Content uploaded by Catherine A. Durham
Author content
All content in this area was uploaded by Catherine A. Durham on May 15, 2014
Content may be subject to copyright.
Journal of Food Distribution Research
Volume 43, Issue 2
July 2012 Volume 43, Issue 2
1
Understanding Markets for Grass-Fed Beef:
Taste, Price, and Purchase Preferences
Lauren Gwin
a
l
, Catherine A, Durham
b l
, Jason D. Miller
c
, and Ann Colonna
d
a
Research Associate, Agricultural and Resource Economics, Oregon State University, Street address here, city,
state, zip Phone: 541-737-1569, Email: Lauren.gwin@oregonstate.edu
b
Associate Professor, Agricultural and Resource Economics, Food Innovation Center, Oregon State University
Street address here, city, state, zip
c
Graduate Research Assistant, Agricultural and Resource Economics, Oregon State University, Street address here,
city, state, zip
d
Sensory Program Manager, Food Innovation Center, Oregon State University, Street address here, city, state, zip
Abstract
We use results of a consumer taste test conducted in Portland, Oregon, and choice-based conjoint
analysis to examine consumer attitudes about grass-fed beef compared to conventional grain-fed:
taste preferences, willingness to pay, and willingness to buy frozen meat in bulk. We consider
the effect of demographic, attitudinal, and shopping location characteristics of consumers. A
baseline, uninformed consumer will pay $0.90-$0.94/pound more for grass-fed ground beef;
knowledge about production and nutritional factors increase the premium. A majority of partici-
pants would buy in bulk if they knew a producer or a friend referred them; 72% will buy frozen
beef.
Keywords: need three to five keywords here
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
2
Introduction
Increased consumer interest in grass-fed, naturally raised, locally produced meats is based on
perceptions and evidence about “healthier” fats, reduced environmental impacts, and increased
animal welfare associated with meats not raised in confinement systems on grain-based diets
(Daley et al. 2010; National Trust 2012; Schmidt 2010; Umberger et al. 2009; Varnold et al.
2011); this interest is also part of the broader local food movement (Martinez et al. 2010). Live-
stock producers who would like to produce and sell grass-fed meats must carefully weigh the
risks of shifting their production and marketing systems, given the significant and often costly
supply chain challenges of getting this type of meat to market. Knowing, in general, that
consumer demand for grass-fed is “up” is not enough: producers require geographically relevant
information not only about consumer demand and price elasticity, but also how and where
consumers will buy the product (such as by the cut or by the carcass, direct or at a store, at main-
stream or natural food retailers).
Achieving the consumer-oriented convenience of conventional meats, sold fresh, by the cut, in
Cryovac packaging, year-round, is neither easy nor cost-effective for many small producers.
Producers report that inventory management logistics and the need for a diversified customer
base to sell the entire animal at a price point that will compensate for extra costs are two critical
barriers to entry into this market niche (Fanatico and Rinehart 2006; Gwin 2009; Gwin, Evans,
and Brewer 2011). Selling animals by the whole, half, or quarter, direct to consumers, in one de-
livery of frozen cuts, is one way small producers can avoid these two problems (Thiboumery and
Lorentz 2009). Some restaurants have chosen to buy whole carcasses direct from farmers and
ranchers; consumer interest in such “bulk” sales appears at least anecdotally to be rising (Jackson
2009). Yet it is unclear how many people, even those who buy local, grass-fed meat by the cut,
are willing to purchase this way.
In this paper, we focus on a few key questions. What are consumer taste preferences and
willingness to pay (WTP) for local grass-fed beef versus conventional grain-fed beef? Do those
differ for consumers who shop primarily at natural food stores versus mainstream food stores?
Are consumers willing to buy beef in bulk, and what demographic, attitudinal, or shopping
location characteristics make them more or less willing? Our answers to these questions come
from results of a consumer taste test we conducted in Portland, Oregon.
Our study provides several new insights into the market for grass-fed beef. First, we examine
consumer interest in buying grass-fed beef in bulk, a valuable and potentially necessary strategy
for direct sales by producers. Second, we expand understanding of WTP for grass-fed beef
relative to conventional beef by exploring the effect of consumers’ prior knowledge and
uncovering the underlying consumer attitudes that result in WTP a premium for grass-fed beef.
Finally, we expand understanding of the impact of taste preference on WTP by incorporating
consumer ratings of the beef they tasted directly into the choice model.
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
3
Background
Understanding consumer WTP is very important to niche markets that rely on premium prices to
compensate for higher production costs. It was once suggested the per pound cost of producing
forage fed beef may be as much as 25% higher than producing conventional beef (Mayer 1999).
Grass-fed production can be more expensive, especially in parts of the country with less year-
round quality forage (Mathews and Johnson 2010). For example, an enterprise budget prepared
for a California grass-fed enterprise notes that pasture can be the limiting factor for grass-fed
systems, and that producers may need to increase grazing acreage, feed harvested forage, or
decrease herd size, all of which add to costs; producers also assume additional risk when they
retain ownership of cattle until finish weight (Larson, Thompson, Klonsky, and Livingston
2004). Yet even in circumstances and regions in which grass-fed beef is less expensive to
produce than conventional beef, post-farmgate supply chain costs are still likely to be high, as for
any smaller-scale, niche product (Gwin and Thiboumery 2012; Hardesty and Leff 2009; King et
al. 2010). Producers must find a customer base that will pay a high enough price for the product
to cover their costs and some profit margin. Identifying the price premium consumers will pay,
as well as the types of consumers who will pay it and who will use a direct sales channel, is
therefore
essential to producers.
Other WTP research around natural and/or grass-fed beef has found varied responses. Studies
that evaluated sensory properties in blind tests have generally found fewer US consumers prefer
grass-fed to conventional beef – for example, Feuz and Umberger (2001) found that only 23% of
consumers sampled in Chicago and San Francisco preferred grass-fed to corn-fed steaks. One
reason is historic: U.S. consumers have long been habituated to the taste and eating experience of
grain-finished beef. Another reason, research-specific, may be that studies have generally
focused on muscle cuts (e.g. steaks) or ground beef made from a single muscle cut, often result-
ing in lower fat content for the grass-fed beef among other sensory differences (Cox et al. 2006;
Feuz et al. 2004; Sitz et al. 2005; Xue et al. 2010). However, Cox et al. (2006) found that during
a home-use test, the percentage of consumers preferring grass-fed beef increased and the
significantly different preference for grain-fed was eliminated. The authors suggest that this is
because home cooking eliminated some sensory differences. It is also important to keep in mind
that even 23% of the national consumer base far exceeds the demand that can be satisfied by cur-
rent domestic grass-fed supply. To satisfy annual market demand for 23% of the population
would require about 6 million head of cattle/year. Yet the grass-fed sector currently harvests
between 150,000 and 170,000 head of cattle per year (Williams 2010).
