Southwest Business and Economics Journal/2010
Hormone Induced Milk and Product Labeling – Empirical Evidence from Kansas
Kalyan (Kal) Chakraborty
Emporia State University
Abstract
The study investigates the importance of mandatory labeling for hormone induced
milk and milk products in the light of Senate Bill 595 passed by Kansas Legislature in
December 2008. The study is based on an empirical research done by the author in 2005
using mail-in questionnaire survey data from Kansas households. The economic value of
milk market segmentation based on mandatory labeling and consumers’ willingness to
pay for hormone-free milk far outweighs voluntary labeling. The information on
willingness to pay would help the dairy farmers to decide whether or not to adopt bST
technology (a growth hormone for enhancing milk production in dairy cows) and its
management practices to increase milk production. The empirical results show if the
consumers are willing to pay a higher price for bST-free milk, social benefit would be
maximized from mandatory labeling for milk and milk products.
I. Introduction
Mandatory labeling of Milk and milk products from cows treated with growth
hormone (commonly know as bST, rbST, or rbGH) has been a debatable issue
nationwide since FDA approved its use for milk production in 1994. Currently labeling
milk and milk products is voluntary in the US. More and more U.S. consumers are
demanding for more information on the content and characteristics of milk and milk
products from mandatory labeling. Several states such as Vermont, Maine, Pennsylvania,
Indiana, and New Jersey have introduced mandatory labeling. Similar bill is on the table
in Ohio awaiting approval from its legislature. In December 2008 Kansas legislature
passed Senate Bill-595 that would require any producer labeling their milk (voluntarily)
as coming from cows free from artificial hormone need to include a disclaimer stating
that the milk is not different from cows treated with hormone. For example, it will allow
the producers of milk, milk products, and dairy products to claim on their product label,
“this milk is from cows not supplemented with rbST” as long as the label contains a
disclaimer statement, “the FDA has determined that no significant difference has been
shown between milk derived from rbST-supplemented and no-rbST supplemented cows.”
(KDA, 2008). The above regulation is effective from January 1, 2010, for all one-time
use containers for milk and dairy products. This bill is far short of what the consumer
associations, environmental groups, and animal activists in Kansas are demanding for – a
mandatory labeling for all milk and milk products that use growth hormone. Consumers
perceive bST-treated milk as unsafe and a health risk because of its unknown long term
effect on human health.
The objective of the current study is to evaluate the impact of Kansas Senate Bill595 on the use of growth hormone for milk production and review the results obtained
from an empirical study done by the author using Kansas household’s data on the
economic impact of mandatory labeling for milk and milk products (Chakraborty, 2006).
A brief survey of the literature is provided in the next section followed by a
discussion on the empirical study on willingness to pay by the Kansans for bST-free
milk. The third section provides a discussion on the dataset and empirical results from the
29
Hormone Induced Milk and Product Labeling
– Empirical Evidence from Kansas
economic impact of mandatory labeling for bST-induced milk and milk products. The
last section provides the summary and conclusions.
II. Brief Review of the Literature
Recombinant bovine somatotrophin (rBST or bST) is a growth hormone that
occurs naturally in cows but is also produced by genetically altered bacteria in the
laboratory. The US Food and Drug Administration (FDA) approved the product in 1994
for its use among dairy cows. The product was originally produced and marketed by the
US biotech giant Monsanto in the commercial name of ‘Posilac’ but has been sold to Eli
Lilly & Co. in August 2008. When injected into dairy cows every two weeks, it increases
lactation by 15-20 percent and feed efficiency by 10-15 percent raising the profit per cow
per year on average $100. Most of the scientific community such as, FDA, World Health
Organization (WHO), American Medical Association (AMA), American Dietetic
Association (ADA), and other US regulatory agencies concluded that milk produced from
bST-induced cows are safe for human and animal consumption and allow voluntary
labeling of such milk and milk products.
However, over the past 16 years consumer associations and animal rights activists
expressed strong opposition against the use of hormone for milk production. Independent
researchers and critics say that the milk from bST-treated cows contains higher levels of a
different hormone – insulin-like growth factor (IGF-1) which has been linked to an
increased risk of cancer in people. Further, the use of hormone increases the risk of udder
inflammation among the dairy cows (a disease called mastitis) which is treated with
antibiotics causing high residue of antibiotics in processed milk (Hansen and Perry,
2008).
