NCEE g
NATIONAL CENTER FOR
ENVIRONMENTAL ECONOMICS
Willingness to pay to Reduce a Child's Pesticide
Exposure:
Evidence from the Baby Food Market
Kelly B. Maguire, Nicole Owens, Nathalie B. Simon
Working Paper Series
Working Paper # 02-03
May 2002
sy
srAr
U.S. Environmental Protection Agency
National Center for Environmental Economics
i 0,\
o « 1200 Pennsylvania Avenue. NW (MC 1809)
il* ^ Washington DH 90460
^ Washington, DC 20460
'%i proi^ http://www.epa.gov/economics
-------
Willingness to pay to Reduce a Child's Pesticide
Exposure:
Evidence from the Baby Food Market
Kelly B. Maguire, Nicole Owens, Nathalie B. Simon
USEPA National Center for Environmental Economics
Corresponding Author:
Kelly Maguire
US Environmental Protection Agency
1200 Pennsylvania Ave. NW (MC 1809T)
Washington, DC, 20460
maguire .kelly @epa. gov
NCEE Working Paper Series
Working Paper # 02-03
May 2002
DISCLAIMER
The views expressed in this paper are those of the author(s) and do not necessarily represent
those of the U.S. Environmental Protection Agency. In addition, although the research described
in this paper may have been funded entirely or in part by the U.S. Environmental Protection
Agency, it has not been subjected to the Agency's required peer and policy review. No official
Agency endorsement should be inferred.
-------
Willingness to pay to Reduce a Child's Pesticide Exposure:
Evidence from the Baby Food Market
Kelly B. Maguire, Nicole Owens, Nathalie B. Simon1
USEPA, National Center for Environmental Economics
Abstract:
In this paper we estimate the price premium associated with organic baby food by
applying a hedonic model to price and characteristic data for baby food products
collected in two cities: Raleigh/Durham, North Carolina and San Jose, California. We
use price per jar of baby food as the dependent variable and control for a number of baby
food characteristics (e.g., brand, type, and stage) as well as store characteristics (e.g. type
of retail establishment). We find the price premium associated with the organic
characteristic to be approximately 12 cents per jar. To the extent this premium reflects
parents' preferences regarding the reduction of their baby's exposure to pesticide
residues, our results could be paired up with risk data to estimate the value of the health
benefits associated with reduced exposure.
Subject Area Classifications:
63 Children's Health
62 Valuation
9 Toxic Substances
Keywords:
Organic Foods, Hedonic Model, Willingness to pay, Children
1 Acknowledgments: The authors wish to thank Mark Dickie, Jim Hammitt, Carol Mansfield, Chris
Dockins, Ron Cummings, Jennifer Bowen, and Matt Massey for comments on earlier versions of this paper.
In addition, we would like to thank the participants of the workshops and conference sessions at which
earlier versions were presented for many useful suggestions. Please address all correspondence to the
authors, US Environmental Protection Agency, 1200 Pennsylvania Ave., NW (MC 1809), Washington, DC,
20460 or maguire.kelly@epa.gov. The views in this paper are those of the authors and do not necessarily
reflect those of the US Environmental Protection Agency. In addition, although the research described in
this paper may have been funded wholly or in part by the U.S. Environmental Protection Agency, it has not
been subjected to the Agency's required peer and policy review. No official Agency endorsement should be
inferred.
-------
Willingness to pay to Reduce a Child's Pesticide Exposure:
Evidence from the Baby Food Market
by
Kelly B. Maguire, Nicole Owens, Nathalie B. Simon
USEPA, National Center for Environmental Economics
I. Introduction
The use of agricultural pesticides is beneficial to the quantity and quality of
agricultural commodities produced by reducing damage from pests. However, because of
their inherent toxicity, most pesticides create some risk to humans, animals, or the
environment. Exposure to pesticides has been linked to numerous health problems such
as birth defects, lymphoma, reproductive tract cancer, Hodgkins' disease, leukemia, brain
cancer and infertility (Blair et al., 1985; Colborn et al., 1993; Blair et al., 1997; Zahm and
Ward, 1998). Individuals may turn to organic foods as a way to avoid these residues.
Recently, the baby food market has expanded to include more organic varieties,
suggesting that individuals are in fact consuming more of these products. We measure
the value of avoiding pesticide residues by using hedonic methods to measure the
premium consumers pay for baby food that is designated as organic.
While exposure to pesticides comes from a variety of sources, diet is one of the
most important sources of exposure and is perhaps the source of greatest concern for the
general public.2 The Environmental Protection Agency (EPA) regulates approximately
600 pesticides by setting tolerances, or legal limits for allowable pesticide residues on
foods. The Food Quality Protection Act requires EPA to consider the susceptibility of
2 Other exposure pathways include hand to mouth contact with exposed flora and fauna and consumption
of contaminated groundwater.