Providing consumers with information on production practices and/or nutritional properties
affects WTP. Earlier studies in which consumers were informed about beef production method,
specifically, what the cattle were fed, did not find a higher WTP for grass-fed. Yet in more re-
cently conducted studies, providing production information increased WTP for grass-fed beef
relative to grain-fed beef (Lusk and Parker 2009; McCluskey et al. 2005; Thilmany, Umberger,
and Ziehl 2006; Umberger, Boxall, and Lacy 2009; Xue et al. 2010). McCluskey et al. (2005)
found that consumers would choose beef with higher Omega-3 levels given information about its
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
4
healthful properties. Umberger, Boxall, and Lacy (2009) also found that providing health
information increased WTP. Conner and Oppenheim (2008) found that providing environmental
and welfare information did not change WTP for pasture-raised meats, but as they noted, their
study participants had pre-existing knowledge of and interest in environmental values and
pasture-raised products.
Allowing consumers to taste samples also influences WTP. When included in consumer studies
the sensory rating is generally the most important contributor to WTP. In several series of exper-
iments auctioning two wines (Combris et al. 2009) it WTP results could be quite different from a
blind tasting versus for origin labeled products without tasting, but that WTP including tasting
with origin labels closely followed the WTP in the blind tasting, thus consideration of origin was
of little importance relative to their sensory evaluation. However, origin is primarily a signal of
sensory quality and consumer’s valuation of grass-fed beef could be quite different even with
equal liking as animal welfare or health properties might result in higher WTP. Marin and
Durham (2007) found that sensory liking overrode quality perception perceived in natural corked
versus screw-cap wines. The experimental auctions conducted by Umberger, Boxall, and Lacy
(2009) to evaluate grass-fed beef premiums or discounts also included a tasting opportunity.
They found that total bids and premiums, when positive, were lower when the participants tasted
grass-fed steaks before bidding. Yet it is clear that at least a sub-population of consumers prefers
the taste of grass-fed beef (Cox et al. 2006), which creates an opportunity to develop a niche
market for that taste preference in addition to the value consumers place on production practices
and possible health benefits. This is important information in evaluating a niche market for
grass-fed beef.
Our study further explores the sensory ratings of consumers and also looks more carefully at the
potential for a niche market for grass-fed beef by simulating product choice under a variety of
conditions to differentiate market segments. Finally, our analysis accounts for a participant’s
prior knowledge of grass-fed beef: our WTP findings, higher than found in other, earlier studies,
likely reflects an overall increase in consumer knowledge about grass-fed beef production and
nutritional qualities. Consumers in earlier studies may truly have had no prior knowledge. We
attempt to clarify the impact of prior and additional information.
Methods
A sensory consumer test was conducted at the Food Innovation Center in Portland, OR.
Participants were selected using an on-line screener, which they found by word of mouth or
through a Craig’s List advertisement. Only consumers who eat ground beef in the form of
burgers were selected. Through screening questions, a sample was recruited that was about
equally split between people that shop in mainstream food stores and those that shop primarily at
natural food stores and/or food cooperatives. This stratification of the sample was employed to
ensure that sufficient variation was obtained in the characteristics of the consumers to evaluate
how those characteristics might influence WTP for grass-fed beef. This information and related
consumer characteristics are used in the econometric model to analyze WTP for grass-fed and
conventional ground beef.
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
5
As is standard in consumer tests, the sensory evaluation took place first to avoid biasing the taste
impression due to consumers guessing test intention and product source. On the test day, ground
beef patties were prepared by weighing out approximately 3 oz. of raw product, which was then
formed into a patty and refrigerated until just prior to cooking. Each patty was salted evenly
across the top with 0.2g of salt. Ground beef samples were baked on parchment paper on a sheet
tray of 10 samples per tray, in a 450° F oven to an internal temperature of approximately 160° F.
The samples were served to the consumer one to two minutes after portioning.
Participants were given two coded beef samples simultaneously on a tray. The first instruction
was to cut the samples and give a color rating of the each on a nine point scale; they were then
asked to write down their taste preference (with a no preference option) and then to rate various
quality characteristics (color, juiciness, tenderness) and their liking of each sample on a nine
point scale (“dislike extremely” to “like extremely”). They recorded their liking rating according
to its code before proceeding with the rest of the questions. This step allowed for the blind
tasting aspect and normal sequence of the sensory test to be preserved while enabling the
participants to refer to their liking rating later in the survey.
After the sensory questions were completed participants were asked several questions about their
beef purchasing experience to gauge their willingness to buy beef directly from producers. They
then answered six choice questions about which of the two samples they would choose to buy at
various price combinations (with an option not to buy either). Before answering these choice
questions they were told that one of the samples they tasted was grass-fed and the other
conventionally produced. They were provided with the sample codes so that they could match
each sample with their overall liking of that sample from the sensory evaluation. The partici-
pants were also split into two groups to receive either a low or a high information explanation of
grass-fed and conventional production practices. The first briefly explained the forage diet of
grass-fed beef and the second added information about production practices and nutritional
characteristics. Only then did participants answer the six questions in which prices varied be-
tween the offerings. When those were completed, participants were presented with a paper ballot
with Likert scale questions to assess attitudes on the environment, health, animal welfare, food,
and nutrition.
This procedure allowed us to analyze participants’ propensity to choose grass-fed or
conventional beef given (a) differences in both price and information about beef production
methods, and (b) participants’ individual characteristics, including their liking for the samples
they tasted. Consumer variable statistics and demographic information for our sample is
presented in Table 1.
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
6
Table 1. Model Variable and Demographic Information
Consumer
characteristics
variable
Description
Population (Std. Dev.)
