i -ional Agricultural Trade and Policy Center
PRICE ASYMMETRY IN THE UNITED STATES FRESH
Napaporn Girapunthong, John J. VanSickle, and Alan Renwick
JRTC 04-1 January 2004
JOURNAL REPRINT SERIES
Institute of Food and Agricultural Sciences
INTERNATIONAL AGRICULTURAL TRADE AND POLICY CENTER
MISSION AND SCOPE: The International Agricultural Trade and Policy Center
(IATPC) was established in 1990 in the Food and Resource Economics Department
(FRED) of the Institute of Food and Agricultural Sciences (IFAS) at the University of
Florida. Its mission is to provide information, education, and research directed to
immediate and long-term enhancement and sustainability of international trade and
natural resource use. Its scope includes not only trade and related policy issues, but also
agricultural, rural, resource, environmental, food, state, national and international
policies, regulations, and issues that influence trade and development.
The Center's objectives are to:
Serve as a university-wide focal point and resource base for research on
international agricultural trade and trade policy issues
Facilitate dissemination of agricultural trade related research results and
Encourage interaction between researchers, business and industry groups,
state and federal agencies, and policymakers in the examination and
discussion of agricultural trade policy questions
Provide support to initiatives that enable a better understanding of trade and
policy issues that impact the competitiveness of Florida and southeastern
agriculture specialty crops and livestock in the U.S. and international markets
PRICE ASYMMVIETRY IN THE UNITED STATES
FRESH TOMATO MARKET
Napaporn Girapunthong, John J. VanSickle, and Alan Renwick1
This paper analyzes pricing relationships between the producer, wholesale and retail
levels of the U.S. fresh tomato industry. The results indicate that price transmission was
unidirectional from producer to retail. There was no asymmetric response for the producer-retail
price relationship. Asymmetric price response was exhibited between wholesalers and both
producers and retailers. Retail prices respond more to rising wholesale prices than to falling
prices. Wholesales prices, however, responded more to declining producer price than to rising
This article appeared in Journal of Food Distribution Research 34(3), 51-59.
1 Girapunthong is a graduate student, Food and Resource Economics Department, University of Florida,
Gainesville. VanSickle is a professor, Food Resource and Economics Department and Director,
International Agricultural Trade and Policy Center, University of Florida, Gainesville. Renwick is lecturer,
Farm Management Department, University of Cambridge, UK.
The issue of market structure and the flexibility of prices came to the fore at the time of
the depression of the 1930s when classical economists were struggling to address the failure of
the market as evidenced by the high level of unemployment. Earlier studies noted that firms in
industries characterized by oligopolies tended to change prices less frequently than theory might
predict. This is known as the "administered price hypothesis Means's work led to analyses of
the relationship between structure and pricing, in particular to analyses of structure and the speed
and extent of changes in cost and demand on prices.
A related issue, which is the focus of the empirical sections of this paper, is the
relationship between market structure and asymmetry in pricing. Asymmetry can be defined as a
difference in the reaction of firms to cost increases compared to cost decreases. Asymmetry has
clearly long been a feature of economic theory, for example Keynes' theory was in part related to
money wages being sticky downwards. Asymmetry can be seen to take two forms, one relating to
the speed of adjustment and the other to the extent of adjustment. Industrial economists have
undertaken a large number of empirical studies with conflicting results.
The issue of asymmetry in pricing has received considerable attention in the agricultural
economics literature (e.g., Ward, 1982; Kinnucan and Forker, 1987; Schroeder, 1988; Willet,
Hansmire, and Bernard 1997). Much of this work on asymmetry followed the work of Ward, who
examined price transmission for a number of fresh produce items in America. For those products
where significant asymmetry was discovered, he found that price rises were not passed on to the
same extent as were price falls. He suggested a number of possible reasons for this, including the
perishability of produce, which meant that retailers would be unlikely to raise prices and risk
stock moving more slowly and deteriorating. Kinnucan and Forker (1987) undertook a detailed
study of margins in the U.S. milk sector and concluded that there was asymmetry, but in the
opposite direction to that found for fresh produce by Ward. Kinnucan and Forker postulate three
reasons why asymmetry in pricing in the agricultural sector may occur: the existence of market
power, government intervention in the pricing system, and differential impacts of shifts in retail
demand as opposed to producer supply.