1
-------
children to exposure and/or to adverse health effects when setting limits. While limits
are set under a "reasonable certainty of no harm" standard to minimize risks, residues
below the tolerance levels often remain on the harvested crops.3 In fact, Kuchler, et al.
(1996, 1997) report pesticide residue detection rates between 40% to 60% for a variety of
fruits and vegetables commonly consumed in the average U.S. diet.4
Infants and children may be particularly at risk from pesticide exposure because
they are growing and developing. As such, their body composition, metabolism and
physiological and biochemical processes differ from those of adults, resulting in greater
vulnerabilities to the toxic effects of pesticides (National Research Council, 1993; Zahm
and Ward, 1998). Additionally, children's diets are much more specialized than are those
of adults. Per unit of body weight, children consume more food and water than do adults.
For example, the average 1-year old consumes approximately eight times as many apples
and four times as many peaches as the average adult (Kuchler et al., 1996, 1997).
Perhaps in part because of concern regarding exposure to pesticide residues in
recent years, the general public has expressed a desire for safer foods (Huang, et al.,
1990; Weaver, et al., 1992). Organic foods are generally seen as being pesticide free and
therefore safer than conventionally grown foods. New USDA guidelines go into effect
October 2002, requiring producers and handlers to be certified by a USDA-accredited
agent to sell, label, or represent their products as "100 percent organic," "organic," or
"made with organic (specified ingredients or food groups)" (USDA, 2002). Even though
the new guidelines are not yet binding, a wide and growing variety of organic foods, both
3 Food Quality Protection Act of 1996 (FQPA; Public Law 104-170)
4 Only 1% of the samples were above legal limits. Prior to testing, foods were handled as consumers
would. Produce was washed, peeled and cored as appropriate before measuring pesticide residues.
2
-------
processed and fresh, are available. Baby food is one notable recent addition to the
organic market.
As the food marketplace changes, it is interesting to examine how consumers
value these changes, particularly in light of new regulatory guidelines. Several studies
have investigated the willingness to pay for organic produce (Hammitt, 1990; Estes and
Smith, 1996; Fu, 1999), other food characteristics, such as nutrition (Stanley, et al.,
1991), and organic products (Nimon and Beghin, 1999). In general, results show that
individuals are willing to pay more for organic products or more healthful products,
though the heterogeneity in products makes it infeasible to generalize to other markets.
Little work has been conducted, however, analyzing baby food specifically. Claims have
been made that consumers are willing to pay a premium for organic baby food; but, the
supporting analysis to date has been limited (Harris, 1997). This paper attempts to fill
this gap. We estimate the price premium associated with organic versus conventional
baby food, controlling for other baby food characteristics (e.g., brand) as well as store
characteristics (e.g. type of retail establishment). To the extent this premium reflects
parents' preferences regarding the reduction of their baby's exposure to pesticide
residues, our results could be used to estimate the value of the health benefits associated
with reduced exposure. We rely upon a hedonic framework and data collected in the
Raleigh/Durham, North Carolina and San Jose, California areas to estimate this premium.
The remainder of the paper is organized as follows: Section II provides the
history of the organic baby food market as well as pertinent details regarding the
packaging of baby food products. Section III describes the methodology employed in
this paper while section IV focuses on the data used in the analysis and summary
3
-------
statistics. Results of our hedonic analysis are presented in section V and section VI
concludes.
II. The structure of the Baby Food market
Most parents purchase jarred baby food for their child as a convenient method for
introducing solid, table food, although some parents opt for making their own baby food
using food grinders or mashers or using frozen varieties. Jarred baby food is offered by
"stage" which is directly related to the developmental stage or age of the baby.
Generally, there are three stages within each brand. Stage 1 baby food consists of simple,
single flavor foods, such as peas or peaches that serve as a baby's first introduction to
"solid" food. Stages 2 and 3 often combine flavors (e.g., blueberries and pears) and offer
increasingly complex flavors by combining food groups (e.g., beef and pasta). As stage
increases, so does the texture of the food, with stage 3 providing chunkier foods for
children with teeth.
Within stage, baby foods can be categorized according to seven types: cereal,
fruit, vegetable, fruit-vegetable combination, meat, dinner, dessert. Not all types are
offered in each stage. The meat category consists of jars with single ingredients (e.g.,
beef), whereas the dinner category consists of more traditional dinner-like flavors (e.g.,
beef noodle dinner). The other categories are self-explanatory. There are a variety of
flavors available within each stage and type. For example, common stage 2 fruits
include, pears, plums with apples, and apple-blueberry.
In the United States, there are five major brands of baby food available at retail
outlets: Beechnut, Gerber, Earth's Best, Heinz, and Organic Baby. Beechnut and Heinz
4
-------
offer conventional baby food only, Earth's Best and Organic Baby are exclusively
organic baby foods, while Gerber offers both conventional and organic varieties. In
2000, Gerber had a 70 percent market share, Beechnut a 13 percent market share, and
Heinz an 11 percent market share of the total baby food market (US Business Reporter,
2001) ,5
Introduced in 1988 in Vermont, for many years Earth's Best was the only
nationally available organic baby food and initially was only available in health food
stores. By 1996, Earth's Best was sold in approximately 45 percent of supermarkets in
the United States (Harris, 1997). In the late 1990's, Gerber introduced Tender Harvest,
an organic line of baby food. While initially available only in stage 2 varieties, Tender
Harvest is now available for all three stages of a baby's development.