→ Mean after Adjustment
for Mainstream Shopper
LIK6
Liking (9 point scale) – 6 (like slightly)
6.07 (1.94) → 0.07
PRICE
Prices for beef, $2.50,$3.50,$4.50;$0 do not buy
2.33 (1.78)
MOR
More information (0=less information, 1=more)
0.509
NAT
Regular natural food store or food co-op shopper
0.554
KNB
Prior Knowledge of Grass Fed (0=Not at all 16.9%,
1=Somewhat 67.0%, 2=Very well informed 16.1%)
0.991 (0.575)
ORU
Organic buying loyalty (% of produce purchases) (ad-
justed to make ORU=0 for non-natural store shopper)
47 (27) → 18.9
AGU – age
Distribution of Age Range Selected by Individual
41.4 (14) → -2.96
18-
24
8.9%
25-29
15.2%
30-34
16.1%
35-39
8.0%
40-45
10.7%
45-49
10.7%
50-
54
8.9%
55-59
8.9%
60-64
7.1%
65-69
3.6%
70+
1.8%
INU - income
in $10,000
units
Distribution of Income Range Selected by Individual
5.26 (3.3) → -0.047
<$19,999
$20-$29,999
$30-39,999
$40-49,999
16.1%
10.7%
15.2%
13.4%
$50-
59,999
$60-79,999
$80-99,999
$100,000+
10.7%
16.1%
8.0%
9.8%
Factors (adjusted → to make factor value=0 for non-natural store shopper)
SEU
Seasonal and Local Buyer
0 (1) → 0.430
ENU
Environment
0 (1) → 0.095
FDU
Nutrition Ingredient Concerns
0 (1) → 0.222
FRU
Farm Preservation
0 (1) → -0.067
DMU
Domestic Animal Welfare
0 (1) → 0.344
Non-Model Demographic Sample Distribution
Gender
Female
0.500
Race or
Ethnicity
White
0.917
Black
0.009
Asian
0.037
Hispanic
0.018
No answer
0.018
Education
High school
0.045
Current student
0.098
Two-year degree
0.205
Bachelor’s degree
0.411
Advanced degree
0.241
Children
Presence of children in household = 1, No = 0
0.464
* Variables are adjusted to equal 0 for the baseline consumer who is a mainstream (not natural food store
or food cooperative) shopper, age 41.3, income $53,100.
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
7
Results and Discussion
Taste Preferences
There was not a statistically significant preference for either the conventional or grass-fed
ground beef. However, unlike similar studies, there was a slight, insignificant preference for
grass-fed: 54% preferred the grass-fed ground beef, 44% preferred the conventional ground beef,
and 2% had no preference. The two types of ground beef were rated similarly for overall liking.
Whether participants primarily shopped at natural food stores or mainstream food stores did not
have a statistically significant effect on their taste preferences, though mainstream shoppers were
found to have directionally higher overall liking for the grass-fed beef and rated it higher than
conventional beef in terms of sensory attributes. This is interesting, because it suggests that
natural food store shoppers aren’t necessarily a target market for grass-fed producers.
Willingness to Pay
Factors influencing consumer preference for grass-fed or conventional ground beef are evaluated
as a choice-based conjoint analysis (CBCA), following Lancaster’s theory of value and random
utility theory (Cohen 1997; Lancaster 1966; Louviere and Woodworth 1983; McFadden 1974).
According to Lancaster’s theory, a product’s utility is an additive utility based upon the utility of
the products’ attributes. To measure that utility, a random utility model is typically used which
assumes that the utility Uij of an individual i for a product j is composed of systematic and ran-
dom components. The systematic component vij is observable and a function of the product at-
tributes and individual characteristics. The random component εij is unobservable influences.
The utility of good j for consumer i can be expressed as:
(1)
Since only the systematic component of the model above is observable, it can be specified as a
function of product attributes and individual characteristics:
(2)
where αj is the marginal utility obtained due to the attributes of choice j, β is the change in mar-
ginal utility due to price pj, and γj is the change in marginal utility of an attribute due to
individual characteristics.
We built two models to evaluate WTP for grass-fed beef and to examine the consumer
characteristics that contribute to WTP a premium for grass-fed beef. The base model examines
whether providing information about grass-fed beef production methods and potential health
benefits increases WTP a premium for the grass-fed product. It also allows us to test whether
self-assessed prior knowledge of grass-fed beef production (not at all, somewhat, or well-
informed), regular food shopping location (natural store or mainstream), as well as price and
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
8
liking could provide a simple way to examine grass-fed WTP. We developed a second,
expanded model to look more closely at consumer attitudes that could lead to a preference for
grass-fed beef.
In both WTP models, the attributes of the choices are production method (grass-fed or conven-
tional) and price. Variables which vary among individuals are considered consumer characteris-
tics. In the choice set for each participant, only relative price changed for each set of questions.
The participants vary in characteristics provided by the screener, sensory liking, and information
they provided (answers to questions) that could affect how they value grass-fed and conventional
beef.
Prior to analysis, the grass-fed versus conventional attribute variable was effects coded, with the
grass-fed variable coded as 1 for the grass-fed choice, 0 for the conventional choice and -1 for
the do not buy choice. A do not buy variable is coded 1 for the do not buy choice and 0 for the
other two choices. This arrangement allows the parameter estimates to be viewed against the
conventionally produced choice as a baseline. The parameter on the ‘Do not buy’ variable can
best be interpreted in this arrangement as representing the utility (or disutility) of not buying: it
captures the utility of the conventional product to the baseline mainstream consumer, but it could
also capture a desire to give an answer and thus cannot be considered as purely the conventional
product value. There are other equally correct approaches to coding the do not buy choice and
product attributes that will produce the same predictions and log-likelihood values. Our primary
reason for using this combination is to emphasize the difference between conventional and grass-
fed. Price is entered in dollars per pound as shown to the participants, and at 0 for the do not buy
choice.
The interactive consumer characteristic variables are created by multiplying the consumer char-
acteristic variables by the effects coded grass-fed variable and additional effects are coded with a
conventional variable (1 if conventional, 0 if grass-fed, and -1 on the do not buy choice). The
effects coded variable for conventional product is used to create the interactive variables for
conventional product for ease of interpretation.
Base Model
The base model (Table 2) restricts consumer differences (z
i
) to how much they liked the respec-
tive beef product (LK6), whether they received high or low information (MOR), whether they
considered themselves knowledgeable about grass-fed beef (KNU), and whether they were a
natural store shopper (NAT).
In the models, explanatory variables are transformed to make the baseline consumer a “main-
stream” consumer with respect to the Portland area: a consumer that declared no prior
knowledge, got the low information treatment, and typically shops at a conventional supermar-
ket. We also transformed the liking score by subtracting 6 (a rating of “like slightly”) from the
score which results in a value of 0 for the baseline consumer. Thus the baseline parameter values
and inferred premiums are based on that individual with a liking level of “like slightly.”