Recently, changes in the U.S. fresh tomato market have led to higher producer-retail
margins. Heien (1980) suggested that margins provide useful information about price linkages in
the marketing system for the agricultural sector. Worth (1999) showed that higher marketing
costs were correlated with higher producer-retail margins. Marketing costs include transportation,
labor and packaging. The role of retailers also has influenced the pricing mechanism, because
retailers can independently increase retail prices and therefore the producer-retail margin (Worth,
1999). Farmers are concerned with two issues in this area (McLaughlin, 1995). First, producer
prices are not fully adjusted when retail market prices increase; this results in retailers absorbing
more revenues at the expense of producers and consumers. Second, supermarkets may not lower
prices when producer prices decrease. When producer price cuts do not pass through to the
consumer, demand remains unchanged. The producers concern is that consumers do not receive
the benefits of falling producer prices when they pay higher prices at the retail level while
producer prices decline. This results in increased margins for retailers at the expense of producers
and consumers. Therefore, producers are concerned that price transmission and structural change
in the U.S. fresh tomato industry have resulted in market imperfections. Price transmission is an
important topic because it provides information about the performance of the marketing system
and changes in market structure.
This paper investigates asymmetric price relationships among the producer, wholesale,
and retail levels in the U.S. fresh tomato market. The general approach of this paper employs
Ward's (1982) price asymmetry model, which is a dynamic model using lagged prices. Granger's
causality test is used to analyze the direction of causality. The following section discusses the
fresh tomato industry in the U.S. The methodology used in the analysis of the U.S. fresh tomatoes
is detailed in the third section. The fourth section presents the empirical results. The final section
presents a brief summary in the last section.
U.S. Fresh Tomato Industry
Tomatoes are an important agricultural commodity in the U.S. because they earn
significant cash crop revenues. Almost two-thirds of the fresh-market tomato production in the
United States comes from California and Florida. Domestic production totaled nearly 1.7 million
tons in 1997, with more than 1.2 million tons produced in these two states. However, due to
location and climate conditions in these two regions, they are not in direct competition with each
other. Florida tends to supply most of its production during the winter months, while California
supplies most of its production in the summer. Florida's main competition comes from Mexico;
Florida and Mexico they supply over 90 percent of the U.S. winter market (Van Sickle et.al.
1994). The competition between Mexico and Florida has its roots in the U.S. embargo of trade
with Cuba, which up to that time was the major import supplier of the U.S. winter fresh-vegetable
market. The competition has been fierce, often referred to as the "Great Tomato War", and has
been subject to much scrutiny by academics (Schmitz et al. 1981; Thompson and Wilson 1997;
VanSickle et al, 2003). At present there is a truce in the on-going litigation following the 1996
suspension agreement following an anti-dumping case instigated by U.S. producers in 1996 (Van
Sickle, 1997). At present there is a truce in the ongoing litigation following the 1996 suspension
agreement following an anti-dumping case instigated by U.S. producers in 1996 (VanSickle
1997). California's competition is more domestic in nature, mainly consisting of producers in
other states and "backyard" production. Increasing competition has also come from greenhouse-
grown tomatoes, but those tomatoes have been judged a different product (U.S. ITC, 2002).
Fresh tomato yields have increased slightly since 1980. Florida yields were highest in
1992, when Mexico suffered from severe flooding. This resulted in prices being significantly
higher for Florida producers, allowing them to harvest higher yields per acre. However, since
1994, imports of fresh tomatoes have increased significantly. Domestic producer prices have
declined, and there has been a reduction in U.S. fresh tomato production. Thompson and Wilson
(1997) estimated that less than one thousand producers in the U.S. are responsible for the vast
majority of production. These producers are generally well represented both politically and
commercially through a number of groups. For example, producers in Florida are represented by
the Florida Tomato Committee (FTC), the Florida Tomato Growers Exchange (FTGE), and the
Florida Farm Bureau Federation. This degree of concentration and organization may suggest that
producers are in a position of countervailing power to the oligopsonistic retailers. However,
attempts by U.S. producers to gain countervailing power are to a certain extent undermined by the
availability of a ready supply of imports.