Ready-to-eat jarred baby food is sold in major grocery stores in the U.S., as well
as smaller convenience stores, drug stores, and other specialty markets. Both organic and
conventional baby foods are sold in most of the larger venues.
III. Methodology/Hedonic Model
In order to estimate the value, or willingness to pay for a reduction in exposure to
pesticides from organic baby food, we rely on the hedonic framework. This framework
has been applied to a vast number of scenarios to assess the implicit price associated with
specific characteristics of differentiable market goods. For instance, hedonic models of
the labor market have been extensively employed to examine price premiums associated
5 This includes all baby food products, including cereals and juices, and therefore does not represent the
market shares for jarred foods exclusively. It is likely that the market shares for jarred foods differ
slightly since the smaller baby food manufacturers offer less in the way of non-jarred choices.
5
-------
with safer occupations (e.g. Thaler and Rosen, 1976 one of the first studies of this
nature). Hedonic models of both housing markets and labor markets have yielded
premiums associated with better air quality (Palmquist 1984; Portney 1981; Bayless
1982), while models of automobile markets have resulted in implicit values of improved
quality and safety (Griliches 1988). Hedonic models have also been applied to
agricultural products to assess implicit prices associated with eco-labels in the apparel
industry (Nimon and Beghin 1999) and various food product characteristics (Price et al.
1990; Estes 1986; Estes and Smith 1996). To our knowledge, this study is the first to
examine the implicit price of the organic component of jarred baby foods using a hedonic
model.
The hedonic price function shows the relationship between the price of a good
and its component characteristics. The partial derivative of the hedonic price function
with respect to a component gives the implicit price associated with that characteristic,
the premium or value of interest. A formal derivation of the hedonic price function can
be found in Freeman (1993, 1995). Briefly, this function is generated through the
intersection of demand and supply equilibrium points.
On the demand side of the market, consumers maximize utility, w, subject to a
budget constraint, where utility is a function of a composite good, x, and a differentiated
good of interest, y, which is composed of components, qr The maximization problem
yields first order conditions, as follows:
(1)
6
-------
wherepy is the price of good>\ Equation 1 shows that utility is maximized when the
marginal rate of substitution between ^ and the composite good (left side of equation 1)
is equal to the marginal price of q; (right side of equation 1). That is, individuals will
consume a component characteristic (as revealed through their purchase of good >') up to
the point where the relative value of that characteristic is equal to its marginal price.
On the supply side of the market, producers maximize their profits, n, such that
the marginal cost of producing a characteristic, qn is equivalent to the price of that
characteristic, as follows:
dPy C
T^= (2
q j y
where ( \4! is the marginal cost of producing qr Equation 2 states that the profit
maximizing level of production occurs where the per-unit marginal cost of producing ^
(right side of 2) is equal to the marginal price of that component (left side of 2). We can
now see from equations 1 and 2 that equilibrium occurs when the value or willingness to
pay for a characteristic is equal to the marginal cost of that same characteristic, as
The hedonic price function for good y, differentiate in its j characteristics, is
given by
follows:
(3)
7
-------
Py = Kqiy'Vly,-,
-------
relationship for estimation purposes. We estimate this model using data collected in two
metropolitan areas as described below.
IV. Data Collection and Descriptive Statistics
Baby food price and characteristic data were collected from a total of 83 retail
establishments in Raleigh, North Carolina and San Jose, California over two three-day
periods in February 2001 and August 2001, respectively. Stores in each city were
randomly selected from a list of all retail food establishments generated from current
local on-line consumer yellow pages.7 We stratified our samples across establishment
type based on the distribution of food purchased by location for consumers in the U.S.
(ERS, 2000). Although specific information is not available on the distribution of jarred
baby food sales by location, we use the ERS (2000) data as a basis for our stratification
and reallocate the sample to more accurately reflect likely baby food retail venues.8
Table 1 summarizes the distribution of establishments as reported by ERS and the
distribution used in our two-city sample.
In each of the stores in our two-city sample, we recorded data on all jarred baby
food offered for sale. Because price varies only rarely across flavors within a stage/type
cell, we chose baby food type (e.g. fruit, vegetables, dinner) within stage and brand as
our unit of observation.9 For example, Beechnut stage 1 fruits is an observation, as is
7 The consumer yellow pages can be found at http://yp.yahoo.com, accessed on January 29, 2001 and again
on July 31,2001.
8 For example, specialty stores and mass merchandisers in Raleigh, North Carolina and San Jose,
California did not sell jarred baby food, while drug stores and baby super centers did sell baby food, but
were not part of the ERS (2000) distribution. We re-allocated the sample accordingly.