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
9
Table 2. Base Model
Variable
Parameter
Estimate
Std.
Error
Premium =
Parameter /
Price Par.
Variable Description
GRS
0.995
***
0.329
0.94
***
Grass-fed versus conventional
NONE
-2.444
***
0.690
Not buy choice
PRICE
-1.061
***
0.107
Price
Interactive Variables-Consumer Characteristics
CNV*LK6
0.783
***
0.101
0.74
***
Liking (Like slightly=0) on Conventional
CNV*MOR
-0.054
0.189
-0.05
More production information on conventional
CNV*KNH
-0.445
**
0.185
-0.42
**
Knowledge of grass-fed on Conventional
CNV*NAT
0.315
0.196
0.30
Natural Store shopper on Conventional
GRS*LK6
0.511
***
0.064
0.48
***
Liking (Like slightly=0) on grass-fed
GRS*MOR
0.588
***
0.157
0.55
***
More information on grass-fed
GRS*KNH
0.620
***
0.146
0.58
***
Knowledge of grass-fed
GRS*NAT
0.169
0.160
0.16
Natural Store shopper on grass-fed
Log-Likelihood Constants only =-510.7
Log-Likelihood Model =-346.2
***Significant difference in impact between grass and conventional at 1% level, ** at 5% level, * at 10% level.
A significant positive (negative) parameter estimate for a variable means that the variable in-
creases (decreases) the probability that the baseline mainstream consumer will choose that prod-
uct. We estimate how much more or less a consumer will pay for the chosen product by dividing
the characteristic parameter by the parameter estimate for PRICE. Because our effects coding
makes the conventional product the baseline, the first value in the fourth column of Table 2,
from equation (2), is how much more the unknowledgeable, less informed, mainstream
store shopping consumer is WTP for the grass-fed product than for the conventional product giv-
en equal liking. Premiums listed for interactive variables tell us how the grass-fed or conven-
tional beef WTP value varies with those consumer characteristics.
Prior knowledge clearly matters. The consumer with some prior knowledge about grass-fed beef
would pay $0.55 per pound in addition to the $0.94 cent premium for grass-fed and $0.42 per
pound less for the conventionally produced beef. If the knowledge variable is excluded from the
model, a larger significant parameter for the natural store shopper would result, indicating that
they are willing to pay significantly more for the grass-fed beef. Thus it appears that it is primar-
ily the knowledge that the natural store shopper has that makes him/her willing to pay more. The
consumer who received more detailed information about the two production practices would pay
$0.59 per pound more for the grass-fed and no more for conventional beef. As noted earlier, oth-
er studies have usually found information about production practices and health to increase WTP
for grass-fed beef. As in other studies, we varied the information given to consumers and noted a
WTP premium associated with more information.
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
10
Expanded Model
We developed a second, expanded model (Table 3) to look more closely at attitudes that could
lead to a preference for grass-fed beef. This model allows us to examine which consumer charac-
teristics change the utility of grass-fed versus conventional beef and thus identify which consum-
ers are more likely to pay a premium for grass-fed.
Table 3. Expanded Model
Variable
Parameter
Estimate
Std.
Error
Premium =
Parameter /
Price Par.
Variable Description
GRS
1.073
***
0.358
0.90
***
Grass-fed versus conventional
NONE
-3.074
***
0.745
Not buy choice
PRICE
-1.199
***
0.119
Price
Interactive Variables-Consumer Characteristics
CNV*LK6
0.716
***
0.105
0.60
***
Liking (Like slightly=0) on conventional
CNV*MOR
0.006
0.210
0.01
More information on conventional
CNV*KNH
-0.382
*
0.211
-0.32
*
Knowledge of grass-fed on conventional
CNV*AGU
0.034
***
0.009
0.03
***
Age in years on conventional
CNV*INU
-0.111
***
0.036
-0.09
***
Income on conventional
CNV*ORU
0.000
0.005
0.00
Organic purchasing level (%) on conventional
CNV*SEU
-0.016
0.107
-0.01
Seasonal/Local on conventional
CNV*FDU
-0.039
0.116
-0.03
Environmental buyer on conventional
CNV*FDU
-0.140
0.110
-0.12
Nutrition/Ingredient concern on conventional
CNV*FRU
-0.145
0.106
-0.12
Farm Preservation Concern on conventional
CNV*DMU
0.031
0.093
0.03
Animal Welfare concern on conventional
CNV*GBFU
0.013
0.008
0.01
Frequency of buying gr. beef on conventional
GRS*LK6
0.663
***
0.075
0.55
***
Liking (Like slightly=0) on grass-fed
GRS*MOR
0.515
***
0.178
0.43
***
More information on grass-fed
GRS*KNH
0.456
***
0.176
0.38
**
Knowledge of grass-fed on grass-fed
GRS*AGU
-0.007
0.008
-0.01
Age in years grass-fed
GRS*INU
0.080
***
0.030
0.07
***
Income ($10,000) on grass-fed
GRS*ORU
0.007
*
0.004
0.01
*
Organic purchasing level (%) on grass-fed
GRS*SEU
0.222
**
0.092
0.19
**
Seasonal/Local on grass-fed
GRS*ENU
0.124
0.088
0.10
Environmental buyer on grass-fed
GRS*FDU
-0.086
0.090
-0.07
Nutrition/Ingredient concern on grass-fed
GRS*FRU
-0.080
0.090
-0.07
Farm Preservation Concern on grass-fed
GRS*DMU
0.298
***
0.082
0.25
***
Animal Welfare concern on grass-fed
GRS*GBFU
0.002
0.006
0.00
Frequency of buying ground beef on grass-fed
Log-Likelihood Constants only =-510.7
Log-Likelihood Model =-312.0
***Significant difference in impact between grass and conventional at 1% level, ** at 5% level, * at 10% level.
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
11
The consumer attitudinal score variables for this model were developed using the Likert scale
questions asked at the end of the consumer test. Those questions were reduced to representative
scores using principal components analysis (PCA). The benefit of using PCA to assess attitudes
is that multiple questions are used to measure an individual’s level of concern or interest about
an issue rather than relying on a single question, which might measure an individual’s attitude
less accurately due to the wording or context. PCA essentially distills multiple variables into a
smaller number of related components. For this analysis the majority of the questions have been
used and developed in previous studies. Durham (2007) used PCA to incorporate health concerns
and environmental attitudes into analysis of what motivates organic purchases. That study drew
upon Roberts (1996), who segmented consumers for their environmental orientations, and Kraft
and Goodell (1993) who did the same for health conscious consumers. McCluskey, Durham, and
Horn (2009) extended the question set to assess food interest and nutritional attitudes as well as
concerns about animal welfare and farm preservation. For this study, some additional Likert
questions were added to enhance assessment of domestic animal welfare concerns and food in-
terest. See the appendix for the questions used and details on the methodology. The factor scores
produced by the PCA and utilized in the analysis are defined as Seasonal and Local Buyer
(SEU), Environmentalism (ENU), Nutrition Ingredient Concerns (FDU), Farm Preservation
(FRU), and Animal Welfare (DMU) based on the questions contributing most strongly to the
score.
