o n s u m e r s w h o participated
in t h e s u r v e y by providing positive return responses w e r e
b e h a v i n g irrationally. Although i n c o m e w a s an important
variable in predicting w h e t h e r an individual would consider
the i n v e s t m e n t analysis, it played n o statistically significant
r o l e in e x p l a i n i n g t h e l e v e l of the d i s c o u n t r a t e u s e d b y
i n d i v i d u a l s . T h e m e a n discount rate observed in the study
w a s 2 2 . 5 p e r c e n t and does not vary significantly with inc o m e . This finding contrasts with earlier work by H a u s m a n
( 1 9 7 9 ) , w h i c h found very high discount rates used by the
poor.
[Received
June 1982. Revised
June
1983.]
APPENDIX
T h e i n d i c e s of e x p e r i e n c e , EXP, a n d i n t e n t i o n s , INT,
were d e v e l o p e d from each r e s p o n d e n t ' s a n s w e r s to the following question:
Check below those things which you or your landlord have
done in the last twenty-four months (two years), if any, AND
those which you or your landlord are definitely planning to
do in the next twelve months, if any. Check ALL that apply.
Individuals w e r e given 10 specific options of things they
m a y h a v e d o n e or i n t e n d e d to d o . F o r e a c h , t h e y c o u l d
c h e c k " h a v e d o n e in the last 24 m o n t h s , " "definitely p l a n n i n g to d o in the n e x t 12 m o n t h s , " or b o t h . T h e l a t t e r
w o u l d b e the appropriate response, for e x a m p l e , if the r e s p o n d e n t h a d recently added ceiling insulation and intended
to increase the thickness within the year. T h e 10 specific
options w e r e :
1. Add ceiling insulation
2. Add wall or floor insulation
3. Put insulation around a water heater
4. Install a clock thermostat
5. Purchase storm or insulated windows or doors
6. Adjust the usual temperature setting, either up
in the summer or down in the winter
THE JOURNAL OF CONSUMER RESEARCH
246
7. Add attic or roof fans
8. Install an auxiliary heating unit using coal or wood
9. Add weather stripping or caulking around doors,
windows, walls, or foundations
10. Purchase a more energy-efficient heating unit, air
conditioner, or hot-water heater
A n eleventh option w a s " o t h e r major energy conservation
efforts (please s p e c i f y ) , " and several blank lines followed
for a r e s p o n s e . B a s e d on the n u m b e r of checked responses
out of a m a x i m u m of 11 for either past or intended action,
the indices EXP and INT w e r e d e v e l o p e d . T h u s , for exa m p l e , INT = 3 w o u l d state that t h r e e of the p o s s i b l e
actions listed w e r e planned for the c o m i n g year.
REFERENCES
Becker, Gary S. (1965), " A Theory of the Allocation of T i m e , "
Economic Journal, 75 (September), 493-517.
Consumer Reports (1978), "The Overselling of Insulation," 43
(February), 6 7 - 7 3 .
Ferber, Marianne A. and Bonnie G. Bimbaum (1977), "The 'New
Home Economies': Retrospects and Prospects," Journal of
Consumer Research, 4 (June), 19-28.
Ferber, Robert (1973) "Consumer Economics, A Survey," Journal of Economic Literature, 11 (December), 1303-1342.
Gately, Dermot (1980), "Individual Discount Rates and the Purchase and Utilization of Energy-Using Durables: Comment,"
The Bell Journal of Economics, 11 (Spring), 373-374.
Hausman, Jerry A. (1979), "Individual Discount Rates and the
Purchase and Utilization of Energy-Using Durables," The
Bell Journal of Economics, 10 (Spring), 33-54.
Kahneman, Daniel, and Amos Tversky (1972), "Subjective Probability: A Judgment of Representatives," Cognitive Psychology, 3 (July), 430-454.
Kirzner, Israel (1973), Competition and Entrepreneur ship, Chicago: University of Chicago Press.
Lancaster, Kelvin (1975), " T h e Theory of Household Behavior:
Some Foundations," Annals of Economics and Social Measurement, 4(1), 5 - 2 1 .
Menefee, John A. (1982), "The Demand for Consumption Time:
A Longitudinal Perspective," Journal of Consumer Research, 8 (March), 391-397.
McKelvey, Richard D. and William Zavoina (1975), " A Statistical Model for the Analysis of Ordinal Level Dependent
V a r i a b l e s , " Journal of Mathematical
Sociology,
4(1),
103-120.
Nunnally, Jim C. (1978), Psychometric Theory, New York:
McGraw-Hill.
Theil, Henry (1981), Principles of Econometrics, New York: John
Wiley.
Tobin, James (1958), "Estimation of Relationships for Limited
Dependent Variables," Econometrica, 26 (January), 24-36.
Tversky, Amos and Daniel Kahneman (1973) "Availability: A
Heuristic forjudging Frequency and Probability," Cognitive
Psychology, 5 (April), 207-232.
Willis, Robert J. (1973), " A New Approach to the Economic
Theory of Fertility Behavior," Journal of Political Economy,
81 (March/April), 514-564.
Wright, Peter L. and Frederic Barbour (1977), "Phased Decision
Strategies: Sequels to an Initial Screening," in North-HollandlTIMS Studies in the Management Sciences, Vol. 6, eds.
Martin K. Starr and Milan Zeleny, Amsterdam: North-Holland.