9 When price did vary within observation we took a weighted average of the prices of all flavors to derive
an observation price.
9
-------
Heinz stage 2 dinners.10 Prior to our data collection efforts, we developed a template for
each brand of baby food that allowed us to record the price of each observation, the
number of flavors offered for sale, as well as the shelf space allocated to the observation
and relevant store characteristics. In addition, we recorded within brand variations in
product labeling that had potential for influencing price. For example, Beechnut offers a
' Simple Recipes' line, which has fewer additives than its regular line. Furthermore,
jarred baby food is sometimes provided in type-specific "multi-packs" (e.g. four jars of
stage 2 fruits packaged together). Recognizing that multi-packs could be priced
differently than single jars, we chose to treat multi-packs as separate observations, noting
the number of jars packaged together and the baby food type and stage. As data
collection progressed, we developed several detailed categories for grocery store type,
including ethnic groceries, small groceries and upscale groceries.
Our data collection effort resulted in 1,697 useable observations, with 933
observations from Raleigh, NC and 764 observations from San Jose, CA. Detailed
descriptions of relevant variables are provided in table 2, whereas summary statistics are
provided in table 3. The data consist of price and organic designation, other product
characteristics, and store characteristics. As noted in Table 3, there are significantly
more organic observations in San Jose (28%) compared to Raleigh (15%). In addition,
the prices of both organic and conventional baby foods are higher in San Jose than in
Raleigh. The price of organic baby foods is higher than conventional baby foods in both
Raleigh and San Jose, as expected. Gerber baby foods were the most prevalent in both
10 We noted the number of flavors within each observation, though we have no expectations that this
variable will influence price.
10
-------
markets we sampled. We include dummy variables for each brand of baby food in our
analysis. These variables capture factors such as brand loyalty, quality variations, and
other characteristics individuals may value that we do not measure. We also include
variables for the various types of stores, accounting for the impact of point of purchase
on price. There is a wider variety of store types in San Jose with ethnic and small
grocery stores not represented in Raleigh.
V. Results and Discussion
We now turn to the empirical estimation results. We present the results for four
models to account for alternative specifications of the dependent variable and the model.
Because most of our independent variables are dichotomous variables our selection of
functional form for the hedonic model is limited to a linear model. Model 1 is estimated
as follows:
price = a + (organic) + /?. (store.) + /?. {brandf) +
/?2 (meat) + /?3 (stage 1) + /?4 (stage2) + (sqft) + (label) + (7)
fi7(multi)+ fis(city)+ v
where storet indicates the i'h store type (e.g., grocery store, upscale market, etc.) and
brandj indicates the jth brand of baby food (e.g., Gerber, Beechnut, etc.). All other
variables are defined above in table 2. In this model the dependent variable is the per jar
price of the observation.
The model performs well, with an overall adjusted R2 of 0.79. With the exception
11
-------
of the ORGANIC BABY and MULTI variables, all variables are highly significant.
Among the product characteristics, we find that the price of STAGE1 and STAGE2 baby
food is 21 cents and 16 cents less per jar, respectively relative to stage 3 (the omitted
category). A number of factors could be driving this result, including size of jar and
availability of substitutes. Typically, jar size increases with the age of the child, with
stage 1 jars 2.5 ounces in size, stage 2 jars 4 ounces in size, and stage 3 jars 6 ounces in
size.
In order to examine the effect of jar size more closely, we estimate model 2 in
which we normalize the price of an observation by the number of ounces in the jar. In
this model, we find that stage is still highly significant; however, the magnitude and sign
of the coefficients have changed. Stage 1 varieties are priced at 5 cents more per ounce
compared to stage 3, while stage 2 varieties are offered at a price premium of 1 cent per
ounce compared to stage 3.
This result makes sense when one considers the number of substitutes available
for the jarred baby foods by stage. Stage 1 varieties serve as an introduction to solid
foods for younger babies (once they have mastered cereal) and are only available in two
types fruits and vegetables. As such, the number of varieties available is small
(typically 5 or 6 flavors by type). Stage 1 foods are also processed until extremely
smooth, making them easier for the child to swallow. Parents are encouraged to
introduce one stage 1 flavor at a time and to continue introducing one new flavor
approximately every 3 days. Once eating jarred foods, babies may move past the stage 1
varieties fairly quickly into stage 2, which is comprised of a much larger number of
combinations and types of foods as described in section 2 above. Stage 2 baby foods
12
-------
have more texture than stage 1 varieties. Stage 3 foods, on the other hand, are even
chunkier, and are intended for babies learning to mash and chew with their new teeth. It
is not unreasonable to suspect that many parents simply skip using the stage 3 varieties
altogether, feeding the baby foods that other family members are consuming at meal time
rather than continuing with jarred foods. In short, stage 1 varieties seem to have the
fewest number of readily available substitutes since processing of foods at home to the
desired texture for a young baby may be cumbersome and time consuming. Stage 3
foods, on the other hand, have the largest number of substitutes as children in this
developmental stage could simply consume foods from their parents' plates.