As in the base model these variables are adjusted to a mainstream shopper baseline. This trans-
formation is accomplished by taking the average attitudinal score for shoppers that did not select
natural food supermarkets or food coops when reporting where they shopped (i.e. mainstream
supermarket shoppers) and subtracting that average from the original attitudinal score. In the
third column of Table 1, the first number reported is our study population’s average for the vari-
able followed by its standard deviation in parentheses. If those are followed by an arrow (→), the
next number is the average once the transformation has taken place. For most of the attitudinal
scores the mainstream supermarket shopper was below the sample population mean which was 0
since these are standardized variables. The other variables added to the model, age, income, and
organic buying percentage are also transformed. Age is transformed by subtracting the mean age,
income by subtraction of mean income (in $10,000 units), and organic percentage by its mean
value for mainstream shoppers. The variables entered into the model are the effects coded grass-
fed variable and price for each choice, a do not buy variable, and interactive variables created by
multiplying the effects coded grass and conventional variables by the individual characteristics.
As before the variables for liking of the respective products (LK6), more information received
(MOR), and prior knowledge (KNU) are included, but not Natural Store Shopper; added to these
are interactive with age, organic purchasing level, the five attitudinal scores (SEU, ENU, FDU,
FRU, DMU), and frequency of ground beef consumption (GBFU).
In this model, the impact of knowledge drops off and is replaced by the actual attitudes associat-
ed with knowledge that might impact the grass-fed choice. Having more information is still im-
portant for grass-fed selection. We would note that attitudes about some topics are quite similar
for mainstream and natural shoppers (for example, farm preservation), while these two groups
vary more dramatically on others (for example, farm animal welfare). The two attitude scores
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
12
that are influential include support for seasonal/local food and small businesses, and concern for
farm animal welfare. Demographic variables that are influential include consumer age, which
increased selection of conventional beef, and income, which increased preference for grass-fed
and decreased preference for conventional, indicating wealth effects and/or that grass-fed is con-
sidered a premium product. More loyal organic produce consumption does not significantly im-
pact the choice of either. As expected, liking remained a key and essentially equivalent influence
on WTP for both types of beef.
Interestingly, the grass-fed to conventional WTP differential resulting from greater knowledge
declines somewhat in this model. This indicates that the knowledge variable was encompassing
the information from additional variables, which supports the idea that the attitudinal concern
underlies the knowledge impact. We drop the natural store shopper variable from the expanded
model because it was not significant once the prior knowledge variable was added to this model
or the previous model.
Results indicated that participants were willing to pay a premium for grass-fed ground beef
versus conventional, grain-fed beef. Mainstream food shoppers were willing to pay a premium of
$0.94/pound. Natural food store shoppers could only be identified as willing to pay more for
grass-fed beef in models which did not include the grass-fed knowledge or other more detailed
consumer attitude variables, and even then the natural store shopper variable did not explain
much of the premium. The natural store shopper variable by itself will be associated with many
consumer characteristics, possibly including age and income as well as individual attitudes and
beliefs. The consumer attributes more strongly associated with natural store shoppers than with
mainstream store shoppers are those that were associated with higher willingness to pay for
grass-fed beef. Other studies of WTP have noted that store type became less significant when
attitudinal information was included in the model (McCluskey, Durham, and Horn 2009). Con-
ner and Oppenheim (2008) found lower mean WTP for pasture raised livestock at grocery store
versus food co-operatives, associated with lower scores on health, animal welfare, and environ-
mental concerns.
Figure 1 graphically depicts the relative impacts of consumer characteristics on WTP for the two
types of beef sampled. Liking had the largest impact on WTP, yet the size of that impact did not
differ significantly based on whether a sample was grain-fed or grass-fed.
WTP calculations can sometimes be questioned when consumers are not actually buying. How-
ever, research has shown that while the baseline total WTP may be biased upwards, the marginal
difference between related goods is not (Lusk and Schroeder 2004). The possibility of bias in the
baseline WTP is the reason for coding the variables in order to look at differences between con-
ventional and grass-fed and what shifts the value of each rather than look at the overall price of
either. The premiums and variation in value due to consumer heterogeneity are our focus.
While the relative premiums are of great interest, it may be more informative to look at how the
basic model predicts how consumers will choose among grass-fed and conventional beef based
on price and across the knowledge and information variables. These results are shown in Figures
2 and 3.
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
13
Figure 1: Impact of consumer characteristics on WTP
Figure 2 compares three of the possible price combinations used in the WTP choice set: conven-
tional at $1 less than grass-fed at $3.50 per pound, both samples at $3.50 per pound, and grass-
fed at $4.50 per pound with conventional at $2.50 per pound. When grass-fed is priced at $4.50
per pound and conventional at $2.50 per pound, a $2 per pound price difference, the model pre-
dicts that 48% would choose grass-fed, 37% would choose conventional, and 15% would not
buy. This sizeable preference for grass-fed at a significantly higher price is likely influenced by
the fact that our consumer sample had a high proportion of natural food shoppers, and half of
them received additional information on grass-fed before making their choices.
Figure 3 shows the impact of information and prior knowledge. Consumers having some prior
knowledge of grass-fed has a larger effect than if they are given more information in increasing
the probability they will pay a premium for grass-fed beef. In the left hand pie chart we see that a
smaller proportion would select the grass-fed at a $2 per pound price difference than in our unad-
justed sample in the right hand pie chart in Figure 2, because for the pie chart on the left hand
side of figure 3 the consumer depicted is our baseline mainstream shopper with no prior
knowledge, who did not receive additional information, at the same price difference.