Producers funnel their production through a relatively small number of packing houses.
The shipment of produce is accounted for largely by grower-shippers. Thompson and Wilson
(1997) estimated that nine grower-shippers in Florida accounted for 80 percent of the State's
shipments. Competition in this sector appears fierce, with grower-shippers operating to increase
their supply windows by expanding across regions and across countries. Co-ordination between
shippers and producers varies considerably, ranging from loose contracts to fully integrated
A feature of the production of field-scale tomatoes is their susceptibility to climatic
variations. In Florida, for example, untimely frosts can devastate production; this has occurred a
number of times in recent years. In addition, the tomato war between Mexico and Florida has also
had an impact on the continuity of supplies at certain times. These factors must be considered
when the pricing relationships are examined. In summary, the supply chain for tomatoes in the
U.S. is characterized by increasing concentration at all levels. However, the degree of
competition appears to vary considerably.
Methodology for Empirical Analysis
The fundamental asymmetric model was suggested by Tweeten and Quance
(1971) and was modified by Wolffram (1971) and Houck (1977). These models are static
models and do not completely explain price changes over time. Ward (1982) elaborated
on the existing dynamic asymmetric model by incorporating lagged prices. The attraction
of formulating this empirical study of pricing in the form of an asymmetry model is that it
will allow a number of issues to be investigated simultaneously. These include the extent
of price transmission, the speed of the transmission, and the existence of differences in
reaction to rising as opposed to falling markets.
Prices at one level of the marketing chain are related to prices at another by Rt
f(W), where R may be the price at the retail level and W the price at the wholesale level.
The general case is where R is related to W through a distributed lag function,
(1) R, = ao, + a Wt +E
If asymmetry occurs then aj differs depending on whether Wt is less than or
greater than Wt 1. Using Young's (1980) framework, the Wvariable can be split into two,
one section capturing price rises and the other price falls. Equation (1) can be rewritten as
(2) R, = + (a'W,'l + d" W,',)+ E,
z_, I1 -1} and
0 o 1i i %, i I'
IV W < W
Z ;' i '_
0 oth 1,i\ Ji
Equation (2) can be simplified into (3) by using Gollnick's derivation.
(3) R? =-10 +V kL, j+1 WO +1 (7 -1) aw" \11+ -
Here the estimate of a" a' gives a direct test of the asymmetry condition.
Polynomial lags can be incorporated into this general model. Under the assumption that
the peak response to a price change is immediate, the use of a first-degree polynomial is valid.
Letting pj be some weighting of the lags then equation (4) can be incorporated into equation (3).
This results in only four unknown parameters, rather than 2k unknown parameters in equation (3).
a' = A'o + (P
In Ward's model, the data used are seasonal in nature and the model is developed to take
account of this. However, continuous data are used in this study, which allows us to bypass some
of the problems in estimation. Putting equation (4) into equation (3) produces a general model for
estimating the price linkage:
(5) Rt = Ao(t) + AHI(t) + H0 2(t) + A H3(t) + 'H 4(t) + E,
H2(t) W J) W J
H3(t) =[kW :J, and
where R is the producer price, Wis the wholesale price, W" is the falling wholesale price, and s,
is a random error term. The asymmetric hypothesis tests whether price changes at one level are
symmetric in response to increases and decreases in prices at other levels of the market system. If
the null hypothesis that retail price response is symmetric to both increases and decreases in
wholesale prices is rejected, then we conclude there is asymmetric behavior between wholesale
and retail price levels.
Two further issues need to be resolved before estimation can take place: the direction of
causality and the nature of the lag system to be used. Generally, two approaches are adopted in
the literature with respect to establishing causality. Either Granger causality tests are used or
direction of causality is simply assumed. Clearly, the use of empirical tests should be preferable,
but the validity of these tests is subject to considerable debate (see Kinnucan and Forker 1987).