Looking at both models 1 and 2, we find that other characteristics are significant
factors in determining price. Brand name, for instance, is significant. Relative to
Beechnut (the omitted category), GERBER and EARTHS BEST are priced slightly
higher, while HEINZ is priced slightly less. MEAT flavors of baby food, defined above,
are also significant predictors of price, commanding a price premium of 26 cents per jar
(9 cents per ounce) compared to other types of baby food.
We also examine the impact of several store characteristics on the price of baby
food. We control for the venue in which observations are sold via dummy variables for
store types (e.g., grocery store, ethnic food market, etc.).11 We tried controlling for other
variables in our models, including store name, under the notion that consumers may favor
a particular chain of stores. Although these latter models performed well, model 1
performed slightly better. In addition, features of the store types are likely to result in
more variation in prices than features of a particular chain. For example, upscale grocery
11 An F-test rejected the hypothesis that all store coefficients, except grocery, were equal to zero.
13
-------
stores tend to have wider aisles, fewer quantities of each product, and more elaborate
displays as compared to traditional grocery stores. We omit the OTHER STORE type
(representing uncategorized stores, such as drug stores and supercenters) and find that
prices are slightly lower in GROCERY stores, ETHNIC markets, and DEPARTMENT
stores. As expected prices are higher in UPSCALE stores, SMALL GROCERIES, and
CONVENIENCE stores.
Regional differences in pricing are also apparent as we find that baby food is
$0.03 less per jar in Raleigh, NC than San Jose, CA. This difference could be due to cost
of living differences in the two regions and/or differences in the structure of the
grocery/food markets. One cost of living index
(http://cgi.money.cnn.com/tools/costofliving/costofliving.html) indicates that groceries,
in general, are 21% higher in the northern California area as compared to Raleigh, NC.
This could in part be driven by the fact that there are more small neighbor stores and
ethnic stores in San Jose, which do not enjoy the same economies of scale as the larger
stores.
Turning to the variable of ultimate interest, ORGANIC, we find that organic baby
food is 12 cents more per jar (3 cents more per ounce) than conventional baby food. As
discussed earlier, this represents both individual willingness to pay for the organic
characteristic, as well as the marginal production costs that may be associated with
organic baby food.12 Previous research has indicated that individuals pay 21 cents per jar
12 We recognize that it is likely that organic baby food is more costly to produce than conventional baby
food. However, assuming the hedonic framework applies, this premium represents willingness to pay
for a characteristic, regardless of the additional production costs. If individuals valued organic baby
food at less than the market price, then we would likely see changes in the market, such as producers
exiting the market. In fact, the opposite has occurred. Per Price, et al. (1990) a disequilibrium between
the supply price and individual values is possible, but not in a long-run equilibrium. The authors
14
-------
more for organic baby food (Harris, 1997) and Earth's Best itself states that its baby food
costs approximately 50% more than conventional baby food
(http://www.earthsbest.com/news_faq.asp). It is unclear how these results were derived
and in fact, our analysis finds a much lower premium than previously cited.
In order to test the robustness of this premium, we also estimate 2 additional
models. Model 3 estimates equation 7 by city in order to more readily identify regional
differences associated with these results where they occur. It could be the case that
variations in preferences or costs exist between the east and west coasts. While modest
differences exist in some of the product and store characteristics across the two cities, we
find virtually no difference in the organic premium between Raleigh and San Jose. The
premium in both cities is between 12 and 13 cents per jar.
Finally, we also estimate the model for Gerber observations only. Gerber is the
only brand available in both organic and conventional varieties. For the non-Gerber
brands, the brand and organic variable are collinear. An issue, then, is the degree to
which brand loyalty is driving the results. In order to investigate this issue, we estimate
our model using only Gerber observations, in which the brand loyalty issue is held
constant. Indeed, we find that the premium is identical among the Gerber observations to
what it is when all data points are included. The premium for Gerber organic baby food
(i.e., the Tender Harvest label) is 12 cents per jar. This is not necessarily surprising,
given that Gerber organic observations represent 79% of the total organic observations.
It is difficult to discern if Gerber is able to exert market power and drive the price above
suggest that the premiums should be "interpreted as...individual consumer valuations of characteristics."
(p. 11).
15
-------
marginal cost. Nevertheless, we can be comforted by the stability of the price premium
across all brands.
VI. Conclusion
For parents interested in reducing their baby's exposure to pesticide residues,
organic baby food offers a natural alternative. While federal regulations regarding
organic labels for all foods are not yet binding, there are several organic baby food
producers offering "certified" organic varieties. It is interesting to analyze the extent to
which individuals value these organic designations. Previous research has focused on
values for organic produce (Hammitt, 1990; Estes and Smith, 1996) and clothing (Nimon
and Beghin, 1999), but it is infeasible to extend these results to the baby food market
given the heterogeneity in products and consumers across these markets.