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
14
Figure 2: Impact of price: percent of consumers choosing grass-fed, conventional, or neither, at
given prices
Our WTP results compared with other geographic areas
Figure 3: Impact of information and prior knowledge: percent of consumers by product choice
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
15
We expected to find WTP results that matched or exceeded those found from similar taste tests
conducted elsewhere in the country, because our study area, Portland, Oregon, is a “leading
trend” market in terms of natural, local/regional, and sustainable food, a characterization
consistent with what we learned about our study participants. Most were aware of and interested
in “sustainable” food production, including grass-fed meats, regardless of typical shopping
location: 72% of mainstream food store shoppers said they were at least somewhat informed
about the possible benefits of grass-fed beef and cattle production. Mainstream shoppers ex-
pressed a very favorable view of grass-fed beef: 86% perceived it as healthy, 60% as more hu-
mane than conventional beef, 62% as better for the environment than conventional beef, 50% as
flavorful, and 56% as safe. A large majority – 64% of mainstream shoppers and 84% of natural
food store shoppers – said that food safety concerns had had an impact on them, and a slight ma-
jority of all participants (51%) have switched to natural or organic beef in the last few years due
to those concerns.
Given this consumer base, a higher-than-average WTP would not have been surprising. Indeed,
compared with earlier studies, our results are much more favorable to grass-fed. Our study esti-
mated WTP a premium of $0.90-94 per pound for grass-fed ground beef, approximately 35-40%
higher than WTP for conventional at equal sensory liking. A decade ago, Feuz and Umberger
(2001) found quite the opposite: WTP was 26% lower for grass-fed beef compared with grain-
fed beef.
Yet when compared with more recent studies, our WTP results are fairly average. We expect this
is due to an increase, over the last decade, in general consumer knowledge of grass-fed meat
production and exposure to grass-fed meat products. More recent studies have found WTP pre-
miums for grass-fed ranging from approximately -37% (Sitz et al. 2005) to about 180% (Evans,
Brown, Collins, D'Souza, Rayburn, and Sperow 2011). WTP results also vary based on whether
the test is done with ground beef or muscle cuts (steaks): comparisons using steaks (Cox et al.
2006) found a smaller premium for grass-fed than comparisons using ground beef, as this study
did. This is consistent with the fact that most U.S. consumers expect steaks to be marbled, for
both flavor and marbling’s relationship to tenderness; grass-fed steaks tend to be less marbled
than grain-fed steaks. The difference is much less noticeable in ground beef, and our WTP re-
sults for grass-fed are much higher than when steaks were used (-9.5% and 3.0%; Cox et al.
2006). That our results were lower than those of other ground beef research (Evans et al. 2011)
may in part be due to the other ground beef research being conducted not in a large metropolitan
area but in a smaller, university town where participants may be more educated (even more so
than in Portland) about alternative food choices in general and grass-fed specifically. It is also
important to remember when comparing the size of WTP effects that there has been a great deal
of heterogeneity in the WTP and preference literature related to alternative versus conventional
meats, with respect to participants’ demographic profiles, sample selection methodologies, and
how premiums have been calculated.
The most distinguishing feature of our study, in comparison with others, is that we found prior
knowledge, in lowering the value of conventional as well as raising the value of grass-fed beef,
to be more influential than information provided during the test. Furthermore, few studies have
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
16
considered whether interest in farm animal welfare is influencing consumer choice. While other
studies have more clearly differentiated between production information and nutrition infor-
mation, they have not examined consumer attitudes regarding production practices. In our model
the factor representing concern about farm animal welfare is associated with a higher premium
for grass-fed. Somewhat to our surprise, the nutrition and ingredient concern variable was not so
associated. This may be explained by a more direct response through the information and
knowledge premiums or because ground beef is not generally considered a health food.
Willingness to Buy in Bulk
To understand market potential, grass-fed beef producers and marketers need to have at least an
estimate of how many consumers will pay how much of a price premium for their product. Yet
information about demand based on purchase format – specifically, how many consumers are
willing to buy frozen cuts in bulk (by the whole, half, or quarter carcass) – is almost as critical as
price premium information, especially for small-scale operations selling fewer than 100 head per
year. The significant cost associated with processing, packaging, distribution, inventory man-
agement, and retailer-required margins for small-volume, unconventional meats can drive up the
overall cost of production, hence the price for consumers. Selling in bulk, direct to consumers,
can lower these costs. But how many consumers are willing (and able) to buy in bulk?
Nearly a quarter of our study participants (24%) responded that they had previously purchased
beef, as a whole, half, or quarter carcass, direct from a rancher, at least once, a surprisingly high
result, even though the question was about beef in general and not only grass-fed beef. Of the
other 76% who had not purchased beef this way, 69% would consider purchasing bulk beef if
they knew a producer that sold it or if a friend recommended a source. Price matters, too: 73%
would consider purchasing this way if it were less expensive than the beef they are currently
buying. Also notable is that 72% of all participants were willing to purchase frozen beef. This
creates more options for producers; fresh product is more perishable and must be sold much
more quickly, which complicates distribution logistics.
Participants who had not yet purchased bulk beef were asked why (by selecting all that applied
from a given set of potential reasons). The most chosen reason, selected by 58% of respondents
to that question, was that there is too much meat associated with such a purchase; the second,
selected by 55%, was that they lack the freezer space for so much meat. The “too much meat”
problem is a serious challenge for producers: 50% of participants said they purchase three
pounds of beef or less per month. At that rate, even a quarter of a beef (more than 100 pounds)
would be fairly overwhelming.
Whether participants primarily shopped at natural food stores or mainstream food stores did not
have a statistically significant effect on whether they had purchased beef in bulk or were willing
to do so. Natural food store shoppers were more likely than mainstream shoppers to consider
buying in bulk if they knew the producer. Only a few other factors were significantly related to
willingness to buy in bulk: current bulk buyers tended to be older and shop at warehouse stores.
Prospective bulk buyers tended to be older and eat more beef than average.
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
17
Conclusions
Our study of consumer preferences and WTP for grass-fed beef has four primary findings. First,
our WTP results for grass-fed beef are within the bounds of those found elsewhere in the coun-
try, when compared only with more current studies. We suggest recent WTP estimates are higher
than older studies due to an increase in general consumer knowledge about grass-fed over the
years. Second, if participants in this study are representative of the Portland Metro region, there
is significant interest in the region in buying beef in bulk, i.e. sacrificing some convenience to
purchase grass-fed beef. Third, we confirm other research findings that whether a consumer typi-
cally shops at natural food stores or mainstream stores does not matter to WTP or willingness to
buy in bulk. Fourth, we find that knowledge about production and nutritional qualities, and also
attitudinal variables, are what matter instead. For example, the premium consumers are WTP for
grass-fed beef increases when consumers know something about possible health benefits associ-
ated with it. When knowledge and attitudinal variables are known and included in the model, the
effect of shopping location drops away.