This relates not only to the fact that the results are susceptible to the number of included lags, but
also to whether filters should be used.
The choice of lag structure for this study was undertaken on the basis that it needed to be
flexible to allow for possible changes over time. Previous studies have used a number of lag
structures, including Almon and Shiller lags. The polynomial specification adopted by Ward
when analyzing tomatoes implicitly assumes that peak effects occur at the outset followed by a
smooth decline. Ward chose a value for j, which represented a geometric-decay function.
Different types of lag structures were considered for this study; however, the function used by
Ward appeared to fit the data well and was chosen. The lag structure chosen was (p = j.
A simple problem in the asymmetric-price model is determining whether
movements in one variable are caused by movements in another. A causality test is one
approach that answers this problem. There are several approaches in which the direction
of causality can be analyzed; however, the Granger causality test is used in this paper.
The basic idea of Granger (1969) causality theory is to test the null hypothesis that
changes in one variable are not able to predict the other. For example, if the null
hypothesis "producer price does not cause wholesale price" is examined, The Granger
causality test is performed by regressing wholesale price on lagged values of itself and
lagged values of producer prices and also regressing wholesale price on lagged values of
itself as follows:
(6) W= a, + +
1= i i
(7) w= a W, E
From these results, an F-test is used to test the null hypothesis that lagged producer prices are
significant in determining wholesale price. If the null hypothesis is rejected (not rejected), it
implies that producer prices do not cause (cause) wholesale price. On the other hand, it is
necessary to test whether wholesale price leads producer price by using the same procedure.
Previous studies have examined the direction of causality in the agricultural food
industry. For example, Gujarati (1995) suggested "the direction of causality may depend critically
on the number of lagged terms included". Pindyck and Rubinfeld (1998, 244) argued that the
causality test should be performed with different lag values to make sure that the empirical results
are not sensitive to the lag length.
Market prices are set daily, and producers respond to these prices in determining their
harvest, packing, and shipping schedules. However, given data availability for the periods
necessary to allow examination of structural changes in the industry, the use of monthly data was
the only practical possibility. Monthly data for producer, wholesale and retail prices for the U.S.
fresh tomato industry are used in this study. Producer and retail prices are available from the
United States Department of Agriculture (USDA) and contain price data from January 1960
through April 1998, a total of 448 monthly observations. Collection of wholesale prices was more
problematic. Monthly price data from Chicago and New York terminal markets were obtained
from the Agricultural Marketing Service (AMS) in Washington. A simple average of the two sets
of prices was taken to represent wholesale prices. It proved difficult to collect a meaningful series
for this data before 1970. Therefore, this analysis is restricted to the period from January 1970 to
The estimation of equations (6) and (7) used different lags for each of the variables
because Pindyck argued those tests of causality may be sensitive to the lag length. Moreover,
Kinnucan and Forker (1987) suggest that Granger's causality test might be inconclusive.
Davidson and MacKinnon recommend that Granger causality tests should use more rather than
fewer lags. We used the results of different lags to judge the confidence in the causality results.
Gujarati (1995) indicates that the direction of causality could be confident if the causality test is
not sensitive to the lag length.
The Granger causality F-tests are summarized in Table 1. The results indicate that
changes in the producer price of U.S. fresh tomatoes clearly lead changes in wholesale and retail
prices. Granger causality tests for retail-to-wholesale prices and the retail-to-producer prices do
not yield strong evidence of causality because the results are sensitive to the length of the lags
used. In other words, the direction of causality from retail price to wholesale price and from
Table 1: F-test for Granger causality tests for U.S. fresh tomato market, May 1975 -
Market Relationship Lag 2 Lag 10
Producer ->Wholesale 50.94* 10.66*
Wholesale ->Producer 8.04* 1.3
Wholesale ->Retail 90.38* 14.52*
Retail >Wholesale 1.81 0.97
Producer ->Retail 178.18* 34.49*
Retail-> Producer 5.82* 1.67
*indicates significantly different from zero at 5 percent level.
retail price to producer price depends on the number of lagged terms used in the model.