We combine a unique data collection method with a tried and true model to
estimate how individuals (specifically, parents of babies) value reductions in pesticide
exposure, as evidenced through the organic baby food market. We find that, in fact, the
value associated with these pesticide reductions is less than the value, or premium for
organic foods, asserted in previous work. Individuals are willing to pay 12 cents per jar
more for organic baby food as opposed to conventional varieties. Previously, research
has asserted that this premium is 21 cents per jar (Harris, 1997) and 50% more than
conventional varieties (www.earthsbest.com).
It is important to note that the premium for organic baby food may not be
equivalent to the value for reducing exposure to pesticide residues. It could in fact be the
case that individuals value other features of the organic baby food than just its health
16
-------
effects. For example, individuals may purchase organic foods because of a preference for
environmentally friendly farming practices or a concern for farm worker exposure to
pesticides. Within this research, we are unable to discern these competing effects; we
merely assert that the premium reflects a desire to avoid pesticide residues. To the extent
that individuals do in fact value organic baby food for its health effects, these results
could be used to infer the overall value of health benefits associated with reduced risks
a subject for future research.
17
-------
Table 1: Distribution of Sample Data Across Venues*
Venue Type
ERS
Percent of Sample
Grocery store
69
71
Mass merchandiser
8
11
Specialty Food Stores
4
Convenience store
4
11
Drug store
5
Baby supercenter
2
Other"
13
* Table provides the percent of food purchased for home consumption by type of
establishment as provided by ERS, 2000 and the distribution of stores in the sample.
** Consists of warehouse clubs, home delivery, farmers, processors, and miscellaneous
locations.
18
-------
Table 2: Variable Description
Variable
Description
ORGANIC CHARACTERISTIC
ORGANIC
= 1 if an organic product, = 0 otherwise
PRODUCT CHARACTERISTICS
MEAT
= 1 if a meat product, = 0 otherwise
LABEL
= 1 if a special label within brand, = 0 otherwise
STAGE1
= 1 if a stage 1 product, = 0 otherwise
STAGE2
= 1 if a stage 2 product, = 0 otherwise
STAGE3
= 1 if a stage 3 product, = 0 otherwise
GERBER
= 1 if a Gerber product, = 0 otherwise
BEECHNUT
= 1 if a Beechnut product, = 0 otherwise
EARTHS BEST
= 1 if an Earth's Best product, = 0 otherwise
HEINZ
= 1 if a Heinz product, = 0 otherwise
ORG BABY
= 1 if an Organic Baby product, = 0 otherwise
MULTI
= 1 if sold in a multi-pack, = 0 otherwise
STORE CHARACTERISTICS
GROCERY
= 1 if sold in a grocery store, = 0 otherwise
UPSCALE
= 1 if sold in a specialty grocery store, = 0 otherwise
ETHNIC
= 1 if sold in an ethnic grocery store, = 0 otherwise
DEPARTMENT
= 1 if sold in a department store (e.g. Target, Kmart), = 0
otherwise
SMALL GROCERY
= 1 if sold in a small neighborhood market, = 0 otherwise
CONVENIENCE
= 1 if sold in a convenience store, = 0 otherwise
OTHER STORE
TYPE
= 1 if sold in other store type (e.g. drug store, super store), = 0
otherwise
SQFT
number of square feet of shelf space
RALEIGH
= 1 if sold in Raleigh, NC, = 0 is sold in San Jose, CA
19
-------
Table 3: Descriptive Statistics1
Variable
All Data
Raleigh
San Jose
PRICE AND ORGANIC CHARACTERISTIC
ORGANIC
0.21 (0.40)
0.15 (0.36)
0.28 (0.45)
PRICECONV
0.57 (0.12)
0.55 (0.12)
0.60 (0.12)
n=1356
n=794
n=552
PRICEORGANIC
0.70 (0.11)
0.67 (0.07)
0.72 (0.12)
n=351
n=139
n=212
PRODUCT CHARACTERISTICS
MEAT
0.05 (0.21)
0.05 (0.22)
0.04 (0.21)
LABEL
0.22 (0.42)
0.20 (0.40)
0.25 (0.43)
STAGE 1
0.19(0.39)
0.21 (0.41)
0.16(0.37)
STAGE2
0.56 (0.50)
0.53 (0.50)
0.60 (0.49)
STAGE3
0.25 (0.43)
0.26 (0.44)
0.23 (0.42)
GERBER
0.61 (0.49)
0.41 (0.49)
0.48 (0.50)
BEECHNUT
0.26 (0.44)
0.37 (0.48)
0.13 (0.33)
HEINZ
0.09 (0.28)
0.06 (0.25)
0.11 (0.32)
EARTHS BEST
0.04 (0.20)
0.02 (0.14)
0.07 (0.25)
ORG BABY
0.003 (0.06)
0.003 (0.06)
0.004 (0.06)
MULTI
0.08 (0.27)
0.09 (0.28)
0.06 (0.25)
STORE CHARACTERISTICS
GROCERY
0.82 (0.38)
0.89 (0.31)
0.74 (0.44)
UPSCALE
0.03 (0.17)
0.01 (0.12)
0.05 (0.21)
ETHNIC
0.03 (0.17)
n/a
0.07 (0.25)
DEPARTMENT
0.05 (0.21)
0.06 (0.24)
0.03 (0.17)
SMALL GROCERY
0.02 (0.15)
n/a
0.05 (0.21)
CONVENIENCE
0.01 (0.12)
0.01 (0.11)
0.02 (0.13)
OTHER STORE TYPE
0.07 (0.26)
0.03 (0.18)
0.12 (0.32)
SQFT
1.16(0.93)
1.24 (0.99)
1.08 (0.86)