We acknowledge that our results are drawn from a very small sample. Furthermore, our results
may not extend far beyond the Portland metro region, except for other, similarly progressive,
food-oriented areas. Most participants in our study had previously tried grass-fed beef, suggest-
ing that this product is available and familiar locally and possibly that the participants are adven-
turous when it comes to food. Finally, our WTP/premium estimates, though relatively large, are
not necessarily high enough to assure a profit for grass-fed producers. Producers will need to find
consumers that are not only willing to buying grass-fed beef for its taste, production practices,
and potential nutritional benefits, but are willing to pay enough of a premium to cover the addi-
tional production and supply chain costs for this unconventional product.
Acknowledgements
Research conducted at the Food Innovation Center, Portland Oregon and partially supported by
the USDA-CSREES Multicommodity Research Special Grant. The authors would like to thank
Carman Ranch for the contribution of the grass-fed beef used in this study.
References
Cohen, S. H. 1997. "CBCA Marries the Best of Conjoint and Discrete Choice Models."
Marketing Research 9:12-17.
Combris, P., P. Bazoche, E. Giraud-Héraud and S. Issanchou. 2009. "Food Choices: What Do
We Learn from Combining Sensory and Economic Experiments?" Food Quality and
Preference 20(8):550-57.
Conner, D. S. and D. Oppenheim. 2008. "Demand for Pasture-Raised Livestock Products in
Michigan: Results of Consumer Surveys and Experimental Auctions." Journal of Food
Distribution Research 39(1):1-20.
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
18
Cox, R. B., C. R. Kerth, J. G. Gentry, J. W. Prevatt, K. W. Braden, and W. R. Jones. 2006.
"Determining Acceptance of Domestic Forage or Grain Finished Beef by Consumers
from Three Southeastern Us States." Journal of Food Science 71(7):S542-S46.
Daley, C. A., A. Abbott, P. S. Doyle, G. A. Nader and S. Larson. 2010. "A Review of Fatty Acid
Profiles and Antioxidant Content in Grass-Fed and Grain-Fed Beef." Nutrition Journal
9(1):10.
Evans, J. R., C. Brown, A. R. Collins, G. E. D'Souza, E. B. Rayburn, and M. Sperow. 2011.
"Determining Consumer Perceptions of and Willingness to Pay or Appalachian Grass-
Fed Beef: An Experimental Economics Approach." Agricultural and Resource
Economics Review 40(2):233-50.
Fanatico, A. and L. Rinehart. 2006. "Beef Marketing Alternatives." National Center for
Appropriate Technology: Butte, MT. 28 pages. https://attra.ncat.org/attra-
pub/summaries/summary.php?pub=200.
Feuz, D. M. and W. J. Umberger. 2001. "Consumer Willingness-to-Pay for Flavor in Beef
Steaks: An Experimental Economics Approach." Cornhusker Economics. University of
Nebraska: Lincoln.
Feuz, D. M., W. J. Umberger, C. R. Calkins and B. Sitz. 2004. "Us Consumers' Willingness to
Pay for Flavor and Tenderness in Steaks as Determined with an Experimental Auction."
Journal of Agricultural and Resource Economics 29(3):501-16.
Gwin, L. 2009. "Scaling-up Sustainable Livestock Production: Innovation and Challenges for
Grass-Fed Beef in the U.S." Journal of Sustainable Agriculture 33(2):189 - 209.
Gwin, L., D. Evans and L. Brewer. 2011. "Marin Sun Farms: A Case Study in Multi-Species,
Multi-Market, Alternative Livestock Production." Small Farms Technical Report. Oregon
State University: Corvallis, Oregon.
Gwin, L. and A. Thiboumery. 2012. "Processing Meat and Poultry for Local Markets."
Economic Research Report. U.S. Department of Agriculture, Economic Research
Service: Washington DC.
Hardesty, S. D. and P. Leff. 2009. "Determining Marketing Costs and Returns in Alternative
Marketing Channels." Renewable Agriculture and Food Systems 25(1):24–34.
Jackson, E. 2009. "Home from the Range." Oregonian, Portland, OR: September 8, 2009.
http://www.oregonlive.com/foodday/index.ssf/2009/09/this_is_the_local_meats_story.ht
ml.
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
19
King, R. P., M. S. Hand, G. DiGiacomo, K. Clancy, M. I. Gómez, S. D. Hardesty, L. Lev, and E.
W. McLaughlin. June 2010. "Comparing the Structure, Size, and Performance of Local
and Mainstream Food Supply Chains." Economic Research Report ERR-99. United
States Department of Agriculture: Washington, DC. 81 pages.
http://www.ers.usda.gov/media/122609/err99_1_.pdf. (Accessed July 9, 2012).
Lancaster, K. 1966. "A New Approach to Consumer Theory." The Journal of Political Economy
74(2):132-57.
Larson, S., C. Thompson, K. M. Klonsky and P. Livingston. 2004. "Sample Costs for a Cow-
Calf Grass-Fed Beef Operation." University of California Cooperative Extension: Davis,
CA. http://coststudies.ucdavis.edu/files/beefgfnc2004.pdf. (Accessed July 9, 2012).
Louviere, J. J. and G. Woodworth. 1983. "Design and Analysis of Simulated Consumer Choice
or Allocation Experiments: An Approach Based on Aggregate Data." Journal of
Marketing Research 20(4):350-67.
Lusk, J. L. and N. Parker. 2009. "Consumer Preferences for Amount and Type of Fat in Ground
Beef." Journal of Agricultural and Applied Economics, 41(1):75-90.
Lusk, J. L. and T. C. Schroeder. 2004. "Are Choice Experiments Incentive Compatible? A Test
with Quality Differentiated Beef Steaks." American Journal of Agricultural
Econonomics, 86(2):467–82.
Marin, A. B. and C. A. Durham. 2007. "Effects of Wine Bottle Closure Type on Consumers’
Purchase Intent and Price Expectations." American Journal of Enology and Viticulture
58(2):192-201.
Mathews, K. H. and R. J. Johnson. 2010. "Grain and Grass Beef Production Systems." In
Livestock, Dairy, & Poultry Outlook, Available at
http://www.ers.usda.gov/publications/ldp/2010/06Jun/ldpm192.pdf. (Accessed July 9,
2012).
Mayer, J. 1999. "“Will Organic and Natural Beef Provide Growing Markets?" Angus Journal
(October):192-94.