Price transmissions for U.S. fresh tomatoes are clearly unidirectional from wholesale to
retail. Furthermore, the results for price linkage from the wholesale price to the producer price are
inconclusive because results are more sensitive to the number of lags included. In summary, the
Granger causality tests suggest that directions of causality for U.S. fresh tomatoes are:
Producer > Wholesale > Retail
where the arrows represent the causal effect.
Price Asymmetry Results
The asymmetry model must consider both the lag length and weight structure. Based on
previous studies and empirical evidence (Ward and Myers 1979, 5); Ward (1982); and Willett,
Hansmire and Bernard 1997, 657), this study uses the weighting of a first-degree polynomial
equal to 3j (for j=0,1,2,3,4 months) where j is the lag length. Following Kinnucan and Forker
(1987) the lag length was determined by adding statistically significant lagged variables. Pick,
Karrenbrock, and Carmen (1990) and Worth (1999) concluded that the maximum lag length
should be 4 months for fresh vegetables. Green (1995, 718) reasoned that determining the lag
length and weight structure simultaneously was problematic. The weight structure should be
arbitrarily picked first and used to select the length of the lag, based on the Schwartz criterion.
The parameter estimates of the pricing-asymmetry models in equation (5) are presented
for each specific function in Table 2. In general, the signs of the estimated parameters were as
expected. All the significant tests on H1 and H2 (rising effect) at 95-percent confidence suggest a
price linkage in the U.S. fresh tomato market. The significant tests on H3 and H4 (asymmetry
effect) for both the producer-wholesale and wholesale-retail price relationships suggest that there
is significant evidence of asymmetry. In contrast, the insignificant tests on H3 and H4 for the
producer-retail price relationship indicate that there is no evidence to support asymmetry between
the producer and retail sectors. Furthermore, each of the intercepts is positive and significant at
the five-percent level, indicating the range of the marketing margins between the two market
The cumulative effects are presented in Table 3. The empirical evidence indicates that the
retail price response to both increases and decreases in producer prices is symmetric, implying
that there is no significant evidence of market distortion between producer and retail markets. The
producer-wholesale price relationship has positive asymmetry. This implies that wholesale prices
of U.S. fresh tomatoes respond more to declining producer price than to rising price, which is
asymmetric. One possible reason wholesalers respond more to price decreases is that they are
trying to maintain their customer base and market share in the wholesale sector. If wholesale
prices responded more to rising producer prices than to falling prices, wholesalers would
jeopardize their customer base, because buyers could buy from other wholesalers or use direct-
procurement contracts. The negative asymmetry at the wholesale-retail price level indicates that
retail prices respond more to rising wholesale prices than to falling wholesale prices. In other
words, the retail prices for U.S. fresh tomatoes increase more in response to increases in the
wholesale price than to similar price decreases in wholesale prices. This result is counter to
Ward's (1982) result that retail prices tended to reflect more of a wholesale price decrease than a
wholesale price increase. Ward reasoned that rising prices might reduce retail sales and increase
the incidence of spoilage. However, it has been argued by Renwick and VanSickle (1998) that
better post-harvest handling practices, direct procurement, and extended shelf life (ESL) varieties
have reduced perishability problems. Furthermore, they argue retailers may be exercising market
power to support this pricing relationship. These arguments support the conclusion that responses
of retail prices to wholesale price increases are quicker than the response of retail prices to
wholesale price decreases.