Observations
1697
933
764
1 Mean (standard deviation) displayed for each variable.
20
-------
Table 4: Results
Model 1
Model 2
Model 3
Model 4
Variable
(y=price)
(y=price/oz)
(y=price)
(y=price)
Raleigh
San Jose
Gerber ONLY
Intercept
Q
0.13***
q (54***
0.76***
0.75***
(0.01)
(0.004)
(0.02)
(0.01)
(0.01)
ORGANIC
0.12***
0.03***
0.13***
0.12***
0.12***
(0.01)
(0.003)
(0.01)
(0.01)
(0.01)
MEAT
0.26 ***
0 09***
0.26***
0.26***
q 23***
(0.01)
(0.003)
(0.01)
(0.01)
(0.01)
LABEL
0.01**
0.005**
0.01
0.03***
0.02*
(0.01)
(0.003)
(0.01)
(0.01)
(0.01)
STAGE 1
-0.21***
0.05***
-0 19***
-0.25***
-0 23***
(0.004)
(0.002)
(0.01)
(0.01)
(0.01)
STAGE2
-0.16 ***
0.009***
-0.15***
-0.18***
-0.16***
(0.004)
(0.002)
(0.01)
(0.01)
(0.005)
GERBER
0.03***
0.01***
q 04***
-0.002
(0.004)
(0.002)
(0.005)
(0.01)
HEINZ
-0 04***
-0.004
-0.03***
-0.06***
(0.006)
(0.003)
(0.01)
(0.01)
EARTHS BEST
0 04***
0.02***
0.02
0.03**
(0.01)
(0.005)
(0.02)
(0.02)
ORG BABY
-0.02
0.002
-0.12***
0.03
(0.03)
(0.01)
(0.05)
(0.03)
MULTI
0.01
-0.01***
0.01*
0.01
0.01
(0.01)
(0.003)
(0.01)
(0.01)
(0.01)
RALEIGH
-0.03***
-0.009***
-0.02***
(0.003)
(0.001)
(0.004)
GROCERY
-0.02***
-0.01***
-0.004
-0 04***
-0.02**
(0.01)
(0.004)
(0.02)
(0.01)
(.01)
UPSCALE
q 04***
0.002
0.10***
0.02
0.03**
(0.01)
(0.005)
(0.03)
(0.01)
(0.02)
ETHNIC
-0.02**
-0.01***
-0 04***
-0.01
(0.01)
(0.005)
(0.01)
(0.01)
21
-------
DEPARTMENT
-0.02**
-0.005
-0.003
-0 04***
-0.02*
(0.01)
(0.005)
(0.02)
(0.02)
(0.01)
SMALL GROC
Q
0.03***
0.16***
0.16***
(0.01)
(0.01)
(0.01)
(0.01)
CONVENIENCE
q 29***
0.06***
0.35***
0.25***
0.28***
(0.01)
(0.01)
(0.02)
(0.02)
(0.02)
SQFT
-0.01***
-0.003***
-0.01***
-0.01**
-0.01***
(0.002)
(0.001)
(0.002)
(0.003)
(0.003)
R2
0.79
0.64
0.75
0.83
0.76
Observations
1688
1688
931
758
1129
Parameter estimates (standard errors) are displayed for each variable and model. Significance is indicated
as follows: *** 99 percent level of significance, ** 95 percent level of significance, * 90 percent level of
significance.
22
-------
References
Bayless, Mark. 1982. Measuring the Benefits of Air Quality Improvements: A Hedonic
Salary Approach. Journal of Environmental Economics and Management 9: 81-99.
Blair, A. and D. White. 1985. Leukemia Cell Types and Agricultural Practices in
Nebraska. Archives of Environmental Health 40: 211-214.
Blair, A, M. Francis, and S. Lynch. "Occupational Exposures to Pesticides and Their
Effects on Human Health. 1997. Proceedings of the Third National IPM
Symposium/Workshop. S. Lynch, C. Greene, and C. Kramer-LeBlanc (eds.). United
States Department of Agriculture, Economic Research Service, Miscellaneous
Publication Number 1542, Washington DC: 59-75.