McCluskey, J. J., C. A. Durham and B. Horn. 2009. "Consumer Preferences for Socially
Responsible Production Attributes across Food Products." Agricultural and Resource
Economics Review 38(3):345–56.
McCluskey, J. J., T. I. Wahl, Q. Li and P. R. Wandschneider. 2005. "Us Grass-Fed Beef:
Marketing Health Benefits." Journal of Food Distribution Research 36(3):1.
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
20
McFadden, D. 1974. "Conditional Logit Analysis of Qualitative Choice Behavior." In Frontiers
in Econometrics. P. Zarembka, ed. New York: Academic Press 105-42.
National Trust. 2012. "What’s Your Beef?": Swindon, UK. 15 pages.
http://www.nationaltrust.org.uk/servlet/file/store5/item842742/version1/What%27s%20y
our%20beef.pdf. (Accessed 6/ 27/ 2012).
Schmidt, J. 2010. "Trends in the Production and Marketing of Grass-Fed Beef." Sustainable
Agriculture Conference: Connecting Cows, Carbon, & Carrots. Emporia, KS.
Sitz, B., C. R. Calkins, D. M. Feuz, W. J. Umberger and K. M. Eskridge. 2005. "Consumer
Sensory Acceptance and Value of Domestic, Canadian, and Australian Grass-Fed Beef
Steaks." Journal of Animal Science 83(12):2863.
Thiboumery, A. and M. Lorentz. 2009. "Marketing Beef for Small-Scale Producers." Iowa State
University Extension: Ames, IA.
http://www.extension.org/mediawiki/files/0/00/Marketing_Beef_for_Small-
Scale_Producers.pdf. (Accessed July 9, 2012).
Thilmany, D. D., W. J. Umberger and A. R. Ziehl. 2006. "Strategic Market Planning for Value-
Added Natural Beef Products: A Cluster Analysis of Colorado Consumers." Renewable
Agriculture and Food Systems 21(3):192-203.
Umberger, W. J., P. C. Boxall and R. C. Lacy. 2009. "Role of Credence and Health Information
in Determining Us Consumers' Willingness-to-Pay for Grass-Finished Beef." Australian
Journal of Agricultural and Resource Economics 53(4):603-23.
Umberger, W. J., D. D. Thilmany McFadden and A. R. Smith. 2009. "Does Altruism Play a Role
in Determining U.S. Consumer Preferences and Willingness to Pay for Natural and
Regionally Produced Beef?" Agribusiness 25(2):268-85.
Varnold, K. A., C. Calkins, B. L. Gordon and W. I. Umberger. 2011. "Marketing Source-
Verified Beef to Restaurant Patrons." In 2011 Nebraska Beef Cattle Report. Available at
http://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1653&context=animalscinbcr.
(Accessed July 9, 2012).
Williams, A. 2010. Allen Williams, Ph.D., of Livestock Management Consultants to Lauren
Gwin. Sept. 14, 2011.
Xue, H., D. Mainville, W. You and R. M. Nayga Jr. 2010. "Consumer Preferences and
Willingness to Pay for Grass-Fed Beef: Empirical Evidence from In-Store Experiments."
Food Quality and Preference 21(7):857-66.
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
21
Appendix
The following questions, listed by the score they contributed to most strongly, were included in
the principal components analysis.
Domestic Animal Welfare: “It is important to treat farm animals humanely,” “I’m concerned
about the welfare of domestic farm animals,” and “I buy free range chicken eggs.”
Environmental: “I have switched products for environmental reasons,” “I will not buy a product
if the company who sells it is ecologically irresponsible,” “I have purchased products because
they cause less pollution,” “I do not buy household products that harm the environment,” “I have
convinced members of my family or friends not to buy some products that are harmful to the
environment.”
Farm Preservation: “I’m concerned about the survival of family farms in the United States,” “I
am concerned about the loss of family farms in my region,” “I would vote for referendums or
initiatives to preserve farmland.” Health Concern (dropped from analysis because not
significant): “I read more health-related articles than I did 3 years ago,” “I worry that there are
harmful chemicals in my food,” “I’m concerned about my drinking water quality.”
Nutrition/Ingredients: “I avoid foods from animals produced with hormones or antibiotics,” “I
am interested in information about my health,” “I avoid foods containing nitrites or
preservatives,” “My daily diet is nutritionally balanced.”
Seasonal Food/Local Business: “I buy ‘environmentally friendly’ products even if they are
more expensive,” “I buy from small and local businesses,” “I seek out seasonal and local
ingredients,” “I like to eat out in restaurants that feature local and seasonal foods,” “I buy food
from local farms and ranches whenever I can.” A few questions contribute to more than one
category: “I buy free range chicken eggs” contributes to Seasonal/Local; “I buy food from local
farms and ranches whenever I can” contributes to Farm Preservation. The PCA was performed
SAS 9.3 using the factor procedure, retaining eigenvalues greater than 1, with varimax rotation
and Kaiser normalization. Additional details are available from the authors.
Endnotes
i
Actual text: “The sample labeled Grass-fed came from an animal whose diet was only grass and
forage. These animals cannot be fed grain or grain byproducts. The sample labeled Conventional
means the beef was purchased at a retail supermarket from conventional sources.”
ii
Actual text: “The sample labeled Grass-fed came from an animal whose diet was only grass
and forage. These ‘animals cannot be fed grain or grain byproducts and must have continuous
access to pasture during the growing season.’ Grass-fed beef has also been found to have higher
levels of Omega-3 fatty acids which have been shown to reduce the risk of certain cancers and
brain disorders. These cattle were also raised without antibiotics or hormones. The sample la-
Gwin, Durham, Miller and Colonna Journal of Food Distribution Research
July 2012 Volume 43, Issue 2
.
22
beled Conventional means the beef was purchased at a retail supermarket from conventional
sources. In the United States beef cattle are typically finished with a grain diet in a feedlot.
Ground beef can come from a variety of sources and may come from dairy as well as beef cat-
tle.”
iii
The paper ballot was used to save time on these questions.
iv
Note that because of the effects coding the full base ‘utility’ of the ‘Do not buy’ choice is the
sum of its own parameter and the negative of the parameter for grass fed beef.
v
We calculated the -37% premium from domestic conventional WTP of $3.95 versus WTP of
$2.48 for an Australian grass-fed sample (Sitz et al. 2005). We calculated the 180% premium
from (Evans et al. 2011) ground beef estimates of $0.44 for grain-fed versus $1.23 for grass-fed.