McLaughlin (1995) suggested a number of possible reasons why retailers may not adjust
their prices downward in times of oversupply. These relate to the perception that consumers will
not buy more if the price is lowered, either because the fall is so small it might not be registered
as a fall or even if it is large enough to be noticed it may not induce increased sales. In addition,
inelastic demand for tomatoes suggests that a price cut will lead to a fall in revenue for the retailer
when price falls, even if the good is elastic in demand, perishability still places a limit on
purchases by consumers. Finally, the product is only one of many products sold in the store, and
pricing needs to be related to an overall strategy for the store, including such factors as shelf
The mean lag shown in Table 3 indicates how long, on average, it takes for the effect of a
price change to be observed. It is an indication of the average speed of price transmission. Mean
lags in this study are small, ranging from 0.125 to 0.468. Ward (1982) concluded that a small
mean lag indicates quick decays. In this study, the mean lags related with the rising producer
price variables are larger than the corresponding mean lags of the falling producer-price variables,
implying that wholesale prices adjust more quickly to falling prices than to rising prices at the
producer level. However, the mean lag of wholesale-retail price changes occurred in the opposite
Table 2: Estimated coefficients for pricing asymmetry in the U.S. fresh tomato market between May 1975 and February 1998.
Adjust R2 0.855
F test 101.814
Number of Observation 274
* significant at the five percent confidence level.
Note: t-values in parentheses
Table 3: Asymmetric price response for U.S. fresh tomato market.
Mean Price Mean Lag Cumulative
($/pound) Rising Falling Rising Falling
Producer-Retail Producer=21.63 0.430 0.468 1.684 1.615
Producer-Wholesale Wholesale=31.48 0.4391 0.282 1.550 1.6042
Wholesale-Retail Retail=69.00 0.125 0.228 1.100 0.829
1Mean lag of rising effect is significant different from falling effect.
2 Cummulative of falling effect is significant different from falling effect.
Conclusion and Summary
This study found unidirectional causality from the producer level to the retail level.
Producer price led both wholesale and retail prices. These indicate that price transmissions in the
U.S. fresh tomato market flow from producer to wholesale to retail levels. However, Ward (1982)
found the lead linkage from wholesale level to both retail and shipping-point levels. The results of
the analysis suggest that causality has changed because of structural changes in the U.S. fresh
tomato market over time. Furthermore, the results of the asymmetric price response model
indicate that wholesale prices respond more to falling producer prices than to rising producer
prices. These results suggest wholesalers compete to keep their customer base and market share
in the wholesale sector. Therefore, wholesale prices adjust more quickly to falling prices than to
rising prices at the producer level. Retail prices of U.S. fresh tomatoes respond more to rising
wholesale prices than to falling wholesale prices. This result differs from Ward, who found retail
prices reflect more wholesale-price decreases than wholesale price increases. Ward reasoned that
rising prices might decrease retail sales and increase the incidence of spoilage. However,
Renwick and VanSickle (1998) suggested that better post harvest handling practices, direct
procurement and extended shelf life (ESL) varieties reduce the perishability problem. These
changes may be contributing to retail prices increasing more quickly in response to wholesale
price increases than to wholesale price decreases. Finally, retail price responses to increases and
decreases in producer prices were symmetric. This indicates that there is no significant evidence
of market distortion between these two markets.
Finally, knowing price linkages among market levels will aid in the evaluation of
the potential impacts of agricultural policy on producers and consumers. For example,
supporting programs to help reduce the cost of production may not benefit consumers if
retail prices do not decrease because of decreasing producer prices. The results obtained
from price transmission and price-asymmetry tests give an indication of efficiency in the
market. In this study, the existence of asymmetric behavior between producer-wholesale
and wholesale-retail price relationships carries important policy implications. They imply
that there is significant evidence indicating market distortions in these relationships. In
addition, the causality test suggests prices are set at the producer level and passed to
wholesale and retail levels. These results suggest that producers can use market strategies
such as direct procurement to avoid problems with price transmission in the vertical
market system, thereby offsetting some of the problems associated with market power.
These results are significant for understanding the pricing behavior between
market segments in the produce industry. The results of this analysis are dependent on the
assumption of stationary transaction costs. It should be noted that some of the results may
be due to shocks occurring in that sector which are not identified in this analysis. Use of
scanner data and collection of marketing-sector cost data could be used to augment these
analyses and provide further insight into price transmission in the fresh produce sector.
Greene, W. H. 1995. Econometric Analysis, 4th ed. New Jersey: Prentice Hall Book Co.