Colborn, T., F. vom Saal, and A. Soto. 1993. Developmental Effects of Endocrine-
Disrupting Chemicals in Wildlife and Humans. Environmental Health Perspectives 101:
378-384.
Economic Research Service (ERS). 2000. "Sales of Food at Home by Type of Outlet."
http://www.ers.usda.gov/briefing/CPIFoodAndExpenditures/Data/tablel6.htm. accessed
on 1/19/01.
Estes, Edmund A. 1986. Estimation of Implicit Prices for Green Pepper Quality
Attributes Using an Hedonic Framework. Journal of Food Distribution Research,
September: 5-15.
Estes, E.and V. K. Smith. 1996. Price, Quality, and Pesticide Related Health Risk
Considerations in Fruit and Vegetable Purchases: An Hedonic Analysis of Tucson,
Arizona Supermarkets. Journal of Food Distribution Research 62: 59-76.
Freeman, A. Myrick. 1995. Hedonic Pricing Methods, in The Handbook of
Environmental Economics, Daniel Bromley (ed.). Cambridge, MA: Blackwell Publishers,
Inc.
Fu, Tsu-Tan, Jin-Tan Liu, and James K. Hammitt. 1999. Consumer Willingness to Pay
for Low-Pesticide Fresh Produce in Taiwan. Journal of Agricultural Economics 50(2):
220-233.
Griliches, Zvi. 1988. Hedonic Price Indexes for Automobiles: An Econometric Analysis
of Quality Change, in Technology, Education, and Productivity: Early Papers with Notes
23
-------
to Subsequent Literature. New York: Basil Blackwell, Inc.
Hammitt, J. 1990. Risk Perceptions and Food Choices: An Exploratory Analysis of
Organic-Versus Conventional-Produce Buyers. Risk Analysis 10: 367-373.
Harris, J. M. 1997. Consumers Pay a Premium for Organic Baby Foods. Food Review,
May: 13-16.
Huang, Chung L., Sukant Misra, and Stephen L. Ott. 1990. Modeling Consumer Risk
Perception and Choice Behavior: The Case of Chemical Residues in Fresh Produce. In
Mayer, Robert N. (Ed.). Enhancing Consumer Choice: Proceedings of the Second
International Conference on Research in the Consumer Interest, Snowbird, Utah,
August, 1990.
Kuchler, F., K. Ralston, L. Unnevehr, and R. Chadran. 1996. Pesticide Residues:
Reducing Dietary Risks. U.S. Department of Agriculture, Economic Research Service,
Agricultural Economic Report Number 728.
Kuchler, Fred, Katherine Ralston, and Laurian J. Unnevehr. 1997. Reducing Pesticide
Risks to US Food Consumers: Can Agricultural Research Help? Food Policy 22(2):
119-132.
National Research Council. 1993. Pesticides in the Diets of Infants and Children.
Washington, D.C.:National Academy Press.
Nimon, Wesley, and John Beghin. 1999. Are Eco-labels Valuable? Evidence from the
Apparel Industry. American Journal of Agricultural Economics. 81: 801-811.
Palmquist, Raymond B. 1984. Estimating the Demand for the Characteristics of
Housing. The Review of Economics and Statistics, 56(3): 394-404.
Portney, Paul R. 1981. Housing Prices, Health Effects, and Valuing Reductions in Risk
of Death. Journal of Environmental Economics and Management 8: 72-78.
Price, D.W., V.A, McCracken, R.C. Mittelhammer, D.Z. Price, S. Lutz, J.E. Lenz, and J.
Kim 1990. Perspectives on Analyzing Consumers' Valuations of Food Product
Characteristics, in Enhancing Consumer Choice: Proceedings of the Second
International Conference on Research in the Consumer Interest, Robert N. Mayer (ed).
Columbia, MO: American Council on Consumer Interests, August.
24
-------
Stanley, Linda R., John T. Tschirhart, and Jennifer Anderson. 1991. A Hedonic Price
Analysis of Nutritionally Labeled Breakfast Cereals: Implications for Nutrient Labeling.
Journal of Nutrition Education 23(5): 231 -23 8.
Thaler, Richard, and Sherwin Rosen. (1976). "The Value of Saving a Life: Evidence
From the Market." In Nestor E. Terleckyj (ed.), Household Production and Consumption.
Cambridge: NBER.
US Business Reporter. 2001. "Food Industry Market Shares."
http://www.activemedia-guide.com/babyfood_mrk.htm accessed 1/29/02.
USDA. 2002. "Frequently Asked Questions."
http://www.ams.usda.gov/nop/nop2000/nop2/FAQ.htm accessed 1/17/02.
Weaver, Robert D., David J. Evans, and A.E. Luloff 1992. Pesticide Use in Tomato
Production: Consumer Concerns and Willingness to Pay. Agribusiness 8(2): 131-142.
Zahm, S. and M. Ward. 1998. Pesticides and Childhood Cancer. Environmental Health
Perspectives, 106 (Suppl 3): 893-908.
25
------- |