A considerable number of studies in the literature found consumers are willing to
pay a premium for bST-free milk and have expressed their intention to reduce milk
consumption if bST-free milk is not available in the market. The National Marketing
Institute’s 2006 Health & Wellness Trend Database found that 53 percent of primary
grocery shoppers said they were looking for dairy, meat, and poultry product that are free
from antibiotics (SFGate, 2007). In response to the growing consumer demand for bSTfree milk some of the major retailers in the US have stopped buying milk from dairy
farmers and milk-cooperatives that use growth hormone. (i.e., Wal Mart, Dillions,
Kroger, Safeway, Food-4-Less, Fred Meyer, Smiths, Starbucks Coffee, Save Mart’s
Sunnyside Farms and Bayview Milk, and Ben and Jerry Ice Cream). According to the US
Department of Agriculture, the number of dairy cows injected with bST nationwide had
dropped from 22.3 percent in 2002 to 17.2 percent in 2007. In early 2008 the largest
dairy cooperative in Michigan, Michigan Milk Producers Association, banned use of
hormone for all its 1,515 dairy farmers in Southern Michigan.
In January 2007, California Dairies’ Board voted to ask its members not to inject
synthetic bovine growth hormone into their cows, if they do their milk had to be
segregated and they will pay a surcharge. The vast majority of milk sold in Northern
California is rbST-free. Several countries in the world banned the use of bST for milk
production such as, Japan, Australia, New Zealand, Canada, and all 27 nations of the
European Union. Codex Alimentarius, the United Nations main food safety body, twice
decided that it could not endorse the safety of bST for human health.
30
Southwest Business and Economics Journal/2010
III. The Dataset and the Survey Results
The consumer’s willingness to pay for bST-free milk was estimated by the author
using survey data collected from 644 Kansans in 2002. A detail discussion on the dataset,
questionnaire, and methodology is beyond the scope of this paper, hence, only the
definition of the variables used to estimate the model is briefly discussed here. However,
interested readers might consult Chakraborty (2005).
The information obtained from the survey reported in Table 1 indicate that about
70 percent of the respondents have no knowledge on bST technology and 28 percent have
some knowledge.
Table 1: Definition and Descriptive Statistics of the Variables Used in the Model
About 74 percent of the respondents strongly support in favor of labeling milk
and milk products produced from cows treated with bST and 76 percent perceive bST
produced milk and milk products are health risk. Initially, 38 percent of the respondents
were unwilling to pay a premium, 26 percent were willing to pay a premium, and 36
percent were uncertain. However when these “uncertain” respondents were further
exposed to a different set of questions including price discounts for bST-milk and
premium for bST-free milk, 19 percent of the ‘don’t know’ respondents changed their
31
Hormone Induced Milk and Product Labeling
– Empirical Evidence from Kansas
decision in favor of bST-treated milk and 61 percent changed in favor of bST-free milk.
After rearranging the respondents according to their willingness to pay the final
percentage for each group in the entire sample turned out as this: 45 percent favored bSTtreated milk, 47 percent favored bST-free milk, and 8 percent still remained undecided. 1
IV. Consumer’s Willingness-to-Pay Model and the Empirical Results
A binomial probit model for consumers’ willingness to pay for bST-free milk is
estimated. Assuming Z i represents unobserved response variable which is determined by
a vector of independent variables x i by the regression relationship:
(1)
Z i = β′xi + ε i where ε i ~ N (0,1)
However, the unobserved variable Z i is related to a binary variable y = 1 if a respondent
is willing to pay a premium for bST-free milk (yes), and y = 0, otherwise (no). Hence, Z i
y = 1 if Z i > 0
is observable when
y = 0, otherwise
z
1 i −s2 / 2
Pi = Pr( y = 1) = Pr( Z i ) =
e
ds
2π −∫∞
(2)
Where Z i is computed from the cumulative normal probability function where s ~ N (0,1)
To obtain an estimate of Z i , the inverse of the CDF is used Z i = F −1 ( Pi ) = β′xi the
parameters of β can be estimated by maximum likelihood method using the log
likelihood function.
Estimated coefficients from the Probit model are reported in Table 2. The
coefficients on most of the variables, except for CHILD and AGE, have expected signs
and are significant at 0.05 or 0.20 levels. Positive and significant coefficients on
MILKLABEL and FOODLABEL imply that a consumer would most likely be willingto-pay a higher price if he/she strongly supports mandatory labeling on milk and milk
products. Negative and significant coefficient on EDUCATION implies an increase in
educational level would reduce the probability of a consumer’s willingness to pay for a
higher price for bST-free milk.
32
Southwest Business and Economics Journal/2010
Table 2: Results from the Binomial Probit Model of Willingness-to-Pay
An intuitive explanation for this might be that the consumers with more formal
education probably have a better understanding of the true risk associated with bSTtreated milk. This suggests that more educated consumers would be less willing-to-pay a
higher price for bST-free milk. The results are consistent with other willingness-to-pay
studies in food safety literature (Wang et al. 1997; and Misra et al. 1991).