Gujarati, D. N. 1995. Basic Econometrics, 3rd ed. New York: McGraw-Hill.
Granger, C. W. 1969. "Investigating Causal Relationships by Econometric Models and
Cross-spectral Methods." Econometrica 37:424-438
Heien, D.M. 1980. "Markup Pricing in a Dynamic Model of the Food Industry."
American Journal of Agricultural Economics 62:10-18.
Houck, J.P. 1977. "An Approach to Specifying and Estimating Non-Reversible
Functions." American Journal of Agricultural Economics 59:570-72.
Kinnucan, H.W., and O.D. Forker. 1987. "Asymmetry in Farm-Retail Price Transmission
for Major Dairy Products." American Journal of Agricultural Economics 69:285-
McLaugnlin, E.W. 1995. "Buying and Selling Practices in the Fresh Fruit and Vegetable
Industry in the USA: A New Research Agenda." International Review of Retail,
Distribution and Consumer Research 5:37-62.
Pick, D.W., J. Karrenbrock, and H.F. Carman. "Price Asymmetry and Marketing Margin
Behavior: An Example for California Arizona Citrus." Agribusiness 6(1990): 75-
Pindyck, R. S., and D.L. Rubinfeld. 1997. Econometric Models and Economic Forecasts,
4th ed. Boston: McGraw-Hill.
Renwick, A. and J. J. VanSickle. 1998. "Issues Surrounding the Assessment of the
Impact of Increased Retail Concentration on Marketing Margins." Staff Paper,
Food and Resource Economics Department, University of Florida.
Schroeder, T. 1988. "Price Linkages between Wholesale and Retail Cuts of Meat."
Thompson, G. D., and P.N. Wilson. 1997. "The Organizational Structure of the North
American Fresh Tomato Market: Implications for Seasonal Trade Disputes."
Tweeten, L. and L. Quance. 1971. "Techniques for segmenting independent variables in
regression analysis: Reply." American Journal of Agricultural Economics 53:359-
United States Department of Agriculture, Agricultural Marketing Service, Fruits and
Vegetable Division, Marketing News Branch. Monthly Summary, various issues.
United States International Trade Commission (U.S. ITC). 2002. "Greenhouse Tomatoes
from Canada." ITC Pub. 3424, Inv. No. 731-TA-925 (final) April.
VanSickle, J. J. 1997. "A Compromise in the Fresh Tomato Trade Dispute." Florida
Journal of International Law 11:399-408.
VanSickle, John J., E. Belibasis, D. Cantliffe, G. Thompson and N. Oebker.
1994"Competition in the U.S. Winter Fresh Vegetable Industry." USDA ERS
Agr. Econ. Rep. 691. July.
VanSickle, J.J., E.A. Evans, and R.D. Emerson. 2003. "U.S.-Canadian Tomato Wars: An
Economist Tries to Make Sense Out of Recent Antidumping Suits." Journal of
Agricultural and Applied Economics 35:283-96.
Ward, Ronald W. 1982. "Asymmetry in Retail, Wholesale, and Shipping Point Pricing of
Fresh Vegetables." American Journal of Agricultural Economics 64:205-212.
Ward, R.W. and L.H. Myers. 1979. "Advertising Effectiveness and Coefficient Variation
Over Time." Agricultural Economics Research 31:1-11.
Willett, L.S., Hansmire M.R., and J.C. Bernard. 1997. "Asymmetric Price Response
Behavior of Red Delicious Apples." Agribusiness 13:649-658.
Worth, T. 1999. "The F.o.b.-Retail Price Relationship For Selected Fresh Vegetables."
Vegetables and Specialties, Situation and Outlook Report. ERS: U.S. Department
of Agriculture, November:26-31.
Wolffram, R. 1971. "Positivistic Measures of Aggregate Supply Elasticities: Some New
Approaches--Some Critical Notes." American Journal of Agricultural Economics
Young, T. 1980. "Modelling Asymmetric Consumer Response, With an Example."
American Journal of Agricultural Economics 31:175-86.