The marginal effects for each independent variable, holding other variables at
their mean, are reported in column 4 of Table 2. For example, the marginal probability on
the variable EDUCATION is -0.1101 which implies that remaining all other things
constant, a person with a college degree or higher would have 11.01 percent lower
probability of buying hormone-free milk. In order to estimate the economic value (sum
of consumer and producer surplus) from milk market segmentation, elasticity of demand
for both bST-induced and bST-free milk was estimated from the respective demand
curves.
The empirical estimation of the demand curves are based on the following linear
specification:
Demand bST / non−bST = α + β1 Pr ice + β 2Child + β3 Family + β 4 Race
+ β5 Married + β 6 Edu + ε
(3)
Estimated coefficients of the demand equations using ordinary least square
regression are reported in Table 3. The F statistics for both equations are significant
although the coefficients of determination are low but reasonable considering most of the
explanatory variables in this model are either binary or categorical and the data is crosssection. Most of the coefficients have expected sign and are significant, except for the
coefficient on PRICE. One of the possible reasons for insignificant PRICE coefficient
might be the aggregation bias.
33
Hormone Induced Milk and Product Labeling
– Empirical Evidence from Kansas
Table 3: Estimates of Linear Demand Equations for bST and non-bST milk
(Dependant variable = quantity of milk consumed)
Since the real price for bST-induced milk is not existent in the market at the time
the survey was conducted, the information on this variable is interpolated from the
consumers’ responses on their willingness to pay a higher price for such milk at
incremental prices. The negative coefficient on PRICE implying $1.00 increase in price
would reduce quantity demanded for bST-treated milk by 0.37 gallon and for bST-free
milk by 0.18 gallon. The point price elasticity of demand for bST-treated milk and bSTfree milk are calculated as -0.20 and -0.12, respectively. This suggests that for a one
percent increase in price the demand for bST-treated and bST-free milk would decrease
by 0.20 and 0.12 percent, respectively. Own price elasticity suggests bST-free milk is
more inelastic than bST-induced milk. Consistent with the literature a linear supply curve
with unit elasticity of supply is assumed for simplicity and convenience measuring
producers’ surplus. According to a study by Blayney (2002), the total quantity of milk
produced and supplied in Kansas for 2002 is 1,520 million pounds at a price of $12.61
per thousand weights (cwt). This information is used in the current study.
V. Economic Impact of Mandatory labeling
The producer and consumer surplus are measured under various scenarios based
on the estimated linear demand curves and hypothetical supply curves under constant
elasticity of supply assumption. The economic impact of mandatory labeling for bSTmilk is analyzed under two hypothetical market scenarios. In Table 4 Scenario-1, it is
assumed that only bST-treated milk is available in the market (voluntary labeling). As a
result, 47 percent of the consumers who preferred bST-free milk would stop buying milk
reducing the demand by the same percentage (observed from the current survey).2 It is
also expected that the use of bST technology would increase milk supply and reduce
price. However, in Scenario-1, it is assumed that although the supply increases by 10
percent price remains same. In order to calculate the producer and consumer surplus, the
first step is to determine the equilibrium price and quantity. From the supply and demand
curves reported in Table 4 Scenario-1, the equilibrium price for bST-induced milk is
$5.72 and the quantity is 758.85 mill pounds. The estimated consumers’ surplus,
producers’ surplus, and the total surplus for Scenario-1 are $26,510, $2,170, and $28,680,
respectively.
34
Southwest Business and Economics Journal/2010
Table 4: Mandatory Labeling and Economics Value for Milk Market Segmentation
In Scenario-2 it is assumed that (under mandatory labeling) both bST-treated and
bST-free milk are available and two distinct milk markets exist. Under this situation the
total surplus from the segmented milk market would be the sum of the consumers’ and
producers’ surpluses generated from the two markets. Scenario-2A represents the market
for bST-treated milk, assuming supply increases by 10 percent and price decreases by 15
percent. Scenario-2B represents the market for bST-free milk, assuming supply does not
change and the price increases by 20 percent. Following the same procedure (explained in
Scenario-1) the surpluses generated from bST-treated milk and bST-free milk under
Scenario-2A and 2B are $31,670.84 and $41,215.16, respectively. The total surplus under
Scenario-2 is $72,886.00. This total surplus (the economic value) from Scenario-2 is 250
percent higher than the total surplus generated from Scenerio-1. It is evident from the
empirical analysis that under certain assumptions, the milk market segmentation would
generate higher social benefit.
The effectiveness of market segmentation based on real or perceived product
differentiation largely depends on how long the market would be separated. So far, it is
assumed that there is no real difference in the marketing cost between single and
differentiated markets. If it does, then further empirical research should be directed
measuring the social cost and benefit of market segmentation under mandatory labeling.
VI. Summary and Conclusions
Although conventional wisdom suggests that consumer’s perceived risk over
genetically modified food would decrease over time due to lack of evidence of adverse
effects from the use of such products, but that is not true according to this study. In fact,
as more and more consumers around the world are exposed to GM food products
including milk, we have more evidences now than ever before on the adverse impact of
such products on human health. Even after sixteen years of its introduction in the US
consumers’ negative attitude and perception about bST-treated milk have not changed.
35
Hormone Induced Milk and Product Labeling
– Empirical Evidence from Kansas
By rejecting to accept bST-induced milk the nation’s leading grocery chains and
producers of milk products are sending a clear and strong signal to the regulating
authorities about the consumers’ sentiment and their right to know what they buy. The
empirical evidence from this study suggests that the producers of bST-free milk would be
able to sell their product at a premium if mandatory labeling is enforced.
In total disregard to consumers’ demand for mandatory labeling the US Food and
Drug Administration (FDA) is set to approve the meat and milk from cloned animals as
food for human consumption by the end of this year. Very soon these products would
make their way to the supermarket shelves. The FDA does not require food made from
cloned animals and their offspring to be labeled but the producers could apply for the
right to label their foods as “clone-free.” In an opinion poll survey on Food and
Biotechnology by Pew Initiative in 2006, it was found that 64 percent of Americans were
uncomfortable with animal clones and strongly believe on mandatory labeling for such
products (ABC-News, 2008). A research by the International Dairy Foods Association
estimated that the $20 billion dairy market could fall 15 percent if cloned milk is
introduced (CBS4, 2008). Consumer groups, including the Consumer Federation of
America have argued that more precautions should be made to protect the safety of food
supply and food products made from cloned animals which must be separated and
labeled. In the word of Jennifer R. Thornley (2001):
“….there is considerable evidence indicating that mandatory labeling is necessary
to maximize the choice consumers have over the products they purchase….. As long
as a state is careful in its choice of words on the label - asserting truthful, nonmisleading information - there is quite possible enough evidence for it to emerge
victorious in convincing the court that mandatory labeling appeases a legitimate
consumer concern, places the choice in the proper hands, and is not too restrictive in
accomplishing its goals.”
36
Southwest Business and Economics Journal/2010
End Notes
1
For simplicity and convenience these consumers were later arbitrarily assigned to the
group that is “willing to buy bST-treated milk.” As a result, the total surveyed consumers
willing to buy bST-treated milk increased to 53 percent.
2
The proportion of Kansans unwilling to buy bST-treated milk in this study is much
higher than the national average (20%) from some of the studies done in the later part of
1990’s. The argument that these consumers might buy organic milk if bST-free milk is
not available, resulting milk demand as a whole would not shrink is not sustainable. This
is because organic milk is still not a viable alternative due to its high price and current
low market share (3%).
37
Hormone Induced Milk and Product Labeling
– Empirical Evidence from Kansas
References
ABC-News. (2008). Many US Consumers Oblivious to GM Food Fears.
<http://www.aseanfood.info/Articles/11021198.pdf>, January 17, 2008.
Blayney, D. P. (2002). The Changing Landscape of U.S. Milk Production. Statistical
Bulletin. No. 978. United States Department of Agriculture, Economic Research
Services.
CBS4. (2008). Meat from Cloned Animals Approved by FDA.
<http://www.cbs4.com/cosnumer/cloned.food.animal.2.630258.html>, January
16, 2008.
Chakraborty, K. (2006). Product Labeling and Economic Value of Milk Market
Segmentation. Journal of International Food and Agribusiness Marketing,
18(1/2), 9-27.
Chakraborty, K. (2005). Consumers’ Attitude Toward Hormone-free Milk: What Have
We Learned? Applied Economics Letters, 12(10), 633-637.
Hansen, M. & Perry, R. (2008). Lobbyists Try to Keep Milk Information from
Consumers.
Kansas Department of Agriculture (KDA). Fact Sheet – Dairy Labeling Regulation –
December 2008.
Misra, S. K., & Clem, K. D. (1998). Demand for Milk Produced With and Without
Recombinant Bovine Somatotrophin. Journal of Agribusiness, 16, 129-40.
SFGate – Got rbST in Your Milk? Dairy Cooperatives Bow to Pressure to Stop Use of
Hormone. <http://www.sfgate.com>.
Thornley, J.R. (2002). Got Hormone-Free Milk? Your State May Have Enough Interest
to Let You Know. Indiana Law Journal, Vol. 76.
Wang, Q., Halbrendt, C., Kolodinsky, J., & Schmidt, F. (1997). Willingness to pay for
the rBST-Free Milk: A Two-limit Tobit Model Analysis. Applied Economic
Letters, 4, 619-621.
38